Fog Creek Software
Discussion Board




When will .NET achieve significant market

penetration?

This post is actually an opinion poll. I am simply curious as to where on the opinion spectrum various JOS forum participants currently stand.

* Adoption is simply a matter of time (most Microsoft developers are simply going to have to wait until everybody from internal IT departments to home computer users have a Windows OS where the .NET framework comes pre-installed).

* I don't know or care to speculate.

* Many companies might do one or two pilot projects, but most will never adopt this development platform like they did with Java?


Btw, I am talking about the business data processing sector of the Information Technology industry (i.e. primarily companies with some sort of internal IT department rather than companies that make a living producing software). I am firmly entrenched in the "I just don't know what the future holds for .NET but wish I did" camp.

While I have read/heard many diverse opinions from both sides of the fence, such as,

"... If Microsoft was interested in getting it right this time around then there wouldn't be a need/reason for the Open Source Mono project ..."

to

"... the fact that .NET comes from Microsoft will mean many companies will eventually choose to adopt ..."

I currently don't feel confident predicting whose opinion will turn out to be right.

Right now, the apparent slow adoption rate of .NET can probably be blamed on the state of the economy (spend less on IT not more) and the fact that most companies like to wait a few years before making a significant investment in what they consider to be "bleeding edge" technology.

One Programmer's Opinion
Wednesday, December 03, 2003

Adoption is simply a matter of time

ted
Wednesday, December 03, 2003

Resistance is futile.

Locutus
Wednesday, December 03, 2003

In 2008 more than 800 million people worldwide will be developing .NET applications.

Gartner
Wednesday, December 03, 2003

If you consider that .NET will completely replace VisualBasic; it'll have huge marketshare no matter what.

Microsoft is not going to let up on .NET; it's too advantageous for them.  I've even heard that certain Longhorn features are only accessible from .NET.  As part of the natural development tool upgrade cycle I imagine pretty much everyone who is currently doing Windows development w/ COM or VB will be using .NET in the future.

Almost Anonymous
Wednesday, December 03, 2003

"If you consider that .NET will completely replace VisualBasic; it'll have huge marketshare no matter what."

You actually point out a big danger of Microsoft's .NET initiative - To a lot of firms, a switch to .NET represents a wholescale change. Given the shift of such a change, and the fact that Microsoft has basically end-of-lined some traditional tools and technologies (at least from a marketese perspective), a lot of organizations have taken the opportunity to re-analyze their platforms, giving Java and other solutions a possible-win when otherwise it would have been a continuation of the status quo and the platform wouldn't have been questioned.

As an example, I was at Microsoft convention quite a few years back (seriously at least 4 years ago) when the Microsoft presenter proclaimed that ActiveX was dead and was being replaced, and everyone had better get their ass in gear preparing on the new technologies (sidenote -- The technology sounds a hell of a lot like what I'm now hearing about Avalon). This was to a hall full of corporate programmers that had invested a lot of blood and sweat in ActiveX controls for intranet and application purposes. I suspect that more than a few of them picked up Java that night. Microsoft is absolutely notorious for doing things like that (again, at least from a marketing perspective, and from the mouths of the talking heads. In reality they keep compatibility for quite a long time), and pretending that the only company that competes with Microsoft is Microsoft.

"I've even heard that certain Longhorn features are only accessible from .NET"

I've been hearing nonsensical bullshit trying to fear people onto the .NET train for _years_ now. I heard several years back, for instance, that the new Office (just released), would be entirely built with .NET. A lot of it sounds notoriously like the Java FUD that came out for years.

Dennis Forbes
Wednesday, December 03, 2003

I'm not moving to .Net until ALL of the COM classes that I need are moved so I don't have to use those silly "compatibility libs".

I mean, the thing that really sucks is that I have a complete toolkit/application framework for COM/VB6 that I'd have to re-do for .Net.  Also, all the 3rd party tools I already own for VB6.

So until I CAN'T use VB6, I'll use VB6.  Besides, I tell all of our clients to stay away from Win2k3 like the plague.  What does it have that you really need?

Wayne
Wednesday, December 03, 2003

If Windows is still the major OS that is installed by OEMs in 2006 then .NET will have deep penetration.

.NET is aparantly a "bet the company" strategy for MS, but there is a chicken-egg thing here. Which will die first Windows or .NET?

Gp
Wednesday, December 03, 2003

<yawn>

Rick Childress
Wednesday, December 03, 2003

Does it matter?  Client side development has gone the way of the dodo.


Wednesday, December 03, 2003

quick check on Dice shows:

.net  2605
j2ee  4546

not insignificant

doobius
Wednesday, December 03, 2003

Interesting comment Rick Childress.  Now, was that a sarcastic <yawn>, or are you really sleepy and just felt the need to tell us about it?  Thanks.

And what is this about client side development going the way of the dodo?  What has this got to do with the price of beans in Argentina?  You can do client or server sided stuff with .Net, what's your point?  Besides, do you really think that people want to be tied to a server to run every single app they need?

Last but not least, doobie-us, this is insignificant because of all of the unknowns.  A search for "visual basic" on monster produced 3158 jobs and most of them were not .Net jobs.

Wayne
Wednesday, December 03, 2003

Seeing the same topic over and over and over and over and over and (well, you get the idea) makes me sleepy.

This is just like the 'vb.net vs c#' thread over in the JoS .net discussion board.

These topics are tired and should be moderated out.

Rick Childress
Wednesday, December 03, 2003

There have been some good points raised here.  I personally have Visual Studio.net installed and an individual install of VB6.  I really only have .NET installed to play around with it -- all my actual work has been done working on older software written in VB6.

However, the VB6 line is over.  There will be no alternative (if I want to continue to develop Windows applications) but to move to .NET.  I think that means that alot of people will move to .NET.

The main problem with Java is that it's still not usable for building real client-side applications.  Microsoft's advantage is that have essentially a java-like platform that can be used to create real client-side applications.

As for client-side development going the way of the Dodo...  I can, at least temporarily, agree with that.  Hell, I spend 99% of my day building web applications.  But the browser interface sucks and shows no sign of sucking less in the future.  If Microsoft can create a hybrid platform taking the best of the client-side and browser deployment they will be pretty well off.  Since nobody else appears to even be attempting this, Microsoft has a good chance.

Almost Anonymous
Wednesday, December 03, 2003

One Programmer's Opinion - are you the guy who did that pretentious article about outsourcing and then posted a link here?

And now you're going to do one on .NET and, as with the last one, are getting thinkers here to give you ideas which you a) don't acknowledge b) ignore if they disagree with your own point of view?

Getting tired of crap
Wednesday, December 03, 2003

(As with most things) browser-based interfaces haven't quite delivered what they are supposed to. The benefits are clear in terms of how easy it is to roll out upgrades and the like, but there are times when the web UI seems too clunky.

Part of me thinks that too many people still don't have their head around how to build a web UI. Last week I was in the bank and watched as the bank guy signed me up for a new account - they were using a mainframe system, but the workflow was very much in the style of web request/response - the system seemed to do everything it needed to and would work really well as a web-based app.

Still, there are times when people do need a more accomodating interface. .NET seems like the best bet for writing desktop apps (as long as people are using Windows, I'm not sold on Mono yet) - I think Borland's market share seems a bit to small to go with a non-.NET version of Delphi; Sun are futzing around with Swing and still not delivering (I don't really think SWT is all that either); Cocoa is supposed to have great UI APIs, just so long as management approves replacing all those PCs with Macs; Linux doesn't seem to be there yet in terms of one viable desktop GUI to rule them all (see Redhat's recent move). So while there is a demand for desktop apps there will be a demand for a toolset/framework to develop desktop apps.

Walter Rumsby
Wednesday, December 03, 2003

Java not usable for building client-side applications???  Are you kidding me?  What are you calling "real" then?

I think that the oft-noticed Microsoft attitude Dennis Forbes mentioned may have more than a little backlash at this point.  We started feeling a bit of that at our site when our Access 97 in-house apps suddenly broke when we got new PCs with Win 2000 (and thus, Access 2000); we had an abysmal time getting Access 97 re-installed along with 2000, and some things never did work completely again; so we invested some time in learning PHP and some light javascripting; the end result was that our users noticed a lot less frustration, and smoother operation, and we can run PHP on our webserver til the cows come home, no matter about changes in data types in MS Access.

We don't do a lot of VB development, but I heard that those on our site who do, when discussing the future and how to bring .NET in, voted unanimously that it was time to move to java.

Lesson learned: keep making people jump through hoops, and at some point they'll get tired of jumping.  I think .NET will gain some decent ground, but that many will also take this .NET supersedence as an opportunity to give other avenues a try.  And I think a lot will be pleased with what they find.

Final prediction: .NET will end up with a smaller percentage of the pie than VB had, but still a big chunk.

van pelt
Wednesday, December 03, 2003

> Lesson learned: keep making people jump through hoops, and at some point they'll get tired of jumping.

This is a frequent comment, and there's some validity to it, but I think it's derived primarily from the grass is greener syndrome.

Jumping through hoops is the price you pay for innovation.  If Java's going to stay the same from here on out, good luck.  I for one won't be buying any Sun stock.  If, on the other hand, they choose to compete with .NET or technology x, they're going to release new bits that will require hoop jumping panache from their consumers.

On the other end of the spectrum you've got the folks who complain about how browser technology has utterly stagnated since MS won the war.  I, for one, am happy that we've been able to get some real work done for a change, rather than endlessly spinning our wheels to support 500 different revisions of browsers from 5 different shops, but that's another issue.

My point is, there's an optimum middle ground here, between providing cool new features and providing enough stability that development dollars go toward end-user value, not just keeping up.  The vendor who best strikes this balance will win.

My opinion: it's only a matter of time.  Sell your Sun stock now.  It's not getting any better. 

Ian Olsen
Wednesday, December 03, 2003

You must consider that both Windows XP SP1 and Windows 2000 SP4 contained the framework.  Penetration is already at a significant level.

Many programmers wonder why MS wants to install this bulky framework, when you can rather work in VB6, Java, or C++.  What they fail to see is they look at it post-penetration, when Windows ships everything needed to run something in these languages.  Even C++ has common runtimes and MFC.  They only difference is they're native code.  .NET will reach this level of saturation.  We're just not there yet.

The .NET languages and framework are offering some great building blocks to write new applications.  You may find that many opponents are not opponents of .NET, but really of Microsoft, and would use the tools whole-hog if they had a Sun sticker on them instead.

If you are worried about the use (or install) of .NET among your customers, roll out a small app where only a portion is actually .NET.  This leaves you the ability of getting out if anything adverse happens.

I believe that .NET will succeed, and studies are showing a good number of new projects are being done in it.  No new language or framework is worth a re-write of existing code, except if the original is really junk.

We have been worried about the same things.  We are on the verge of releasing a new product that will help us guage our customer's reaction.  It has a .NET gui on the top, and C++ underneath.  If .NET doesn't work out, we just change the top gui without ruining the important stuff below.

If we see no problems, we have a larger scale project using .NET down the road early next year.  Again, if successful, we plan to shift some significant development in that area.

Walt
Wednesday, December 03, 2003

Does 2000 SP4 really contain the framework? That's the first I heard of it.

inex
Wednesday, December 03, 2003

I just installed SP4 the other day, went back to Windows Update and THEN installed the .NET framework.

Walter Rumsby
Wednesday, December 03, 2003

"Jumping through hoops is the price you pay for innovation. "

Here's the problem though -- people are willing to jump through the hoop with you if they believe in what you're doing, but if you're pursuing change for the sake of change, or significant change for marginal gain, a lot of people are going to question whether it's progress, or simply a make work project.

This is not stating that .NET isn't worthwhile (it is absolutely worthwhile, particularly for web applications/web services), but if you're a Visual Basic/Delphi programming with a wide assortment of stable, powerful components, and you're generating pure, efficient Wintel code, what exactly is .NET offering to you? A lot of people haven't been sold on the hand waving and smoke machines that have been used to sell .NET (because almost all of the advantages, apart from actually being cross-platform*, have been around for years in Java -- if they didn't hop on the Java train, then why are the same advantages from Microsoft worth jumping now?)

One of the most interesting thing in the computing world remains the fact that the majority of the world's most important applications are "Cross-platform" C++.

Dennis Forbes
Wednesday, December 03, 2003

Most of the comments here focus on client-side stuff.

We are a Windows DNA shop that focuses on server-side stuff.  Moving to .NET is just good business sense. 

All the graduates coming out of school will be familiar with the new technology, and if you don't move your company will be less attractive. 

The inertia moving to .NET is great, and the productivity benefits are very real, especially in traditional line of business applications where "performance" 99% of the time means "database performance" and _almost_ nothing else.

Rick Watson
Wednesday, December 03, 2003

"Client side development has gone the way of the dodo."

Try telling that to the huge multinationals. From what I can guage most online applications are only used for selling books and sunglasses etc, anything that involves REAL money is nearly always client server.

Realist
Wednesday, December 03, 2003


"Does 2000 SP4 really contain the framework? That's the first I heard of it."

No, but it did make /pae work as advertised since the day Windows 2000 debuted.  Not bad 3 years to get what was promised originally.

Mike
Wednesday, December 03, 2003

Well,  I work for a bank,  and all our new POS systems are either:

1) Web applications written in java
2) Java/swing applications talking XML/SOAP to Servlets written in Java.


We're multi-national, and spending big money.  So while it's not true that there'll be no more client side apps, Office 2007 will not be written in (D)HTML.  There's a huge market for code that isn't tied to the specific version of windows you're using.

It's only a small step from there to code that isn't tied to windows at all.

Michael Koziarski
Wednesday, December 03, 2003

"We are a Windows DNA shop that focuses on server-side stuff.  Moving to .NET is just good business sense. "

Indeed, this is very true, but virtually all of the same benefits apply to alternative platforms as well. Indeed when previously I talked about companies reassessing (I have consulted on several such decision processes...and contrary to what you might think I am usually the "pro-.net" voice), I was speaking specifically of the server side: With the "protocols", be they SOAP, HTML, whatever, standardizing, suddenly when your client apps are talking a standard protocol (instead of say DCOM) and you're no longer coupled, and suddenly you have to do a big shift anyways, a _lot_ of companies have made the shift to Java (or other platforms). This is simple reality. These are organizations that were in Microsoft's pocket, but they suddenly felt the seismic shift and after analysis chose Java.

"All the graduates coming out of school will be familiar with the new technology, and if you don't move your company will be less attractive. "

Very few schools teach .NET. In any case, it's ironic again because the majority of professional software developers (and an overwhelming majority of open source developers) work on C/C++ (grab a copy of Dr. Dobbs -- they do a job survey), often on standard libraries and POSIX. This idea that everyone is clamoring for .NET is absolutely proposterous.

"The inertia moving to .NET is great, and the productivity benefits are very real, especially in traditional line of business applications where "performance" 99% of the time means "database performance" and _almost_ nothing else."

Given that Microsoft is behind it, the inertia behind .NET could best be called "a subtle gravitational pull". There is, quite literally, no one who would call .NET a stunning success at this point, or even really a success. What success it has achieved is simply _given_ success -- ASP shops naturally will migrate to ASP.NET. Etc.

Dennis Forbes
Wednesday, December 03, 2003

It already has significant market but I wouldn't assume that it will dominate any platform or type of application either now or in the future.

There's this feeling that if Microsoft develop something and bring it to market that it will succeed, that its bound to be bad and that its bound to make life worse and not better.

Microsoft products don't always succeed, you could easily argue that all of the network platforms up to this point delivered by them have failed,  DCOM, DNA, Hailstorm, etc, etc.

.NET in itself isn't a bad framework, but its not the only way to do things and its not the best delivery model for every kind of distributor.  Personally, I find the apps I have used with .NET to be clumsy and slower than they need be, a little better than a Java Swing app, but not by much.

Simon Lucy
Thursday, December 04, 2003

800 million? 10 percent of the whole population of earth?
gee, that hurts

na
Thursday, December 04, 2003

I personally have not heard of anyone going .NET for a trial project that is suitable for a 1.0 technology and not liking it. In fact, most are raving about it.
Is it ready for the realy big stuff? It is getting there, but so far I would say no. There is still too much that has to go through interop. I looks like by the time 2.0 is ironed out it will be ready to go.

I don't get the point about there being abundant .Net trained people rolling out of university soon. I our part of the woods at least university CS departments are "Anything But Microsoft" indoctrination camps.

Just me (Sir to you)
Thursday, December 04, 2003

Have 10 years of VC++/COM desktop development under my belt. When we needed to develop three new products from scratch I decided to go with .NET/C# even through we had lots of COM components. No one in our team (6 guys) had the .NET stuff installed on their machines, ever. Bought books...started a little slowly...After 6 months deployed three industrial products that among other things include 100 deployed Palm Pilots talking Web services to 2 ASP.NET/MS SQL servers.

Love the platform and love the fact that there is one set of tools that covers pretty much everything. Love the robustness and component orientation of .NET. Love that it is compatible with COM. And I KNOW that it is applicable to real life projects and that the learning curve is not really THAT tough.

Serge
Thursday, December 04, 2003

To Getting tired of crap,

No, I am not the person who wrote that outsourcing article. In fact, I mentioned in my post that I pretty much disagreed with the author's beliefs/opinions.  I posted the article link here on JOS because I happen to like format in which the article was written (i.e. that forum is setup so that you can interact with an author and not worry about your post falling off the board).

I rarely post in this forum anymore simply because this place wasn't designed to be a generic discussion forum. Joel simply allows people to post NON Joel on Software article discussion topics here.

Getting tired of crap wrote, "And now you're going to do one on .NET and, as with the last one, are getting thinkers here to give you ideas which you a) don't acknowledge b) ignore if they disagree with your own point of view?"

Nope. Where in my post did you notice me asking for ideas or advice? My post is what it is.  Now having said this, I (like a few others JOS posters here) have spent a considerable amount of my free time reading and playing with C#, VB.NET, ASP.NET, ADO.NET, etc. and I am currently uncertain how deep down the rabbit hole I should continue to travel.  For me learning new technologies/programming languages is a ROI issue. My free time is very precious to me, so if there isn't a good chance that sufficient development work is going to be available in this particular area...

One Programmer's Opinion
Thursday, December 04, 2003

"I rarely post in this forum anymore simply because this place wasn't designed to be a generic discussion forum"

BIG FLIPPING DEAL!

Utopia is a place where no moron exists
Tuesday, December 09, 2003

*  Recent Topics

*  Fog Creek Home