|
Sponsored Links
Resources
Enterprise Java
Research Library
Get Java white papers, product information, case studies and webcasts
|
News
News
News
|
Messages: 139
Messages: 139
Messages: 139
Printer friendly
Printer friendly
Printer friendly
Post reply
Post reply
Post reply
XML
XML
XML
|
 |
IBM in talks to buy Sun Microsystems
From the article:
IBM is in talks to buy Sun Microsystems Inc for at least $6.5 billion, The Wall Street Journal reported, in a deal that could bolster their computer server products against rivals such as Hewlett-Packard Co.
That would translate into a premium of about 100 percent over Sun's Nasdaq closing price Tuesday of $4.97 a share, the paper said, citing people familiar with the matter.
Sun, which was not immediately available for comment, has long been cited as a takeover target for International Business Machines Corp, HP, Dell Inc or Cisco Systems Inc, which this week unveiled its plan to start making blade servers that power corporate computer networks.
[snip]
Link to Reuters Article
What would this mean for the future of Sun and Java technologies? Would IBM be a faithful steward of JCP and openness? What would happen with the Sun JVM and the IBM JVM?
While we all tire of "is Java dead?" discussions, I thought this one was interesting enough to post and generate discussion.
|
|
|
Message #306182
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
IBM has done some good things for Java, so I don't see how this is a bad thing. I do wonder if IBM will continue to support two JVMs though.
|
|
|
Message #306185
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
IBM has done some good things for Java, so I don't see how this is a bad thing. I do wonder if IBM will continue to support two JVMs though.
The bad I see is that having Sun and IBM supporting Java and the JVM gives it some redibility as a cross-platform vendor-neutral platform. While it was never 100% true it is an advantage. If this happens, Java may become completely associated with IBM.
|
|
|
Message #306187
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
<sarcasm> But the really important question is - would IBM take Sun's stock symbol? </sarcasm>
|
|
|
Message #306189
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
emphasis on "some", Calendar api would not be one of those
|
|
|
Message #306199
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Reference implementations
If WebSphere becomes the reference implementation for j2ee going forward, this will be looked back upon as a dark, dark day.
On the other hand, my impression is that IBM is out there using this stuff every day more than Sun was, so I'm hopeful that this will give Java some stronger direction.
|
|
|
Message #306200
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
IBM should buy Sun every week. That's a good way to keep the stock up.
|
|
|
Message #306201
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Netbeans vs. Eclipse
I wonder what this acquisition will mean for the Netbeans vs. Eclipse controversy. While both have their merits, there is indeed a tremendous overlap.
Steve Punte
Test Lens
|
|
|
Message #306203
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Netbeans vs. Eclipse
I wonder. If IBM -owned- Java, could Java SE 8 then possibly incorporate SWT?
|
|
|
Message #306204
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Netbeans vs. Eclipse
I wonder. If IBM -owned- Java, could Java SE 8 then possibly incorporate SWT?
Java SE shouldn't even incorporate AWT and Swing. If anything I would hope IBM would follow through with modularization.
|
|
|
Message #306206
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
my wish list
- adopt eclipse, swt and drop netbeans, swing, awt
- JVM becomes a osgi container
- adopt apache derby, and drop mysql
|
|
|
Message #306207
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
Sun, which was not immediately available for comment, has long been cited as a takeover target for International Business Machines Corp, HP, Dell Inc or Cisco Systems Inc
Well rather IBM than Dell or Cisco systems!
|
|
|
Message #306208
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
Sun, which was not immediately available for comment, has long been cited as a takeover target for International Business Machines Corp, HP, Dell Inc or Cisco Systems Inc
Well rather IBM than Dell or Cisco systems!
Or dead.
|
|
|
Message #306209
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Good and bad
This can be good news and bad news. IBM has been pushing Sun to make Java open source for some time now. Hopefully, under IBM, JVM, JAVA [SE and EE] everything will be open sourced.
A few things to note though:
1. The future of Solaris is unknown. I see it less likely that IBM will support both AIX and Solaris - Fewer choices in the server market and a good OS disappearing don't make me happy
2. The future of Glassfish and Netbeans is unknown. I like both Eclipse and Netbeans and I want both of them to be around. Same goes with Glassfish. I have always liked it more than JBoss or Geronimo (as a free container)
3. Other products such as Sun directory server will probably disappear
4. The future of JBI is definitely at risk. IBM never bought into the idea of the JBI spec. In fact, it came up with SCA
5. Sun has done a pretty good job of keeping the Java/J2EE certifications clean and pure (although I saw inclusion of Sun's Metro stack in the new SCDJWS exam, which I didn't like), I am not sure IBM will be able to keep the certifications clean
And finally, as a long time consultant specializing on IBM platform/products, I *speculate* that IBM considers Java/J2EE a mature and saturated market. Lately, I have been seeing a lot of new IBM products targeted for SOA/BPM market, which is a good thing and something I would expect from a thought leader, but I wouldn't want JVM/Java to get a second class citizen status under IBM because of its focus on SOA/BPM. In fact, I would love to see more standards and specifications to make solutions developed using SCA (BPEL, mediations, human tasks), CBE/CEI, WSRR, etc. portable across multiple containers.
|
|
|
Message #306212
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Good and bad
This can be good news and bad news. IBM has been pushing Sun to make Java open source for some time now. Hopefully, under IBM, JVM, JAVA [SE and EE] everything will be open sourced.
The funny thing is IBM has been pushing on Sun to open source things, but what have they open sourced? Nothing of substance. Essentially this was a brilliant PR move on IBM's part, but there was no substance behind it.
Sure, Eclipse is open source, but IBM has their own closed source products on top of it. Eclipse is essentially the bait and "commodity code" for IBM's expensive commercial products.
The other funny thing is when Sun open sourced Java via GPL IBM ran to Harmony instead (portions of J9 are now based on Harmony). GPL was too open for IBM -- it required them to openly provide all of their changes to Java which they won't do.
|
|
|
Message #306214
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
IBM Informix!
...so what happens to MySQL now?
What happened to Informix, it didn't exactly conquered the world nor became a bleeding edge technology, did it?
|
|
|
Message #306215
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Reference implementations
If WebSphere becomes the reference implementation for j2ee going forward, this will be looked back upon as a dark, dark day.
Couldn't agree more. WebSphere is the most 'complicated' and unusuable piece of software I've worked with in the EJB world.
|
|
|
Message #306216
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
no worries
But then again, isn't this a common concern with technology company acquisitions.
Take for example Oracle's acquisition of BEA. Isn't it the case that Oracle's Applications server (based on OC4J)also have overlaps with BEA's WebLogic(JRockit)? A similar case is with Oracle's HR application and Peoplesoft's. But what eventually happens is that the acquiring company offers both software as flavors of their solution. Although this approach usually spurs another round of debate, this is eventually what happens anyway.
Hopefully, IBM will initially provide the IT community even just high-level information as to their plans with product directions to assuage strong concerns and address
rogue speculation.
|
|
|
Message #306217
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: my wish list
My wish list:
- HP steps in and counter offers to acquire Sun
- HP adds the engineering and vision to Sun that it doesn't have now
|
|
|
Message #306218
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Netbeans vs. Eclipse
I wonder what this acquisition will mean for the Netbeans vs. Eclipse controversy. While both have their merits, there is indeed a tremendous overlap.
Steve Punte
Test Lens
No worries.
This is a common concern with technology company acquisitions.
Take the case of Oracle's acquisition of BEA. Their application server (OC4J based) and BEA's WebLogic AS(JRockit) had significant overlaps too (in fact, the same is true with their SOA products). We find a similar situation in the enterprise applications arena, particularly Oracle's HRMS and Peoplesoft's.
But what eventually happens is the acquiring company simply offers the software entities as flavors of their solution offering. Though this approach spurs another round of debate, this is eventually what happens anyway.
And it will be the market and, broadly, the user community who will ultimately determine if this strategy is indeed effective and pragmatic as it is hoped to be, or just a decision being deferred (instead of just deciding to select which becomes the only solution offering).
This is just my opinion anyway. Cheers.
|
|
|
Message #306219
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: my wish list
My wish list:
- HP steps in and counter offers to acquire Sun
- HP adds the engineering and vision to Sun that it doesn't have now
Your wish list version 2 ...
After HP is able to acquire SUN, Oracle buys HP (or vice versa). This is a more complete package to rival IBM's deep offerings - hardware, software, and strategic architecture & engineering.
|
|
|
Message #306221
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
It's just a beginning
What WSJ didn't say or do not know is other vendors are also in talk to buy Sun. Oracle, HP, DELL, or even Cisco. This IBM-Sun specific news is on purpose.
What actually happened yesterday ?
* NASDAQ:JAVA rose 78%, good to Sun for negotiation.
* Good feedback to IBM's attemp, pressure to other potential buyer, also good to Sun for negotiation.
* Southeastern Asset Management (Sun's largest shareholder) probably have sold lots JAVA shares. good to Southeastern AM.
What will probably happen...
* Jonathan's "strategy" is proven dead, customer feel uncertain on Sun's future. Good to Sun's competitor, especially for IBM, HP.
* Sun will continue to ignore this "rumor" intentionally before the stock price drop back to "normal" value.
* Rumor is always rumor, IBM-Sun deal is just a joke.
|
|
|
Message #306226
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Certainly not a joke -2002 article about IBM wanting to buy
* Rumor is always rumor, IBM-Sun deal is just a joke.
2002 interview with Mills (IBM software exec) when IBM thought about buying it ..At that time it would have cost $40B now they are talking $6B
http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2860393,00.html#talkback
-" In addition to new revenue streams from the licensing fees generated by Java, this is exactly the position in which IBM needs to be."
-As a tool for marginalizing Microsoft, Java is everything OS/2 was not.
-Java is the heart and soul of IBM's WebSphere (the middleware for the rest of IBM's software portfolio).
-Java is the primary software vanguard that keeps Microsoft from penetrating the datacenter. Even better, Java represents the first credible threat to Redmond on the client side.
-If IBM did manage to bring Java into its intellectual property portfolio, some of IBM and Steve Mill's biggest competitors-BEA , Oracle , CA -would be beholden to IBM in the same way that IBM is now beholden to Microsoft.
Considering the price and the abbility to take control of Java and large scale Unix ( which is struggling fighting cheap intel machines) markes its a bargain for IBM . I suspect Solaris will eventualy die after some key features get moved to Aix ( like the newer file systems).
The future for Java is very interesting the goals of IBM are obvious , eg try to get some sort of dominance like IBM had and MS has now and use this as a computer tax.
IMHO IBMs open source efforts have nothing to do with any commitment to it , only to enhance its profit ( eg it failed with its own desktop OS OS2 ( and the follow up project) and used the popularity of Linux as it was cheaper ) .
A lot of Open source projects will struggle without Suns support.
|
|
|
Message #306227
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Sun has no value to anyone but MS Or IBM
Suns revenue line is SUN hardware this has no value to anyone but IBM , they will remove a competitor ( ie a monopoly on high end Unix servers) and get some technology which they can merge with their current offerings .
$7B is not a lot for IBM and MS however it is a lot for a company loosing money with no prospects of making it. Microsoft may come up with that money ( they have that in cash) just to spite SUN and hold a monopoly on software development. However other firms don't have the cash eg HP , Oracle nor can they make a business case from the deal. Hp have a glut of high end UNIX system and have been trying to reduce them , they just bought EDS ( i think) for $20B mainly with debt ( it has 90B in liabilities on its balance sheet) . While IBM also has this issue they want to get control in the software market and hence Java provides a big push , i suspect ( as per the article) that IBM software is pushing for this but IBM p system doesnt want it.
The only company with a war room this week for Sun will be IBM and MS. Take your pick ...
|
|
|
Message #306228
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: no worries
We have not seen the end of the BEA Oracle merge yet. Still the OAS is used as the default server in their suites like the SOA suite.
If you listen to Oracle people of records they still cling on to OAS.
On the BPM, ESB, process (BPEL) side BEA had overlaps that they could not handle (even if they liked to sell all their product to you). Now they have even more overlap.
|
|
|
Message #306229
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Sun has no value to anyone but MS Or IBM
The only company with a war room this week for Sun will be IBM and MS. Take your pick ...
I would be far more concerned from having Sun acquired by Microsoft. This would be a threat to the whole Java ecosystem - taking out the only real competitor besides PHP would be a definite advantage.
IBM on their side have prospered on Java.
|
|
|
Message #306230
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
Fear and terror: if the temple of bureaucracy acquires Sun, the developers tribe will immediately escape from Java, and the lands remaining are a sad comparison to Java or C# :-( . I really hope it is just a joke.
|
|
|
Message #306231
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
what about Glassfish
What will happen to GlassFish and friends: OpemMQ, OpenDS, OpenPortal and Hudson.
These project represent the most interesting evolution to come out of SUN lately, will they die or be killed to save websphere?
|
|
|
Message #306232
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
Fear and terror: if the temple of bureaucracy acquires Sun, the developers tribe will immediately escape from Java, and the lands remaining are a sad comparison to Java or C# :-( . I really hope it is just a joke.
I believe that Sun's wrongdoings with Java (particularly the JCP and the resistance to open source the JVM) cannot be exceeded; on the other hand IBM has contributed a lot with Eclipse and pushing OSGi. Or am I missing something?
|
|
|
Message #306234
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: what about Glassfish
What will happen to GlassFish and friends: OpemMQ, OpenDS, OpenPortal and Hudson.
These project represent the most interesting evolution to come out of SUN lately, will they die or be killed to save websphere?
OpenMQ is certainly overlapping with WebSphere MQ, although IBM might consider OpenMQ the open-source, unsupported alternative to WebSphere MQ.
I personally do not know if OpenDS overlaps with any IBM offering.
OpenPortal has already many competitors, even in the open source market, and I personally do not think it is worth surviving.
Hudson - well, this is an interesting point. I would expect an enterprise version with some degree of integration with the IBM Software Delivery Platform, but it is only a guess.
|
|
|
Message #306235
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Other possible matchs ?
Are possible other matchs ?
Google ? May be interested as a BIG open source+hardware consumer. But does not fit very well with Java on the client side.
Apple ? May be a good match. Apple = strong positions on marketing side+client side. SUN = pretty good technology on the server side, and less good for marketing.
Dominique
http://www.jroller.com/dmdevito
|
|
|
Message #306237
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Other possible matchs ?
Maybe SpringSource? :)
ROFL. You had me going for a minute! I owe you a beer :)
|
|
|
Message #306238
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
do you or have you ever worked for a large corporate where you HAD to use IBM software? You might change your mind if you have.
|
|
|
Message #306239
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
IBM and Mobile World
Sun is a powerhouse for server side, that's for sure. However, J2ME is a pervasive tech on every mobile phone (except Iphone). This is aspect of the potential deal is being missed out. This might be a very indicative for what IBM is actually aiming at. Google and others who want to take a share on the mobile market might find themselves against a giant unexpectedly. See similar discussion at experimentxero.blogspot.com.
|
|
|
Message #306240
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
do you or have you ever worked for a large corporate where you HAD to use IBM software? You might change your mind if you have.
I am presently an IBM employee. But <legal>my opinions do not necessarily represent the ones of my employer</legal>
|
|
|
Message #306251
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
I am presently an IBM employee.
I believe that Sun's wrongdoings with Java (particularly the JCP and the resistance to open source the JVM) cannot be exceeded; on the other hand IBM has contributed a lot with Eclipse and pushing OSGi. Or am I missing something?
Due to limitations caused by English not being my native language I cant exactly describe what you did there, but if you were pushing agenda, then I admit you are doing great job, otherwise if you truly believe what you just said, you need reality check.
|
|
|
Message #306253
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Do we need another AIG?
AIG is such a big company that nobody can let them go bankrupt because it would impact the whole world! I have heard some people are suggesting to create a kind of regulatory to prevent creation of such giants in future!
Shouldn't we learn lessons from what's going on in financial market and apply them to IT? Do we really need such a big giant that would have control on all aspects of client's business?
|
|
|
Message #306256
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Sun has no value to anyone but MS Or IBM
Microsoft may come up with that money ( they have that in cash) just to spite SUN and hold a monopoly on software development.
Ahem... if you want to see a beeline of lawyers to Washington DC, sure :), MS can come up with the money.
|
|
|
Message #306257
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
This is great news! Finally, IBM will GPL ZFS & Linux will get a genuine ZFS port, instead of having to run it via FUSE!
|
|
|
Message #306258
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Not sure if AIG analogy applies
AIG is such a big company ...Do we really need such a big giant ...
The logic with AIG was, if AIG fails some other big banks would fail and that would impact the entire financial industry and everything else that depended on it.
While I am against big giants for monopoly and innovation reasons, I am not sure if the AIG analogy applies here. What do you think would be the implications of IBM or Microsoft failing? And are they really comparable to those of AIG failing?
C
http://ChintanRajyaguru.com
|
|
|
Message #306259
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Analogies
I am not sure if the AIG analogy applies here
Not really, but the overall moral of "don't become dependent on any supplier" is a good one to stick to; or in other words stick to software implementing standards and if the supplier disappears (or its buyer discontinues that software) then you can just switch to an alternate implementation of the same standard. There's plenty of evidence for what can happen to software from recent takeovers and the subsequent dropping of some well-used software, so any self-respecting project would be bearing all of this in mind in their choices
|
|
|
Message #306265
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Really bad news
It means we will loose a lot of very good products like Glassfish, Netbeans, OpenDS, Husdson, and must move to really bad products like Websphere application server, Websphere portal, RAD, etc ...
If this is true, i will quit java.
(Please excuse my English)
|
|
|
Message #306266
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Really bad news
It means we will loose a lot of very good products like Glassfish, Netbeans, OpenDS, Husdson, and must move to really bad products like Websphere application server, Websphere portal, RAD, etc ...
If this is true, i will quit java.
(Please excuse my English)
I am surely biased here but I believe that you, like any other here, will follow direction the market is heading to.
|
|
|
Message #306268
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Get Real
OK, if WebSphere and RAD are so bad, why does IBM have market share on Application Servers? Time to get back to reality. Sorry for venting, but I have been around I/T for a long time and worked for several top notch companies where we look at solutions from all angles including viability of the company staying in business, overall architecture including the solution meeting functional and non-functional requirements, product support, and function/feature. Many products we thought were cool at the time, ie. Smalltalk have gone to the sunset. Why? Many reasons, including the product never archieving and momentum in the marketplace, customer concerns about buying from a company that may not be in business in next 6 months, etc. You have to put on your business hat in addition to your technology hat today. This is especially important today with the economic situation we are in.
|
|
|
Message #306272
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Get Real
Microsoft has a big market share in operating systems but its product is not really good! Same applies to IBM, so many non technical reasons Ok, but i am very close to people selling IBM products and at least here, many colleagues agree with me: working with IBM products is like have a night in the hell.
|
|
|
Message #306273
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Get Real
viability of the company staying in business
Isn't that something like vendor lock in. Instead, if the company / project adopted open standards such that the vendor is removed from the equation then wouldn't that pose lesser risk in any case. Plus it will be a more robust implementation that would be more flexible.
|
|
|
Message #306274
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
No other matches
There are no other matches , SUN will collapse with the current economic down turn corporates wont buy SUN servers ( instead cheaper Linux etc), if it goes on they will burn through their cash.
Now most IT companies will loose money big time in 2009 (especially google) , how can you make a business case on buying a loss making company with little prospects. Note SUNs business is hardware which is nothing to do with Google , the Java view slants the business case.
|
|
|
Message #306275
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Get Real
because their salesladies wear shorter skirts? no seriously, i'm curious too...
|
|
|
Message #306276
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
There is always C++. It is also worth noting C# is not owned by MS the language is now independent( the .NET framework on the other hand is MS) , the only major language to be owned by a Company is Java/SUN (Cobol , C,C++ , Pascal , Modula2, Basic , C#,F# , Python, Ruby).
Java is almost unique because it grew so much with only one owner under the guise of OpenSource , in most cases an Academic (C/C++) or Company(C#) creates the language wanting it to be big and hence releases the standard as public. Once it is big there is no reason to do this.
eg You can use C# under Mono with no MS involvement and its multi platform , however you dont get the huge commmunity you get with Java ( with Mono).
|
|
|
Message #306277
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Get Real
Its worth noting Java is one of the few languages which is not open. While a lot of the code has been released you cant use it as the logo , name and language are SUN ( and soon to be IBM) IP. eg MS (who negotiated a licence) got sued by SUN for adding a few win32 APIs to the IE JDK to make it easier for developers to use and hence it did not comply to the standard. So while you can use the Open source Java if you make changes that cause it to stray from the official API you are liable ( unless you call it something else and dont use the Java name etc).
Compare
C# - Language standard in the public domain and anyone can implement it. OPen source implementation Mono.
C++ - Same as C#.
Java - Language standard controlled by Sun , but implementation is Open source. So when Java v3 language standard is released the open source released may be modified to be v3 compatible however it may not get the licence to be called Java v3 unless Sun Agrees.
If IBM strayed too far people may create J3 or J# as their own (Open) standard but this would be hurtful and fragment the market.
Note im not 100% on all these legal matters.
|
|
|
Message #306279
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Analogies
I am not sure if the AIG analogy applies here
Not really, but the overall moral of "don't become dependent on any supplier" is a good one to stick to; or in other words stick to software implementing standards and if the supplier disappears (or its buyer discontinues that software) then you can just switch to an alternate implementation of the same standard. There's plenty of evidence for what can happen to software from recent takeovers and the subsequent dropping of some well-used software, so any self-respecting project would be bearing all of this in mind in their choices
I think the point is "Big is bad for everyone but those who are big".
|
|
|
Message #306280
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
Fear and terror: if the temple of bureaucracy acquires Sun, the developers tribe will immediately escape from Java, and the lands remaining are a sad comparison to Java or C# :-( . I really hope it is just a joke.
I believe that Sun's wrongdoings with Java (particularly the JCP and the resistance to open source the JVM) cannot be exceeded; on the other hand IBM has contributed a lot with Eclipse and pushing OSGi. Or am I missing something?
Have you used IBM's JVM's? How about WebSphere? Have you notice the unavailability of source code for these and all other IBM product? Have you noticed how far they are behind everyone else in terms of adopting and implementing new Java versions and standards?
If you feel that Java is too open, that the pace of Java development is too fast, and want Java 7 in 2012 instead of 2010, then IBM's just what you want. Otherwise...
|
|
|
Message #306281
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
IBM see the value, but Sun can't, why?
In my previous reply to this subject, I said it is just a joke.
The reason is the motivation behind this "news" is unclear.
From IBM's point of view, if in the talks, no reason to leak such information. In contrast, Sun and its #1 shareholder SEAM could take advantage by spreading such "rumor".
I also believe Sun has been seeking a buyer for a while, but it is a hard one to buy. Sun is a collection of too many products, too many IP, too many assets , and too many high-paid employees.
As a ex-Sun working on middleware software delivery. It is the first time to hear from outsider saying the value of Java. Although many people still use Server Market Share to figure out the pro/con, but I think Java and several key software technologies is the important factor to make Sun as a good acquisition target.
But during my time in Sun, hardware business occupies almost all resources. Software team (especially the middleware) and its support/consultant/delivery feel like 2nd class citizen. They only get respect on Sun's stock symbol:JAVA.
Sales are all hardware oriented, selling Sun software product is only for selling more hardware. In many deals, 70% $$ go to servers/storage, 20% $$ go to SW license, 10%$$ go to poor professional service who will cover all product, and implementation issue. Software and SW Service people at fields are dying. Now most employee stay at field are Server Sales. To them, software means Solaris, they do not know shit about Java. They are sinking with the giant boat.
Even though Jonathan's "free for more adoption" strategy is right, nobody will "EXECUTE" it. The quarter by quarter revenue goal make people leave sw/ps deal as possible as they can. If they cannot sale big box, then try to sell many small boxes, if not, then cheat. They do not want to touch sw/ps because they cannot do customization and cannot solve issue for customer. SW/PS revenue cannot be recognized immediately like HW. SW/PS project bring lots of post-service issue Sun has no resource to take care (eventhough it is an opportunity).
OK, back to the "rumor"
The acquisition won't happen soon, who the buyer is ...unclear. But let's guess what gonna happen to Sun's products once IBM buy Sun.
* IBM will certainly be the driving force of JDK development and Spec.
* IBM will not throw resource to Glassfish, Netbeans, and many other Sun's opensrouce project such as OpenDS, OpenSSO, OpenPortal.
* IBM will try to keep the key developers of these opensource projects. but ask them to do more for Websphere or Tivoli products.
* IBM will make ZFS and DTrace available on Linux and AIX.
* IBM will abandon Solaris. Then its recipe become: AIX+Linux+Win.
* Hardware ? Storage ? Those Sun sales will be wiped out.
* SPARC will be terminated or sold out as soon as possible. No more ROCK cpu.
* Some technical follow will be kept.
* Open Office ? Star Office ? hm.. keep the talent but drop the product.
That's what I guess
|
|
|
Message #306282
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Hell?
working with IBM products is like have a night in the hell.
Hmm... hell from what perspective? Compared to what? I would think you would know hell only when you know what is not hell. I have always been an IBM consultant (not an employee but it just happened that I ended up working on projects involving IBM products) and my experience with IBM tools is pretty good. Yeah I have had my frustrations with them but that goes for open source too.
Naming names, when I had to find some information on Informatica PowerCenter and CA's Autosys, I ended up with a new found respect for IBM for providing great documentation around its products. Good luck finding good documentation on PowerCenter and Autosys (I know BEA/Oracle are different stories).
so many non technical reasons Ok
But then you are giving more non-technical reasons.
but i am very close to people selling IBM products...
I am not saying you are wrong. I am just curious as to what technical reasons made you feel like hell working with IBM products?
Just looking for enlightenment, not flames...
C
http://ChintanRajyaguru.com
|
|
|
Message #306286
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Worse than Hell actually
I have had the benefit of starting on a Solaris/Oracle/Weblogic for a few years, before I had to move to a AIX/DB2/WebSphere (And I have been working on this stack for the past few years as well) stack. All I can say is working with IBM has been worse than hell.
IBM really can't write good software even if their life depended on it. Give you a simple example. We wrote an application, deployed it to the production server and it was working. Then we told the WebSphere admin to restart the server to make sure everything is configured properly to ensure the application can start if they ever needed to restart the app server. Lo and behold, the application could not start up and errors were being thrown out which did not make any sense. We asked the WebSphere Admin to restart the app server one more time, and the app was working again. We kept trying this and we noticed the problem was occurring intermittently. When we called IBM Support they told us it was a problem with our application, and it was not J2EE compliant. To prove them wrong, we deployed the application on WebLogic, JBoss and Glassfish and the application did not ever encounter any errors. When we bring this up, their response to us was that we need to develop the application using WSAD. I'm sorry, but I don't know in which section of the J2EE specification did it say that an app must be written in WSAD for it to be considered a J2EE application. Till this date, the official recovery procedure provided to the WebSphere administrator is to restart the app server if the problem occurred, coz IBM could not provide us with a solution to the problem.
On the AIX front, we have a minor ETL job that has to deal with a legacy backend to update into a DB2 database. The legacy app does a full dump of all it's data into flat files and the ETL job does a diff with the file unloaded the day before and processes any deltas that were encountered. The entire process was running on Solaris server for years. Eventually, due to management decisions, it was decided to migrate this functionality to an AIX box. Once moved over, the diff command on AIX failed on certain files. Albeit these were big files, the diff command still failed. bdiff was not a viable option due to the manner in which it operates. We raised this issue to IBM and all they could tell us was that we should use bdiff and did not provide a reasonable explanation of why it would work on Solaris and not on AIX. Due to some digging by one of my technical leads, he managed to find some documentation (in the Tivoli Access Manager documentation), about setting the LDR_CNTRL parameter before the process to allocate more data segments before the process. We tried that and amazingly some of the files that the diff command was failing on previously, started working. There was still one file that was failing even with that setting. We had to be the ones to tell IBM about the LDR_CNTRL option, and it took them about a week to find and respond to us with someone who actually knew what the LDR_CNTRL setting was about. However, they still could not help us solve the problem with the single file that was still failing. Eventually, we took the Linux Tools for AIX package, and use the diff command from there, and it worked for all files. Honestly, the diff command has been there for so long. It does a very basic function. How the hell does someone actually screw it up when there are so many open sourced implementations which can act as a reference.
Informix support provided by IBM sucks as well. We had performance issues with certain functionality on Informix which did not make sense. Just out of curiousity, we switched from using the SHM option to the TCP option. We immediately got a 50% performance increase. We raised this to IBM and till today, they have yet to provide us an explanation as to why the shared memory option is slower than the TCP option when everything is running on the same box, or a viable workaround for the situation, even after visiting the data center to take a look with their engineers.
Don't get me started on DB2. We had so many issues with DB2 that make no sense whatsoever, I don't even know where to begin. Worse, when they could not provide a solution, we put in a temporary workaround while waiting for a solution. Eventually, the DB2 version was EOL/EOS, with no solution provided by IBM. And when we follow up with them, their response now is that our version is EOL/EOS and we have to upgrade before they can support. However, the idiotic part about it is we raised the issue whilst the product was still under support, and due to the inability to provide support, we now have a problem with no solution and we are being asked to upgrade? When we ask them whether they are willing to guarantee that upgrading the version would solve the problem, they cannot answer the question. So why should we upgrade when we get such crappy support ? In fact, my DB2 Administrator Team sits beside my Oracle Administrator team. And whilst my DB2 Administrator team is constantly busy with priority 1 tickets, my Oracle team has pretty much nothing to do and focus on improving the performance of the database and planning for upgrades, migrations, and other activities.
Even on the hardware front, our SAN storage from IBM was having some problems, and our engineers were pretty confident that it was a hardware fault. IBM took over 2 weeks to actually admit it was a hardware fault and replace the faulty hardware. And only did so after we got our infrastructure managers to send an e-mail to our account manager from IBM and include the IBM GM in the e-mail.
I hope that gives you an example of why I hate IBM and working with IBM has been worse than hell for me. The only reason I am currently stuck with working on the IBM platform is due to the direction set by the head office to use IBM products and solutions (as far as I understand from them, it's only coz IBM is cheaper). Pretty much my entire infra team would love to move away from IBM, and we are even recommending Wintel boxes where possible to avoid having to deal with the issues from IBM.
I have been working as a solution architect for about 10 years now. When I first started with the Solaris/Oracle/Weblogic stack, most of my time was spent designing solutions and coming up with new and innovative products and services. Today, I spend most of my time trying to figure out why simple things which should work don't simply coz it is being run on an IBM platform and why errors occur on applications that were working perfectly fine and suddenly stop working for no apparent reason.
Therefore, to me, if IBM really does buy out SUN and decides to take over the Java Platform, I am quite confident I will just move into the MS space and work on .NET as I have little confidence in the abilities of IBM.
I am sorry about the long post, but I have so many horror stories to share with regards to IBM, and I truly hate having to work with IBM solutions at the moment. I would much rather spend my time on more productive activities versus finding workarounds simply coz it was an IBM platform.
Regards
Jega
|
|
|
Message #306288
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Its worth noting Java is one of the few languages which is not open.
Looooooooool. C# more open than Java. Seriously, now I have seen it all.
|
|
|
Message #306294
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Other possible matchs ?
Maybe SpringSource? :)
ROFL. You had me going for a minute! I owe you a beer :)
LOL
|
|
|
Message #306295
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Hell?
My experience:
10 years developing java SE/J2EE/JEE products, deployed on Tomcat, Jboss, Glassfish and no problem at all, and when i found some bugs or problems, a lot of community help comes to me.
I am a consultant right now, and i have been working on many customers using variated SW/HW stacks, no matter what kind of product the are developing, in general if they are working with products like Jboss or Glassfish they are happy people like me, but if they are using something like Websphere they are highly frustrated: old specs, unreasonable complicated install procedures, expensive and bad support.
And have you ever installed WebSphere Portal? first you need a Supercomputer to get it started (if you get it successfully installed)
Or have you ever used RAD? last time i installed it, it occupies 1GB in my hard disk, and needed 2GB of RAM to run so slow.
Maybe some people loves IBM products, it is possible, but my experience is that sales departments are happy selling IBM products because they get high incomes in licensing renovation each year, but technical teams hate it.
|
|
|
Message #306296
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
I will second and third that sentiment. I've worked for IBM in the past and have used several different versions of Websphere over the last 12 years. Once you get Websphere working, it's stable, but getting there is like a spanish inquisition. Thing that should be straight forward are hard and things that are hard are horribly painful. Take DB2 UDB for example. Large complex correlated subqueries that compile fine in Sql Server, and Oracle will fail in DB2. Often times, you have to increase the compiler buffer to get DB2 to compile large queries. Then there's the stupid query compiler in DB2 UDB, which doesn't like extra line breaks. Who in the world wrote the sql compiler in DB2 and made it so that it doesn't follow standard compiler theory and practice? DB2 GUI tools strip out extra line breaks to get around that stupid bug. I for one hope IBM doesn't buy SUN.
IBM has produced some great products, like eclipse and SWT, but websphere and DB2 aren't user friendly products. Less choice isn't good for the long term health of the IT industry. My bias 2 cents.
peter
|
|
|
Message #306299
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
IBM has produced some great products, like eclipse and SWT, but websphere and DB2 aren't user friendly products. Less choice isn't good for the long term health of the IT industry. My bias 2 cents.
peter
I think we need to remember how bad Eclipse was before it was open sourced. I think it's only become a decent IDE in the last few versions. I think getting pretty good now but I don't really credit IBM for that. The parts that are still crappy can be traced back to the original version that was open sourced. For example, the errors and problems panes are stupid. When my project can build, I can't just hover over it in the navigator. I have to go into the 'problems' pane, skim past a bunch of nonsense that I don't care about and find the issue.
And SWT is OK but it has the flaw that so many IBM APIs have: a gazillion int constants. I use IBMs MQSeries APIs and when there is an exception, you don't get any text description. You get an int error code. Then you need to go and find the meaning of that code. What I normally do is use the compiled binary of the exception class to find the number and look at the name of the constant it's assigned to. This is usually enough to determine what's wrong. Think about that for a moment. They took the time to put each of these names in their source but didn't put it in the exception message. There's no excuse for this kind of stupidity.
It seems like a lot of IBM devs think that Java is C or they are coding everything in C and porting it to Java. In either case, this pattern of behavior frightens me.
Ultimately this all comes down to IBM's biggest problem: arrogance. If you can't accept that what you are doing is not the best possible approach, you can't improve.
|
|
|
Message #306300
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
IBM has produced some great products, like eclipse and SWT, but websphere and DB2 aren't user friendly products. Less choice isn't good for the long term health of the IT industry. My bias 2 cents.
peter
I think we need to remember how bad Eclipse was before it was open sourced. I think it's only become a decent IDE in the last few versions. I think getting pretty good now but I don't really credit IBM for that. The parts that are still crappy can be traced back to the original version that was open sourced. For example, the errors and problems panes are stupid. When my project can build, I can't just hover over it in the navigator. I have to go into the 'problems' pane, skim past a bunch of nonsense that I don't care about and find the issue.
And SWT is OK but it has the flaw that so many IBM APIs have: a gazillion int constants. I use IBMs MQSeries APIs and when there is an exception, you don't get any text description. You get an int error code. Then you need to go and find the meaning of that code. What I normally do is use the compiled binary of the exception class to find the number and look at the name of the constant it's assigned to. This is usually enough to determine what's wrong. Think about that for a moment. They took the time to put each of these names in their source but didn't put it in the exception message. There's no excuse for this kind of stupidity.
It seems like a lot of IBM devs think that Java is C or they are coding everything in C and porting it to Java. In either case, this pattern of behavior frightens me.
Ultimately this all comes down to IBM's biggest problem: arrogance. If you can't accept that what you are doing is not the best possible approach, you can't improve.
I agree with you eclipse and SWT have benefited tremendously from being open source. Having written plugins for eclipse 2.x, 3.0.x and 3.3.x, it has improved a lot over the years.
I couldn't agree with you more about IBM's general handling of error codes. Personally, I find IBM's documentation for Websphere, MQ and DB2 completely user unfriendly. It's a mountain of information, but organized in an unusable format. Back when I worked at IBM, I had to setup DB2 Database Partitioning Feature and it was a complete pain. IBM has a lot of smart engineers, but the IBM process has a way of turning out user unfriendly products.
peter
|
|
|
Message #306302
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Hell?
But, the thought of working with WSAD and WebsFear is much deeper than Hell!
Notice R&D and innovation are terms of extinction; can't the US bail out SUN to protect citizens from IBM, Oracle, Microsoft and others?
Perhaps, the EC could make SUN another Airbus, folks this is very bad news for us "open" minded...
|
|
|
Message #306304
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
I couldn't agree with you more about IBM's general handling of error codes. Personally, I find IBM's documentation for Websphere, MQ and DB2 completely user unfriendly. It's a mountain of information, but organized in an unusable format. Back when I worked at IBM, I had to setup DB2 Database Partitioning Feature and it was a complete pain. IBM has a lot of smart engineers, but the IBM process has a way of turning out user unfriendly products.
peter
IBM documentation is usually fairly complete, if you can find it. I have bookmarks to pages that I spent hours finding on IBM's websites (apparently they need a domain for every location) and when I go back to them, I regularly get a 'page not found' response. The same thing goes for links I find from forums via google. I can't tell you the number of times I've found a forum where someone responds on the exact issue I am trying to solve and provides an IBM URL which likely has the information I need and the page has moved with no information about where. Who f*#king does that? Move the URL, fine but make sure you forward the old one. It's not rocket surgery.
|
|
|
Message #306306
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Hell?
But, the thought of working with WSAD and WebsFear is much deeper than Hell!
Notice R&D and innovation are terms of extinction; can't the US bail out SUN to protect citizens from IBM, Oracle, Microsoft and others?
Perhaps, the EC could make SUN another Airbus, folks this is very bad news for us "open" minded...
It's called RAD (Rational Application Developer) now. Same shit, different name. You do get useless UML diagrams for free though.
I was working with a contractor and I said he could use Eclipse straight-up or RAD. I didn't care. He wanted to use RAD because it had extra features. So after a while we were talking and I asked something about him using RAD. He said he switched to Eclipse because he was having too many problems with RAD.
The really sad thing is that if IBM would just sell plug-ins (OSGI bundeles) for Eclipse they might actually be worth paying for. But RAD (at least what we have) is based on an old version of eclipse. This not only causes it to be not as good as Eclipse, it means a lot of other plug-ins for things like source control don't work properly.
|
|
|
Message #306305
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
I couldn't agree with you more about IBM's general handling of error codes. Personally, I find IBM's documentation for Websphere, MQ and DB2 completely user unfriendly. It's a mountain of information, but organized in an unusable format. Back when I worked at IBM, I had to setup DB2 Database Partitioning Feature and it was a complete pain. IBM has a lot of smart engineers, but the IBM process has a way of turning out user unfriendly products.
peter
IBM documentation is usually fairly complete, if you can find it. I have bookmarks to pages that I spent hours finding on IBM's websites (apparently they need a domain for every location) and when I go back to them, I regularly get a 'page not found' response. The same thing goes for links I find from forums via google. I can't tell you the number of times I've found a forum where someone responds on the exact issue I am trying to solve and provides an IBM URL which likely has the information I need and the page has moved with no information about where. Who f*#king does that? Move the URL, fine but make sure you forward the old one. It's not rocket surgery.
I think everyone that has used Websphere, Db2 or RAD has come across that problem. Personally, when that happens, I find myself cursing up a storm. Looking at that problem from the inside, it's the corporate culture of IBM. Plus, making their products hard to use drives revenues for the service division. 1/3 of IBM's revenue is from services, so there is a method to the madness. As a end user, I find that frustrating and avoid IBM products when I can.
peter
|
|
|
Message #306307
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Plus, making their products hard to use drives revenues for the service division. 1/3 of IBM's revenue is from services, so there is a method to the madness. As a end user, I find that frustrating and avoid IBM products when I can.
That's what so mind-boggling to me: how they get away with that. And it's not just IBM. Sun's guilty of this too; they just never made any real money off of it. But somehow, this doesn't result in people moving away from their software. I don't understand why the people purchasing this software can be so naive about it.
|
|
|
Message #306310
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Plus, making their products hard to use drives revenues for the service division. 1/3 of IBM's revenue is from services, so there is a method to the madness. As a end user, I find that frustrating and avoid IBM products when I can.
That's what so mind-boggling to me: how they get away with that. And it's not just IBM. Sun's guilty of this too; they just never made any real money off of it. But somehow, this doesn't result in people moving away from their software. I don't understand why the people purchasing this software can be so naive about it.
probably stating the obvious, but it's because the execs probably went to school together or play golf together. it's got very little to do with technology, but I might just be overly cynical.
|
|
|
Message #306317
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
probably stating the obvious, but it's because the execs probably went to school together or play golf together. it's got very little to do with technology, but I might just be overly cynical.
I think it has a lot to do with critical mass. Because software companies are so volatile, knowing that IBM will be there in a few years is more important than the quality of their product (from a business perspective.) I've been studying game theory and this is a common situation. Because of what everyone else is doing, your options are limited to one choice and everyone else is in the same situation. The game (in the game theory sense) needs to change before the other option becomes viable. Some of this has to do the perception of the situation, though.
Having said that, I think companies tend to under-estimate the impact that poor quality software has on their business. It's not valued properly. I think if more managers understood how much value good software can provide, they'd be less worried about the cost and more worried about what they are getting.
|
|
|
Message #306319
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
This is bad news people, but the solution is simple. No cussing and bellyaching required. Just start doing your development work using Microsoft tools and platforms.
I know that's what I'll be looking at doing if this deal pulls through. Microsoft must be smiling now as they know this uncertainty over the Java landscape can't be good news to millions of Java developers, and may look at wooing them over to the .Net side.
IBM taking over Sun is like putting Java and its derivative technologies in a corporate straight jacket. Jonathan Schwartz lied to me. He told me everything was ok at Sun.
And to think that I just bought the OpenSolaris Bible book on Amazon. What a waste. I'm working on this new app and was planning on going all out with Glassfish, Mysql, OpenSolaris, and OpenMQ 4.3 for the deployment. Now this news item just knocked the wind out of my sails.
|
|
|
Message #306356
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Wow !! Really great experience !!
Just my 2 cents.
I installed Websfear in 2001 (3.0 if I remember well) on
IBM B50 machine with AIX 4.1 and DB2 some version.
100% pure IBM prods.
The result ?
SIGBUS error without **any** app deployed with a totally idle machine.
Switch to JBOSS without any problems.
For the sake of completeness, SIGBUS disappeared when I used HSQLDB as backend, even if configuring WAS to use it required a massive dose of voodoo.
I could be wrong, but, if I remember well, IIOP for servlet-to-ejb invocation was tunnelled in HTTP packets.
Guido
|
|
|
Message #306358
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
probably stating the obvious, but it's because the execs probably went to school together or play golf together. it's got very little to do with technology, but I might just be overly cynical.
I think it has a lot to do with critical mass. Because software companies are so volatile, knowing that IBM will be there in a few years is more important than the quality of their product (from a business perspective.) I've been studying game theory and this is a common situation. Because of what everyone else is doing, your options are limited to one choice and everyone else is in the same situation. The game (in the game theory sense) needs to change before the other option becomes viable. Some of this has to do the perception of the situation, though.
Having said that, I think companies tend to under-estimate the impact that poor quality software has on their business. It's not valued properly. I think if more managers understood how much value good software can provide, they'd be less worried about the cost and more worried about what they are getting.
Managers cannot be blamed because something doesn't work: they choose a well-known brand, they have telephone number to call.
It doesn't matter it support does not solve problem.
It saves manager's ....back
@Peter
Yes, maybe they play golf together, but I think that there could be (at least I would not be surprised) other than sports....
Here Doug Schmidt's editorial on IT managers fits great.
Guido
|
|
|
Message #306374
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Managers cannot be blamed because something doesn't work: they choose a well-known brand, they have telephone number to call.
It doesn't matter [if] support does not solve problem.
I think this is true. Part of my reasoning is that if managers were actually responsible for the quality of software and support they buy, a lot of software would never be purchased. The quality of software and support is so poor for most COTS and shrink-wrapped software that the only reasonable option would be to built the software in house (unappealing as that is.) Of course, if that happened, then the quality of software and support would surely improve but for the time-being there's no incentive for big companies to provide great software or support. They make just as much money selling barely usable crap.
|
|
|
Message #306379
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
This is bad news people, but the solution is simple. No cussing and bellyaching required. Just start doing your development work using Microsoft tools and platforms.
I know that's what I'll be looking at doing if this deal pulls through. Microsoft must be smiling now as they know this uncertainty over the Java landscape can't be good news to millions of Java developers, and may look at wooing them over to the .Net side.
IBM taking over Sun is like putting Java and its derivative technologies in a corporate straight jacket. Jonathan Schwartz lied to me. He told me everything was ok at Sun.
And to think that I just bought the OpenSolaris Bible book on Amazon. What a waste. I'm working on this new app and was planning on going all out with Glassfish, Mysql, OpenSolaris, and OpenMQ 4.3 for the deployment. Now this news item just knocked the wind out of my sails.
That's a strange attitude. Many people choose Java because they know they have a choice of which vendor to use (or which open source ecosystem). You know if one implementation of a spec. or one company or project goes down the tubes - you're likely to be able to pick up an alternative (that implements the same API or something similar). Java presents a huge amount of choice - there's absolutely no reason for you to be tied to the fortunes of one company.
Java decoupled itself from Sun years ago - it's bigger than Sun, bigger then IBM; bigger than both together.
MS and any other non-standard, proprietary ecosystem continue to be the bigger risk. There you're dependent on the continued innovation and success of a single vendor. Your success is inextricably linked to theirs.
Caveat Emptor.
Rich Sharples
JBoss, a division of Red Hat
http://blog.softwhere.org
|
|
|
Message #306381
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Hope NetBeans/Java survive as a spun off consortium led projects
NetBeans - I enjoy using NetBeans for my personal projects so I hope NetBeans survive as a spun off consortium led project(akin to eclipse).
Solaris - I guess will survive as a legacy unix flavor like Digital UNIX
Java - I pray is spun off into a more vendor neutral hands.
I think IBM top honchos are smart enough to realize that Java adoption will grow more only if it has a vendor neutral image.
MySQL - I think IBM or any buyer that already has a DB will sell off that unit to someone else. Red Hat/Dell? Who has money on their books?
“In the modern world of business, it is useless to be a creative original thinker unless you can also sell what you create. Management cannot be expected to recognize a good idea unless it is presented to them by a good salesman.” - David Ogilvy
Sad to see the SUN setting. Wish just technological innovation led to more money making but nothing ain't that easy.
|
|
|
Message #306382
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Therefore, to me, if IBM really does buy out SUN and decides to take over the Java Platform, I am quite confident I will just move into the MS space and work on .NET as I have little confidence in the abilities of IBM.
I suggest you try it before you decide to move. You'll change your mind.
|
|
|
Message #306383
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM see the value, but Sun can't, why?
Open Office ? Star Office ? hm.. keep the talent but drop the product.
IBM wraps OpenOffice in Eclipse and gives it for free as Lotus Symphony. It can be used as a plugin for other rcp applications.
|
|
|
Message #306384
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
This is bad news people, but the solution is simple. No cussing and bellyaching required. Just start doing your development work using Microsoft tools and platforms. LOL. I use Microsoft tools and platforms. The grass is no greener. I do PLENTY of cussing and bellyaching when using them. In fact, I do tons more.
|
|
|
Message #306387
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
This is bad news people, but the solution is simple. No cussing and bellyaching required. Just start doing your development work using Microsoft tools and platforms. LOL. I use Microsoft tools and platforms. The grass is no greener. I do PLENTY of cussing and bellyaching when using them. In fact, I do tons more.
I'll second and third that too. Often I find myself re-inventing stuff for .NET that already exists in J2EE. Many things in .NET are good, but like all things it's got plenty of negative also. If microsoft's XSD.exe wasn't so horrible, I wouldn't have written my own open source xml schema compiler for .NET, so it's far from greener pastures.
here's to hoping IBM doesn't buy SUN.
peter
|
|
|
Message #306396
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Hi,
While I do agree with you, that Java does provide us with a substantial amount of choice on vendors and platforms we wish to use, rarely is it the case from my experience.
Platforms are decided by management due to strings being pulled at the higher levels.
There was a banking project on which I was consulting on about a year ago. Their requirements weren't exactly complicated or overly difficult in my opinion. I suggested they use the Flux Job Scheduler as it was able to meet most of their requirements and provided added functionality that would have made the overall solution more complete.
However, as Flux was a relative unknown in the environment they were in, they asked me to do an evaluation of other more reputable vendors. I evaluated BMC's Control-M, CA's AutoSys, and IBM Tivoli Workload Scheduler. Of these solutions which I evaluated, IBM's was the most expensive and the one that would require the largest amount of customization to achieve the objectives of the client. In the end, I did put up my proposal to them that Flux was still the better option and at a fraction of the cost of the alternatives. IBM was at the bottom of my recommendations simply due to the amount of customization required.
In the end, they decided to go with IBM. Reason was simple. Some of the higher ups in IBM called the higher ups in the bank, and the decision was to go with IBM. It was simply that. It was not a technical decision call, but a business call. It also had nothing to do with the support element or even cost. It was simply strings being pulled.
The project failed horribly in the end due to the amount of customization required when Flux could have done the job simply with minimal work required. Today, they are still back to using shell scripts that the operator triggers to get the batch processes running. And when things screw up, they spend hours on end trying to recover, wasting valuable time and resources.
In my current engagement, as long as we want to use JEE, we have to use IBM products. This is the mandate from management. If I want to use UNIX, it has to be AIX. If I need a database, it has to be DB2. If I need an App Server, it has to be WebSphere. Even for UNIX boxes, I have to buy IBM. I can only change this if I choose to go Wintel, whereby I am allowed to use HP Boxes with SQL Server as a database backend.
There is a funny story to add to this though. We have a sales application running on the notebooks of salespeople. It's a JEE application running off Tomcat (do not ask me why we use Tomcat for an application intended to be used locally on a notebook. I was not the architect for this particular solution, and the implementation was already done when I got involved.). At one point in time, the management and security teams wanted us to use WebSphere even on these notebooks as that was the "official" platform for JEE. They only backed off when we explained to them the cost implications and the machine requirements to use WebSphere instead of Tomcat. They did force us to use DB2E on the notebooks though. We eventually had to drop DB2E due to the problems we encountered using it, and switched to HSQL. It's been working perfectly now, on a non-IBM platform.
Therefore, decisions on platforms are not always made by the people in the know. And in some cases, even when suggestions and recommendations are made, they are overridden due to strings.
In such environments, I do not see how it differs much from the vendor lock in one would encounter on a MS based platform.
Regards
Jega
|
|
|
Message #306415
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Really bad news
It means we will loose a lot of very good products like Glassfish, Netbeans, OpenDS, Husdson, and must move to really bad products like Websphere application server, Websphere portal, RAD, etc ...
If this is true, i will quit java.
(Please excuse my English)
+1
I am surely biased here but I believe that you, like any other here, will follow direction the market is heading to.
-1
|
|
|
Message #306428
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
Or it might further your resolve.
IBM software isn't like it used to be. Within the past couple of years they've made tremendous strides in performance and usability.
|
|
|
Message #306429
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Hell?
But, the thought of working with WSAD and WebsFear is much deeper than Hell!
Notice R&D and innovation are terms of extinction; can't the US bail out SUN to protect citizens from IBM, Oracle, Microsoft and others?
Perhaps, the EC could make SUN another Airbus, folks this is very bad news for us "open" minded...
With an attitude like that, don't count yourself among the open-minded.
IBM isn't a panacea, but it's certainly not a demon, either.
(Not to mention that you talk about "WSAD" which hasn't been a current product for several years).
|
|
|
Message #306430
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
People Amaze Me
You know, nobody is ever happy with anything. It is human nature to complain. But I get tired of the hyprocracy for people who have never developed commercial software in their life and complain about IBM, Microsoft, Oracle, Sun, you name it. If you can do so much better, go do it! Ever vendor has to make tradeoffs to get their software out the door on time, within budget, and meeting the project requirements, but they all try to make their software better and I mean everybody, IBM, Microsoft, Oracle, etc. Software is written by human beings and humans do make mistakes, but most importantly they learn from those mistakes and do better next time. So if you are so perfect and can do everything 100% correct the first time, then by all means go for it.
|
|
|
Message #306435
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: People Amaze Me
You know, nobody is ever happy with anything. It is human nature to complain. But I get tired of the hyprocracy for people who have never developed commercial software in their life and complain about IBM, Microsoft, Oracle, Sun, you name it. If you can do so much better, go do it! Ever vendor has to make tradeoffs to get their software out the door on time, within budget, and meeting the project requirements, but they all try to make their software better and I mean everybody, IBM, Microsoft, Oracle, etc. Software is written by human beings and humans do make mistakes, but most importantly they learn from those mistakes and do better next time. So if you are so perfect and can do everything 100% correct the first time, then by all means go for it.
I actually worked for IBM a few years back. Yes, everyone makes mistakes and there's always trade-offs. The thing is, the process at IBM makes it nearly impossible to produce user friendly software. My friends at IBM do their best to produce good software, but often they are powerless. I'll give you a concrete example. When I worked at IBM, we had to test the software on a variety of platforms. In my case, it was windows, linux, aix, hpux, zos and solaris for the OS. Add to that the supported databases: sql server 2000/2005, oracle 9/10, and db2. Multiply the two and you get the number of full regression tests we had to run.
In a reasonable world, the developers would get the proper hardware resources to do their work. Well, IBM executives didn't agree with that, so rather than having decent systems to test on, we were given out dated machines with 100GB hard drives. That meant sometimes we couldn't run our tests, because there wasn't enough free space. If you wanted to run AIX against sql server and oracle at the same time, you couldn't. IBM has tons of these types of idiotic processes and rules. That's one of the reasons why IBM is so slow supporting new specs.
Compared to weblogic and jboss, Websphere is horribly painful. My bias take is IBM has the developers to achieve the goal. The problem is the management. If IBM buys SUN, the management will squash things.
|
|
|
Message #306436
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: People Amaze Me
You know, nobody is ever happy with anything. It is human nature to complain. But I get tired of the hyprocracy for people who have never developed commercial software in their life and complain about IBM, Microsoft, Oracle, Sun, you name it. If you can do so much better, go do it!
People do. All the time. That's what's so galling about the situation. Superior solutions are available but because they aren't sold by a big company, they aren't used. I just had an experience where I was involved in evaluating a vendor. Their software (AFAICT) would address our needs better than anything else on the market. It was approved all the way up to the board where it our decisions were overruled because the company wasn't big enough.
There is a valid reason for this and it's a concern that the company won't be around to support the software in the future. But, valid as it may be, it creates a terrible situation. If no one is willing to take a risk on a company for fear it will fail, it fails. The big vendors can continue to sell crap because "at least they will be around." I think open source is a way around this trap and is why open source companies are able to go from blips on the radar to widely used very quickly (hibernate, spring.)
|
|
|
Message #306441
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
This is bad news people, but the solution is simple. No cussing and bellyaching required. Just start doing your development work using Microsoft tools and platforms.
I know that's what I'll be looking at doing if this deal pulls through. Microsoft must be smiling now as they know this uncertainty over the Java landscape can't be good news to millions of Java developers, and may look at wooing them over to the .Net side.
IBM taking over Sun is like putting Java and its derivative technologies in a corporate straight jacket. Jonathan Schwartz lied to me. He told me everything was ok at Sun.
And to think that I just bought the OpenSolaris Bible book on Amazon. What a waste. I'm working on this new app and was planning on going all out with Glassfish, Mysql, OpenSolaris, and OpenMQ 4.3 for the deployment. Now this news item just knocked the wind out of my sails.
That's a strange attitude. Many people choose Java because they know they have a choice of which vendor to use (or which open source ecosystem). You know if one implementation of a spec. or one company or project goes down the tubes - you're likely to be able to pick up an alternative (that implements the same API or something similar). Java presents a huge amount of choice - there's absolutely no reason for you to be tied to the fortunes of one company.
Java decoupled itself from Sun years ago - it's bigger than Sun, bigger then IBM; bigger than both together.
MS and any other non-standard, proprietary ecosystem continue to be the bigger risk. There you're dependent on the continued innovation and success of a single vendor. Your success is inextricably linked to theirs.
Caveat Emptor.
Rich Sharples
JBoss, a division of Red Hat
http://blog.softwhere.org
Good point. Many folks at .NET are using a lot of open source frameworks like nHibernate, Spring.NET etc and these ideas originated in java communities. Only now, MS has been serious about an MVC for .NET, so you can imagine how people had to rely on ASP.NET webforms for everything, whether or not webforms were really the right tool for the problem. IMO, .NET solutions lack the elegance of Java and I don't think many Java developers will like the idea of working in a controlled environment like MS.
|
|
|
Message #306444
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
This is bad news people, but the solution is simple. No cussing and bellyaching required. Just start doing your development work using Microsoft tools and platforms.
I know that's what I'll be looking at doing if this deal pulls through. Microsoft must be smiling now as they know this uncertainty over the Java landscape can't be good news to millions of Java developers, and may look at wooing them over to the .Net side.
IBM taking over Sun is like putting Java and its derivative technologies in a corporate straight jacket. Jonathan Schwartz lied to me. He told me everything was ok at Sun.
And to think that I just bought the OpenSolaris Bible book on Amazon. What a waste. I'm working on this new app and was planning on going all out with Glassfish, Mysql, OpenSolaris, and OpenMQ 4.3 for the deployment. Now this news item just knocked the wind out of my sails.
That's a strange attitude. Many people choose Java because they know they have a choice of which vendor to use (or which open source ecosystem). You know if one implementation of a spec. or one company or project goes down the tubes - you're likely to be able to pick up an alternative (that implements the same API or something similar). Java presents a huge amount of choice - there's absolutely no reason for you to be tied to the fortunes of one company.
Java decoupled itself from Sun years ago - it's bigger than Sun, bigger then IBM; bigger than both together.
MS and any other non-standard, proprietary ecosystem continue to be the bigger risk. There you're dependent on the continued innovation and success of a single vendor. Your success is inextricably linked to theirs.
Caveat Emptor.
Rich Sharples
JBoss, a division of Red Hat
http://blog.softwhere.org
Good point. Many folks at .NET are using a lot of open source frameworks like nHibernate, Spring.NET etc and these ideas originated in java communities. Only now, MS has been serious about an MVC for .NET, so you can imagine how people had to rely on ASP.NET webforms for everything, whether or not webforms were really the right tool for the problem. IMO, .NET solutions lack the elegance of Java and I don't think many Java developers will like the idea of working in a controlled environment like MS.
You'd be surprised how many java developer work in the .NET world. Just look at the number of users of NHibernate and you'll see lots of java developers use both Java and C# on a daily basis. It's one of the reasons why open source java projects get ported to .NET.
peter
|
|
|
Message #306446
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
This is bad news people, but the solution is simple. No cussing and bellyaching required. Just start doing your development work using Microsoft tools and platforms.
I know that's what I'll be looking at doing if this deal pulls through. Microsoft must be smiling now as they know this uncertainty over the Java landscape can't be good news to millions of Java developers, and may look at wooing them over to the .Net side.
IBM taking over Sun is like putting Java and its derivative technologies in a corporate straight jacket. Jonathan Schwartz lied to me. He told me everything was ok at Sun.
And to think that I just bought the OpenSolaris Bible book on Amazon. What a waste. I'm working on this new app and was planning on going all out with Glassfish, Mysql, OpenSolaris, and OpenMQ 4.3 for the deployment. Now this news item just knocked the wind out of my sails.
That's a strange attitude. Many people choose Java because they know they have a choice of which vendor to use (or which open source ecosystem). You know if one implementation of a spec. or one company or project goes down the tubes - you're likely to be able to pick up an alternative (that implements the same API or something similar). Java presents a huge amount of choice - there's absolutely no reason for you to be tied to the fortunes of one company.
Java decoupled itself from Sun years ago - it's bigger than Sun, bigger then IBM; bigger than both together.
MS and any other non-standard, proprietary ecosystem continue to be the bigger risk. There you're dependent on the continued innovation and success of a single vendor. Your success is inextricably linked to theirs.
Caveat Emptor.
Rich Sharples
JBoss, a division of Red Hat
http://blog.softwhere.org
Listen up. Everybody knows that things concerning Java usually moves at a glacial pace where IBM is concerned. Also, many persons know why IBM is able to sell their crap to the big corporate IT shops. The point I'm trying to make here is that I don't think IBM has the vision, agility, and commitment that Sun and other open source members have brought to the Java space. I fear that Java may go the way of COBOL under IBM's stewardship.
If you take Sun and other big Java vendors out of the picture and look at the software match up between IBM and Microsoft, I would hazard a guess to say that Microsoft's software stack across all categories, including database, office applications, OS, web, and application development, etc. beats those of IBM up and down the street. That's maybe why they eventually went the route of becoming a big consultancy services giant. They couldn't compete on the software front. And also, because of their lack of vision in the software technology space, they just tagged along the coat tails of Sun and the open source community (including Linux) just to package and re brand the derivatives under their world renowned corporate brand.
Sun's problems stem from the fact that they're more good at making things rather than articulating and selling them. I think Jonathan Schwartz's move to get deeper into the open source business model and build stronger ties with the developer community was a good one albeit it was five years late and that Sun got caught in this global economic crisis.
By having IBM suck up yet another big Java tech vendor like what Oracle did with Bea Weblogic, means that the Java ecosystem will eventually lose it's diversity and innovativeness through competition. The ties with the open source community will become more strained and may eventually whither and die. You don't need WORA if you're the only Java vendor or one of two (i.e. Oracle being the next), especially if you own most of what Java needs to run on anyway. You eventually start looking like a Microsoft/.NET sell. You may even start seeing the accelerated fragmentation of Java (i.e. IBM JVM, and Oracle JVM).
Do you really believe that IBM and Oracle give a damn about JCP and open source? Just look at how IBM held out on accepting JBI, why? - because they have tonnes of their own proprietary EAI products that they want to still maintain and keep on selling.
Sun is the company that keeps Microsoft, IBM, Oracle, and the others, honest, and thus maybe the overall IT industry balanced.
Sun was again a true pioneer in the vision they were executing in having a large technology vendor deciding to go deep diving into the open source technology space on both the software and hardware fronts.
I can tell you that you will not see that happen from the other big tech vendor companies again. IBM, Microsoft, and Oracle don't know what the open source model really is and they don't really need to. These are companies that really only care about the bottom line, and you gotta pay upfront to play. End game for the open source business model, and back to business as usual (i.e. Old School). Sun may just have been the last of the maverick disrupting big tech vendor companies.
Less competition in the Java space means less innovation, and more inefficient pricing (i.e. Higher prices), and selling (e.g. $50,000/cpu core, and 30 day trial periods for IDEs, etc.) of tech products. Once the precedent is set with Sun, then the last of the rebels will be snapped up for chump change (i.e. Jboss, Redhat, etc.).
Surely this can't be that difficult of an argument to accept?
|
|
|
Message #306447
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
LOL. I use Microsoft tools and platforms. The grass is no greener. I do PLENTY of cussing and bellyaching when using them. In fact, I do tons more.
The grass is always greener on the Java side with the tools, frameworks, IDEs.....until you end up having to actually program in Java (the language). Since Java (the language) is going into maintenance mode, maybe IBM can port C# 3.0 over to the JVM already.
|
|
|
Message #306448
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re:People Amaze Me
"But I get tired of the hyprocracy for people who have never developed commercial software in their life and complain about IBM, Microsoft, Oracle, Sun, you name it. If you can do so much better, go do it!"
Well, define "better."
My opinion of oracle/microsoft/and oracle is that they do things "good enough" and then send in marketing to make up the difference. sun and bea were/are marginally better then those 3.
But I doubt that by "better" you mean do what has the highest margins.
|
|
|
Message #306449
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
.....until you end up having to actually program in Java (the language).
+1.
And (unfortunately) Java is still the only mainstream language and will be the only mainstream language on the Java server platforms in the foreseeable future.
|
|
|
Message #306515
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
LOL. I use Microsoft tools and platforms. The grass is no greener. I do PLENTY of cussing and bellyaching when using them. In fact, I do tons more.
The grass is always greener on the Java side with the tools, frameworks, IDEs.....until you end up having to actually program in Java (the language). Since Java (the language) is going into maintenance mode, maybe IBM can port C# 3.0 over to the JVM already.
Frank, it seems that you never miss a chance to bash on Java. Serious question, though: since C# is almost identical to Java, what's the big gain there?
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
|
|
|
Message #306517
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re:People Amaze Me
Andy -
My opinion of oracle/microsoft/and oracle is that they do things "good enough" and then send in marketing to make up the difference. sun and bea were/are marginally better then those 3.
But I doubt that by "better" you mean do what has the highest margins.
Speaking from what I've seen from working alongside BEA for 10 years and being at Oracle now for 2 years, the reality is quite different from what you stated. Obviously, the proof will be in improved products, better quality and more clear technological leadership, but if the latest jRockit results (new World Record benchmark publications on SPECjbb and SPECpower based on JRockit, in collaboration with NEC, Dell, HP, and Fujitsu-Siemens) are anything to go by, the increased investment by Oracle is paying dividends:
http://www.oracle.com/us/corporate/press/018046_EN
http://www.oracle.com/us/corporate/press/018045_EN
http://www.oracle.com/us/corporate/press/018043_EN
http://www.oracle.com/us/corporate/press/018044_EN
You're right in a way though: It is a lot easier for large companies to get by with "good enough" products, and that's a trap that all large companies have to watch out for.
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
|
|
|
Message #306524
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
LOL. I use Microsoft tools and platforms. The grass is no greener. I do PLENTY of cussing and bellyaching when using them. In fact, I do tons more.
The grass is always greener on the Java side with the tools, frameworks, IDEs.....until you end up having to actually program in Java (the language). Since Java (the language) is going into maintenance mode, maybe IBM can port C# 3.0 over to the JVM already.
Frank, it seems that you never miss a chance to bash on Java. And part of the problem with "Java" is that there's no game plan for the future of "Java".
Cameron you never miss a chance to bash on a lot of things. But Java (the language) rightly deserves to be bashed on. The future of Java probably wouldn't be so bleak if there was some successor to Java (the language) on the drawing board as Java (the language) goes into maintenance mode.
Serious question, though: since C# is almost identical to Java, what's the big gain there?
So you haven't even looked at C# lately, have you;)
|
|
|
Message #306527
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
I have.
What's so different about C#? It's not like we are talking about c++ to Java here. I like the out operands and some other minor details, but there are also other quirks.
It is certainly not very different...
|
|
|
Message #306528
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
I have.
What's so different about C#? It's not like we are talking about c++ to Java here. I like the out operands and some other minor details, but there are also other quirks.
It is certainly not very different...
Properties
Automatic Properties
Extension Methods
Value types
Delegates
LINQ (lambda syntax, comprehension syntax,Closures)
Local type inference
The "using" construct for automatic disposal
as you mentioned, out parameters along with ref parameters
Indexers
Collection Initializers
Operator overloading
Yield
Preprocessor Directives
pinvoke
Object Initializers
and so on, and so on...
They're really much different languages at this point.
Java shouldn't try to play catchup though. IBM, Google, someone besides dead-man-walking Sun, is going to have to take the bull by the horns and come up with a successor to Java.
|
|
|
Message #306549
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
IBM doesn't "take the bull by the horns"
Java shouldn't try to play catchup though. IBM, Google, someone besides dead-man-walking Sun, is going to have to take the bull by the horns and come up with a successor to Java.
How is Sun a dead man walking? They made an operating profit last quarter and their software division is growing quickly in revenue. Their strategy is correct, but the recession came at exactly the wrong time.
As for "taking the bull by the horns", well, that's like calling Kate Moss fat in comparison to Rosie O'Donnell. IBM took a year after Sun to come out with IBM JDK 5. Their software development is glacially slow, and aside from meeting the bullet points they use to sell their stuff to upper management, their software really doesn't work that well.
As for the Java language, Sun has plenty of irons in the fire to replace it:
JRuby
Jython
Groovy
Scala
JavaFX Script (actually a full-featured programming language with closures, properties, binding, and a saner syntax for generics)
|
|
|
Message #306554
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM doesn't "take the bull by the horns"
As for the Java language, Sun has plenty of irons in the fire to replace it:
JRuby
Jython
Groovy
Scala
JavaFX Script (actually a full-featured programming language with closures, properties, binding, and a saner syntax for generics)
None of those are replacements for Java (especially JRuby, Jython), except for maybe Groovy. But Groovy is dynamically typed (even when you declare a type). I believe Groovy 1.6 does have the new joint compiler, so you could mix Java proper and Groovy in the same project, but I think that opens up a whole new can of worms for management.
It's all a moot point anyway, unless Sun (for now) and others backs a successor to Java.
|
|
|
Message #306555
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM doesn't "take the bull by the horns"
JavaFX Script is statically typed and allows you to use all the Java libraries. It's a brand-new language, so it needs to be ironed out, but I speculate that it's the eventual replacement for the Java language.
Due to backward compatibility reasons, the Java language will always be around. I just believe that Sun will maintain the language instead of updating it.
|
|
|
Message #306556
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Cameron you never miss a chance to bash on a lot of things.
Other people bash; I just state opinions .. ;-)
But Java (the language) rightly deserves to be bashed on. The future of Java probably wouldn't be so bleak if there was some successor to Java (the language) on the drawing board ..
There is.
Serious question, though: since C# is almost identical to Java, what's the big gain there?
So you haven't even looked at C# lately, have you;)
I've been working in C# for the past few weeks. I love typing "virtual" everywhere. And the triple slashes .. gotta love those ;-)
Trust me, I like C# a lot better than I like C++.
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
|
|
|
Message #306557
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Cameron you never miss a chance to bash on a lot of things.
Other people bash; I just state opinions .. ;-)
But Java (the language) rightly deserves to be bashed on. The future of Java probably wouldn't be so bleak if there was some successor to Java (the language) on the drawing board ..
There is.
Serious question, though: since C# is almost identical to Java, what's the big gain there?
So you haven't even looked at C# lately, have you;)
I've been working in C# for the past few weeks. I love typing "virtual" everywhere. And the triple slashes .. gotta love those ;-)
Trust me, I like C# a lot better than I like C++.
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
p.s. Seriously, how hard is it to warn that the tags don't match up?
|
|
|
Message #306582
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
How come?
Current Java is somewhat like C# 1.0, albeit C# had more syntactic sugar from the very start. C# 3.5 is a league, maybe 2, above Java. C# 4.0 is coming soon while what is coming soon for Java is discussed in this thread.
Type erasure in generics alone is enough to start laughing over Java.
Next, JVM. The Sun one JIT compiles what the VM thinks to be critical for performance. C# is compiled to native code 100%. I guess this is why the very first mini test I ran revealed that Java can be 50+ times slower than C#, while C# can easily beat C on performance.
Web frameworks, libraries, and application servers may be comparable, but Java the language is plain nothing compared to C#.
|
|
|
Message #306593
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
language vs APIs
C# vN may be great, but MS APIs will always be MS APIs
I may suffer Java the language, but I enjoy all the APIs and frameworks I can use ...
I´d rather use awkward languages with great tools and APIs... Otherwise I´d be using ____
As for IBM ... Java will be the next Cobol.
|
|
|
Message #306594
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Summing it all up...
IBM buying Sun would be nothing short of a nightmare.
C# seems better than Java because microsoft creates really lousy APIs and infrastructure which requires much more detailed programming steps to use and thus there is a lot of C# that is not in Java to support the microsoft way. The original licensing issues came about specifically because of people in Microsoft wanting to make significant changes in Java to support the Microsoft nonsense.
See how much of C# you'd need to use compared to the Java APIs to write code that was just server code. Sun, and the Java community in general, has failed to understand that there are like 100,000 computers on the planet that are higher performance servers and 100,000,000 that are desktops and 1,000,000,000 that are cell phones and other dedicated devices which need completely different APIs.
If IBM buys Sun, I'll be moving all of my development efforts towards Apple.
If Apple buys Sun, I might finally get Java on my iPhone and that would make writing software there much easier, and it might actually be more stable given the Objective-C memory model and how fragile it can be in new development.
|
|
|
Message #306597
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
But Java (the language) rightly deserves to be bashed on. The future of Java probably wouldn't be so bleak if there was some successor to Java (the language) on the drawing board ..
There is.
I agree, C# ported to the JVM.
I've been working in C# for the past few weeks. I love typing "virtual" everywhere. And the triple slashes .. gotta love those ;-)
Trust me, I like C# a lot better than I like C++.
Peace,
Cameron Purdy
If Java 7 doesn't at least get local type inference, it'll be even a bigger joke than it is now.
Oracle Coherence: Data Grid for Java, .NET and C++
p.s. Seriously, how hard is it to warn that the tags don't match up?
|
|
|
Message #306598
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Worse than Hell actually
Seriously, how hard is it to warn that the tags don't match up?
By my previous post, you can see I agree with that too.
|
|
|
Message #306599
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Summing it all up...
IBM buying Sun would be nothing short of a nightmare.
As opposed to...?
C# seems better than Java because microsoft creates really lousy APIs and infrastructure which requires much more detailed programming steps to use and thus there is a lot of C# that is not in Java to support the microsoft way. The original licensing issues came about specifically because of people in Microsoft wanting to make significant changes in Java to support the Microsoft nonsense.
See how much of C# you'd need to use compared to the Java APIs to write code that was just server code. Sun, and the Java community in general, has failed to understand that there are like 100,000 computers on the planet that are higher performance servers and 100,000,000 that are desktops and 1,000,000,000 that are cell phones and other dedicated devices which need completely different APIs.
That has be one of the most nonsensical paragraphs of a post ever on TSS.
|
|
|
Message #306602
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
C# APis are not bad
C# APis are not that bad.. Nor is the Java language , there is very little practical difference between Java and C# ( except for purists) . The only thing is C# is evolving which Java is not..( Good native generics in V2.0 , LINQ and functional programming in v3.0). IBM may help here , internet committees tend to freeze things.
Ok win 32 API and Windows forms ( which Rad developers use and is a very thin wrapper over the win32 API) have Bad APIs ... very bad and a LOT of people use them. MS try to ween people to WPF but the learning curve is hard.
Most RAD devs havent even made the jump to OO.
ASP.NET - average..Im still not sold on the concept of sending view states to the client. Silverlight is pretty impressive ,though the jury is out.
Most of the .NET framework ,especially newer systems like WPF and WCF are however very good .. If WPF and WCF worked with Java they would be the dominant GUI/Service host very quickly. Arguments on what is better in the framework come down to very fine details and in some cases MS make sub optimal choices so developers don't shoot themselves in the foot.
|
|
|
Message #306604
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Sun is dying
Sun is dying read the earlier posts they have maybe 3-6 months before they are out of cash.
http://www.google.com/finance?q=NASDAQ:JAVA
Note for Q ending Dec2008 ( before corporates cut their new projects , and move of SUN servers so they dont get those massive maintenance fees)
Net income -209 M
Net profit margin -6.49
Operating margin -6.18%
And now they are in a recession.. You can see they are burning cash at about 250M a quarter and this rate will increase.
Software is like 20% of the business and i would expect a lot of that to be Solaris licenses.
Not good. Announcing a profit is meaningless as there are many accounting tricks to revalue assets look at the cash flow.
|
|
|
Message #306606
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Current Java is somewhat like C# 1.0, albeit C# had more syntactic sugar from the very start. C# 3.5 is a league, maybe 2, above Java. C# 4.0 is coming soon while what is coming soon for Java is discussed in this thread.
Type erasure in generics alone is enough to start laughing over Java.
Next, JVM. The Sun one JIT compiles what the VM thinks to be critical for performance. C# is compiled to native code 100%. I guess this is why the very first mini test I ran revealed that Java can be 50+ times slower than C#, while C# can easily beat C on performance.
Web frameworks, libraries, and application servers may be comparable, but Java the language is plain nothing compared to C#.
I have an open source RETE rule engine. The C# version on .NET 3.5 is slower than java 1.4, 1.5 and 1.6. There are plenty of things that are faster in C# and plenty that are slower. For enterprise applications, .NET is sorely lacking and missing a lot. I'm constantly re-inventing stuff in .NET because microsoft doesn't provide it and there's no third party tool for it.
Just my bias opinion, but there's no clear winner. Take for example Generics. Java generics blow ass, while C# generics rock. Hot code replacement while debugging in .NET works way better than java debugging.
In contrast, threading support in Java is more mature and generally easier to me. One thing is clear though, IBM buying SUN is a horrible idea.
peter
|
|
|
Message #306608
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Hot code replacement while debugging in .NET works way better than java debugging.
As of VS.Net 2005 it is not. I am not sure about other IDEs but Eclipse is tons better than VS.Net on this.
|
|
|
Message #306612
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Hot code replacement while debugging in .NET works way better than java debugging.
As of VS.Net 2005 it is not. I am not sure about other IDEs but Eclipse is tons better than VS.Net on this.
Since Eclipse is my default IDE, I gave it a try with 3.3.1 and it doesn't work like I expect. I started up a plugin project I have and tried to hot replace a mouselistener by adding a few lines of code during debugging. The end result is I get an exception. If I restart the plugin, it does what i expect and lets me step over the new lines I added. I can do the same type of thing with VSS.NET 2008 and it "just works".
I love eclipse and think it rocks, but hot code replacement isn't one of its strengths. In many areas, I feel eclipse is superior to VSS.NET, but that's my bias opinion. I don't know if other java IDE handle hot code replacement better. Does anyone know? Maybe IntelliJ?
peter
|
|
|
Message #306614
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Sun is dying
SUN may be dying, but IBM taking over it is kind of uncomfortable feeling. I wish there is a different way for SUN to overcome financial problems and go on by itself. The SUN's way of promoting Java/JEE technologies has been really clean and healthy for developer and vendor community. Last 10-12 years of Java dominance and the satisfaction developers attained of Java is largely due to extreme hard work put by SUN team into specs, reference implementations and JCP. They have served the Java community selflessly and professionally. I doubt the same level of effort from IBM. In India, their development centers do not have same level of quality as other product companies like Adobe, MS and SUN have maintained. They have become usual service companies where things are mostly unorganized, only external projection is focused upon. I would love SUN to continue the way it has been.
|
|
|
Message #306616
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Sun is dying
Yes but SUN will also be dying as Rational Software Company did, remember what happened to the Company after it was bought by IBM. The Software Quality was so bad from the new products like RAD, that you can not work with it (updates with 1GB and more...), the old products like Rational Modeler was nice to use (only problem just on windows).
There was also another thing, Rational had a very good Support documentation, there is no comparable one for the RAD.
I hope too SUN will stay as it is, a standalone company.
|
|
|
Message #306657
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Andrey -
Current Java is somewhat like C# 1.0, albeit C# had more syntactic sugar from the very start. C# 3.5 is a league, maybe 2, above Java. C# 4.0 is coming soon while what is coming soon for Java is discussed in this thread.
Dude, it's awesome that C# makes you that happy. Honestly.
I work in Java and C# and frankly I don't notice any big differences. They each irritate me in their own unique ways, and I can usually get the job done in either.
I also have to work in C++, and just be glad you're getting to use Java and/or C#, because I'd almost rather bang my head against concrete than use C++.
Type erasure in generics alone is enough to start laughing over Java.
Part of your not-hidden personality problem is that you are using your choice of language as a substitute for self confidence.
Would it help solve the problem if I told you that you're better than that? Smarter than that? Too witty for that? ;-)
Next, JVM. The Sun one JIT compiles what the VM thinks to be critical for performance. C# is compiled to native code 100%. I guess this is why the very first mini test I ran revealed that Java can be 50+ times slower than C#, while C# can easily beat C on performance.
I am surprised that the tale of your single micro-benchmark hasn't settled the matter once and for all.
Unfortunately, there is a large body of work that shows both your basic premises as well as your conclusions to be wrong. Highly optimizing the parts of the code that actually get run is far more important than creating poorly optimized native code for all of the classes. Nonetheless, you can have your cake and eat it too: Oracle's jRockit JVM immediately produces 100% native code (like the CLR JIT), then uses runtime profiling information to do a second pass optimized compile producing far better code than HotSpot or the CLR. FWIW it also holds all the performance records:
http://www.oracle.com/us/corporate/press/018046_EN
http://www.oracle.com/us/corporate/press/018045_EN
http://www.oracle.com/us/corporate/press/018043_EN
http://www.oracle.com/us/corporate/press/018044_EN
Web frameworks, libraries, and application servers may be comparable, but Java the language is plain nothing compared to C#.
Comparable? Yeah, the few frameworks that got ported from Java to C# are comparable, but other than that .NET's ecosystem is extremely shallow and poor. Dude, the IDE blows almost as badly as the circa-1997 Visual Cafe for Java. Without Resharper, I'd almost rather use Notepad.
And don't even get me started on build and test automation tools -- .NET is just a patented graphical way of giving developers Carpal Tunnel Syndrome. Click. Click. Click.
Nonetheless, like I said, C# isn't that bad a clone of Java. Hopefully Sun will get better at taking ideas from C# without ruining them in the process.
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
|
|
|
Message #306691
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: C# APis are not bad
Soyatec products looks very interesting.
Again E4 are going for a html/css style .. These are Terrible .
The good think MS did with WPF is throw out html/css and Windows Forms - it was not designed for the things we use it for . Instead they said if you were to do a rendering engine from scratch how would you do it.
|
|
|
Message #306693
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Cameron ,
While i very much agree with your first post that the differences between the languages are academic ( though the jury is out on the functional extensions). I also am very scared for the C++ code i had to maintain 2 years ago.. mem leaks and pointers are frightening beasts.
The rest of your post i disagree with and it was rude to the poster. Ok he maybe one of the purist/academic types which infest software but its wrong to take a personal line and says more about you then him. You have a personal preference for keyboard driven UIs which date back to vi.
Im very happy with VS you can customize it as much as you like. If you don't like clicks use hotkeys , snippets though you have to spend time learning it. However the majority of devs cant be bother learning shortcuts anymore and use GUis and mouse . I was shocked how few people use bookmarks ,Ctrl - etc in C# to navigate. Lastly VS gets significant improvements every release as well eg VS2008 is a huge improvement over 2005).I find the built in Testing ok but do use an add on to do TDD ( and its worth noting the current Beta of VS10 has a massive improvement here) .
My only gripe with VS is the designers are lacking in hand held development ( they exist but are annoying) and the WPF designer does not support rendering DataTemplates from a resource file so you can do MVVM(instead of MVC) which means you want Expression Blend.
In terms of libraries /addons there are plenty. WPF/WCF/WF are superior to any Java lib I know of in those fields. Enterprise lib is solid and easy to use and extend ( once you bother learning it) and arguably superior to a number of Java libs like log4. I think LINQ for SQL / Linq for entities is superior to nHibernate ( or Hibernate) with an easier learning curve.
The best part is most of the addons come pre-installed or are included and hence a lot of developers know them. I dont think it matters if its shallow or not ..Here are the shallow ones i can think of
Logging , .NET framework loggging , Log4Net , Entlib
Testing Nunit and MSTest ( with a few VS gui add ons)
Honestly do these libs really matter in your final product and the debates over what is better? For more important libs like UI and DataAccess there are tons in the .NET space also.
Java has a lot of good libs that have few users and are a good breading ground for ideas , in the past .NET got all their ideas from Java but i think that has changed with WPF , Silverlight and WCF.
Maybe IBM can invent some good things as they have the name and size to push it ,it will be good for the industry.
|
|
|
Message #306694
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
I have and tried to hot replace a mouselistener by adding a few lines of code during debugging. The end result is I get an exception.
I just did it and it works. I've used eclipse since ... and it has always worked. If I change the method signature, then it will ask me if i want to continue (with old code or restart (with new).
I can do the same type of thing with VSS.NET 2008
I have not tried 2008. But are pretty sure it doesn't work in web apps in 2005. I will have to test this again. I have not touched .Net in a few months (yaaaaay :) )
|
|
|
Message #306695
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Sun been shopping for buyers for weeks Intel CEO
http://news.moneycentral.msn.com/provider/providerarticle.aspx?feed=OBR&date=20090325&id=9729490
"I can tell you that Sun was shopped around the valley and around the world in the last few months," Otellini said in comments detailed on Wednesday in a filing with the U.S. Securities Exchange Commission.
"A lot of companies got calls or visits on buying some or all the assets of the company. It looks like IBM is in the hunt now. And at a hundred-and-some-odd-percent-premium, I suspect they'll get it."
Sources with knowledge of the matter told Reuters IBM was in exclusive talks with Sun and IBM was examining Sun's businesses as part of a due diligence process.
So it will happen unless Sun have cooked there books to much ( in which case its in deep trouble).
|
|
|
Message #306697
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Cameron ,
While i very much agree with your first post that the differences between the languages are academic ( though the jury is out on the functional extensions). I also am very scared for the C++ code i had to maintain 2 years ago.. mem leaks and pointers are frightening beasts.
The rest of your post i disagree with and it was rude to the poster. Ok he maybe one of the purist/academic types which infest software but its wrong to take a personal line and says more about you then him. You have a personal preference for keyboard driven UIs which date back to vi.
Im very happy with VS you can customize it as much as you like. If you don't like clicks use hotkeys , snippets though you have to spend time learning it. However the majority of devs cant be bother learning shortcuts anymore and use GUis and mouse . I was shocked how few people use bookmarks ,Ctrl - etc in C# to navigate. Lastly VS gets significant improvements every release as well eg VS2008 is a huge improvement over 2005).I find the built in Testing ok but do use an add on to do TDD ( and its worth noting the current Beta of VS10 has a massive improvement here) .
My only gripe with VS is the designers are lacking in hand held development ( they exist but are annoying) and the WPF designer does not support rendering DataTemplates from a resource file so you can do MVVM(instead of MVC) which means you want Expression Blend.
In terms of libraries /addons there are plenty. WPF/WCF/WF are superior to any Java lib I know of in those fields. Enterprise lib is solid and easy to use and extend ( once you bother learning it) and arguably superior to a number of Java libs like log4. I think LINQ for SQL / Linq for entities is superior to nHibernate ( or Hibernate) with an easier learning curve.
The best part is most of the addons come pre-installed or are included and hence a lot of developers know them. I dont think it matters if its shallow or not ..Here are the shallow ones i can think of
Logging , .NET framework loggging , Log4Net , Entlib
Testing Nunit and MSTest ( with a few VS gui add ons)
Honestly do these libs really matter in your final product and the debates over what is better? For more important libs like UI and DataAccess there are tons in the .NET space also.
Java has a lot of good libs that have few users and are a good breading ground for ideas , in the past .NET got all their ideas from Java but i think that has changed with WPF , Silverlight and WCF.
Maybe IBM can invent some good things as they have the name and size to push it ,it will be good for the industry.
Seriously, LINQ isn't a substitution for ORM. Although LINQ is useful, it's got plenty of limitations. Just look at all the criticism .NET people are throwing at it. WCF is nice, but honestly it's not better. It's just different. There are plenty of limitations with WCF. Microsoft still doesn't have an enterprise level ESB solution. There are plenty of people in .NET world that don't like WPF and think it's a failure. Better is always going to be a subjective measurement.
my bias 2 bits
peter
|
|
|
Message #306698
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
I have and tried to hot replace a mouselistener by adding a few lines of code during debugging. The end result is I get an exception.
I just did it and it works. I've used eclipse since ... and it has always worked. If I change the method signature, then it will ask me if i want to continue (with old code or restart (with new).
I can do the same type of thing with VSS.NET 2008
I have not tried 2008. But are pretty sure it doesn't work in web apps in 2005. I will have to test this again. I have not touched .Net in a few months (yaaaaay :) )
I should clarify what i tried. In my plugin, I have a FormPage, which has a button. The button has a listener, done the usual way.
button.addMouseListener(new MouseListener() {.....});
If I try to hot replace the logic in mouseUp() method in that MouseListener, it doesn't give me a warning, since I didn't change the method signature. Instead it just gets an exception.
I did some digging yesterday and in many cases it doesn't work. For a normal method, hot replace does seem to work. Though in the past, I've come across cases where it doesn't. My own experience, the Hot code replace does "seem" to work better in VSS2008. That's a non-scientific opinion and isn't proof of any sort.
peter
|
|
|
Message #306702
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Ben -
The rest of your post i disagree with and it was rude to the poster.
Yes, you are right. I was bothered by his technical inaccuracies and the unnecessarily condescending manner in which they were delivered, but I probably should have used a more diplomatic tone.
You have a personal preference for keyboard driven UIs which date back to vi.
Au contraire .. I never cared for emacs or VI. I use IDEA for Java, but the nice thing is that all the Java tools can also be automated (command line, scripting, etc.) which means that you have the choice of clicking or typing (or automating). With .NET, we just don't have that choice, so I have to launch VS (each of 2003, 2005 and 2008) in order to build the various versions of our software.
Lastly VS gets significant improvements every release as well eg VS2008 is a huge improvement over 2005).
True.
I think LINQ for SQL / Linq for entities is superior to nHibernate ( or Hibernate) with an easier learning curve.
I like LINQ from what I've seen. It seems generally well-designed and clever, but I haven't had much time to actually do anything with it. I keep hoping to get some time to use it with Coherence, but something always comes up.
I wouldn't compare LINQ to Hibernate, though. You could compare Hibernate to NHibernate .. ;-)
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
|
|
|
Message #306779
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: IBM in talks to buy Sun Microsystems
I wonder exactly what IBM could do with the JCP. The JCP is an open organization where many companies and individuals participate but Sun has a great deal of control within the JCP. Would IBM use this to their competitive advantage? Why would IBM do this? The JEE application server market is in the billions just for the software let alone the consulting side of that industry. Then we have J2ME on mobile, etc. In this lousy economy I have a great deal of trouble envisioning IBM as a nice guy that would turn away opportunity and let things keep running as they are currently.
|
|
|
Message #306780
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Where's Larry?
I may have missed something but is Ellison thinking of jumping into this and bidding? He sure has alot invested in Java...
|
|
|
Message #306803
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Netbeans
NetBeans - I enjoy using NetBeans for my personal projects so I hope NetBeans survive as a spun off consortium led project(akin to eclipse).
>Almost certainly. Its not in IBMs interest to let it die , it will provide a lower end intro into Java development. I expect Eclipse will get a lot of new features .
Solaris - I guess will survive as a legacy unix flavor like Digital UNIX
> Probably. There is not room for AIX and Solaris , the best features will move to AIX.
Java - I pray is spun off into a more vendor neutral hands.
I think IBM top honchos are smart enough to realize that Java adoption will grow more only if it has a vendor neutral image.
> No chance see article i posted above. IBM wants control and its the main reason they are buying SUN.
MySQL - I think IBM or any buyer that already has a DB will sell off that unit to someone else. Red Hat/Dell? Who has money on their books?
> Yes this will probably be sold .. However IBM may convert it into a commercial SQL product ( better tools , better multi threading etc). There DB2 line has not done that well in PC land.
“In the modern world of business, it is useless to be a creative original thinker unless you can also sell what you create. Management cannot be expected to recognize a good idea unless it is presented to them by a good salesman.” - David Ogilvy
Sad to see the SUN setting. Wish just technological innovation led to more money making but nothing ain't that easy.
|
|
|
Message #306804
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: How come?
Seriously, LINQ isn't a substitution for ORM. Although LINQ is useful, it's got plenty of limitations. Just look at all the criticism .NET people are throwing at it. WCF is nice, but honestly it's not better. It's just different. There are plenty of limitations with WCF. Microsoft still doesn't have an enterprise level ESB solution. There are plenty of people in .NET world that don't like WPF and think it's a failure. Better is always going to be a subjective measurement.
my bias 2 bits
peter
Note 2 types of LINQ which are used for data access , LINQ for entities and LINQ for SQL. The Main complaints are LINQ for entities is too heavy ( its more like an ORM) and they prefer the really easy to use LINQ For SQL however MS are pushing people to LINQ for entities. And like all ORMs there are issues with nTier scenarios.
WCF provides a lot more than Axis , eg Transactions , Soap over tcp or udp , extentions , logging/tools etc . Best not get into a discussion on ESBs were to OT already, i do a lot of work with SOAs and im not a big fan of "big" ESB solutions - you need to define the term first as it means different things (UDDI ? , the new .NET services ESB , WS-Management etc) .
WPF complaints are mainly from older VB /RAD developers who still scream at OO. It has a massive learning curve especially for forms developers and is actually easier for web designers to pickup. However if you want to provide a GUI that looks better you have a product thats easy to write , super easy to maintain , is logical and renders consistently. RAD and win forms/datasets still has its place as a quick and dirty ( very) and for this WPF ( or the JAVA UIs) cannot compare.
Complaints dont make a product bad there are lots about Eclipse and NetBeans (mainly from each other).
The one thing MS does well which the Java community does not is to provide support for the majority of programmers eg "B programmers" - those people who dont touch it outside of their job. However MS has been dragging them kicking and screaming since VB.NEt was released.
|
|
|
Message #306824
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Comparing Product & Technologies
as I understood from the replies, the following is difference in two company offerings:
Sun IBM
------------------------
Sparc Power
Solaris AIX / Linux
MySQL DB2
GlassFish WAS
Netbeans Eclipse
JBI SCA
Swing SWT
JavaFX ?
did I miss some thing? can you add more?
|
|
|
Message #306837
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
MSBuild automates the .NET build process
I never cared for emacs or VI. I use IDEA for Java, but the nice thing is that all the Java tools can also be automated (command line, scripting, etc.) which means that you have the choice of clicking or typing (or automating). With .NET, we just don't have that choice, so I have to launch VS (each of 2003, 2005 and 2008) in order to build the various versions of our software.
As you are aware Cameron, your opinions run heavy on the Java world, that's why I feel troubled when you say something like "I have to launch VS ... to build the various versions of our software". The .NET SDK has, since 2005, MSBuild (http://msdn.microsoft.com/en-us/library/wea2sca5.aspx) exactly for automating the build process, even in the absence of Visual Studio. For even more control, you can throw Powershell in the mix.
|
|
|
Message #306839
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Why do you have to type "virtual" so frequently?
I love typing "virtual" everywhere.
That makes me thing that you are using class inheritance more frequently than interface implementation. Even if you are overriding a lot, then Intellisense will help you before you get to the "t" in "virtual", supposing you're not using code snippets... :-)
|
|
|
Message #306857
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: Why do you have to type "virtual" so frequently?
I love typing "virtual" everywhere.
That makes me thing that you are using class inheritance more frequently than interface implementation. Even if you are overriding a lot, then Intellisense will help you before you get to the "t" in "virtual", supposing you're not using code snippets... :-)
I have not used 2008, but to display objects in a Table/List in WinForms, I had to subclass instead of use interfaces, otherwise I would get a class cast exception. If they fixed issue (ie - Have renderers like Swing and JFace), i would be overjoyed!
|
|
|
Message #306858
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: MSBuild automates the .NET build process
The .NET SDK has, since 2005, MSBuild
I think a lot of Cameron, but I was gonna get him on that too. In fact, we had a mix language project (VB.Net and C#) that was almost impossible to get a good build by checking out of VSS and then compiling with VS.Net. I used NAnt to check it out and build it. Worked perfectly. (FYI, this was with 2003, me thinks).
|
|
|
Message #306871
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Swing rendereres ~~ XAML templates ?
If they fixed issue (ie - Have renderers like Swing and JFace), i would be overjoyed!
I'm not an expert on Swing but I think XAML (that is, WPF and Silverlight) databinding and templates do what renderers do. So, overjoyed you should be! :-)
|
|
|
Message #306979
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: MSBuild automates the .NET build process
.. when you say something like "I have to launch VS ... to build the various versions of our software". The .NET SDK has, since 2005, MSBuild (http://msdn.microsoft.com/en-us/library/wea2sca5.aspx) exactly for automating the build process, even in the absence of Visual Studio. For even more control, you can throw Powershell in the mix.
I haven't checked out Powershell. It's on my list.
There were two problems that I am aware of:
* We still have to support (i.e. build in) VS '03, '05 and '08 ..
* The Microsoft doc tools were resistant to automation.
At any rate, for our next release (Coherence 3.5) we have incorporated a number of open source tools (including for doc) which have allowed us to finally automate the entire build and test environment. The amount of effort to automate .NET builds and tests compared to Java is quite extraordinary.
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
|
|
|
Message #306980
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: MSBuild automates the .NET build process
The .NET SDK has, since 2005, MSBuild
I think a lot of Cameron, but I was gonna get him on that too. In fact, we had a mix language project (VB.Net and C#) that was almost impossible to get a good build by checking out of VSS and then compiling with VS.Net. I used NAnt to check it out and build it. Worked perfectly. (FYI, this was with 2003, me thinks).
I think the difference is that we do a lot more in our product build than just compile the .cs files ;-)
Peace,
Cameron Purdy
Oracle Coherence: Data Grid for Java, .NET and C++
|
|
|
Message #307000
Post reply
Post reply
Post reply
Go to top
Go to top
Go to top
|
|
Re: MSBuild automates the .NET build process
The .NET SDK has, since 2005, MSBuild
I think a lot of Cameron, but I was gonna get him on that too. In fact, we had a mix language project (VB.Net and C#) that was almost impossible to get a good build by checking out of VSS and then compiling with VS.Net. I used NAnt to check it out and build it. Worked perfectly. (FYI, this was with 2003, me thinks).
I think the differenc
| |
