.Net:... 3 Years Later

Ashcrow writes "EWeek has posted an article on Microsoft's .NET initiative. It's been three years since we were first introduced to .NET and virtually none of the promised advantages have come true. Is it time for Microsoft to move on?"

Comments

Reality is quite nice though

[mccalli]

The marketing hype surrounding .Net evaporated, true. However as a means of developing for Windows in virtual machine which supports multiple languages, the actually technology is still going strong.

And so it should - it's better than the alternatives which preceded it. It's just important to divorce the .Net marketing cloud from the actual technology on the ground.

Cheers,
Ian

Re:Reality is quite nice though

[Ooblek]

I would have to agree. One of the problems is that even Microsoft is still releasing products that are based on BizTalk and Commerce Server, which seems sort of counter-intuitive. If you look at some of the new products being released by Microsoft Business Solutions (aka Great Plains), you have to wonder what they were thinking. Their business portal product is based on BizTalk, their .NET CRM application talks to the financial application through BizTalk, and they still have their e-commerce packages that are based on plain-old ASP and COM+.

I will say though that I have recently been working on a project to allow a unix legacy system talk to a web service to do real time credit card authorizations from a COBOL application. Using GCC 3.3, libxml2, libxml++, and libwww to post to a web service, it appears to be transacting quite nicely. I can see a lot of legacy application adapters being developed in this manner in the future. Now if only some of the documentation of these libraries were better....

Re:Reality is quite nice though

[FatRatBastard]

Agreed about technology vs marketing hype, but there's something about .NET that has bothered the hell out of me. They technology (or at least the hype around it) is at odds with the business reality at MS.

MS claim that .NET will be open and cross platform, but the only way this can happen is if cross platform means "across *our* platforms."

Currently MS makes the bulk of their money from the OS and Office. If they truely made .NET cross platform (or let something like Mono take hold) then that starts to eat into both their server and desktop base. I mean, why would anyone pay MS $$$ for each desktop / server if you could choose between *BSD / Linux / VMS / Un*x / et al? For instance, if I had cycles to burn on an IBM mainframe it would make sense to host my .NET services on it, assuming it was truely cross platform.

So basically I fail to see how MS could inplement a businees plant such that .NET would generate more money than the potential loses from the hit they'd take on server / desktop licenses.

Again, MS makes (prints???) money by selling OSs and Office (everything else is just a rounding error). You can be damn sure they're not going to do anything to threaten that cash cow. The interesting thing will be how MS ties .NET to its own OSs. The big draw about web services is that they're supposed to facilitate easier communication / data sharing between disparate systems.

Re:Reality is quite nice though

[Chester K]

They technology (or at least the hype around it) is at odds with the business reality at MS.

MS claim that .NET will be open and cross platform, but the only way this can happen is if cross platform means "across *our* platforms."

.NET is about interoperability, not cross-platform execution. The big reason Microsoft is behind .NET is to get Windows a foothold in shops currently based around Unix.

NO tolerance for standards wars

[mekkab]

GM's Scott issued a strong warning to Microsoft, Sun and the other players in the Web services industry, that enterprises will not tolerate the standards wars of the past. "We have no appetite for it," he said

Exactly, so he and everybody else is sitting back and waiting for a clear winner with mature functionality to materialize.
In other words, he's saying "Screw .net, let some other schmuck take the cost of developing it. WE got screwed on ISO networking and Token ring! Twice bitten, 3 times shy!"

It takes insight to notice these things take time.

[Sheetrock]

Very few of today's Internet standards were recognized even within three years as standards. Usenet took seven before it became ubiquitous, IRC took at least four (with DCC still not part of the spec), and even the WWW took six. Remember, it was fundamentally a revision of Gopher technologies, which in turn were an iteration of something else (Archie?)

Most of .NET was puffery, to be sure (I read a piece on MSDN more or less admitting this), but that's largely because it was a working title given to a number of next-generation technologies that may or may not pan out, many of which haven't been released. You can't really consider C# or Hailstorm to have been around and competing for three years, can you?

fun with fud.

[x0n]

Woe betides us once more: brace yourselves for another flood of misinformed, biased and downright incorrect assertions from both sides of the fence. Please, no "c# is java", ".net is slower than java" or other such empty statements. If you've worked with .NET for 6 months plus (remoting/asp.net/interop/ado.net), great. We welcome your comments. Perl monkeys need not apply.

Likewise for you "java" programmers out there who in actuality have only ever compiled one applet, and it was a recompilation of a decompiled shareware scroller that you removed the copyright notice from. Well done. On the other hand, if you've solid experience developing beans, rmi and other such projects, we also welcome your comments.

The rest of you shut up and learn.

Rant over.

- Oisin

What about Linux

[fudgefactor7]

All the Linux vendors out there pretty much said that they were going to take over about 3 years ago too...is it time for them to move on as well?

all about the Benjamins

[AssFace]

say the words "dot net" and you get to add so much to the cost of projects that it immediately makes it worth it to switch over.

that is the only reason I could see why .NET might ever catch on.
I'm not saying it is a useless bit of technology, I'm just personally partial to using any number of existing technologies that do the same thing and are cheaper to implement.

my current employer is retarded when it comes to computers and they paid someone to do a very basic web project in "dot net" because there was a general misunderstanding in the difference between the domain and the programming structures.

In the end it cost them a ton and now it is costing them more to maintain. I am trying to get them to port it all over to a much lighter system (php on linux or freebsd), but they are currently not interested.

.Net a complete success

[tanguyr]

.Net was (and still is) a marketing ploy to counter the sudden gains in mindspace being made first by Sun with J2EE and later by "web services" in general. Judging from the fact that most PHBs have heard about it it seems to have worked quite well - the fact that they (or, it seems, almost anybody) have no idea what it does it besides the point. As long as MS is still getting column inches ("comparing .Net to Crack Cocaine" or whatever) then it's working for them just fine, thanks. This isn't anything new - MS practically invented the word "vapourware" back in the 90's. I'm not saying .Net does nothing, i'm saying that the engineers got there after the marketing department and the advertising budget.

/t

Speaking for myself

[m00nun1t]

As a pretty experienced web developer, I've worked at some level (some more than others) with most of the popular platforms: ASP, PHP, Cold Fusion, JSP and ASP.NET (very little perl, which I've always regretted if just for completeness).

From that perspective, ASP.NET just totally rocks my world. I can debug more easily. Performance is better. It encourages good architectural practices. And my productivity has gone through the roof - I haven't done any formal tests but based on personal experience I'd say I can develop at *least* 30% faster with ASP.NET compared to any other platform, possibly more. The difference is most pronounced in more complex systems where it really shines. For less than, say, a thousand lines of code it probably doesn't save as much time, but I rarely work on that anyway.

So, maybe .NET has "failed" and maybe not, but for me, ASP.NET has improved my working world radically. Don't knock it till you've tried it.

Re:Speaking for myself

[truthsearch]

Other than Cold Fusion I've also worked with all of these. The largest learning curve definitely goes to ASP.NET. My overall preference is easily PHP. One factor is many things can be done with far less code in PHP than ASP.NET. The only advantage ASP.NET has over anything else is the tool, VS.NET. It's not the technology that's saving development time. It's the tool helping to write the code and debug that's the real time saver. So from a business point of view it may be the right choice since the tool's good enough to make up for some overly complex platform requirements. But if you get good at manually typing PHP it's far and away superior based on my experiences and from others I've read.

Re:Speaking for myself

[m00nun1t]

You are right about the IDE being a big part of the advantage of ASP.NET (and I should have mentioned it), but it certainly can't take all the credit.

IMHO the biggest thing in ASP.NET that leaves PHP for dust is the separation of code from layout. The other big one (and closely related) is easy componentisation. These two make life so much easier, and speak to much of the architectural niceties I mentioned in my original post. Not only can it be done, but it's easy to do and the flow *encourages* you to do it. I love a tool that makes it easy for me to do things the right way.

I do agree ASP.NET has a steeper learning curve than PHP (or any of the others listed, with the possible exception of JSP). Based on my experience, it's a price well worth paying.

For a small project, PHP would usually be my first choice, but anything medium to large, IMHO, ASP.NET is just miles better. Not trying to start a religous war as I do respect PHP, but I thought this was interesting, a comment from a respected member of the PHP community: http://www.edwardbear.org/blog/archives/000189.htm l

.Net was never clearly defined

[cait56]

Three years in and I believe it is fair to say that most people do not understand exactly what .Net is -- other than a vague "trust me" monolithic solution.

Which I believe is the core of its problem. While there are some fools who will buy anything that fill in the name of their favorite supplier offers, more of the market wants to make decisions for themelves.

From the little I've had time to study .Net, there were a few aspects of it that were indeed superior to what had proceeded it on the market. But the information to make a cohesive strategy was just missing. What if I liked the characteristics of the run-time engine, but needed to stick with CORBA interfacing?

The most telling flaw in the strategy, for me, was that you could find entire racks of books on .Net. But absolutely none that explained the basic wire protocols used. They were all "How to Program a .Net application inside one box using language Y".

When I'm designing a system, the language used on each box is the last detail that I consider. I want to understand the interactions of the remote systems, how dependent they are on each other, how they evolve seperately, how the failure of one will affect the others, etc.

Re:.Net was never clearly defined

[johnmckeon]

Three years in and I believe it is fair to say that most people do not understand exactly what .Net is -- other than a vague "trust me" monolithic solution.

It seems to me that this has been a problem whenever MS introduces a new technology (COM, COM+, ActiveX). I can find plenty of people using these terms, but no one can give me a two or three sentence summary of what they are. Unfortunately, it seems like .NET is having the same problem.

Re:.Net was never clearly defined

[cait56]

On a point-by-point comparison, .Net frequently is superior to Java. It falls short on the fundamental points you raise: interoperability, and more importantly seperability. Using Java you know exactly which technologies you are embracing, and which you are leaving out. Java/XML, Java/RMI, Java/Corba... It's all your decision.

The other feature that .Net has is superior native execution, it was designed to be translated to native code. The .net virtual machine is better defined than the JVM is. But I agree that on whole, the tradeoff is not worthwhile.

Re:You are kidding, right?

[Michael Hunt]

You, Sir, are a troll :)

Albeit, a very good troll in that you ALMOST had me going until I read point 4. Upon rereading your other points:

1) 'Single-source logons' are a function of AD/Kerberos under 2000/2003. In a corporate environment, they give you all the benefits you're claiming that .net does. The '.net passport' stuff hasn't really taken off (is anybody apart from hotmail and msn using it?)

2) How does this have anything to do with .net? Remote access is a function of authentication (AD/KDC as above in a 2000/3 environment) and security (leased line or 'VPN'.) .net has nothing to do with the latter part of the equation.

3) Since the various .net RPC mechanisms use a more verbose protocol than traditional MS/DCE RPC calls, I fail to see how this could be the case, unless you're using the 9/10 of your TCO saved in (5) to buy bigger pipes.

4) My windows 2000 servers at work usually only get a reboot when someone installs a hotfix. Since the patch lifecycle is test->uat->production, we have ample warning for this. Uptime, on average, is around 5-6 months. These machines are everything from AD controllers supporting thousands of users, to RDP/MS TS boxes with 50-odd users each.

5) correct me if i'm wrong, but isn't .net more expensive, being subscription based? I realise that this isn't the whole of the TCO equation, but windows servers are windows servers, and no amount of point and click window dressing is going to reduce the amount of manpower required to run systems well.

I'm no Windows apologist (check my posting history,) but surely your argument is bunk :P

IHBT. IHL. HAND.

.NET = Windows API 2.0

[Anonymous Coward]

.NET has little to do with anything .NET. It's a new Windows API designed to turn Windows into a virtual machine like Java so it can be architecture independent. That's what CLR and C# and all the rest of that stuff is about. It's about MS getting off x86-32 and into a larger world of ia64, amd64, and maybe even ppc64. CLR is the new Windows runtime. Once the move is complete, Windows will be able to run on anything and apps will not have to be recompiled at all. This will make Windows more portable than *nix.

If Sun didn't invent Java would .NET exist?

[wukie]

If Apple didn't implement Xerox's windows would Microsoft have created a version?

If Apple hadn't invented multi-media for micro-computers would Microsoft have it's own implementation?

Microsoft haven't done any (apart from Word for Mac, then later Windows) inventing of their own, and what they have done, has always been a poor copy! .NET is a perfect example.

They must be doing SOMETHING right

[chia_monkey]

Think about it...for three years they've been talking about this amazing .NET thing. And every year the masses go "what the hell is this?" and each year it gets a feature here dropped and a feature there dropped. And yet, after three years, people still talk about it. People still want to develop for it. People are still holding out from developing with any of the other options because of .NET.

So...it may not DO anything just yet, but in terms of stalling development on other platforms and continuing to put MS in the news, I'd say it's a success.

In my Experience . . .

[Badgerman]

The results are really odd. .NET adaption went slower than I expected. It was crammed down our throats . . . and no one really seemed to care.

Then, recently (last year) I've seen a real explosion in .NET. Literally I think 75% of Microsofts pushing the tool was useless or even backfired. Time seemed to be needed.

As a developer who has worked in a variety of languages, OVERALL, .NET has some good ideas mixed in with some unexpectedly lame ones. In general I'm able to develop faster and more efficiently (In some cases I've developed ASP.NET applications over twice as fast as ASP, yet with far less ASP.NET experience), but there are moments of strange and odd roadblocks.

Do I think .NET will rule the world? Not really. It's just one of many options. Web development and related technologies seem to be in a phase that's a mix of overcautious and overenthusiastic, and I don't think anyone is sure where things are going right now.

Will Microsoft give up on .NET? I don't expect they will - too much of an investment, too much behind it. It'll get altered and poked and prodded and integrated, but it'll be around in some form for awhile.

Re:Yes

[mingot]

Without a doubt, the latter of the two was far superior in every aspect, INCLUDING EASE OF USE. PHP has got to be the easiest freakin language ever

A lot of things are "easier" than ASP.NET/ADO.NET coded using an OOP language. For simple things you're better off using something like PHP or ASP/VBS. Of course when project complexity reaches a certain point you'll start to find real advantages to going with a modern approach that seperates the presentation layer from the business layer. Of course taking this approach can make writing a simple application seem daunting, but in the long run it pays off.

It has a lot to do with simply knowing what sort of application you're going to be writing and picking the proper tool for the job.

Apache trumps IIS with the ability to do the majority of configuring with one file, instead of having to browse through a maze of tabbed windows with options, checkboxes, pop-up boxes, etc.

Totally. 100% agreed. Much easier to administer Apache via it's text configuration IMO.

I have to agree

[uradu]

I don't know how much platform independence has been a consideration, but they probably just got sick and tired of plain old Win32 and MFC. If nothing else, it gave them a chance to finally bring out a decent framework, just like everybody else already has. Must be so liberating for them to finally be able to code a dialog box dynamically without having to fool with resources and message map macros. Microsoft have finally discovered proper OOP and class frameworks. Welcome to the '90!

Re:It actually outperforms J2EE by a lot

[Inf0phreak]

Consider this: Microsoft's EULA states that you may not publish a review of .Net's performerance without their explicit written permission. Now tell me if you really think that there will be any negative reviews of it?

MS knows it's a dog. It's as simple as that.

Silly

[Overly Critical Guy]

It is far from time for them to move on. Longhorn will be entirely .NET based. The latest betas already have explorer.exe running as .NET managed code. The old, crufty Win32 that Slashbots loved to bash is finally being replaced, and all Slashbots can do is find new ways to complain.

This is just Slashdot getting its weekly naysaying in. .NET is coming and will be here to stay with Longhorn, and enough people like .NET to have started work on a version for Linux.

Re:J2EE

[Get Behind the Mule]

If its time from Microsoft to move on from .NET then its time for Sun, IBM, Oracle, etc to move on from J2EE.

That's one company with the one technology, and three companies plus "etc." with the other. Wouldn't it make more sense for Microsoft to drop .NET and join everyone else with J2EE?

.NET a definite upgrade, good competition for Java

[shodson]

If you've been building Windows apps for a while you have welcomed .NET because it makes building Windows apps much simpler than the complexities of VC++ and rescues us from having to deal with the hoakiness of VB. As a long-time Java developer as well I am glad to have a full set of OOP features in a VM-like environment like there is in Java available to me. If Java supported the Windows desktop more elegantly and efficiently then .NET wouldn't matter as much, but Swing is dismally slow and cumbersome for Windows apps, though JDK 1.4.2 is supposed to be better. But look at the rift IBM's SWT has caused in the Java/desktop community.

And I don't agree .NET is just about desktop apps. It makes building distributed apps easier as well, if you want to use web services. I do believe, however, J2EE is still a stronger alternative for large-scale distributed apps. But let's face it, nobody cared much about web services until .NET. Not that a lot of people care too much now, but it's seen as the future of distributed computing, from an internet-scale basis, by just about everybody. What else is there, CORBA? RMI? EJBs? Puhleeze. Firewall unfriendliness is the biggest challenges for these protocols. And the Java camp has been working feverishly to add web services support to their platform and developers have been demanding it. See J2EE 1.4, Apache Axis, Sun's WSDP, BEA's "as-easy-as-VB" WebLogic Workshop IDE for building web services, etc.

The best thing .NET has done perhaps is light a fire under the pants of those in the Java camps. Since .NET's release Sun and the major Java vendors have been scrambling to "answer" some of the advantages of .NET and the cool features of C#. The JCP is trying to respond more quickly. The upcoming JDK 1.5 will have most language changes since 1.1 (generics, foreach iterations, attributes) in an attempt to meet or beat some of C#'s strengths over Java, etc. And the prospect of open-sourcing Java is becoming more of a reality as Sun's stranglehold on the standard has slowed the pace of Java's improvement and started to cause some splintering among some previously strong supporters of Java (aka, IBM, creating SWT, not showing up at this year's JavaOne, etc.)