Microsoft is boasting that it has surpassed Java usage with .Net and is preparing to make .Net more conducive to Web services, but offered no proof. A Sun official commented that "I'm certainly extremely skeptical of .Net passing Java. There is a difficulty in measuring the market because [.Net] is all bundled together."

The article can be read here.

With what criteria were they measuring surpassed? Number of developers, number of products, number of users, number of active web services? You can prove anything if you pick the right metric!

It's quite easy to say .Net has more users that Java. I switched on my Windows box this morning and hey presto....some .NET components got used! Hardly rocket science.

Seriously though, does it really matter who has the highest usage? Surely what should be more important is picking the best tool for the job and improving the interoperability between the two platforms. Some of my customers are more comfortable with Java, some prefer .NET and most don't care as long as I deliver quality software on time and on budget!

Don't see any evidence that even remotely hints at this being true.

Try searching the job boards, any job board, (unless of course its a microsoft job board), you'll find more jobs that require java. In fact, recently the gap seems to be widening between Java vs C#.

Obviously its very sparse on details of how they came to this claim because its very outlandish and a bit hilarious.

.

LOL

Maybe they got their numbers from Rolf!

-Nick

FUD

Enough is enough. Time to fight back...

So far OSS has been quite effective in creating top quality enterprise products like Linux, JBoss and Eclipse. However, one thing that has not been done (or at least been done well) is Open Marketing. I think with the sheer number of OSS developers and even vendor support (IBM, Red Hat etc.), we could create a model for a superb marketing machine. Doing so, we could expose the risks of using Microsoft products, in addition to promoting and supporting the OSS model.

I know there are more OSS contributors in the world than Microsoft marketing staff...and the majority of us have teeth for debate that could chew through a telephone pole. It's time to put it to some good use.

We need to get off the forums (no offense TSS) and into writing articles, advertisements, open case studies and public presentation. The Microsoft FUD on your boss' desk MUST be drowned in a pile of alternative articles, reports, events and promotions.

Thoughts?

Cheers,
Clinton

Agreed.

But to make sure this does lead us somewhere better we will have to define what we want to bring into the heads of those "bosses with MS FUD on their desk".

Is there a list of, say, TOP 5 criterias that OSS generally performs better in than closed source software ?

I guess depending how you use OSS your most important reason for using OSS differs a lot. So these TOP 5 criterias should not be ordered but be a set of the most important criterias from people looking from different angles at OSS.

Clinton, if you start something, let me know, I'll be interested.

Ciao,
Tobias

Microsoft will never get 7.1 billion downloads, so JBoss alone is far more popular that .NET. ;-)

(Note: While there is some sarcasm in the above statement, it has to be said that JBoss and Tomcat in particular and other Java open source projects have done more to expand the Java marketplace and take the wind out of the .NET sails than just about everything else combined.)

Peace,

Cameron Purdy
Tangosol, Inc.
Coherence: Easily share live data across a cluster!

>
> (Note: While there is some sarcasm in the above statement, it has to be said that JBoss and Tomcat in particular and other Java open source projects have done more to expand the Java marketplace and take the wind out of the .NET sails than just about everything else combined.)

If this isn't hyperbole, I don't know what is. Did OSS project do more to expand Java's market place? sure, that stands to reason. The other comment "wind out of the sails" is no better than FUD that MS is always accused of spreading in this forum.

--Dilip

Dilip: If this isn't hyperbole, I don't know what is. Did OSS project do more to expand Java's market place? sure, that stands to reason. The other comment "wind out of the sails" is no better than FUD that MS is always accused of spreading in this forum.

Well, it is my opinion. I wouldn't call it hyperbole, but you are free to call it whatever you'd like ;-)

The reason that OSS has helped Java significantly is pretty simple:

1. Sun/IBM/BEA/Oracle etc. are "high end" ... relatively few sales for relatively large dollars. They've always been this way.

2. Microsoft is "volume" ... relatively many sales for relatively fewer dollars per piece.

(If we stopped here at #2, the obvious end result is Microsoft continuing its expansion into the companies listed in #1.)

3. OSS (including Linux, Apache Foundation, etc.) provides an even more attractive quantity/price proposition, with the license cost per server at $0. In the US, this has meant that about half of the new servers are Linux or FreeBSD based, for example. Outside of the US, it is even more lopsided, with Microsoft software being viewed as expensive or even unaffordable.

Before you think I'm going off against Microsoft, please re-read the above. I am not saying that Microsoft software is bad or expensive. It is just *more* expensive than OSS software in terms of cost-of-acquisition.

The fact that decent Java OSS is available, including Tomcat and JBoss etc., means that a lot of applications that would have been done on Windows (arguably the next cheapest software license) are now being done on other technologies, and often that is Java.

Once a market starts to tilt, it has its own momentum. HTTP/HTML, Linux, Java, PHP, Python, etc. ... these technologies / platforms / languages have their own critical mass, and have spun up entire markets.

IMHO, J2EE without JBoss and Tomcat (etc.) would have been easy pickings for Microsoft. Microsoft is just too good and too effective at competing with companies that have to make a profit. OSS doesn't suffer from that little problem.

Peace,

Cameron Purdy
Tangosol, Inc.
Coherence: Easily share live data across a cluster!

"it has to be said that JBoss and Tomcat in particular and other Java open source projects have done more to expand the Java marketplace and take the wind out of the .NET sails than just about everything else combined."

ROTFLOL!!!!!
gotta agree with that...there was some concern during the hype period, but .net has pretty much disappeared off the radar screens as a real threat to java, and the reason is not because of sun, but because of the fact there are so many open source tools for java around (and you CANNOT discount the continuing rise of Linux as a driver of java growth on the server). plus, there's actually more hype over J2ME right now than .NET - LOL.

we've had:

- companies planning a .net system then dropping it for J2EE (eg. eBay);
- companies using .net then dropping it in favor of J2EE (e.g. Cerner);
- job banks showing java jobs continuing to overwhelm .net (and one pseudo-study showing .net job postings SHRINKING in monster.com):
http://www.javalobby.org/thread.jsp?forum=61&thread=8595
- microsoft developers complaining about being on a never-ending threadmill:
http://www.angrycoder.com/article.aspx?cid=1&y=2003&m=7&d=17
- c# developers asking for c# to java converters (notwithstanding the fact there are very few mature c# projects around to actually build a market for such a product);
- apache throwing in its lot with j2ee completely with the announcement of a new open source j2ee app server
- vb developers continuing to move to java (31% of disgruntled vb programmers) instead of .net

with the help of open source (and especially linux), it looks like microsoft is in grave danger of being chained forever to its desktop.

ROTFLOL!

There are at least two ways this could be true.

(1) They're actually referring to Web Services. Web services aren't very big in the Java world -- Java has several other communication techologies that are more appropriate for most applications. .NET tends to emphasize web services for everything (applicable or not), so it shouldn't be a surprise if .NET is bigger in web services than Java.

(2) They could be counting deployments of Visual Studio.NET under the assumption that if you use Visual Studio.NET, then you're a .NET programmer. This is a false assumption since Visual Studio.NET can be used to write MFC and Win32 applications. Programmers who move to Visual Studio.NET from Visual Studio 6, most likely want to take advantage of the better C++ standardization and extended MFC and ATF.

Maybe they have partnered with AOL to distribute their 120-day evaluation .net cd's instantly claiming millions of unsuspecting developers overnight. I'll bet the marketing people were using a new version of Excel based upon .net and due to bugs it miscalculated the actual number through multiplication rather than division.

Only a fool would base their future IT business on silly thoughts and I'm sure the media has nothing to do with this.

3. Microsoft is comparing the amount of client software written in any of the various .NET languages with the amount of client software written in Java. I'm sure that there are some folks out there writing WinForm stuff to do client/server access of databases, but relatively fewer folks doing the same with Swing. That Microsoft chose to trumpet .NET's popularity with respect to "Java" instead of with "J2EE" indicates to me that Microsoft is trying to play a numbers game and include client software in the count instead of eliminating client software from the count and providing the more interesting measure of server software popularity. I haven't seen or heard of any evidence that .NET has become popular as a server side development technology.

4. Microsoft is counting users of its for-money Visual Studio .NET IDE as an indication of .NET's popularity, but is not counting users of for-free IDE's such as Eclipse (let alone EMACS) as an indication of Java's popularity. This is a problem that analysts such as Gartner have when making claims about the popularity of Linux. The analysts don't seem to know how to measure the popularity of free software, and make claims based only upon un-meaningful metrics such as license revenue.

Given the economic difficulties of the past few years, there have been relatively few server side development projects started within the various corporations. Therefore, even if all of the new server side projects started since .NET arrived have been .NET projects, it is probably not the case that .NET could have overtaken J2EE in the number server side projects started since the arrival of J2EE.

Now, if all of the new server side projects started since .NET arrived were in fact .NET projects, Microsoft could make some sort of claim about .NET popularity, but Microsoft couldn't necessarily make any claim about .NET's suitability for server side software. After all, the popularity of PHP+MySQL is far greater than the popularity of J2EE+MySQL, but few would argue that PHP+MySQL is suitable for server side software just because of its relative popularity. And if .NET does not prove to be suitable for server side software, then, even *if* it is popular now, it won't remain popular for long.

Until server side projects reach deployment, they are potential failures. Microsoft's claim doesn't seem to be about deployed applications. So, even if Microsoft was able to make a claim about the popularity of .NET for server side development, they would still have to be able to provide a few success stories before they could really talk about the goodness of .NET. EBay pumps millions of transactions through EJBs. Who's pumping millions of transactions through .NET? Without some significant success stories, even *if* .NET is popular now, it won't remain popular for long.

Given that there are almost no free server side libraries/frameworks available for .NET and that there are many (e.g. Cocoon, Lucene) free server side libraries/frameworks available for J2EE, it seems unreasonable to believe that cash-strapped IT departments would opt for .NET over J2EE for server side development.

Moreover, given that there are no and may never be any fully functional CLI / Class Library / Enterprise Services implementations available for Linux, and given the rising popularity of Linux within the cash-strapped IT departments, there's at least one factor that limits the potential popularity of server side .NET, regardless of .NET's goodness.

In short, until Microsoft explains how it arrived at its .NET popularity conclusions, I don't see any reason to believe that its conclusions have any meaning with respect to server side software and J2EE.

The sad thing is that there are plenty of people who only know M$ tools technologies and platforms, that will take comments like this, and repeat them as if they were the truth, to people (Management) who have less of a clue than they have... and Microsoft know this.

They are probably counting installations of Windows in the figure.

Silly. I do not know one developer doing C#

One very good clue of the non-usage of .Net is to browse Souceforge's software map (http://sourceforge.net/softwaremap/trove_list.php?form_cat=160). There's 906 C# projects. Less than assembly. And if you browse thru them, many of them are only ports of Java projects (JUnit, OJB, Log4J).

I remember hearing from .Net a few years ago. It was "the next big thing (tm)".
Since then I've never seen a single project using .Net. I don't even know a single .Net developer. And during the last few years I've seen quite a number of projects (in banking, telecos, e-procurement).

Just my 0.02 Euros.

Julien Dubois.

Hi,

Security is major issue in todays WebServices world .Microsoft technologies has to prove itself first in OS market its security credibility and then talk about webservices and .NET ???? Im tiered of updating my desktop with security patches.System Admins will have fun time in updating .NET servers every week :-)

if you are tired, so why you don't use LINUX.

I'm sorry, I don't understand your argument.
I've been using Linux for several years, with great pleasure & success. What's the point?

> software map (http://sourceforge.net/softwaremap/trove_list.php?form_cat=160).
> There's 906 C# projects. Less than assembly. And if you browse thru them, many
> of them are only ports of Java projects (JUnit, OJB, Log4J).

Well, although I believe .Net usage is far beneath Java usage, I don't think counting SF.net projects is a good measure. Java developers generally make use of a lot of OSS, increasing, of course, the chance of developing OSS projects themselves. Most .Net developers, I suppose, don't feel very comfortable in OSS communities, which are generally anti-Microsoft, "Linux freaks". :)

More stats (jobs in the UK) :
http://www.jobstats.co.uk/jobstats.d/Details.d/Trends.d/SKILL/JAVA.d/index.html
http://www.jobstats.co.uk/jobstats.d/Details.d/Trends.d/SKILL/C.hash..d/index.html

".Net? Does this thing still exist?"

ROTFLOL! indeed it does, but microsoft has to shout about it once in awhile before people forget....

on a side note, if we count j2me phones and smartcards, there will shortly be (or probably already is) more overall java clients than windows (even with addition of embedded win ce i would think).

Before drawing conclusions on these stats, be sure to look at not only the trends, but also the scale and the dates as well...

Claiming an increase of 93% sounds good until you read that it's from "0.161" to "2.3". Java rose as much, only the scale was "9" -> "11"!

To create the amount of momentum that Java saw in the last 7 years, Microsoft would basically have to encore the .COM boom....only rather than relying on clueless investors, they'd have to instead rely on their own clueless executives.

Okay, a bit harsh, but warranted in the face of this charade... ;-)

Cheers,
Clinton

Yes, But please note that all projects are getting offshored to asian countried like india. Try to search there...in asia, .NET jobs are like flooding out...

Where does one look for online job listings in India? None of the sites I found in my google search provided enough data to get anything like a statistically significant comparison.

Check the job listing for all the medium to large companies in the USA? That is where they are headed.

http://www.monster.co.in/

Java - 2361
J2EE - 1366
.Net - 498
C# - 288

I feel pretty safe as a Java developer...in India or otherwise.

Cheers,
Clinton

try to search j2ee in google (1,400,000)
and .net (222,000,000)...

so never depend on web site search results. This is totally dependent on search engine intelligence.

do we have any thing similar to http://www.google.com/microsoft.html

for Sun ?

Yeah, including:

  - www.java.net
  - www.sourceforge.net
  - java.net (as in the java package)
  - Net Nanny!

...and actually anything containing the word "Net". Have a look at the first page of results google returns for ".net" --NOT ONE HAS ANYTHING TO DO WITH MICROSOFT .NET!

Whereas the J2EE search yields 99% real J2EE sites (I doubt anything else is called j2ee).

You're funny. ;-)

Not very scientific obviously, but an interesting index non-the-less. According to this, Java is far far ahead of C# in terms of interest expressed by web references.

http://www.tiobe.com/tiobe_index/tekst.htm

Its funny I have never seen in the M$ website or discussion that Java is kicking their ass or Java is doing much better in mass market than .Not. I guess TSS gets a lotta mula publishing this kinda crap. Futher more I agree that Opensource Java has lot to do with Java popularity these days. I know in our company Eclipse, TomCat and JBOSS is being used in shrinking IT market where we cannot afford Weblogic or WAS.

And finaly the best technology is rewarded !

So Microsoft claims .Net has surpassed Java?

Well, today, I am proclaiming I am surpassing Anakin Skywalker as the greatest Jedi. I am also claiming that Jabba the Hutt is my love child as a result of my affair with Grand Moff Tarkin.

I offer no evidence of these claims - I merely pronounce them as being true - Therefore, they are.

I suggest we deal with MS's gratuitous assertions logically - gratuitously dismiss them.

I am sorry but Luke was the greatest Jedi.

:-)))

If you want to get an idea of how popular .NET has
become, see a listing of the skill sets in most demand at

 http://www.theserverside.com/discussion/thread.jsp?thread_id=18821

just because u installed .net pkg doesn't mean that your webapp is .net.

same way, just b/c you have tomcat plug-in to apache doesn't mean that your webapps is J2EE compliant.

O.K. well it seems that Microsoft appear to be counting all of those .NET evaluation edition CDROMS that are sent out in there Millions. I wonder how many developers actually install them ? Ours is still sitting on the desk AKA coaster.NET

This information was not intended for developers. It was intended to influence management types. It's sad but many managers hear something like this and think, "Wow! That was fast. C# seems to have a lot of [insert buzzword phrase here]. We should use it."

That information is only for Microsoft .Net users, not for java developers. Microsoft will release a new service pack to upgrade it's user.

Number of transactions processed daily by servers running platform X

    multiplied by

Total dollar amount of transactions processed.

It says so right in the top corner of the "Windows Messenger" window on my screen right now.

Maybe Sun should count all desktops that have any form or version of a JVM installed.

>
> Maybe Sun should count all desktops that have any form or version of a JVM installed.

Try to count JAVA applet users too :)

> >
> > Maybe Sun should count all desktops that have any form or version of a JVM installed.
>
> Try to count JAVA applet users too :)

And all the pda's and phones with J2ME support from companies such as:
Motorola, Nokia, Sony/Ericsson, Ericsson, Siemans, Panasonic, and Psion.

This is the alliance that MS has to fight against. I'love my Nokia phone - powered it up some 10 months ago (been charging it once in a week). Not a _single_ 'reboot'.

No. There's no way in this world that MS could ever accomplish anything like that. They lost the mobile market. They're losing on the server market. On desktop they're stalling.

I hate MS. I don't hate their products (some of which are pretty good) but I hate their marketing style. I hate the fact that they use their money to trash competitors. They don't just compete. They compete unfairly.

MS is the only one amongst the big IT-players that creates so much hatred. Something like that cannot last very long. If they continue in the path they've chosen they'll most certainly lose.

Remember - most of us use windows becouse we have to. All linux users use linux becouse they _want_ to. There's a big difference.

Phew. End of rant.

This article songs like, another article that says:
"TSS was bought by Microsoft"

HA HA HA HA HA HA HA

Hello,

According to a recruter (friend of mine), currently there are
aproximately 3:1 java:c#(dot.net) demand. Still, he observes
tendency: new and migration projects go to c# in
"more than half of cases". Considering also recent
TSS test results, it is stupid to call net as "non-existant".
MSFT is wrong (liar) to claim surpass java, but it is not FAR
behind.

AV

Micro$oft simply counted all .net sites on the internet as resources using .Net. They thought that if the site domain name ends with ".net", it's got to be using .Net.

I'll say, right now Java and .NET are competing and Java seems to have the upper hand. And all these guys spelling Microsoft with '$' probably are good for java too: I mean, they won't go for the nicer thing is that thing's coming from MS. But still, they always make me thing of dinosaurs. Denial. Java can be used but it's clumsy. .NET is nicer and can do the same things as Java, cheaper and faster. There isn't one thing Java can do that .NET can't, except running on refrigerators (read UNIX' es). Intel/AMD is gonna win anyway and you should all know that by now.

What about mobile? will Intel/AMD win there too? :-)

People spell MS as Micro$oft because that's what it is about: big $$$ to counter big ideas. As someone said, the problem is not their products, but the way they push things over us. If you accept that, it's ok. But most people don't like it.

PS: what if windows fails?

>People spell MS as Micro$oft because that's what it is about: big $$$ to counter big >ideas. As someone said, the problem is not their products, but the way they push >things over us. If you accept that, it's ok. But most people don't like it.

>PS: what if windows fails?

Mobile is a huge market and Sun was very smart to target it early. But it's not _the market_ for development platforms. Tha's always gonna be the microprocessor and the rich enviroment of a modern OS.

If Windows fails, than Microsoft is gone. But Windows cannot fail, because it has already succeeded. It's up to others to fight the up hill battle agains it (Linux).

"I just want to say LOVE YOU SAN!! Billy Gates why do you make this possible? Stop making money and fix your software!!"

"I just want to say LOVE YOU SAN!! Billy Gates why do you make this possible? Stop making money and fix your software!!"

NO DOUBT!!! Since my work computer was upgraded to Windows XP, I've been infected with 2 worms/viruses (in 4-5 months). With NT4, I hadn't been infected in 4 years of use. Don't be fooled --Windows security is getting MUCH worse.

The interesting thing about the MSBlast virus (from which the text above was derived), is that it's a DCOM flaw --a popular Microsoft interface for developing distributed components (competes with EJB). In a sense THIS IS the foundation .NET is built on! I can't wait until .Net servers/products start exposing their interfaces as web services..."wow look, we can tunnel THROUGH firewalls on port 80 --yay!".

The only thing secure in a Microsoft world is their grasp on your wallet.

Cheers,
Clinton

I'm not a Microsoft fan, I'm not a _fan_ of any software, but posts like these make me reply (from time to time).

<quote>
"I just want to say LOVE YOU SAN!! Billy Gates why do you make this possible? Stop making money and fix your software!!"
</quote>
A statement like this should not be taken literally by any serious person. Making money and making software go hand in hand, they don't exclude each other. This is one of the biggest industries in the world and that's why we get payed so well! This phrase can only be the product of a delusional kid which thinks that the world of software is built on perl and bash scripts.

<quote>
NO DOUBT!!! Since my work computer was upgraded to Windows XP, I've been infected with 2 worms/viruses (in 4-5 months). With NT4, I hadn't been infected in 4 years of use. Don't be fooled --Windows security is getting MUCH worse.
</quote>
Yeah ? In the time of NT 4, not many worms were around. Still, i remember some pretty bad cracks for that OS. We could easily check securityfocus.org or something for stats concerning bug/OS/year.

<quote>
The interesting thing about the MSBlast virus (from which the text above was derived), is that it's a DCOM flaw --a popular Microsoft interface for developing distributed components (competes with EJB). In a sense THIS IS the foundation .NET is built on! I can't wait until .Net servers/products start exposing their interfaces as web services..."wow look, we can tunnel THROUGH firewalls on port 80 --yay!".
</quote>
You shouldn't speak if you don't know. For one, DCOM is antique. It's from the NT 4 days. Right now there's COM+ version 1.5 or something. Second, .NET is NOT based on COM. You can build huge, enterprise applications without COM just as you can in J2EE without EJB. Many people recommend doing stuff without EJB (JDO, rmi and the rest as a 'replacement'). Trust me, the equivalent can be done in .NET. I've been there.
As for how actually you've made the connection between an exposed interface and a firewall crack, that's beyond me. And you know the port, too. It's 80.

this is my take on the whole .net & microsoft & "secure" computing. very few things in the microsoft world have been truly backwards compatible between major revision numbers. meaning that they introduce a ton of new source code which opens themselves up for new bugs, many of which are security based. they don't seem to care because they want to make a new product in order to generate revenue to keep investors happy and their monster fat, even though at the same time they break a ton of stuff causing negative stigmas with their software, and driving customers/developers away to a more stable platform... linux & java. true, i do not have any experience with .net, the couple articles i've read have mentioned that they've done a much better job with backwards compatibility between their two .net releases of it (1.0 - 1.1) however i've heard of some bad cases where it is broken ( i could be wrong ) but i pose this question. how can you have a stable "enterprise" environment if there's always a chance of installing something (a much over-hyped .net widget) that breaks existing code.? sounds like a lot of sleepless nights to me. i am personally diverse in both unix & window worlds (os, software, development tools, libraries) and prefer the linux/unix worlds + java + even php because the approach that they take to software engineering seems to be more thought out.. in regards to developing a specification, standards, openness, interoperability, (insert catch-phrase here), etc. i know, i know the same issues pop up from time to time, but i bet it's less frequent :). that is why i haven't migrated to .net at all. i DECIDED to not buy any new tickets for that roller-coaster, tired of puking my guts out!


enjoy.

>> Right now there's COM+ version 1.5 or something.

If you think DCOM will go away anytime soon, you will be sadly surprised. Further, if you think COM+ doesn't use DCOM, you are again sadly mistaken.

Please refer to the following website:

http://www.microsoft.com/com/presentations/default.asp

Download the zipped presentation titles COM+ internals.

Refer to statements such as: "Remoting architecture doesn’t change (DCOM)"

I'm sure there are more technical documents to reference for more information...if you really want to take it any further.

>> Second, .NET is NOT based on COM. You can build huge,

See above. If you think Microsoft is just going to up-and-throw-away one of it's largest technological investments (and possibly one of it's only innovations), think again. Even if you can't see it (i.e. wrapped with COM+), trust that DCOM is there and will be for a long time.

>> As for how actually you've made the connection
>> between an exposed interface and a firewall crack,
>> that's beyond me. And you know the port, too. It's 80.

Microsoft has been advertising (nay, bragging) at the click of a button developers (including Microsoft) can expose major APIs arbitrarily via Web Services (which typically run via HTTP; default port 80). I can't wait to see what holes the cracker community will find therein.

I hope that clears things up for you.

Cheers,
Clinton

Leaving COM/DCOM/COM+ out of the picture, I wanted to look up .Net Remoting to see what it had to offer in improving security over DCOM. One document I found had an interesting quote in discussing the differences between DCOM and .Net Remoting. This was stated:

"DCOM is very secure. .NET Remoting has no built-in security mechanisms (except when hosted by IIS)."

Three things on this:

  1) We now know how (in)secure DCOM is...yet they consider it "very" secure.
  2) .Net Remoting has NO security??? Perhaps the MS approach to solving their security problems is by avoiding it altogether?
  3) Correction, it's only secure when hosted by IIS? Does ANYONE really consider IIS a star of security?

http://www.osborne.com/products/0072224436/0072224436_ch27.pdf

Wow.

Clinton,

> 2) .Net Remoting has NO security??? Perhaps the MS approach to solving their security problems is by avoiding it altogether?

I just about fell out of my chair when I read this. Good One! Made my afternoon.

Chris

If you read up on remoting,

you might as well read up on code access

security and role based security.


If you make fun of something, you better be right. Or else
you look foolish :-)

chameleon,

Don't be offended, I'm being more sarcastic than anything. I'm not a .Net expert (I think anyone who knows me, knows that). In case anyone didn't notice, this thread has taken a farcical turn that is matched only by the original assertion made by Microsoft (which I've found quite amusing).

But your point is valid: Just because Microsoft can do anything and say anything it wants about the competition, doesn't mean I can...right?

I guess I'll need to work on my stockpile of billions of dollars before I gain that right.

PS: More sarcasm.

Cheers, ;-)
Clinton

<quote>
.NET is NOT based on COM. You can build huge, enterprise applications without COM just as you can in J2EE without EJB.
</quote>

Edward,

Please represent MS technology (however flawed it may be) well and research before you speak here. I deal in both worlds (MS and J2EE) and have to defend MS decisions when made. This is very difficult when I must fight foolish statements like this.

You should really go back and understand everything from WOSA, DDE, OLE1.x, OLE2.0, COM, DCOM, COM+ to now (Chris Sells and Don Box are excellent references). Please educate yourself with the following libraries:

mscoree.dll - a COM wrapper dll that is the entry point to the CLR (delegates CLR startup to one of the following)
mscorwks.dll - a COM dll that IS the CLR for one-way boxes
mscorsvr.dll - a COM dll that IS the CLR for multi-way boxes

Just open up a new console project in VC6 and use some raw COM programming to load the CLR: CoInitialize, CoGetClassObject, CLSID_CorRun, IID_CorRun, ICorRuntimeHost, IUnknown, etc

Maybe, your statement implied that .Net DEVELOPMENT is not based on COM techniques. This is mostly true but the runtime is MORE THAN DEPENDENT on the SCM, COM, et al.

Please just be specific next time...

Thanks,
Lance

<quote>
Edward,

Please represent MS technology (however flawed it may be) well and research before you speak here. I deal in both worlds (MS and J2EE) and have to defend MS decisions when made. This is very difficult when I must fight foolish statements like this.

You should really go back and understand everything from WOSA, DDE, OLE1.x, OLE2.0, COM, DCOM, COM+ to now (Chris Sells and Don Box are excellent references). Please educate yourself with the following libraries:

mscoree.dll - a COM wrapper dll that is the entry point to the CLR (delegates CLR startup to one of the following)
mscorwks.dll - a COM dll that IS the CLR for one-way boxes
mscorsvr.dll - a COM dll that IS the CLR for multi-way boxes

Just open up a new console project in VC6 and use some raw COM programming to load the CLR: CoInitialize, CoGetClassObject, CLSID_CorRun, IID_CorRun, ICorRuntimeHost, IUnknown, etc

Maybe, your statement implied that .Net DEVELOPMENT is not based on COM techniques. This is mostly true but the runtime is MORE THAN DEPENDENT on the SCM, COM, et al.

Please just be specific next time...

Thanks,
Lance
</quote>
COM is inside Windows, so you if you use it or you don't you have no choice anyway. It's deep inside. I didn't mean that. What i meant is that you don't have to build your own serviced components to do a decent distributed application, because .net remoting is great. I was addressing the productivity/development issues, not the inner plumbings of windows.

Hope this clears things up.

Edward

Up to now it's not been _the market_, but trends have shown that this kind of platform to gradually take over desktops. Looks like palm and similar devices are going to be the majority platform for everyday work, eventually. And MS is not big in that market, despite their efforts.

>
> If Windows fails, than Microsoft is gone. But Windows cannot fail, because it has already succeeded. It's up to others to fight the up hill battle agains it (Linux).

Just because it has succeeded, doesn't mean it can't fail in the future. Nothing garantees this. Of course it won't disappear overnight, but strong competition will surely shrink it's market more and more, and if (a big IF) linux finally have big acceptance in home and offices, no windows success will stop linux from taking over. Personally, I find it very unlikely to happen, bit it's not impossible.

Google (15.8.2003)

"x programming" (where x stands for Java/C/C++/etc)

1. Java 467,000
2. C 324,000
3. C++ 315,000
4. Perl 154,000
5. PHP 119,000
6. .NET 92,200
7. Visual Basic 68,500
8. C# 43,900
9. Javascript 39,900
10.Python 26,000

"x developer"

1. Java 102,000
2. .NET 99,300
2. PHP 70,800
3. C++ 19,700
4. Visual Basic 11,500
5. Perl 8,400
6. C 5,760
7. C# 3,510
8. Javascript 3,020
9. Python 2,270
10.Ruby 1,470

"x programmer"

1. Java 84,900
2. C++ 34,500
3. C 29,200
4. Visual Basic 19,600
5. Perl 18,400
6. PHP 15,000
7. .NET 9,800
8. Python 6,190
9. Javascript 3,720
10.C# 3,430

"x architect"

1. PHP 25,900
2. Java 7,510
3. .NET 6,830
4. C# 122
5. C++ 106
6. Python 49
7. Javascript 43
8. Visual Basic 25
9. Ruby 12
10.Perl 10

I feel that using ".NET" in searches produces results that favor .NET overmuch.

.NET includes such things as ADO.NET, ASP.NET, Freenet, general networking, all kinds of networked organizations & web-rings, etc..
I'd subtract at least some 20..30% from .NET figures before taking them 'as a fact'. But the figures here are unaltered readings from Google.

P.S.
Man, do PHP guys do any programming at all... it seems that all their time is spent on architecting. ;-)

P.S.II
Trolf - Since you were so keen on statistics, what do these "real life - easy to reproduce" figures tell you?

..Forgot to mention:

There seems to be exactly 10 Perl architects in the whole world... 8-) I guess they're all writing CPAN scripts?

As it seems windowsupdate.com is running on Linux since 15 August.

Check netcraft:

http://uptime.netcraft.com/up/graph/?host=www.windowsupdate.com