Is Java losing to LAMP and .NET?

Discussions

News: Is Java losing to LAMP and .NET?

  1. Is Java losing to LAMP and .NET? (181 messages)

    An article on BusinessWeek Online yesterday titled "Java? It's So Nineties" talks about how Java is rapidly losing ground to other development platforms, both on the open source side to LAMP and on the Microsoft side to .Net.

    The article includes some less than fully informed sounding comments such as:
    The next generation of lighter-weight programming tools, including AJAX and PHP, are immensely popular with the Web 2.0 startups...(and) allow developers with less training to build applications rapidly."

    Ignoring PHP, I wouldn't classify AJAX as a "lighter-weight" tool. The article also points to the fact that book sales of AJAX are up contrasting against flat Java book sales (clearly AJAX is a competitor to Java in his mind).

    Another comment states: "When you write code on Linux there are less layers. You don't need Java for it."

    The insight here seems to be that Linux is an OS and Java is a programming language. One sentence later its noted how fast it is to develop on .Net (applications which of course I'm sure are running on the Linux boxes ;)

    Ignoring some of those comments (and not knocking AJAX or Linux, just how the comments reference them with regards to Java), the article does still recognize significant growth of LAMP and .NET.

    How strong do you consider these competitors to Java in the enterprise space? In what areas is Java losing significant ground? Are there any contrasting articles that refute the points listed in this article?

    Threaded Messages (181)

  2. Oh please, BusinessWeek[ Go to top ]

    Who reads that dish rag :)

    I don't consider that publication credible. Aside from that, Java will eventually be replaced by something else. Whether that is PHP or .NET no one knows. Anyone claiming such is just clueless. If we're going to replace Java, lets really do better.

    peter
  3. Oh please, BusinessWeek[ Go to top ]

    Who reads that dish rag :)I don't consider that publication credible.
    Pointy-haired bosses who decide on their IT strategy based on whatever magazine was available during their last visit to the very private place where they have their deepest thoughts ?
    Aside from that, Java will eventually be replaced by something else. Whether that is PHP or .NET no one knows. Anyone claiming such is just clueless. If we're going to replace Java, lets really do better.peter
    The amount of mental confusion in this article is so alarming that it's hard to know where to start. The guy has obviously received a double LAMP/AJAX brainwash. Java is the new COBOL so it ain't dying any time soon (at least no sooner than mainframes and SAP). Now if some of the horrendously complex ecosystem that has formed around it could just die a quick death... The idea that PHP could somehow replace the 150 WebSphere products is however very entertaining.
  4. Oh please, BusinessWeek[ Go to top ]

    Who reads that dish rag :)I don't consider that publication credible.

    Pointy-haired bosses who decide on their IT strategy based on whatever magazine was available during their last visit to the very private place where they have their deepest thoughts ?
    Aside from that, Java will eventually be replaced by something else. Whether that is PHP or .NET no one knows. Anyone claiming such is just clueless. If we're going to replace Java, lets really do better.peter

    The amount of mental confusion in this article is so alarming that it's hard to know where to start. The guy has obviously received a double LAMP/AJAX brainwash. Java is the new COBOL so it ain't dying any time soon (at least no sooner than mainframes and SAP). Now if some of the horrendously complex ecosystem that has formed around it could just die a quick death... The idea that PHP could somehow replace the 150 WebSphere products is however very entertaining.

    How about the USA start out-sourcing and off-shoring pointy-haired bosses. that would make developers lives easier.

    but hey, atleast the article generated some traffic for TSS, so I guess it serves some purpose.

    peter
  5. Oh please, BusinessWeek[ Go to top ]

    How about the USA start out-sourcing and off-shoring pointy-haired bosses. that would make developers lives easier.but hey, atleast the article generated some traffic for TSS, so I guess it serves some purpose.peter
    It's the next step... Someday shareholders will realize they could save even more money by outsourcing management to India. But, you know, eventually India will become a rich country with high salaries while America will slip further into the third world, and so someday the jobs will come back your way - outsourced by Indian capitalists :-)
  6. Oh please, BusinessWeek[ Go to top ]

    For more craziness, read the current issue of SD Times article on AJAX. Talk about mixed up.

    http://www.sdtimes.com/article/special-20051201-01.html
  7. Oh please, BusinessWeek[ Go to top ]

    The idea that PHP could somehow replace the 150 WebSphere products is however very entertaining.

    this is one is the best .... :)
  8. The article is wrongly phrased. It is not Java that is out of fashion but the Java consultants talk and behavior. No longer are 14 year old girls looking at you with stars in their eyes when you belch forth: "Organize inter-service transfers according to use cases from known domain objects into a coarse-grained Composite".

    Of all arguments nothing is as idiotic as "Java is good for the big and mission critical projects". In fact there exist not a more sure way of creating a real disaster than the deadly combination Java-BigProject and there is no such thing as a high performance EJB system in the whole world.

    Remember the Soviet-state economy that collapsed into itself? Remember the Dutch Tulip craze in 1636? You are not the first ones to be disillusioned.

    It is incredible to me that a whole profession tried to elevate themselves above all others in the industry.

    Best regards
    Rolf Tollerud
  9. As a Java developer, I'm not too worried about .NET. It's legally as close as possible to Java and according to Bruce A. Tate (Beyond Java) : "Adopting .NET would be like overhauling your diet by swearing off McDonalds, and going to Burger King every day.". So it's nothing revolutionary if our customers turn to .NET.

    There's a lot I like in Java (for example open source libraries like Hibernate, Ant, JUnit etc). Also performance, scalability and productivity are good enough for enterprise applications of our customers (despite of what our friend Rolf is telling above). But there are some features in the language itself, that I dont like. Therefore, I'm personally waiting for something like Ruby on Rails to mature and getting widely used. That would truly offer different approach for coding applications.

    Tomi
  10. Therefore, I'm personally waiting for something like Ruby on Rails to mature and getting widely used. That would truly offer different approach for coding applications. Tomi

    I suspect it wouldn't. I believe that RoR will gain at least some of the baggage of JEE in order to do the sorts of things that JEE does. Already RoR has changed as it has got to version 1.0. For example, the initial idea that all you needed to do was to define your data model in the database has been subtly altered: now you can have a ruby representation of the database schema (well, for a limited subset of databases) - One of the things that some of us criticised RoR for (and were often told this was an invalid criticism) has now been 'fixed'. I'm sure there will be other 'fixes' to RoR as it better adapts to enterprise use (I suspect a portable JDOQL-style query language will turn up at some point). Still, if this makes RoR more usable for a wider range of enterprise projects, I will not complain, especially if it encourages the use of Ruby, and especially if RoR can be run on the JVM.
  11. Drop legacy stuff like JAVA, .NET and LAMP. RoR and JavaScript with Ajax will kill them all.
  12. It is a burden[ Go to top ]

    1) EJB is a totally brain dead technology, check
    2) Rich clients with xmlhttp (nowadays called AJAX ) is the future, check
    3) The big Java app server is an outdated and overpriced dinosaur, check
    4) J2EE is made up by well-meaning and impractical theoreticians, check
    5) The Java is constantly loosing to .NET technology little by little every month, check
    6) Java users = loud and vociferous fanatics full of hyperbole and hatred, check
    7) XAML/WinFX with Windows Vista will be the last nail in the Java coffin, check - no wait! (Not yet happened..)

    Brash interviewer from "The naughty nose":
    How does it feel to always be right?
    RT: Nobody realizes what a burden it is.
  13. It is a burden[ Go to top ]

    1) EJB is a totally brain dead technology, check2)

    It's always the most braindead technology that catches on, you know ...
  14. It is a burden[ Go to top ]

    "It's always the most braindead technology that catches on, you know"

    I wouldn't say that if I were you, Things would not have gone so totally wrong if not for the .com boom, IMO.

    Regards
    Rolf Tollerud
  15. It is a burden[ Go to top ]

    "It's always the most braindead technology that catches on, you know"I wouldn't say that if I were you, Things would not have gone so totally wrong if not for the .com boom, IMO.RegardsRolf Tollerud
    Bull, everything has gone totally wrong since the .bat boom!!!
  16. It is a burden[ Go to top ]

    "Bull, everything has gone totally wrong since the .bat boom!!!"

    Not everything Microsoft seems to be doing very well. I like to browse Slashdot and entertain myself over all the resentment and envy over the xBox 360 success. And with Vista in beta..

    Regards
    Rolf Tollerud
  17. It is a burden[ Go to top ]

    And with Vista in beta..RegardsRolf Tollerud
    A year ago I believed that XAML/Avalon is the way to go. I am not so sure now. Not because the technology sucks but because I don't think that client base will be large enough. Heck, I still use Win2K and consider it the best Windows ever. Did not have a need to upgrade to XP and don't have an urge to upgrade to Vista now, especially with its enormous hunger for resources especially for video card. I can live without transparent windows. The strategy worked well in the past when Microsoft doubled or quadrupled requirements for memory, disk size and processor speed for every new OS version. This won't work anymore, the processor speed has not increased for a year or so.

    So, new features require too much from hardware, while cutting back on features does not encourage to upgrade. The deadlock. Considering that XAML stuff will not be backported to Win2K/XP, I do not see the future of new MS technology to be rosy.

    No client base means no wide acceptance. While Ajax and Flash are getting speed. I may not like Flash but it *is* portable (though damn slow). Should not Microsoft pump some more money into Mono to back up possible upcoming issues with Vista sales?
  18. But Avalon/WinFX will be available for XP Michael, and then people will now even know if they are browsing with .NET/C# technology or not.

    No need for AJAX hacks! That is what I wanted in my "programmer-chrismas-dreams" and MS kindly provided it.. with XAML as an extra bonus.

    Regards
    Rolf Tollerud
  19. you do not believe in Vista but..[ Go to top ]

    But Avalon/WinFX will be available for XP Michael, and then people will now even know if they are browsing with .NET/C# technology or not.

    Avalon is one of these cool things that won't live up to its promise. It has nothing to do with the features or how revolutionary it is - all of that is irrelevant.

    We have been through this path many times already. Many technologies have tried to create proprietary extensions for the net but never succeeded in any significant scale. There won't be 'Microsoft's Internet' for the simple fact that NOBODY wants Microsoft's Internet.

    Overall, it will still be about HTML, and little else. HTML is simple, it works in most cases, everyone knows how to click on a web page, and it works on every single browser from any OS. So far that is what Internet is all about.

    You will probably see little 'applets' here and there but Avalon will not do much else.
  20. you do not believe in Vista but..[ Go to top ]

    Well, I do wish it some luck. I would love to see it kill the browser, as we know it, as an application medium. That is my Christmas wish.
  21. you do not believe in Vista but..[ Go to top ]

    That is what I wanted in my "programmer-chrismas-dreams" and MS kindly provided it.. with XAML as an extra bonus.RegardsRolf Tollerud

    Kids who get coal for Christmas usually get it cause they are bad, not cause they want it.
  22. you do not believe in Vista but..[ Go to top ]

    But Avalon/WinFX will be available for XP Michael, and then people will now even know if they are browsing with .NET/C# technology or not. No need for AJAX hacks! That is what I wanted in my "programmer-chrismas-dreams" and MS kindly provided it.. with XAML as an extra bonus.RegardsRolf Tollerud

    Yes, because everyone trusts Microsoft to provide their browsing experience. Hold on...

    http://www.onestat.com/html/aboutus_pressbox40_browser_market_firefox_growing.html

    Oh, looks like they don't!

    Still, I guess all you have to do is insist that everyone uses the latest version of Windows and IE simply to browse your website.

    Great way to please your customers!
  23. what is it?[ Go to top ]

    Well I am disappointed with you Steve, you really must eat more fish.

    The only reason that Mozilla Firefox has gained on MS IE is that MS has not competed but been sitting still for years and even disbanded the IE team.

    Has it never occurred to you that MS (that put 1000 programmers against the old Netscape browser) may have something in mind? Where have you hidden your reason and intellect?

    Is there anything more patetique that a Linux/Java user that is proud of Firefox? What do they have to glee of? That they after 4 years have a browser that is similar to IE in every way?

    "What is it that MS has in their back pocket that makes them unafraid of Mozilla/Firefox?

    Regards
    Rolf Tollerud
  24. what is it?[ Go to top ]

    Is there anything more patetique that a Linux/Java user that is proud of Firefox? What do they have to glee of? That they after 4 years have a browser that is similar to IE in every way?
    Have you used FireFox?
  25. what is it?[ Go to top ]

    Well I am disappointed with you Steve, you really must eat more fish.

    I do my best to keep the omega 3 levels up in my diet. Obviously I am failing.
    Has it never occurred to you that MS (that put 1000 programmers against the old Netscape browser) may have something in mind?

    Not really - Microsoft tend to be very reactive... "Oops! We have lost 10% of market share.. time to re-form our IE team!"
    Where have you hidden your reason and intellect?

    I don't know... but then I guess I would need intellect to find it.
    Is there anything more patetique that a Linux/Java user that is proud of Firefox?

    Patetique = a piece for brass orchestra by Beethoven. Yes, I would say that such a user is harmonious and in tune with things. This must be what you are suggesting...
    What do they have to glee of?

    We have glee of a browser that resists spyware, has tabbed browsing, has popup blocking, is cross-platform, and actually does a reasonable job of handling W3C standards.....
    That they after 4 years have a browser that is similar to IE in every way?

    Apart from resisting spyware, having tabbed browsing, having popup blocking, being cross-platform, and actually doing a reasonable job of handling W3C standards.....
    "What is it that MS has in their back pocket that makes them unafraid of Mozilla/Firefox?RegardsRolf Tollerud

    Oh, I don't think they are unafraid of it... it forced them to make a U-turn and re-establish their IE team. Unless losing market share was part of their cunning plan.....
  26. IMO, here's the client side UI fight that will begin next year:
    - MS with yet-another-proprietary framework -> WinFX + XAML.
    - Mozilla/Firefox with SVG + XForms + ECMAScript open standards

    Given the current AJAX momentum and the growing list of adopters (Google, RoR, some Java Web Frameworks and others, plus the many script frameworks like Dojo, Scriptaculous, Prototype, Rico, etc), I think MS is really pushing itself into a tight corner. Besides, we have had .Net for years already, and guess how many desktop apps are based on it? Yes, very, very few. Why would people start using it, if not because MS pushes it down our throat by tieing it to Windows OS once again?
  27. IMO, here's the client side UI fight that will begin next year:- MS with yet-another-proprietary framework -> WinFX + XAML.- Mozilla/Firefox with SVG + XForms + ECMAScript open standardsGiven the current AJAX momentum and the growing list of adopters (Google, RoR, some Java Web Frameworks and others, plus the many script frameworks like Dojo, Scriptaculous, Prototype, Rico, etc)

    Interestingly (in my view) there is another client side UI fight going on - for dominance of non-web GUIs. And it is being won by Swing:

    http://weblogs.java.net/blog/hansmuller/archive/2005/10/official_swing.html

    "There are more developers building applications using Swing and Java SE than WinForms and .NET."

    Microsoft aren't even winning the battle for GUI development on their own platform, let alone on the web.
  28. you do not believe in Vista but..[ Go to top ]

    Microsoft aren't even winning the battle for GUI development on their own platform, let alone on the web.

    Not only that, but they don't believe in their own technology even. They have not produced one .NET client app I could go and buy, or even get bundled with the OS. All of .NET's traction is in replacement of VB web apps.
  29. "We have had .Net for years already, and guess how many desktop apps are based on it"

    That is because the time for installed desktop applications is gone forever.

    "Why would people start using it"

    Because the "apps" are not installed.

    Regards
    Rolf Tollerud
  30. you do not believe in Vista but..[ Go to top ]

    "We have had .Net for years already, and guess how many desktop apps are based on it"That is because the time for installed desktop applications is gone forever.

    Yes... I mean no-one uses applications like MS Office, do they? No-one uses word processors, spreadsheets, presentation packages, graphics applications, programs to manage their photo collections and MP3s, or financial packages, or CAD packages, or media players or e-mail clients? That time has gone forever.
  31. you do not believe in Vista but..[ Go to top ]

    No Steve don't be difficult. I thought we all are talking about contract consulting here are we not? Of-the-shelf software will never be made in either Java or .NEt so why discuss it..
  32. you do not believe in Vista but..[ Go to top ]

    No Steve don't be difficult. I thought we all are talking about contract consulting here are we not?
    Of-the-shelf software will never be made in either Java or .NEt so why discuss it..

    Ah - so the highly profitable 'off the shelf' mobile games and applications market, which is overwhelmingly dominated by Java must be an illusion....

    And 'off-the-shelf' Java desktop applications like NeoOffice, MoneyDance, Thought Mapper and Omniscope can't possibly exist?

    Oh, and unless you are saying that these are insignificant, MoneyDance is commercial, and was reviewed as comparable to Quicken by the Washington Post... not bad for software that 'will never be made'.
  33. you do not believe in Vista but..[ Go to top ]

    "NeoOffice, MoneyDance, Thought Mapper and Omniscope!"

    :-)
  34. you do not believe in Vista but..[ Go to top ]

    "NeoOffice, MoneyDance, Thought Mapper and Omniscope!":-)

    Oxygen, JEdit, JCad, JDraw, CADViewer, Electric VLSI, JlGui, JAlbum, JPhotoBrush, Jake, Swift, MathEx, Meditor, Jex, nanocad, Fantasia...
  35. It is a burden[ Go to top ]

    Hi Rolf, it was boring on TSS without you.
  36. It is a burden[ Go to top ]

    Hi Rolf, it was boring on TSS without you.

    Well, we managed to discuss things reasonably sensibly, and without too much trolling. I guess it had to end sometime.
  37. It is a burden[ Go to top ]

    As much as would like to come back I think I am not able to keep it up anymore. Having done no web-development in over one year and Java for at least two year my knowledge is rusty. I have to settle for being right in almost everything I said!

    Regards
    Rolf Tollerud
    "Mobile device developer"
  38. It is a burden[ Go to top ]

    As much as would like to come back I think I am not able to keep it up anymore. Having done no web-development in over one year and Java for at least two year my knowledge is rusty. I have to settle for being right in almost everything I said!RegardsRolf Tollerud"Mobile device developer"

    Ah yes, as in your past classic statements such as 'EBay does not use EJB'. Well, I guess you did say 'almost everything'.
  39. It is a burden[ Go to top ]

    As much as would like to come back I think I am not able to keep it up anymore. Having done no web-development in over one year and Java for at least two year my knowledge is rusty. I have to settle for being right in almost everything I said!RegardsRolf Tollerud"Mobile device developer"

    Welcome back Rolf.. Nice to hear from you again..

    Regards
    Surajeet
  40. Did you hear about those planted newspaper stories in Iraq? That how good the on-going war and people dying on streets every day is for the country in the long run?
    It's an official strategy.

    Dunno who paid to get this LAMP crap written. I have interviewed, changed jobs and been active on the job boards for the past year and a half and haven't noticed "LAMP" anywhere. Go figure.
  41. It is a burden[ Go to top ]

    Hi Rolf, it was boring on TSS without you.
    Well, we managed to discuss things reasonably sensibly, and without too much trolling. I guess it had to end sometime.
    I think Rolf is right, if article is silly then comments must be silly too.
  42. It is a burden[ Go to top ]

    Let us discuss it from another angle. I like to think of myself as an intellectual atheist. But still, I can appreciate different religions from the standpoint: "which one would I like to be true if it were possible?" (for the record the old Greek world of Gods is most alluring in my personal opinion).

    Using that viewpoint, which programming environment culture do you prefer? Would you choose the one without charm without style without sensibility or reason or common sense without table manners! Of course not.

    Regards
    Rolf Tollerud
  43. It is a burden[ Go to top ]

    Let us discuss it from another angle. I like to think of myself as an intellectual atheist. But still, I can appreciate different religions from the standpoint: "which one would I like to be true if it were possible?" (for the record the old Greek world of Gods is most alluring in my personal opinion).Using that viewpoint, which programming environment culture do you prefer? Would you choose the one without charm without style without sensibility or reason or common sense without table manners! Of course not.RegardsRolf Tollerud
    It must be better to have knowlege than religion (learn them all). Religion just demonstrates the lack of knowlege.
  44. It is a burden[ Go to top ]

    BTW "atheist" was a kind of religin in Soviet Union too.
  45. It is a burden[ Go to top ]

    BTW "atheist" was a kind of religin in Soviet Union too.
    ... whose god is himself ...
    or
    "... whose god is his belly ... "
  46. It is a burden[ Go to top ]

    BTW "atheist" was a kind of religin in Soviet Union too.
    So? This remark makes less sense than "everyone who eats picles dies".
  47. It is a burden[ Go to top ]

    BTW "atheist" was a kind of religin in Soviet Union too.
    So? This remark makes less sense than "everyone who eats picles dies".
    I am sorry if my statements insult insult atheists and communists.
  48. It is a burden[ Go to top ]

    BTW "atheist" was a kind of religin in Soviet Union too.
    So? This remark makes less sense than "everyone who eats pickles dies".
    I am sorry if my statements insult atheists and communists.
    It was not an insult, because (1) it is just an observation (2) which makes no sense (3) and does not imply anything beyond itself. What changes if you take Soviet Union out of the sentense, like this: 'BTW "atheist" was a kind of religin'. How the idea that atheism is a kind of religion changes when you add Soviet Union or China or the North Pole to it?

    BTW, the third law of Newton was a kind of rule in Soviet Union too. Just in case people in your galaxy did not know about that.
  49. It is a burden[ Go to top ]

    Sorry, idealogy is not a religion. My statement was very stupid.
  50. It is a burden[ Go to top ]

    It must be better to have knowlege than religion (learn them all). Religion just demonstrates the lack of knowlege.
    So, what would be the knowledge of religion demonstrate?
  51. The article is wrongly phrased. It is not Java that is out of fashion but the Java consultants talk and behavior. No longer are 14 year old girls looking at you with stars in their eyes when you belch forth: "Organize inter-service transfers according to use cases from known domain objects into a coarse-grained Composite".
    Oh, it's just that the coming generation is totatlly corrupted, that's ancient knowledge ...
    Of all arguments nothing is as idiotic as "Java is good for the big and mission critical projects". In fact there exist not a more sure way of creating a real disaster than the deadly combination Java-BigProject
    How about LAMP-BigProject?
     and there is no such thing as a high performance EJB system in the whole world.
    Yes there is.
    Remember the Soviet-state economy that collapsed into itself? Remember the Dutch Tulip craze in 1636? You are not the first ones to be disillusioned. It is incredible to me that a whole profession tried to elevate themselves above all others in the industry. Best regardsRolf Tollerud
    Couldn't agree more. Java best of breed philosophy is much more sustainable than the Soviet-state communist .net/MS platform. One size fits all and if you're too tall we'll just chop off your head. Steve Ballmer is the Stalin of the IT industry, no doubt ...
  52. Best regardsRolf Tollerud

    Welcome back!
    and there is no such thing as a high performance EJB system in the whole world.

    It is good to see that your sharp ironic sense of humour hasn't changed.
  53. Of all arguments nothing is as idiotic as "Java is good for the big and mission critical projects". In fact there exist not a more sure way of creating a real disaster than the deadly combination Java-BigProject and there is no such thing as a high performance EJB system in the whole world.

    Rolf,

    So what's your alternative? The complaint is worthless without some alternative, so let's have it!

    From my experience, the language is not on the top of the list regarding what makes projects a disaster vs. what makes them a success. Sucessful projects start with good, solid management, a relatively well understood goal to work towards, and a team of good engineers using the correct tools to achieve the desired results. Remove any of these aspects from this list and the result have a strong potential to be less than satisfactory.

    Thomas
  54. Java over .NET[ Go to top ]

    To me its ridiculous saying Java is losing to .NET, there is no cmparision in the first case. Java is going to be there for some more years and it might loose to some other languages but not .NET , never.
  55. No way.[ Go to top ]

    Just because a lot of small companies use the php and .net.
    If you just from the number, Java becomes smaller, but all the big companies are using Java. It like the OS. In the world, maybe Windows is the first one used by a lot of people, but it just run on the PC.
  56. please use some moderation[ Go to top ]

    There's been a lot of this garbage flamebait showing up here lately. I used to read theserverside for serious news. We have the javalobby for flamefests.

    I mean come on, I'm so tired of these "the end is nigh" type articles about foo killing bar in the enterprise. First of all the enterprise jumps on new stuff on a regular basis. This is the normal mode of operation. Second of all, those trading in foo after adopting bar six months ago do not represent the industry. Third, LAMP is old news and therefore unlikely to displace anything because it doesn't have the hype and backing of stuff that is actually new (Soap, EJB 3.0, .Net 2.0). Fourth, there are no silver bullets. And especially the ones that are presented as simple should be approached with extreme prejudice.

    Specifically, which of ruby on rails maybe hundreds of thousands lines (optimistic, I know) of code map to which tens of millions of lines of java code out there. What's missing. What's not there that you are going to need anyway? Or how trivial is your app that you think you don't need that fancy stuff?
  57. Java vs LAMP and .Net[ Go to top ]

    In my view, in the enterprise space, Java is the domain language for quite a while to go for serious apps. For smaller or noncritical apps, I think .Net will be fine.

    The same arguments: VB is really good for small apps while failed when projects become complex... Sooner or later, the VB apps for large/critical apps will be rewritten with more serious tools...
  58. Java vs LAMP and .Net[ Go to top ]

    In my view, in the enterprise space, Java is the domain language for quite a while to go for serious apps. For smaller or noncritical apps, I think .Net will be fine.

    Apart from the fact that .NET is not really suitable for smaller apps on anything but Microsoft platforms, Java is in fact the de-facto language for the smallest and perhaps least critical apps of all - mobile games!
  59. Java is bad[ Go to top ]

    It is so naughty and complex!
  60. Java vs LAMP and .Net[ Go to top ]

    Apart from the fact that .NET is not really suitable for smaller apps on anything but Microsoft platforms, Java is in fact the de-facto language for the smallest and perhaps least critical apps of all - mobile games!

    Java games are starting to get acceptance on other platforms as well, not just mobile phones. Java has JNI bindings to 3D libraries like OpenGL, and the performance is not bad. This difference in performance comes from the JNI overhead really. If you look at game ports like Jake2 and comparison benchmarks (http://www.bytonic.de/html/benchmarks.html), Java can be even faster in some cases than game engines written in C!

    If Java's productivity over C/C++ can be taken to game development, there is no reason that Java could not become a major force in this arena as well. Not only that, but being able to run the games on any platform that has OpenGL support without changes is a big plus.
  61. Java vs LAMP and .Net[ Go to top ]

    Apart from the fact that .NET is not really suitable for smaller apps on anything but Microsoft platforms, Java is in fact the de-facto language for the smallest and perhaps least critical apps of all - mobile games!
    Java games are starting to get acceptance on other platforms as well, not just mobile phones. Java has JNI bindings to 3D libraries like OpenGL, and the performance is not bad. This difference in performance comes from the JNI overhead really.

    The reason I could see Java being used more and more for games would be growing iterest in Linux. People tend to scoff at Linux desktops as nerdware in the US (which is probably true) but outside of the US and especially China and India, this isn't necessarily the case. If you are marketing to 'normal' users, you can't expect them to compile your code on their machnes. So you need to support n versions of Linux. By developing to the JVM, you simplify you requirements greatly.

    The other thing is that even if Java isn't sexy anymore, there are lots of languages that compile to JVM bytecodes that are. MS didn't start using a VM for kicks. They did it for pragmatic reasons. It's a better way to do things.
  62. Java vs LAMP and .Net[ Go to top ]

    That makes Java both for small and large applications and more versatile leader..NET only covers the medium and small size applicaitons and which then get hooked to the MS platform and have no other option to upgrade than to just restart from scratch.
  63. Is Java losing to LAMP and .NET?[ Go to top ]

    If you ignore the sensational headline, the article itself appears fairly balanced. I think the main point is that we live in a world where there are three major software technology stacks: Java, Microsoft .NET, and LAMP. I wouldn't expect any of the three to disappear any time soon. Some organizations will choose to use one of the three exclusively, and many will choose to use more than one--especially within very large companies you'll see both Java and .NET. It's good that we have these choices, and it would be bad if any one of the three dominated the others.
  64. no way![ Go to top ]

    JEE is losing out to LAMP, etc. The layers are EJB and JSF, relative to SQL based LAMP.

    Java has a lot of resources and Jars and pros know how to avoid the traps. For example iBatis, and Groovy (runs as Java bytecode). We can use Java in SQL centric ways, like LAMP, Ruby and the sucesses.

    Java ALSO works great on client, deploys from browser on Mac, Linux and Widows (ex: pointcast.com)

    Java can be improved:
    1. A realy layout, such as GroupLayout should not be separate download.

    2. Dyanimics and Groovy features, so I do not need to download groovy. With dynamics added ( to keep up w/ ObjectiveC- now to compilte to windows, not just mac and C#-Dot.GNH-Mono), and cleans up collections.(to be more dynamic
    Using Groovy we process and convert (nttp, mail, rss) 40K messages per day, nice way to do heavy lifting. (I initilay tried Groovy when I found Google does Phython on back end, and I wanted similar productivity and features. And... if you do collections in C#, Java Collections seems odd now, and I stoped using beans a long time ago.)
    To me, dynamic is more important then scripting.
    Ex: data[2]["f_name"];
    this gets me 3rd row in a List called data, and in that 3rd row of the List, it gets me a value in a Map of a key "f_name".
    Of course, I dont't hard code '2' and "f_name", I just do 'for each' to enumerate rows/collums.

    So I think all business week is saying:
    Sun PetStore is no good.
    Look, .NET is not stable on Linux yet.
    How rich a user interface can you do on client side w/ PHP?

    otoh: WinFX will be released at same time as Java 6!


    .V
  65. no way![ Go to top ]

    JEE is losing out to LAMP, etc. The layers are EJB and JSF, relative to SQL based LAMP.

    This is a strange comparison. How can JSF possibly be losing out relative to LAMP when it is still not that much used and only just starting to gain wide acceptance? If you had said 'struts' or 'JSP' relative to LAMP, that might at least make some sense....
  66. no way![ Go to top ]

    JEE is losing out to LAMP, etc. The layers are EJB and JSF, relative to SQL based LAMP.
    This is a strange comparison. How can JSF possibly be losing out relative to LAMP when it is still not that much used ...

    Java itself would be compred to ObjecitiveC or C#. Unless you are BW.
    JSF (and EJB) has been out for years, and it's in the LAMP tier, a part of JEE. When someon say layers... I think those 2. That is why its not used much and losing out. Do you see the diference btwn JEE and Java?

     Java can be used in a SQL centric way(like every other platform) and w/ a rich UI(which most can't match for cross platform! So it's great for back end, great for front end.

    .V
  67. no way![ Go to top ]

    JSF (and EJB) has been out for years, and it's in the LAMP tier, a part of JEE. When someon say layers... I think those 2. That is why its not used much and losing out.

    You may think of those two, but most developers surely don't - as JSF as not yet had wide acceptance. Therefore, if Java is losing out (which I dispute) it can hardly be due to JSF.
  68. no way![ Go to top ]

    You may think of those two, but most developers surely don't - as JSF as not yet had wide acceptance. Therefore, if Java is losing out (which I dispute) it can hardly be due to JSF.
    I think JSF is one of the best examples showing why Java is losing ;) It resembles me early EJB days. Have a look at Wicket and you'll see the difference.

    Regards,
    Theodore Kupolov
  69. no way![ Go to top ]

    You may think of those two, but most developers surely don't - as JSF as not yet had wide acceptance. Therefore, if Java is losing out (which I dispute) it can hardly be due to JSF.
    I think JSF is one of the best examples showing why Java is losing ;) It resembles me early EJB days. Have a look at Wicket and you'll see the difference.Regards,Theodore Kupolov

    On the contrary, JSF is good example of how Java is developing successfully - an easy to-use (in contrast to early EJB!) component-based system which is extensible, has multiple vendor support (including open source projects), allows (but certainly does not require) GUI-based design of web interfaces, and has given rise to a significant number of related projects extending JSF to a range of interesting GUI technologies (Flash, WML, XML, SVG etc). If this what you call 'losing', it is certainly not a definition I recognise!
  70. no way![ Go to top ]

    Ditto! Even the deployment problems are looking similar.
    But try to explain that to JCP members :(

    But being honest it is not a reason for J2EE fall. If you dislike something you can simply avoid that. The problem is that the wrong stuff being pushed too hard. Rolf is almost right here. Did you remember the old Sun’s "J2EE without EJB is not J2EE etc."

    Plain JSP with taglibs can beat PHP. So LAMP is not a problem at all. You can develop for MySQL too. Maybe the one problem is the cheaper PHP hosting but in the rest JSP and servlets are much more better.


    Marina
    http://www.servletsuite.com
  71. no way![ Go to top ]

    This is a strange comparison....

    I don't get the comparison at all.

    Let's see L A M P
    Linux- Yep. I use that to run my Java apps.
    Apache - Yep. Use that too for static pages
    MySQL - Yep. That's the database I use.
    PHP(PCP,Python et all.) - Now let's compare this with java and see where we stand for enterprise development.


    My favorite quote from the article

    The situation at Merrill Lynch & Co. (MER) in New York is telling. The investment bank runs many of its newer math-heavy applications -- such as options, futures, and derivatives -- using just Linux and the Apache server.

    These technologies are superior to Java in computational applications, says Andy Brown, chief technology architect at Merrill Lynch. "It's closer to the metal," he says. "When you write code on Linux there are less layers. You don't need Java for it."

    So what exactly is running these computations ? Linux is an OS and Apache is a webserver. If they don't actually have an application, then of course they don't Java.
  72. no way![ Go to top ]

    Vic,

    Again, I would agree with you a couple years ago, but with the annotation capabilities of JSE 5, you can write extremely vanilla code and have frameworks jump on annotations to handle any complexities-- carrying the logical load for you. The EJB 3 stuff is awesome, IMHO, so much convention in there you have to do double takes on the example code to realize its simplicity. I would say in the next year, just as we've come to love POJOs, EJBs will become just as, if not more common with it's interceptor stack, injection, listeners etc. There's no extra work needed, just plain old Java, common semantics that everyone is familiar with and can maintain.

    -- Jacob
  73. no way![ Go to top ]

    I would agree with you a couple years ago, but with the annotation capabilities of JSE 5 ...

    Jacob, it's a moving target, and it's a generation behind still. EJB is where we were a couple years ago.
    Today, EJB is still beans, they know the weakness in Collections, finaly they are droping beans reference and calling it "Entities". Collections!

    And look at this syntax:
    data[2]["f_name"];
    We can do this in Groovy(C#, Objective C, etc), great for "heavy lifting".

    I think anotations make the code less readable, and as such are an example of decay and weakness and age of java. (rember when we added pre-processors to C++, that was the sign of the end). It was only added to be the "xdoclets" to try to save EJB.
    And EJB is ORM. That is the point, it's not SQL centric.
    Ruby, PHP, etc... even Spring JDBC, Apache's jPetStore are all SQL centric. ORM just does not represent business as well as SQL maps and E/R, and it's all set theory, relational aglebra... ie, mathemticticaly inferior. Why the Vendors don't give clients what they want.. who knows. Pride of some sorts.

    So I agree that "vendor" JEE is lesser that other tech stacks(as used by Steve Zara and newbs ;-), Zing)! Look at petstore.
    (Also Friendster dumped Java to go to PHP, as per RiA-SoA article).
    I would recomend any boss to use Drupal and vBulletin over JEE. But... JEE when used by people over 10 years exp... is a whole different thing. (SQL centric, dynamic, etc.)



    But:
    C# is not stable on Linux.
    Objective C (build to Windows) is hard to deploy (no "ClickOnce" for them)
    PHP can't do client side.

    Java strength is on client side. I do not know if you have see how easy deployment is in Java 5.06!
    So Java wins on client side apps by far. I wish Java 6 had the released GroupLayout from Netbeans.

    Java wins on server side, since it CAN and does run on Linux. (No way Steve Zara develops on Linux, he's a VS '05 candidate, :-) zang!)

    Look at how ustable Windoze is, some of my clients servers keep crashing, and look at how much worse SQL 2005 is vs old SQL 2000. OSX on x86 is going to embaras Vista.

    EJB is steps behind.
    JSF ... I have yet to see pull app design I would use. Even the name is "Server" faces.
    UI eye candy is not server, please!
    Let the vendors add protlets to JEE, at least clients want that, then you can compare that stack.

    And... JEE does not address the main architecute need in '06:
    *REMOTING.*
    Shale does, Spring does (have remoting).
    W/ all Ajax, Flash, ClickOnce, WebStart, Objective C for Windowze... we need remoting. JEE? EG? Hani?
    (No Steve, Axis is not remoting, Axis is... PetStore)

    .V

    ot: Have you guys seen how nice the Laszlo Mail sample app is? vs beta of Yahoo mail. Wow!
  74. no way![ Go to top ]

    (No way Steve Zara develops on Linux, he's a VS '05 candidate, :-)

    I have no idea how this is relevant, but I do all my development on Linux, using NetBeans and Eclipse.
    No Steve, Axis is not remoting

    Where did I say it was?
  75. Is Java losing to LAMP and .NET?[ Go to top ]

    When any major IT technology that is dominant shows even the slightest decline (which is statistically inevitable, even when the technology is still in it's growth phase) this sort of story turns up. It makes for good headlines. Perhaps a better (or at least significant contradictory) indication of the continuing growth in Java is that it has just become the most widely used language on Sourceforge:

    http://www.osnews.com/story.php?news_id=12778

    Java may be experiencing healthy competition in some areas, such as in small-scale website development, but it others it is just starting to gain acceptance (high-performance computing) or still experiencing rapid growth (J2ME). Also, a change in the use of one development language/platform does not involve any losing out on the part of supposed competitors. Much of the increased use of .NET is almost certainly as an upgrade to now deprecated Microsoft products such as VB6 and VC++; having little impact on Java in these cases.

    So, to say Java is 'losing' is, in my opinion, ridiculous.
  76. Of course it's losing[ Go to top ]

    Let's throw in the towel now before it gets too late
  77. Of course it's losing[ Go to top ]

    Let's throw in the towel now before it gets too late

    You're absolutely right. Java is a dinosaur and we're all doomed ;-)
  78. Java Rex[ Go to top ]

    Let's throw in the towel now before it gets too late
    You're absolutely right. Java is a dinosaur and we're all doomed ;-)

    Let's not forget that the dinosaurs where very successful and long-lasting. They were adaptable, and many were very agile. They were 'scalable' and occupied all sorts of niches, large and small. They only died off because of an accident. In fact they didn't really die off - their feathered descendants are all around!

    Successful, long-lasting, adaptable, scalable, and likely to be around for a very long time in some form or another no matter what happens - sounds like Java :)

    Perhaps it is not such a bad analogy after all!
  79. Is Java losing to LAMP and .NET?[ Go to top ]

    On what grounds in Java losing?

    It seems to me the comparison is being made on ease and speed
    of development which is very proportional to the complexity.

    I would be curious to know where PHP will be a few years from now when people start putting O/R, AOP, MVC, Patterns, XML config files, Tag libraries and basically repeat what we did in the Java world but apply all that to PHP.

    If you want a fair comparison, lets compare a Model-I JSP page
    to a PHP page.
  80. I would be curious to know where PHP will be a few years from now when people start putting O/R, AOP, MVC, Patterns, XML config files, Tag libraries and basically repeat what we did in the Java world but apply all that to PHP.

    As a little bit of a tangent, what is better for the developer? Where's the right balance for utility languages to grow to take on more features vs. a enterprise language creeping down so that it can be used in more and more scenarios?

    I think most would agree that flexible developers know a variety of technologies - various platform stacks, scripting languages, OO languages, appservers, db servers, etc.

    What are the bloated areas in the java world that the Java community might be better off killing and accepting a different technology fit?
  81. What's wrong with these people? I mean the author of the article who is mixing apples and oranges spreading FUD, and the readers who leave comments like this:
    Review: One of the "new" trends mentioned in the article was forgoing Java and running directly on the metal. Some kind of programming language was used. The financial industry loves C++ because of its speed. As a C++ web developer, I love the language because it catches a lot of the obvious mistakes very early on and saves trouble down the road. Is this progress? You bet. Using C and C++ I can develop a Web app very quickly and be assured that it will perform well. The very best benefit: these applications are very highly portable. The same code runs on nearly any operating systems and all it needs is to be compiled for that operating system/hardware combination. It's no shock that this is the case: C and C++ were designed from the start for porting programs to different computers with few changes.
    Date reviewed: Dec 14, 2005 2:04 PM

    C++ as "portable" alternative to web development? Now we talking...
  82. Different levels[ Go to top ]

    When I speak today with somebody who is interested in Java, we spoke about AOP, JBI, MDA, code manageability, security, runtime and deployment possibilities etc. When you ask theese RubyOnRails/LAMP (replace with most recent hyped technology) guys about this topics they look on you very strange way.
    However comment you found about C++ is winner in my list.
    :-)
    For them Java is real pain in ass and I'm happy that they found some other language which is better than Java. They could live on their island.
    To be honest about Java, today problem is size. Number of technologies, implementations, frameworks, possibilities is HUUUUUGE for every important area. Look on e.g. security, persistent mechanisms, web dev. Until now we welcomed any new framework and technology. I guess is time to get rid of some.
    Now when I'm going to implement something, it's really difficult to decide about framework and technologies. Thinking and testing frameworks is taking more time than implementing. Doing simple things is now very difficult in Java. I know I can do it simple way, but why when here is so much great frameworks.
  83. Different levels[ Go to top ]

    To be honest about Java, today problem is size. Number of technologies, implementations, frameworks, possibilities is HUUUUUGE for every important area. Look on e.g. security, persistent mechanisms, web dev. Until now we welcomed any new framework and technology. I guess is time to get rid of some.Now when I'm going to implement something, it's really difficult to decide about framework and technologies. Thinking and testing frameworks is taking more time than implementing. Doing simple things is now very difficult in Java. I know I can do it simple way, but why when here is so much great frameworks.

    Just perfect. I work for a huge company and we are planning the move from oracle programming tools to java. And that's the unsolved question: witch is the rigth choice among the hundreds of frameworks, aplication servers, development tools, patterns and all the mix involved on enterprise java?

    I've worked before with .net, it just works. Doesn't matter to me if microsoft wnats to take over the world, i just want to do my job. And remember the mentor of .net is the fater of Delphi and so many borland tools, not the precios VStudio Team.

    I *really* want to enter in the java world but there are so many doors, so many ways it's too confusing.

    Sorry for my bad english...
  84. Different levels[ Go to top ]

    Just perfect. I work for a huge company and we are planning the move from oracle programming tools to java. And that's the unsolved question: witch is the rigth choice among the hundreds of frameworks, aplication servers, development tools, patterns and all the mix involved on enterprise java?I've worked before with .net, it just works. Doesn't matter to me if microsoft wnats to take over the world, i just want to do my job. And remember the mentor of .net is the fater of Delphi and so many borland tools, not the precios VStudio Team.I *really* want to enter in the java world but there are so many doors, so many ways it's too confusing.Sorry for my bad english...

    It is true that junior Java programmers need some direction to get started with good and solid frameworks and concepts. With .NET you are just stuck with using what Microsoft offers. While it may seem a clearer path, it's not automatically a better path.

    I am mainly a Java developer, but often work on the .NET side as well. What I know about Java makes .NET look quite bleak and 'behind the times'. I find myself using .NET ports of Java frameworks due to lack of basic infrastructure and support for several things. I know it is easier for me to see the Java space as an opportunity rather than a confusing set of options.
  85. Different levels[ Go to top ]

    I am mainly a Java developer, but often work on the .NET side as well. What I know about Java makes .NET look quite bleak and 'behind the times'. I find myself using .NET ports of Java frameworks due to lack of basic infrastructure and support for several things. I know it is easier for me to see the Java space as an opportunity rather than a confusing set of options.
    Me too and +1.
  86. Different levels[ Go to top ]

    I know it is easier for me to see the Java space as an opportunity rather than a confusing set of options.

    +1
    Java is about choices
  87. Different levels[ Go to top ]

    I know it is easier for me to see the Java space as an opportunity rather than a confusing set of options.
    +1Java is about choices
    Yes everybody aggree. But can you imagine that you would like to switch today from e.g. C++ to Java. Somebody will show J2SE 5.0. You will get language very fast but how much time it will take before you start to understand differences between StringBuffer and StringBuilder. What immutable String mean. How to do synchronisatation right way. Some special topic like how to replace preprocessor with "final static" variable. Then you have in front of you huge library of standard classes. How much time is just to get overview (JDBC, JMX, Reflection). Then J2EE (JMS, EJB, ...). And finally somebody will show how to search for other frameworks. Who will answer your questions? Nobody, because even Java community is diversified about many basic topics and this is very confusing not only for newbies but sometimes also for "old dogs". Isn't it?
  88. Same stuff[ Go to top ]

    But can you imagine that you would like to switch today from e.g. C++ to Java. Somebody will show J2SE 5.0. You will get language very fast but how much time it will take before you start to understand differences between StringBuffer and StringBuilder. What immutable String mean. How to do synchronisatation right way. Some special topic like how to replace preprocessor with "final static" variable.
    A lot of similar stuff, JDK 5.0 even has concept similar to templates - generics.
    Who will answer your questions? Nobody, because even Java community is diversified about many basic topics and this is very confusing not only for newbies but sometimes also for "old dogs". Isn't it?
    No, it is not. By the way, last time I checked Google was still available.
  89. Different levels[ Go to top ]

    But can you imagine that you would like to switch today from e.g. C++ to Java. Somebody will show J2SE 5.0. You will get language very fast but how much time it will take before you start to understand differences between StringBuffer and StringBuilder. What immutable String mean. How to do synchronisatation right way. Some special topic like how to replace preprocessor with "final static" variable. Then you have in front of you huge library of standard classes. How much time is just to get overview (JDBC, JMX, Reflection). Then J2EE (JMS, EJB, ...). And finally somebody will show how to search for other frameworks. Who will answer your questions? Nobody, because even Java community is diversified about many basic topics and this is very confusing not only for newbies but sometimes also for "old dogs". Isn't it?

    I have always found Java and its libraries far, far easier to deal with that C++ and its series of ad-hoc libraries with varying degrees of object orientation. Even things that look simple in C++ can have a huge underlying complexity (I remember battling with copy constructors, and losing, years ago).

    Java has JCP specifications. You may not like them, but at least they are there and well documented. Where are standard cross-platform GUI libraries for C++? How about standards for ORM persistence? Java has that good combination of well-documented recommended standards and also a thriving developer community who can offer interesting and innovative alternatives.

    If you are confused, go to the JCP specifications and the relevant websites. You will see considerable documentation.
  90. Different levels[ Go to top ]

    Java has JCP specifications. You may not like them, but at least they are there and well documented. Where are standard cross-platform GUI libraries for C++? How about standards for ORM persistence? Java has that good combination of well-documented recommended standards and also a thriving developer community who can offer interesting and innovative alternatives.If you are confused, go to the JCP specifications and the relevant websites. You will see considerable documentation.
    Where is JSR for Spring or AspectJ? If something have no JSR it mean it's useless? Which of 300 JSRs overlaps? Which implementation of EJB3 (JSR 220) is best?
    I'm not against choice. I'm using Java from first release available on net. However I guess some points have some weight. Confusion of beginers and average programers is real.
    For example how much you have to study when you would like to find best persistent framework. OK after 2 hours it's clear, best is hibernate. Then you download Geronimo server you will find that OpenEJB is integrated. Do I have to spent time with integrating Hibernate or keep with OpenEJB? Is Hibernate really so good, because just recently I found article which propose some other lightweight and faster framework -> confusion -> I have to study more -> confusion ...
    This are typical things which I'm helping resolve everyday with my friends/colleagues.
  91. Different levels[ Go to top ]

    Where is JSR for Spring or AspectJ? If something have no JSR it mean it's useless? Which of 300 JSRs overlaps? Which implementation of EJB3 (JSR 220) is best?

    If you read my post, I am certainly not suggesting that because something isn't a JSR it isn't useless!
    Confusion of beginers and average programers is real.

    I realise this may sound harsh, but welcome to the real world of IT! A good developer should spend a considerable amount of time investigating, evaluating and reviewing different options, Java or otherwise. It is totally unrealistic to expect to be given a clear and unambiguous choice for any aspect of IT.
    I found article which propose some other lightweight and faster framework -> confusion -> I have to study more -> confusion ...

    In any highly technical area, there will be choices. It is part of the role of a skilled developer/manager to keep up with the latest options and provide a way through the confusion.
  92. Different levels[ Go to top ]

    I realise this may sound harsh, but welcome to the real world of IT! A good developer should spend a considerable amount of time investigating, evaluating and reviewing different options, Java or otherwise. It is totally unrealistic to expect to be given a clear and unambiguous choice for any aspect of IT.
    Not everybody have so good employeer as me, and not everybody have so much time to investigate all possibilities. As result I see lack of know-how (lack of well-know know-how). And you need somebody who can support you (for something or against something). In big organisation you will stay alone and frustrated with choices management is doing.
    In any highly technical area, there will be choices. It is part of the role of a skilled developer/manager to keep up with the latest options and provide a way through the confusion.
    Here are no managers which will really understand technologies like SOA, MDA or AOP. With .NET or PHP is easier to make decision because here is not so much choices. It's very naive to think that when you are educated elite it will help you not to loose your job. Today I know that more important is to stay 1 step before others not 10 steps because then you are alone and stranger without contact and support. Then you have only two choices:
    - Stay on top but alone
    - Keep with other but not on top
    Which one is better?
  93. Different levels[ Go to top ]

    Not everybody have so good employeer as me, and not everybody have so much time to investigate all possibilities.

    Sorry, but I think that investigating choices is a core part of being a developer. Saying this is like saying that not everyone has time to test, or time to debug. At least, that is my view.
    In big organisation you will stay alone and frustrated with choices management is doing.

    This may be the case, but it is a different issue from being personally confused about which technologies you use.
    It's very naive to think that when you are educated elite it will help you not to loose your job.

    I totally disagree. I think that not being educated about current developments in IT, and the possible choices, its a very good way to put your job at risk.
    Then you have only two choices:- Stay on top but alone- Keep with other but not on topWhich one is better?

    I don't believe anyone is limited to these choices.
  94. To be honest about Java, today problem is size. Number of technologies, implementations, frameworks, possibilities is HUUUUUGE for every important area. Look on e.g. security, persistent mechanisms, web dev. Until now we welcomed any new framework and technology. I guess is time to get rid of some.Now when I'm going to implement something, it's really difficult to decide about framework and technologies. Thinking and testing frameworks is taking more time than implementing. Doing simple things is now very difficult in Java. I know I can do it simple way, but why when here is so much great frameworks.

    Just perfect. I work for a huge company and we are planning the move from oracle programming tools to java. And that's the unsolved question: witch is the rigth choice among the hundreds of frameworks, aplication servers, development tools, patterns and all the mix involved on enterprise java?I've worked before with .net, it just works. Doesn't matter to me if microsoft wnats to take over the world, i just want to do my job. And remember the mentor of .net is the fater of Delphi and so many borland tools, not the precios VStudio Team.

    I *really* want to enter in the java world but there are so many doors, so many ways it's too confusing.

    Sorry for my bad english...

    I would agree that .NET just works, as long as you do everything the MS way and only do things MS wants you to do. The minute you have to do something they don't do, then the amount of work you have to do rapidly becomes 4x more than the equivalent in Java.

    Having used both .NET and Java. For large scale transactional applications, .NET is behind. For doing simple ASPX stuff, it's great and productive. Wait 2 years and we'll hear BusinessWeek proclaim SOAP webservices is out of fashion and bloated. It's all part of the natural cycle.

    peter
  95. I would agree that .NET just works, as long as you do everything the MS way and only do things MS wants you to do. The minute you have to do something they don't do, then the amount of work you have to do rapidly becomes 4x more than the equivalent in Java.

    This has been my experience many times over. (and have said so in the past).
  96. I would agree that .NET just works, as long as you do everything the MS way and only do things MS wants you to do. The minute you have to do something they don't do, then the amount of work you have to do rapidly becomes 4x more than the equivalent in Java.

    This has been my experience many times over. (and have said so in the past).

    there's one good thing about .NET and the limitations of MS. it inspired me to write my own Schema compiler for C#, since XSD and XSDObjectGen are sad excuses for schema compilers. Not that I'm biased or anything, but I would think that MS with all it's resources could do better. Especially since generating JAXB style C# classes from schema isn't all that hard to do.

    For me to prefer .NET over Java, a lot more open source Java libraries would need to be ported over, so I totally disagree with the silly BusinessWeek article. if my bias wasn't completely obvious already :)

    peter
  97. To be honest about Java, today problem is size. Number of technologies, implementations, frameworks, possibilities is HUUUUUGE for every important area. Look on e.g. security, persistent mechanisms, web dev. Until now we welcomed any new framework and technology. I guess is time to get rid of some.Now when I'm going to implement something, it's really difficult to decide about framework and technologies. Thinking and testing frameworks is taking more time than implementing. Doing simple things is now very difficult in Java. I know I can do it simple way, but why when here is so much great frameworks.
    Just perfect. I work for a huge company and we are planning the move from oracle programming tools to java. And that's the unsolved question: witch is the rigth choice among the hundreds of frameworks, aplication servers, development tools, patterns and all the mix involved on enterprise java?I've worked before with .net, it just works. Doesn't matter to me if microsoft wnats to take over the world, i just want to do my job. And remember the mentor of .net is the fater of Delphi and so many borland tools, not the precios VStudio Team.I *really* want to enter in the java world but there are so many doors, so many ways it's too confusing.Sorry for my bad english...
    I would agree that .NET just works, as long as you do everything the MS way and only do things MS wants you to do. The minute you have to do something they don't do, then the amount of work you have to do rapidly becomes 4x more than the equivalent in Java.Having used both .NET and Java. For large scale transactional applications, .NET is behind. For doing simple ASPX stuff, it's great and productive. Wait 2 years and we'll hear BusinessWeek proclaim SOAP webservices is out of fashion and bloated. It's all part of the natural cycle.peter

    Couldn't agree more. I remember the early days of Java, it's ease of use, simplicity and eligence. That was the main reason Java over grew all other competing technologies. I remember once (I think )even James gosling mentioned that the biggest strength of Java is not right once run every where, but learn once and apply every where. But some hwo (Alas!)we got so much into open source Frameworks, Aspects and all other BS, we totally lost the real beauty of Java, the productivity as a developer. I still think the CORE Java API is the most elegant framework, any thing beyond that is nothing but fad. It is so evident that the birth of Java was due to it's ease of use and it's death will be becuase of overwhelming complexity, and as always the blame goes to shallow and immature developer commuinty.
  98. To be honest about Java, today problem is size. Number of technologies, implementations, frameworks, possibilities is HUUUUUGE for every important area. Look on e.g. security, persistent mechanisms, web dev. Until now we welcomed any new framework and technology. I guess is time to get rid of some.Now when I'm going to implement something, it's really difficult to decide about framework and technologies. Thinking and testing frameworks is taking more time than implementing. Doing simple things is now very difficult in Java. I know I can do it simple way, but why when here is so much great frameworks.
    Just perfect. I work for a huge company and we are planning the move from oracle programming tools to java. And that's the unsolved question: witch is the rigth choice among the hundreds of frameworks, aplication servers, development tools, patterns and all the mix involved on enterprise java?I've worked before with .net, it just works. Doesn't matter to me if microsoft wnats to take over the world, i just want to do my job. And remember the mentor of .net is the fater of Delphi and so many borland tools, not the precios VStudio Team.I *really* want to enter in the java world but there are so many doors, so many ways it's too confusing.Sorry for my bad english...
    I would agree that .NET just works, as long as you do everything the MS way and only do things MS wants you to do. The minute you have to do something they don't do, then the amount of work you have to do rapidly becomes 4x more than the equivalent in Java.Having used both .NET and Java. For large scale transactional applications, .NET is behind. For doing simple ASPX stuff, it's great and productive. Wait 2 years and we'll hear BusinessWeek proclaim SOAP webservices is out of fashion and bloated. It's all part of the natural cycle.peter
    Couldn't agree more. I remember the early days of Java, it's ease of use, simplicity and eligence. That was the main reason Java over grew all other competing technologies. I remember once (I think )even James gosling mentioned that the biggest strength of Java is not right once run every where, but learn once and apply every where. But some hwo (Alas!)we got so much into open source Frameworks, Aspects and all other BS, we totally lost the real beauty of Java, the productivity as a developer. I still think the CORE Java API is the most elegant framework, any thing beyond that is nothing but fad. It is so evident that the birth of Java was due to it's ease of use and it's death will be becuase of overwhelming complexity, and as always the blame goes to shallow and immature developer commuinty.
    Finally, the competition is knocking some sense into the Java world. The problem is not the number of frameworks and standards that have flooded the market. It is the developers that are chasing after these hypes that have created the confusion. I remember a few years ago, I was working on redesigning a webapp and happened to have mentioned CORBA to my client. When I left for another job, he showed me the job posting for my replacement, CORBA became a "must have" qualification. Managers and business folks don't understand the technology. It is up to us to properly inform them. If we keep telling them unrealistic stories about how Spring, Hibernate, AOP and JSF are going to make the products 10 times better, often without any working experience to base on, they will believe us and start asking new hires for those qualifications. That is how most of these overly complex concoctions got snowballed into the next hottest "technologies". Many developers do it so they can put the buzz words on their resume. Then we end up with all these "architects" who know every latest lingo but can't write solid code to save their life. All of this is fine until some new and simple languages come in to challenge the Java empire.
  99. If we keep telling them unrealistic stories about how Spring, Hibernate, AOP and JSF are going to make the products 10 times better, often without any working experience to base on, they will believe us and start asking new hires for those qualifications. That is how most of these overly complex concoctions got snowballed into the next hottest "technologies".

    But... these APIs can make products better; both for the developer and the user. This not unrealistic - it is based on the experience of thousands of Java developers. For example, use Hibernate can result in the use of high-performance full-featured persistence while allowing the user to select the relational store. This can be a real advantage of the 'use simple SQL and hope' portability used in LAMP.

    Also, these are not 'overly complex' - they are used because they reduce complexity for the developer. Spring, Hibernate allow more powerful and flexible applications to be written with less code.

    These are not complex, and they are not overhyped. They are APIs/products that many LAMP developers would love to be able to have an equivalent of.
  100. If we keep telling them unrealistic stories about how Spring, Hibernate, AOP and JSF are going to make the products 10 times better, often without any working experience to base on, they will believe us and start asking new hires for those qualifications. That is how most of these overly complex concoctions got snowballed into the next hottest "technologies".

    If they are so complex and Java is too complex compare to other technologies, why are there .net versions...

    http://www.springframework.net/
    http://wiki.nhibernate.org/display/NH/Home

    Maybe some .net developpers need those COMPLEX technologies. I am tired of hearing JEE is too complex. It was true before but I have never been that productive since I have all those tools and I think they are quite simple. Hibernate is not difficult compare to using direct JDBC. Spring gives you lot of goodies too. And I think working with JSF is very simple compare to Struts or pure JSP/Servlet approach. And even if I haven't tried it yet, Java 5.0 does seem to ease programming (especially annotations).

    Maybe people are too lazy to learn now? In any technological fields, you have to keep learning, it's not about complexity but the technology speed. Building a big web-based application is a complex task afterall.
  101. They are APIs/products that many LAMP developers would love to be able to have an equivalent of.
    http://www.wiki.cc/php/Object_Relational_Mapping
  102. They are APIs/products that many LAMP developers would love to be able to have an equivalent of.
    http://www.wiki.cc/php/Object_Relational_Mapping

    Which, of course, is unlike like Hibernate, or JDO, or EJB3. If you read my post, you will see I said:
    This can be a real advantage of the 'use simple SQL and hope' portability used in LAMP.

    PHP ORM, like RoR does not have a full-featured truly portable query language like HQL or JDOQL 2.0, that can automatically make extensive use of specific SQL dialects. This is the vital key to portability, and because it is not there, installing the same PHP app (ORM or not) on different relational systems can involve a considerable amount of hacking and SQL (I have personal experience of this).

    We have discussed this in depth before in previous threads.
  103. I do not get it anyway, how OQL dialect is better than SQL dialect ?
  104. I do not get it anyway, how OQL dialect is better than SQL dialect ?

    I don't know why you don't get it, as we have discussed this in great detail in threads earlier this year. However, to summarise:

    It is not a matter of using OQL; it is a matter of using a truly portable query language without the considerable dialect and feature differences of different SQL implementations. A good portable query implementation allows applications to be easily moved between different stores wile also producing high-quality SQL to interact with each store.

    Apart from protecting you investment in code if you need to change database products, it also allows development and deployment on different databases, and gives your customers a lot of choice about their deployment too. It is a good thing.

    The alternative is to use a portable subset of SQL, which can in many cases restrict your use of the rich features of a particular database or persistence method, and can prevent the use of pretty much all non-relational methods of persistence.

    If you really wish to cover this ground yet again, I will look up the URLs of previous discussions on TSS.
  105. We are talking about LA(M)P, do not we ?
    BTW how can JDOQL protect your investments ? You will need to migrate to EJBQL anyway and probaly to some new XYZQL later. I think SQL is more portable and better protects investments, but it is your business, I am not going to change your religion.
  106. We are talking about LA(M)P,

    No, we are talking about the differences between Java and LAMP.
    BTW how can JDOQL protect your investments ? You will need to migrate to EJBQL anyway and probaly to some new XYZQL later.

    You need to research more. There is of course no need to migrate to EJBQL. Apart from the fact that there are many vendors and products who will continue to support JDO in addition to EJB3.0 - often even in the same application - both JDO 2.0 and EJB 3.0 support pluggable query languages - you can use JDOQL, EJBQL... even non-portable SQL if you like (although if you choose the latter, you lose portability). Part of JSR 243 (JDO 2.0) was designed to allow JDOQL to be used with EJB 3.0.

    This is how Java protects your investments.
    I think SQL is more portable and better protects investments, but it is your business, I am not going to change your religion.

    You can think that SQL is more portable and better protects investment, but this sounds more like a religious belief to me, as you have been presented with clear evidence of how it is not portable, and how migration can be difficult and expensive in previous threads, many months ago. Again, if you doubt this, I can provide links to the discussion and evidence. I personally have tens of thousands of lines of highly db-specific SQL written years ago that are very difficult to port to any other database. So I simply don't believe you.
  107. No, we are talking about the differences between Java and LAMP.
    Yes, there is a big difference, LAMP doe's not invent query languages. It looks like MySQL users just do not need to migrate and probably some vendor or OSS community will implement migration tools for LAMP if somebody will need to migrate. I am not SQL portability expert too, I just do not need to migrate SQL and have no portability problems.
  108. No, we are talking about the differences between Java and LAMP.
    Yes, there is a big difference, LAMP doe's not invent query languages.

    Why not?
    It looks like MySQL users just do not need to migrate

    A very broad generalisation which is hard to justify. In fact, I am a MySQL user who does need to migrate! I migrated to PostgreSQL then Oracle.
    and probably some vendor or OSS community will implement migration tools for LAMP if somebody will need to migrate.

    Yes - just like the various other things the 'LAMP OSS community' has supposed to have come up with when demand wa there? Like high-performance VMs, for example?

    The problem with LAMP is the assumption that simple approaches always work, and other issues can be dealt with later.
    I am not SQL portability expert too, I just do not need to migrate SQL and have no portability problems.

    Well, of course you don't have portability problems if you don't need to migrate! However, thousands do need to migrate, or want to option to migrate so that they aren't tied down to the varing licensing and pricing conditions of one vendor....
  109. If we keep telling them unrealistic stories about how Spring, Hibernate, AOP and JSF are going to make the products 10 times better, often without any working experience to base on, they will believe us and start asking new hires for those qualifications. That is how most of these overly complex concoctions got snowballed into the next hottest "technologies".

    If you have not worked with those technologies, you really don't know what you are saying. You would have deep respect for them had you tried to recreate something like that yourself.

    People keep dreaming about a simple and elegant language. Java comes pretty close in my opinion. What people tend to forget however is that once people start writing software with this language X, we end up in the same situation as with Java.

    What is it that makes people think that Java as a language is now so complex, when it used to be simple? Up until recently it has not changed much so I don't get it.

    Ah, but it all is about the bad software that was written with Java that makes it bad now? When a language becomes mainstream, it is inevitable that 'complex' software will be written with it.

    I am sorry but that will be true, no matter what language we are talking about. I am also sorry that beginners can not understand it. It is just the nature of the beast - you do not become an expert in any field over night. It takes years. People who expect a freebie to fall on their lap will surely be disappointed.

    You can still write software with Java the way it was written 10 years ago. You are welcome to go back to the 'simple life' any time you like...but you will find it is far harder than today.
  110. <blockquoteIf you have not worked with those technologies, you really don't know what you are saying. You would have deep respect for them had you tried to recreate something like that yourself... You are welcome to go back to the 'simple life' any time you like...but you will find it is far harder than today.
    I have worked for many years with these so called "technologies" to know that they are nothing but an attempt to move code from one place to another. Code that was originally written in plain Java are now moved into XML or some other forms of code. All of a sudden, they are "configurable" because you don't compile it. That was how EJB's metadata hell broke loose. It takes just as much work to learn and write these "configuration" code as it is to write Java. These XML control files are often cryptic, broken fragments of program logics that only the author of the framework can fully understand and know what their effects are. There are those who always claim to be expert in EJB or AOP or this or that. And there are those who simply claim to be expert in backend server or web-application design, or even just software engineering. When I see something that can really do more with less code instead of just disguising one form of coding for another more cryptic form, I will endorse it whole-heartedly. But I am sure glad that new technologies challenging the empire with exactly the simple elegance that Java started out with.
  111. have worked for many years with these so called "technologies" to know that they are nothing but an attempt to move code from one place to another.

    They are nothing of the sort. They are mostly ways to save huge amounts of code. Comparing the transparent persistence and retrieval of Hibernate with the vast amount of JDBC code that this replaces leaves one in no doubt about this. Compare the simple pages and event handling of JSF with the amount of JSP or Struts that this replaces adds to the argument.
  112. If we keep telling them unrealistic stories about how Spring, Hibernate, AOP and JSF are going to make the products 10 times better, often without any working experience to base on, they will believe us and start asking new hires for those qualifications. That is how most of these overly complex concoctions got snowballed into the next hottest "technologies".

    100% agreed.
    What is it that makes people think that Java as a language is now so complex, when it used to be simple? Up until recently it has not changed much so I don't get it.

    Varargs, Autoboxing are few from pure language point of view. They are useless features that no discipline programming language should allow (my personal opinion). I strongly believe that Java added these features just to compete with .NET hype, so they can satisfy those shallow developers who weight langauge in terms of exotic features instead of deep semantics.

    From framework point of view, 10,000 frameworks, and writing 10,000 tags for every one line of code for each of these framework:-)

    Do you need to know more?
  113. What is it that makes people think that Java as a language is now so complex, when it used to be simple?

    I think more of the masses have seen Java. Including those with no formal OO training. Java is simple for those who understand OO (and for those who wrote C++ previously) it is not simple for those that learned how to program with PHP and Javascript or VB. (Even people brought up w/Pascal and C have trouble with Java)

    Also, I think today we are becomming more ambitious. Java helped us get beyond the "GC is bad" and the "interpreted is evil" myths. So now more experienced developers are looking for the wonders of Lisp and Smalltalk in their new languages.

    I remember some Smalltalk people say that C++ wasn't OO because it was statically typed. By that standard neither is Java. So maybe some people are looking for more dynamic languages now that the cat is out of the bag (since dynamic languages tend to be "simpler").
  114. What is it that makes people think that Java as a language is now so complex, when it used to be simple?
    I think more of the masses have seen Java. Including those with no formal OO training. Java is simple for those who understand OO (and for those who wrote C++ previously) it is not simple for those that learned how to program with PHP and Javascript or VB. (Even people brought up w/Pascal and C have trouble with Java)

    I didn't get your point here; ofcourse People coming from procedural language will find writing Java difficult and vice versa, it has nothing to do with the Java language being difficult it self.
    I remember some Smalltalk people say that C++ wasn't OO because it was statically typed. By that standard neither is Java. So maybe some people are looking for more dynamic languages now that the cat is out of the bag (since dynamic languages tend to be "simpler").

    Developers who make claims like above are the root of all evils in our industry. Static VS dynamic typing has nothing to do with OO, many bread of different lnaguages use these concepts regardless. It's just a trade off between putting the responsibility on compiler vs some flexibility for the programmer.

    Can you tell me rationally why the programmers today are itching to use dynamic languages instead of static languages like C++ or Java. Here is I think why, you add your version. Any thing with "dynamic" is a fad and devlopers love it:-)
  115. I didn't get your point here; ofcourse People coming from procedural language will find writing Java difficult and vice versa, it has nothing to do with the Java language being difficult it self.
    The point is that languajes like PHP and Python are easier to grasp for the OO newbie. Since you aren't forced to do things in a OO way if you don't want to.
    I remember some Smalltalk people say that C++ wasn't OO because it was statically typed. By that standard neither is Java. So maybe some people are looking for more dynamic languages now that the cat is out of the bag (since dynamic languages tend to be "simpler").
    Developers who make claims like above are the root of all evils in our industry. Static VS dynamic typing has nothing to do with OO, many bread of different lnaguages use these concepts regardless. It's just a trade off between putting the responsibility on compiler vs some flexibility for the programmer.
    Well, without taking sides myself. The argument was that the late binding languages like Smalltalk use is part of the intrinsic OO experience. Having static typing would limit you because you are always needing to specify what type of variable you are using.

    Part of the OO experience is to be able to have scaleable code. That is, simple things are as simple as hard things.. just use objects and messages to do everything. Languages like Smalltalk lent themselves to this much moreso that Java does.. Some say that that is what OO should be like, and that static typing is limiting that experience.

    My opinion.. is I dunno.. let's see if this dynamic thing catches on.. and then we can all have a better informed opinion.
    Can you tell me rationally why the programmers today are itching to use dynamic languages instead of static languages like C++ or Java. Here is I think why, you add your version. Any thing with "dynamic" is a fad and devlopers love it:-)
    Because they used dynamic languages in their respective Universities, saw the benefits in terms of development speed. They saw how not having to compile code before running it (just selecting and evaluating it) saved them time. They saw that the code was simpler when you don't have to worry about generics, interfaces, typing in general. And because many of their professors have been touting the benefits of these languages for decades.

    What I wonder is what will happen with maintenance when/if people stop using statically typed languages. Today I rely heavily on the compiler to remind me where I need to edit code when I change code.
  116. The point is that languajes like PHP and Python are easier to grasp for the OO newbie. Since you aren't forced to do things in a OO way if you don't want to.

    I don't agree with you here. If that would be the case C++ would be much easier to pick becuase it is almost a superset of C, and you can do any style (procedural, functional, assembly) of programming in C as you like. But my friend that is exactly what makes a language so difficult to master, and that is what most of the programmers don't understand. This was the main reason why Java quickly became hit, becuase it only allows you to do the programming in one paradigm and that is OO.
    Well, without taking sides myself. The argument was that the late binding languages like Smalltalk use is part of the intrinsic OO experience. Having static typing would limit you because you are always needing to specify what type of variable you are using. Part of the OO experience is to be able to have scaleable code. That is, simple things are as simple as hard things.. just use objects and messages to do everything. Languages like Smalltalk lent themselves to this much moreso that Java does.. Some say that that is what OO should be like, and that static typing is limiting that experience.My opinion.. is I dunno.. let's see if this dynamic thing catches on.. and then we can all have a better informed opinion.

    Late binding is the heart of OO and that is the least minimum requirement of any OO langauge, and all the OO languages including Java and C++ support it any way to provide the polymorphic behavior. When you talk about late binding in the context of static and dynamic typing it is the extreme use of late binding and like any thing in this world extreme use of any thing is nothing but chaos:-)
    Because they used dynamic languages in their respective Universities, saw the benefits in terms of development speed. They saw how not having to compile code before running it (just selecting and evaluating it) saved them time. They saw that the code was simpler when you don't have to worry about generics, interfaces, typing in general. And because many of their professors have been touting the benefits of these languages for decades.What I wonder is what will happen with maintenance when/if people stop using statically typed languages. Today I rely heavily on the compiler to remind me where I need to edit code when I change code.

    I can only hope in future we can come up with a better reason than that to choose between static vs dyniamic languages.
  117. 3 Clear Trends[ Go to top ]

    As an Enterprise Java Developer, I'm seeing 3 clear trends in the industry at the moment.

    1) For Enterprise Software (100+ users, expected lifetime in use of 3 years or more), Enterprise Java appears to be the clear leader.

    2) For Web 2.0 Type Software LAMP (or close relations) appears to have most traction.

    3) For 'Drag and Drop' Programming, I have seen no tool that comes close to Microsoft Visual Studio

    It is interesting to note the AJAX can be added to all three of these solutions, (e.g. in this article I wrote about adding Ajax to Java Struts

    What has changed is that Java is no longer the one and only solution - it is now first amoung equals.

    Paul, FirstPartners.net
  118. I don't agree with you here. If that would be the case C++ would be much easier to pick becuase it is almost a superset of C
    C is not easy for most beginners or most people that have been doing other types of programming. C++ could be easier for C programmers.. but not for newbies or those that have started w/VB or something else.
    Because they used dynamic languages in their respective Universities, saw the benefits in terms of development speed. They saw how not having to compile code before running it (just selecting and evaluating it) saved them time. (other stuff snipped...)
    I can only hope in future we can come up with a better reason than that to choose between static vs dyniamic languages.

    Hmm.. what better reason is there than saving time?
    Time==money..

    I think that's a pretty good reason.
  119. Different levels[ Go to top ]

    To be honest about Java, today problem is size. Number of technologies, implementations, frameworks, possibilities is HUUUUUGE for every important area. Look on e.g. security, persistent mechanisms, web dev. Until now we welcomed any new framework and technology. I guess is time to get rid of some.Now when I'm going to implement something, it's really difficult to decide about framework and technologies. Thinking and testing frameworks is taking more time than implementing. Doing simple things is now very difficult in Java. I know I can do it simple way, but why when here is so much great frameworks.
    Just perfect. I work for a huge company and we are planning the move from oracle programming tools to java. And that's the unsolved question: witch is the rigth choice among the hundreds of frameworks, aplication servers, development tools, patterns and all the mix involved on enterprise java?I've worked before with .net, it just works. Doesn't matter to me if microsoft wnats to take over the world, i just want to do my job. And remember the mentor of .net is the fater of Delphi and so many borland tools, not the precios VStudio Team.I *really* want to enter in the java world but there are so many doors, so many ways it's too confusing.Sorry for my bad english...

    I'll try to help you. You should start by using typical JEE stack. Apache Struts or the new JSF, Spring, Hibernate (Toplink is also good or Apache iBatis). Then you have to choose a build system, typically between Apache Ant and Apache Maven. Personaly I prefer Maven (especially 2.0) because you just have to fill in some data, declare some plugins and voilà, you are ready to go. Finally, usually you need some tests tools. The obvious choice is JUnit for unit tests. Groovy can also be interesting to program test cases faster but it's not essential. You can also take a look at Apache Jakarta JMeter for performance tests. I think with this you have a robust development environment.

    There maybe a lot of tools but development is not just about writing application code but also test code, deployement, blablabla...

    One last advice, take a look at Apache Jakarta Commons project. Lot of good stuff there's. It can avoid you a lot of programming effort.

    Hope it's help!
  120. Different levels[ Go to top ]

    Doing simple things is now very difficult in Java.

    How can this be? A large number of frameworks have been added to Java, but this does not mean that simpler ways of doing things have been in some way abolished! For web development Java has a technology that is as simple as PHP: JSP. (Many of these frameworks increase simplicity. For years I have been using Java + JDBC + SQL. Now I use Java + JDO and it is far simpler, with a dramatic reduction in code size.)
  121. Different levels[ Go to top ]

    Doing simple things is now very difficult in Java.
    How can this be? A large number of frameworks have been added to Java, but this does not mean that simpler ways of doing things have been in some way abolished! For web development Java has a technology that is as simple as PHP: JSP.
    Ajax is a technology orthogonal to Java vs. JSP vs. ASP.NET debate. Considering JSP, it gets better despite JSF traction. JSP 1.2 revised the spec for dynamic inclusion, JSP 2.0 embraced and simplified JSTL. JSP probably has got too good that now it can compete with JSF. No kidding. JSP allows to write component-oriented applications, with Ajax or without. See my blog piece on creating independent JSP fragments without Portal engine.
  122. Are you so sure of what you are saying?

    In my company we use a light c++ framework to do web-apps on apache and the code is absolutely portable between win32 and Linux. We only use portable libraries in development (APR, Boost, etc.) and we avoid any proprietary extensions. The pain is limited at the writing of two different makefiles, one for linux and one for windows.

    Used in a certain way c and c++ are hugely portable among different platform. More so today when a number of portable libraries do exists like wxWidgets for UI developement, the APR for OS access (process manipulation, threads, dynamic loading, etc), apr_dbm for DB access, etc.

    The list of OS in which those libraries are available is in fact quite longer than the list of OS where the JVM is available.

    Surely, it isn't a no brainer cost-free portability as for Java, but it is perfectly doable and in fact is has been done for years and years. If you have the right motivation
    that is.

    You are the one spreading FUD and comparing apples and oranges


     
    What's wrong with these people? I mean the author of the article who is mixing apples and oranges spreading FUD, and the readers who leave comments like this:
    Review: One of the "new" trends mentioned in the article was forgoing Java and running directly on the metal. Some kind of programming language was used. The financial industry loves C++ because of its speed. As a C++ web developer, I love the language because it catches a lot of the obvious mistakes very early on and saves trouble down the road. Is this progress? You bet. Using C and C++ I can develop a Web app very quickly and be assured that it will perform well. The very best benefit: these applications are very highly portable. The same code runs on nearly any operating systems and all it needs is to be compiled for that operating system/hardware combination. It's no shock that this is the case: C and C++ were designed from the start for porting programs to different computers with few changes.Date reviewed: Dec 14, 2005 2:04 PM
    C++ as "portable" alternative to web development? Now we talking...
  123. Are you so sure of what you are saying?In my company we use a light c++ framework to do web-apps on apache and the code is absolutely portable between win32 and Linux. We only use portable libraries in development (APR, Boost, etc.) and we avoid any proprietary extensions. The pain is limited at the writing of two different makefiles, one for linux and one for windows. Used in a certain way c and c++ are hugely portable among different platform.

    I have heard this before. I worked with C++ for 7 years before I permanently moved to Java. It is true that you can write fairly portable software with C++, but the problem is that all the works done in C++ are not available on all platforms. This fragments the developer base and usefulness of writing software with C++. There are Windows developers, Linux developers, Solaris developers, Mac developers, but not really just C++ developers.

    And regardless of what you say, there is a serious maintenance aspect there because you have to test everything on N+1 platforms, which is very costly and slow! Even when you write portable libraries, you are just creating something that a virtual machine can do for you without the cost and effort.

    C++ has lots of cool features but after some time, you can't help but wonder if actually writing applications with is that cool. People tend to trip on coolness too much and the result is that people fight over details of automatic pointers in C++ magazines for years, stuffing the original design with cool template traits and other cool tricks that only a few people even knew about. Very cool but fruitless. If you want 'automatic pointers' to manage your resources, just use Java and get over it.
  124. I stopped developing in C++ at a daily basis about 6 years ago.

    I remember looking at a C++ usegroup 11 years ago, then looking again a few months ago. People are still confused about how the language works. They still can't figure it out.

    Whatever you say about Java being complex.. I think C++ wins in the obscurity-you-must-read-the-spec-to-understand-it war. I've heard Perl 6 is even better.. but don't know.

    Whatever move we do should be towards higher levels of abstraction (example Ruby, Python, Smalltalk) not back to assembly. :)
  125. I still love writing some C++ now and then though..
  126. I still love writing some C++ now and then though..

    You're not the only one still using C++ . Almost every survey shows that there's still a substantial percentage of enterprise applications developed in C++. That's 10 years after the C++ language peaked. There's no reason to think that the adoption cycle for Java will be anything different to other languages (although cycles may accelerate).


    PJ Murray
    CodeFutures Software
    Code Generation for Java Persistence
  127. Is Java losing to LAMP and .NET?[ Go to top ]

    I think JSP 2.0 w/JSTL beats PHP when it comes to speed of development and maintainability hands down! JSTL makes whipping up pages very quick.. you don't even need to write any "code".

    Also, the EL makes templating extremely simple. Tag files can make portions of the code reusable with well defined inputs and outputs. No more nasty includes (includes are the gotos of web programming).

    I've successfully taught this combination to people who don't even really know how to program!
  128. Is Java losing to LAMP and .NET?[ Go to top ]

    I think JSP 2.0 w/JSTL beats PHP when it comes to speed of development and maintainability hands down! JSTL makes whipping up pages very quick.. you don't even need to write any "code". Also, the EL makes templating extremely simple. Tag files can make portions of the code reusable with well defined inputs and outputs. No more nasty includes (includes are the gotos of web programming).I've successfully taught this combination to people who don't even really know how to program!

    I agree. It is very productive and way more intuitive then PHP. Just 4 tags librairies to learn. SQL tag librairies is quite good for replacing PHP.
  129. Nasty includes?[ Go to top ]

    the EL makes templating extremely simple. Tag files can make portions of the code reusable with well defined inputs and outputs. No more nasty includes (includes are the gotos of web programming).
    The remark about includes is totally off base. Maybe this samplewill help you to change your mind. The page source code basically looks like this (changed angle brackets to square ones):

    [script src="ajaxhelper.js" language="javascript"][/script]
    [p]This paragraph is defined directly in the parent page
    and should [strong]precede[/strong] the content of login control.[/p]
    [div id="LoginComponent"]
      [jsp:include page="loginComponent.jsp"/]
    [/div]
    [p]This paragraph is defined directly in the parent page
    and should [strong]follow[/strong] the content of login control.[/p]

    By the way, if you load the page into Netscape 4 or just turn Javascript off, it will still work. Without ajax, of course, but visually and conceptually absolutely the same. Courtesy of JSP dynamic include action + redirection.
  130. Nasty includes?[ Go to top ]

    I think there are places where includes may be the best option (just as there are a few places where gotos might still be ok).

    But in general you are better served by creating an abstraction that represents the component that you want and properly defining its inputs (this is analogous to using procedures versus GOTOs).

    Tag files also have the advantage of letting you stay in the same language (in this case JSP). With just a simple cut&paste you can remove the stuff from your complex page and modularize it. With includes you do the same.. but good luck trying to make it reusable.

    Includes aren't a good way of doing components. Tag files are. In a sense tag files are a way of extending the HTML language on the server side.
  131. Nasty includes?[ Go to top ]

    I think there are places where includes may be the best option (just as there are a few places where gotos might still be ok). But in general you are better served by creating an abstraction that represents the component that you want and properly defining its inputs (this is analogous to using procedures versus GOTOs).
    That is what I am doing, I have independent components with properly defined inputs. I can include these components in any page using a simple include.
    Tag files also have the advantage of letting you stay in the same language (in this case JSP). With just a simple cut&amp;paste you can remove the stuff from your complex page and modularize it. With includes you do the same.. but good luck trying to make it reusable.
    I don't see how with includes I cannot stay in the same language.
    Includes aren't a good way of doing components. Tag files are. In a sense tag files are a way of extending the HTML language on the server side.
    I am including tag files, I mean JSP with tags. And it works out pretty well. Tags and includes do not exclude each other.
  132. In early days Java was the biggest player. Microsoft .Net came out later and LAMP was not mature enough (few tools, skills unavailable etc.). At the same time, Java was hot so market started using Java for everything even when it didn't make sense. Now, .Net, LAMP and Java seem to have their own market segments.

    The statistics sited in the articles don't mean anything as they don't necessarily prove that Java lost its share to .Net or LAMP. The numbers just show that the use of .Net and LAMP increased and that of Java decreased.

    Google and Yahoo use AJAX and PHP on the front end because they want lighter and faster framework. I was recently talking to some one from Amazon.com and he said one of the reasons they use free and open source technologies is that the run time (servers) required to run Java is extremely expensive (given the number of licenses they need). They would rather spend half a million and write their own runtime rather than spend 2 million in licenses. It is quite possible that the same logic applies to Google and Yahoo but it doesn't mean Java is loosing to its competitors. If backing from large companies is any measure then we must as well take into account IBM's recent renewal of Java licensing.

    Details on book sales are the funniest data point in the article. Of course the book sale of a new technology will grow faster.

    In a nut shell, the article does successfully convey that the use of .Net and LAMP is growing but it focuses on a wrong message.

    C
    http://chintanrajyaguru.com
  133. elmirafudd@yahoo.com[ Go to top ]

    You wascally developuhs you! Java is losing! I like wabbit hunting.
  134. For a number of years, Java was the only real choice. Now we are seeing more of a specialization of different stacks suited to different types of problems.

    For most large, complex systems, Java works well. The LAMP stack works well for applications with a simple domain model. Imagine trying to maintain a complex 500kloc system written in Perl? Possible, perhaps, but probably not the best solution. .NET has also clearly established a market segment. .NET still has the best IDE. Period. Also, Microsoft's server applications, like SQL Server, have convenient admin GUIs which are accessible to just about anyone who needs to do routine admin tasks.

    As our collective experience grows,it will become routine to prescribe each of these specialized stacks to the appropriate type of system. We will reach a steady state where the market share simply reflects matching appropriate technology with the types of systems businesses require. Until, of course, the next big thing comes along with enough hype to overwhelm over good judgement.

    Don Morgan, Founder
    http://www.DeveloperAdvantage.com
    Audio Training for Software Developers
  135. Updated graph that I originally posted as part of this "Saving J2EE" article.

    http://www.realmeme.com/roller/page/realmeme?entry=java_vs_lamp_vs_web

    Along with a graph of my thoughts on development environment versus organization size (as a proxy for complexity).
  136. Not FUD Per Se[ Go to top ]

    I don't think the author is trying to spread FUD. I do sort of feel like publications geared towards the business/ops side of things tend to be pro-Microsoft, however and I did feel that this article tended to be a little misleading in its attacks on Java.

    For instance:
    An earlier Jupiter Research report showed that 62% of midsize businesses have adopted .NET, vs. 36% for IBM's WebSphere technology, which is based on Java.

    This is just ridiculuous to compare. How can you compare adoption of the ENTIRE .NET platform to WebSphere users? The best comparison is .NET vs. J2SE or ASP.NET vs. WebLogic, WebSphere, JBoss, and Tomcat users.

    Plus, I think a lot of quotes were taken out of context, including the Linux layers comment.

    In the end, what we are seeing is more of a market correction rather than JArmeggedon. What I honestly feel is occuring is that a lot of people that were using Java because it was an object oriented, advanced programming language are now looking at C# as an option . In some cases, I've been forced to use C# in applications because my clients are more familiar with the Microsoft platform. Plus, .NET is new and that makes it the sexier pick.

    And God forbid, C# is actually a decent language. I know, I expect to be flamed for saying this, but it's not bad. In fact, PHP can be a good language, but it is a few constructs away from being a real competitor for Java. That's my opinion.

    LAMP stacks are easy. Mindlessly easy. Java is hard. And the frankly, when I talk to newbie programmers they tend to feel that the Java community is fairly elitist and judgemental to boot. I would back that up in a lot of cases, especially dealing with forums like the JBoss forums. That's another tangent for another time.

    Anyway, a LAMP stack is really easy to install and design simple websites for. .NET removes a lot of the complexity associated with software development because it does everything for you. You have one IDE that is basically Eclipse + XMLBuddy + Eclipse Web Tools + App Server Integration + Axis IDE Integration all in one. Those of us who program Java are basically accepting that we're going to assemble our own stack everytime we create an app because we choose our data access layer (sometimes we even choose our own JDBC driver) plus our integration layer (Spring?) and so on and so forth. With C#, you start writing code and hope that Visual Studio will let you do what you want without breaking (I could blog for days on stupid quirks of VS2003.)

    So, at any rate, I am a Java bigot and will continue to be because at the end of the end of they day, it's the best community. Linguistically, it's not that much better than C# (it is a little bit), but the fact that there are OS alternatives for everyting I want to do means that I can deliver software quickly, effectively, and I can produce a cheaper project with higher quality.

    OK, that's the end o' me rant.
  137. Another genius[ Go to top ]

    I do not know where to start, look at these statements:

    1. LAMP is used more than Java: a more accurate comparison is to compare LAMP against LAMJ or P against J. Java is used widely with LAM components.
    2. PHP is used more than Java: Well, HTML is used more than both. Counting all the sites that use Java or PHP is meaningless. It is given that there are more basic, simple, or home pages than there are professional complex site. Simple sites are more likely to be written in PHP. This is similar to saying there is more printed material in the USA Today than for Shakespeare, hence USA Today must be better.
    3. Sales Of AJAX books grew more than Java: The term AJAX did not exist two years ago, so selling 10 books compared with 0 books two years ago means that AJAX sales have grown infinitely. The other flaw in this argument is that many of the AJAX frameworks are Java based. This is like saying the sale of JDBC, JMX, … grew more than Java. Look at http://adaptivepath.com/publications/essays/archives/000385.php for the first public article of AJAX, it is dated Feb 18 2005.
    4. Google and Yahoo do not use Java: Both are C++ shops. Their usage of AJAX is with C++/Java at the backend. Remember, Google hired Josh Bloch not for his PHP skills.
    5. Merrill Lynch & Co ... using just Linux and the Apache server: So what language do they use? The author does not say. Clearly you cannot write programs with Linux and Apache, you need a programming language. This is a meaningless statement.
    6. Jupiter Research report showed that 62% of ... .NET, vs. 36% for IBM's WebSphere: This only shows that .Net is used more than WS. It does not count WL, Tomcat, JBoss, ...

    In conclusion, this is not new. I have been seeing this since Java became popular. Most of these claims are made by ignorant people or people with hidden agendas.
  138. Another genius[ Go to top ]

    I do not know where to start, look at these statements:

    Excellent posting.
  139. RE: Another Genius[ Go to top ]

    Couldn't agree more. This article reminds me of something my HR group would write about technology. Totally misunderstood the technology and it's organzation/hierarchy.
  140. Another genius[ Go to top ]

    Google and Yahoo do not use Java: Both are C++ shops. Their usage of AJAX is with C++/Java at the backend. Remember, Google hired Josh Bloch not for his PHP skills.5.

    I know some engineers who work in Yahoo is rewriting all the PHP code to java....
  141. Is Java losing to LAMP and .NET?[ Go to top ]

    To be honest with you I hadn't heard of LAMP until I read this article and I try to keep up with the latest development languages and technologies. His company ActiveGrid is trying to sell Application Server products to replace Java Application Servers for enterprise applications. I would take his opinions with a grain of salt. I would suspect the decline of the use of Java is predominently due to .NET rather than LAMP.
  142. Is Java losing to LAMP and .NET?[ Go to top ]

    I would suspect the decline of the use of Java is predominently due to .NET rather than LAMP.

    Assuming there is any significant decline in the use of Java, I think it is unlikely to be due to .NET. A detailed look at job adverts suggests that .NET and Java are being used for different things. .NET is predominantly client-side, and is rapidly growing as a replacement for older versions of VB and VC++. Java is predominantly server side. Because of this, a significant number of jobs list skills in both technologies.
  143. Is Java losing to LAMP and .NET?[ Go to top ]

    In my organization (a Fortune 100 company) I have seen .NET used for web applications because of the choice to use MS SQL Server as the database for the system. Typically these are smaller scale projects replacing Lotus Notes databases or manual processes using Excel. I have also seen several external job postings for web development of .NET applications over the past year. For the critical web applications at my company Java is always used. I would be interested to know how big the LAMP development community is and if anyone has used it on a real project.
  144. Clearly BusinessWeek is the loser[ Go to top ]

    This is the second article in two days I've seen on BusinessWeek where I have to wonder how this magazine continues to exist when the people writing the articles clearly have no involvement with technology beyond the articles they write.

    I don't think I'm going to bother with anything coming out of BusinessWeek.

    Here's the other piece of garbage.
    http://www.businessweek.com/technology/content/dec2005/tc20051212_622616.htm
  145. This is the second article in two days I've seen on BusinessWeek where I have to wonder how this magazine continues to exist when the people writing the articles clearly have no involvement with technology beyond the articles they write.I don't think I'm going to bother with anything coming out of BusinessWeek.Here's the other piece of garbage.http://www.businessweek.com/technology/content/dec2005/tc20051212_622616.htm
    The funny thing is that, as stated at the bottom of the page, the author works for a salesforce.com competitor. So the agenda is not even "hidden" in this case! There is *some* truth at least to the criticism against über-platforms and vendors pushing their own agendas and trying to take control on customers IT strategies...
  146. biz week is, imo, is fud rag[ Go to top ]

    had subscription to biz week this past year. enjoyed some good reads, but could still sum up its content in general as:

    * china is going to kick our butt
    * india is going to kick our butt
    * here is a list of good b-schools

    as such, this article, especially considering its apple-orange comparisons, is taken to have just nominal value.

    >> hippest of hip software
    not a criteria for gauging quality. that goes for java, php, ajax alike.

    >> as old and out of style
    good grief. were it not for wanting to read the article fully before posting a reply, i would have stopped at this point. this article has the same tone as pre-bubble pop "new economy" bits.
  147. biz week is, imo, is fud rag[ Go to top ]

    btw, there is a much more professional item on roughly the same topic on today's front page of the marketplace section of the WSJ.
  148. Jesus Christ! Why so much worry when is clear as spring water that the author has nothing to do with real software engineering experience, real system architecture and professional hard-programming?! Just the inflations of someone with IT culture built from reading marketing advertisements ... or hidden paid agenda. Point.
  149. Jesus Christ! Why so much worry when is clear as spring water that the author has nothing to do with real software engineering experience, real system architecture and professional hard-programming?! Just the inflations of someone with IT culture built from reading marketing advertisements ... or hidden paid agenda. Point.
    I also typically don't react on such articles. However this will be for sure read by our bosses. We have to react before somebody will replace you. Business decisions aren't technologically or logically driven. For 70% it's just feeling and I wouldn't like to leave their opinion appear in head of my boss untouched by our opinions.
  150. Just the oppinion of an looser-user[ Go to top ]

    I also typically don't react on such articles. However this will be for sure read by our bosses.

    That is my fear. Sadly it is usually the blind leading the blind. Not always.
  151. Just the oppinion of an looser-user[ Go to top ]

    I also typically don't react on such articles. However this will be for sure read by our bosses.
    That is my fear. Sadly it is usually the blind leading the blind. Not always.

    Uhhh...not really. Blind people know their blind and become more reliant on other senses.

    People like this see themselves as visionaries.
  152. Just the oppinion of an looser-user[ Go to top ]

    I also typically don't react on such articles. However this will be for sure read by our bosses.
    That is my fear. Sadly it is usually the blind leading the blind. Not always.
    Uhhh...not really. Blind people know their blind and become more reliant on other senses.People like this see themselves as visionaries.
    Well true. Guess it is more like the blind, deaf and limbless ...

    Anyway, it is a saying (At least in the USA) that means, "people who can't see showing others who can't see the way to go".
  153. Hani has the perfect reply.[ Go to top ]

    http://www.jroller.com/page/fate?entry=web_twopointschmoe
  154. LAMP and .NET are soooooooo old. How possibly they can take over anything ?
    Did you miss big news guys ? Ruby on Rails hits 1.0 release !!
    We should abandon this almost stopped bandwagon called Java and migrate en masse to new ground. They still do not have ERG (Enterprise Ruby Gems) and RSF (Ruby Server Facets) and lot's of other acronyms we are so good at implementing.
    Let's help them ! :))
  155. The problem is not that Java can be replaced; can you replace DB2 or Oracle with MySQL? I do not think so. Java replaced C++ in certain niches. Lamp is going to replace Java in certain niches. That is not a problem. By the way, the author is going to write a book about this?
  156. The problem is not that Java can be replaced; can you replace DB2 or Oracle with MySQL? I do not think so. Java replaced C++ in certain niches. Lamp is going to replace Java in certain niches. That is not a problem. By the way, the author is going to write a book about this?

    Good point. This may be the biggest limiting factor of their solution.
  157. 1) Java now includes so many different application areas that overall I believe it is still growing very fast.

    2) LAMP is probably enough for you average small e-commerce web site, and I expect that many IT managers will choose it over Java. But I expect that many such websites are eventually rewritten with Java and Oracle (at least if they make any money.) FYI, MySQL didn't support XA until very very recently. Imagine using it for transactional messaging.

    3) .NET involves the usual huge risk of lock-in, realized every day, and this acts as a break on it, whereas risk in Java these days is very very low.

    Guglielmo
  158. BusinessWeek On Dope :)[ Go to top ]

    Is Java losing to LAMP and .NET? This is so funny. Peter Yared, CEO of software maker ActiveGrid please do more research before writing such a stupid article.

    Cheers !!
  159. Don't make war, make peace[ Go to top ]

    I often use java as backend, to do the hard work.
    Then i use php for the frontend, the thing that my users see.

    Java is good at working, php is good for making thing cool looking.
  160. Don't make war, make peace[ Go to top ]

    I often use java as backend, to do the hard work.
    Then i use php for the frontend, the thing that my users see.

    Java is good at working, php is good for making thing cool looking.
    I supposed CSS(+JavaScript) is good for making pages cool looking ;)

    And why do you need java+php hosting environment instead of php-like coding using JSP?
  161. php on the frontend of java ...[ Go to top ]

    must havev not stumbled across tapestry or jsf yet
  162. Lamp ? Who's lamp ?[ Go to top ]

    I googled , and found articles about lamps , but also lava lamps . Sounds like it's the hottest thing out there in the development comunity :)
  163. Of course NOT ![ Go to top ]

    Anything compared to Java, must it be a platform. How can u call LAMP a general platform ?! Java platform is clean and elegant. It has a lot of really good apis and it's enterprise ready.
  164. Is Java losing to LAMP and .NET?[ Go to top ]

    Short answer, no.

    These are not analogous things to compare really. .Net and Java are somewhat more analagous than either Java and PHP or .Net and PHP but all in all they are different tools for different jobs.

    Java works well in large enterprise environments. With support for various platforms and OSes it is really the only across the enterprise choice. Java while not open source, also has a somewhat open governing board and sort of embraces the "spirit" of openness - this matters to those concerned about open standards and interoperability and so on (no its not perfect with Sun retaining control, but that s another debate and has pros and cons). Java does have many utilities and libraries at this point which can make things complicated, and in some language constructs is a bit more dated than .Net, but neither of those things is a show stopper. Java is still a great and proven language with lots of advantages.

    .Net is also a great choice if you can commit to a single platform up front. And if you want to leverage different languages (C#, VB, VJ, etc) and run the resulting "bytecode" on the same platform. I dont mean that as an underhanded dig either. I dont personally prefer the MS platform, but if you are an MS shop anyway for whatever reason then .Net makes all the sense in the world.

    PHP (does the OS and database get rolled in with the other ones, arent we really talking about PHP and not LAMP?) has its place as well. For smaller installs, or shared hosting and so on, PHP is untouchable. PHP is also very very nice in thats its so straightforward (in most instances) to use (and itself has a plethora of utils and libs available as well). PHP can be used enterprise also, dont get wrong, but using it in an OO manner and under heavy usage - while it *can* be done - is not as straightforward as it is with Java. (At least not in my experience, but its been a while since I worked on a large PHP job - havent used 5 for example.)

    Its still just a question of the right tool for the job. As always.

    Now all that said I would say that one HUGE advantage for PHP when it comes to very small (and cheap) one off apps-sites over just about everything else on the planet (Ruby, Java, whatever) is that its ubiquitious at hosting providers. There are various reasons why but PHP is there and works at your average 20 bucks a month (or less) provider and Java is either not there or doesnt work very well (limited to 128 or 256 heap - not ideal but doable in most cases - but then CPU is not well controlled in a multi user "virtual" shared hosting env). That fact - even though it only really matters to people wanting to pay very little for hosting (you can get dedicated for $100 bucks a month now) is certainly one of the reasons for the popularity of PHP.
  165. Compare Java and .Net???[ Go to top ]

    Short answer, no. These are not analogous things to compare really. .Net and Java are somewhat more analagous than either Java and PHP or .Net and PHP but all in all they are different tools for different jobs.

    Oh, please, while .Net is a platform Java is a language. If you want to compare, compare Java and C#, or J2EE and .Net.
  166. Compare Java and .Net???[ Go to top ]

    Short answer, no. These are not analogous things to compare really. .Net and Java are somewhat more analagous than either Java and PHP or .Net and PHP but all in all they are different tools for different jobs.
    Oh, please, while .Net is a platform Java is a language. If you want to compare, compare Java and C#, or J2EE and .Net.

    Oh please indeed, you might want to read what I actually wrote - I flat out stated they are "different tools"?

    And I would not call Java just a language and .Net a "platform". If by platform you mean underlying arch to run something, such as including the CLR or JVM. Java CAN compare in that sense, Java does not specifically allow you by specification to run different syntax and compile it to run on the same VM, but it does HAVE a VM which in some ways is analogous.

    Though I wasnt bagging .Net. I am not a .Net guy I admit but I think its a great advance for MS and a great model for software. The entire ECMA CLI thing is a great concept, write once run anywhere, wonder how they thought of that?

    Even with an open source or other CLI implementation for various platforms though I do not think your typical large Microsoft investment shop is going to run C# compiled code via DotGnu (or even other such as a Microsoft other OS CLI) on Solaris. It can be done yes but once you make the big MS investment the real supported "platform" is also MS.
  167. I do not talk about religion[ Go to top ]

    Every language and plataform has its strength and weakness and are good for some situations than others.
    I think its time to shutdown the religion quarrel and drive our thought to stuffs like JSR 223: Scripting for the JavaTM Platform.
    I mean trying to work out a unified portable platform and focus on what we like to do building software solutions.
  168. Embrace change people![ Go to top ]

    Embrace change, face it, you will and are currently always updating your skill set, do whatever it takes, simple does it, life rocks, get a life, .....word
  169. It takes time![ Go to top ]

    Well... YES Java is losing to LAMP - and if not - it will lose to something else - BUT its going to take 10 years before that thing will replace Java from the commercial deployments.

    Consider this, Java got its real boost in the .com boom, but the REAL commercial deployments are happening today. It takes time before an emerging technology become mature enough to be used for first class projects.

    Even if LAMP is gaining - its going to take its own time before it gets of the hands of the niche developers - and becomes a mass technology.

    Cheers,
    Viral
    www.c-sam.com
  170. RE: Is Java losing to LAMP and .NET?[ Go to top ]

    I have a couple LAMPs at home, but I do not have Java 'cause I do not drink coffee. You guess.
  171. If other technologies can do this...[ Go to top ]

    then I wouldn't mind switching to something else.

    The are the technologies I use, together:

    Hibernate, for ORM
    Spring, for business layer
    Tapestry, for pages
    Acegi, for security
    Drools, for business rules (used everywhere)
    Lucene, for search
    log4j, for logging
    JMS, for messaging
    JMX, for remote control

    My sites are extremely complex and require rules logic to be grafted in just about everywhere. But using the above technologies has made the amount of code *I* have to write extremely minimal.

    If there LAMP or anything else can do this better, I'm all for it.

    But I'm not listing my mp3's on some personal website. If that were all that I was doing, I wouldn't use the above technologies.

    Maybe LAMP has just added to the number of code-driven websites, rather than really taken marketshare.
  172. Closer to the steel[ Go to top ]

    These technologies are superior to Java in computational applications, says Andy Brown, chief technology architect at Merrill Lynch. "It's closer to the metal," he says. "When you write code on Linux there are less layers. You don't need Java for it."

    Sure. Amputate your legs and you'll be closer to the ground, but moving will probably be a challenge.
  173. Closer to the steel[ Go to top ]

    These technologies are superior to Java in computational applications, says Andy Brown, chief technology architect at Merrill Lynch. "It's closer to the metal," he says. "When you write code on Linux there are less layers. You don't need Java for it."
    Sure. Amputate your legs and you'll be closer to the ground, but moving will probably be a challenge.

    Indeed. But anyway, the idea that Java imposes any more layers that any other development system is nonsense. You could hardly get 'closer to the metal' that the highly optimised machine code produced by Java's Hotspot system.
  174. Arrogance or Ignorance[ Go to top ]

    Well I must say, after reading some of the comments in response to the Business Week article, I could not think of a better title.

    Loosen up guys, don’t be so stuck up about Java being the best language ever. Why are we not ready to hear a opinion?

    On the article I think its one of mixed opinions, some are based on facts and some hold no argument at all. The AJAX argument as readers have suggested here is baseless. AJAX is not a language/technology that you can take and compare with Java, its a paradigm. Its the way you choose to design your apps.

    On .Net argument, its true that it has made inroads where Java previously existed. Some case its been completely replaced and in some its a case of slow migration and .Net being picked for new apps. BTW...this is true the other way round as well. Lot of MS shops have picked Java/J2EE as the platform of choice.

    And then there’s always the case when a new technology enters the market. There's that buzz about it and everyone loves to give it a shot. Project managers and developers alike like to play around with it. And then in some cases it is rejected and in some it becomes the platform.

    About PHP, its definitely a serious contender for small-to-medium size "Web" applications. I am not talking about enterprise applications. Java definitely will get picked over PHP there. But the fact is that ease of use of PHP against Java (read too many frameworks for everything have convoluted the market) makes it a credible contender for small size applications where quick results are expected at least costs/time possible.

    I guess I mentioned PHP being a good choice for small/medium applications. Take a look at http://www.sugarcrm.com
    Its a decent enterprise app written in PHP, kind of throws my argument out.

    But anyway, the point I am making is that I don’t think its a good idea to throw the argument, presented in the Business Week article, out of the window completely.

    Imagine what the C++ folks would have thought or the numerous(may be similar) debates they had when someone suggested something similar about Java, back in the early mid-90s.

    I think the Java world has its task cut out. They need to simplify. We need to cut out the zillions of framework that has convoluted the Java market. It just makes decision making so much more difficult. Remember the project managers/decision makers/management don’t want to hear zillions of acronym and frameworks. We need to make their job easier. I know choice is good and these frameworks do serve their purpose(hell...I use them a lot), but may be consolidation is the word. They need to be reduced, for a fact and more standardization needs to be brought in.

    EJB 3.0 is a definitely a move in the right direction. Hibernate has served its purpose well too, a lot more Java shops are standardizing on that. Hopefully J2EE 1.5 will make a step forward in the right direction as well. Web Service’s is another area where we need to work a bit.

    I think Java's USP i.e., vendor choice, has also served a bit of negatives here as well. Too many vendors trying to pull away customers on their side of the world. Having said that, I think it has also brought in a lot of innovation as well. JVM/JDK is one area that comes to my mind. JRockit and IBM JVM are good examples of that. "Balance" is a key word here. The vendors need to strike the right balance between choice, innovation and not crowding the Java world with too many acronyms and technologies.

    The vendors also need to work closely with the Open Source world. A lot of good and cool innovation has come from their side. The open source can definitely act as a glue between different vendors. Eclipse is a perfect example of that.

    I also read someone mentioning blasting Business Week on this article - http://www.businessweek.com/technology/content/dec2005/tc20051212_622616.htm

    Let me throw a analogy here, remember when your company bought a big ass AppServer with all the cool technologies/acronyms built in and all you ran on top of it was a JSP/Servlet/Struts/Hibernate application. Well you could have run this damn thing on a JBoss, Tomcat, Resin etc. I guess that’s one of the arguments that the gentlemen is trying to make in that article. Makes sense?

    The big app servers definitely have a space in the market. Some of them bring unique capabilities that are not available in the open source solutions, like integration, high availability, Web Service infrastructure etc., just to name a few.

    Well that brings me to the end of my two cents or may be a little more. I just tried to bring in a different perspective to this discussion. Hopefully I did not piss off too many people :)

    Kapil

    ....yes I come from the Java world :)
  175. Lets look at the stats[ Go to top ]

    TOIBE December 2005 Index:

    http://www.tiobe.com/tpci.htm

    Java still way at the top and gaining stronger than the rest.
  176. Plain JSP with taglibs can beat PHP[ Go to top ]

    "Plain JSP with taglibs can beat PHP"

    -

    Clearly, people making these statements have not worked with PHP much ... if at all.

    JSP cannot come close to touching PHP's productivity - no chance in hell. That combined with the fact that Java web hosting is no walk in the park and that PHP is a breeze to learn, use and set up, PHP will continue to rule the small and medium sized web application market for years.

    -

    Java has been designed and guided by large corporations for years, and the language now feels like a large corporation - too many layers and too complex!

    PHP was designed by a bunch of web nerds who just wanted to get stuff done - and it feels/works accordingly.

    PS: PHP5 has fixed PHP4's crappy OO implementation.

    Stef

    www.killersites.com
  177. JSP 2.0 w/JSTL and tag files can beat PHP in terms of:

    1. Size of code

    2. Simplicity

    3. Maintanability

    4. Modularity

    I have actually used PHP quite a bit.. and I agree that it is very productive. Even though I do program much more in Java than in PHP, I could write many programs much faster in PHP than in Java.

    However, JSTL, EL and tag files can make it very easy (easier than PHP) to write simple dynamic web sites. No Java is actually needed. I think you are refering to the more arcane JSP 1.0 and scriptlets.. in that case I agree that PHP is more productive.

    I do however agree with you that web hosting issues with PHP are much simpler than with Java. I also agree that Java architectures tend to have too many layers and are too complex.. however, you aren't forced to use all that when designing simple web dynamic web sites.
  178. "Plain JSP with taglibs can beat PHP"-Clearly, people making these statements have not worked with PHP much ... if at all.JSP cannot come close to touching PHP's productivity - no chance in hell. ...

    So as PHP matures and becomes more and more like Java, the gap will be closed. You are infact stating that PHP evolves backwards ...
  179. Is Java losing to LAMP and .NET?[ Go to top ]

    My opinion: Yes Java is suffering decreasing popularity compared to Lamp. The reasons I see :

        * the absence of free mutualized web hosting for Java
              o the free availability of the JDK was revolutionary in its time, but now the web is the media.
              o It could have been done by external projects but the Java platform is in tight control by Sun with an unclear mixture of legal and technical ways, and alone Sun does not have the human resources to develop the platform as it should be (think of the startup time of the JVM compared to Python or Perl)
        * the perception that Java is complex
              o shure the Object Oriented technology needs experience; but also Java could be less complex:
                    + suppress the static keyword and make it a pure Object language like Eiffel
                   + make public fields read-only and get rid of the ugly getXXX() naming pattern (again like Eiffel)
                    + simplify the library and its documentation by grouping overloaded methods
              o of course the fail of the EBJ technology has done a lot of harm
  180. Is Java losing to LAMP and .NET?[ Go to top ]

    of course the fail of the EBJ technology has done a lot of harm

    What failure? Have you checked the huge number of projects and jobs that use EJB?

    Just because a technology may not be liked by some people, and because it may now be seen as too complex for some projects in no way means it has been a failure.
  181. LAMP, don't NET and Java are the major platforms in development arena... We can say java is “Grand father”..

    Do you respect your grand father (who obviously faced lot of challenges in his life and developed lot of successor)?

    Java seems grand father but difference is Java will dead but it will bring lot of successor... which will represent java as a young one....

    Think about Java EE's current trends EJB3, spring etc. which reminds the truth...

    So java is a set of successor ... it will never destroy or dead until world is destroyed…..

    Thank you
    NHM Tanveer Hossain Khan (Hasan)
    http://hasan.we4tech.com
  182. A Fundamental Flaw of J2EE[ Go to top ]

    A fundamental flaw of the J2EE stack compared to .NET is that the J2EE services (such as JTA) are only available to applications running *in* the J2EE containers, i.e., Web applications/services and EJB applications. These services are not available to other types of applications such as a standalone Java application (e.g. to be run as a cron scheduled job).

    In .NET, all .NET services are available to all types of .NET applications, be it a console application, a WinForm application, a Web application/services, a Windows service application, or a remote application.