Fog Creek Software
Discussion Board

.net is...

".NET is a feature-filled services platform for building web-based applications and developing rich interactive experiences for users and their systems. The small-w "web" includes both the Internet and intranets. " - Steven B. Levy, Microsoft Technet Editor.

Does that actually help anyone know what .net is? I'm interested in the idea of rich interactive experiences for users *and their systems*. I know my computer was begging me the other day for a rich interactive experience.

But what happens when you try to boot your machine to run Word or CityDesk(!) or Quake one day and it says "Go away, I'm too busy having a rich interactive experience, no login dialogue for you!"

[Quote taken from ]

Robert Moir
Thursday, January 17, 2002

".NET is a feature-filled services platform for building web-based applications and developing rich interactive experiences for users and their systems."

I guess he smoked one joint too many.

But seriously, I like to publicly thank Microsoft for sending so many competitors on a wild goose chase for the holy grail called .NET. It's a bit like Joey described in his Fire and Motion story. Only it's Microsoft providing the cover.

Jan Derk
Thursday, January 17, 2002

Exactly... I loved Joel's horoscope analogy from the "architecture astronaut" articles. Like a horoscope, this stuff is so vague - you can read anything into it.

In fact it's incredibly unhelpful. I teach .NET Framework development (C# etc) and I get intelligent developers saying things like "So I hear it'll cost me $10,000 for the licence for these .NET apps that I'm building"... er, no sorry that licence fee is for integrating with the .NET My Services .NET Passport stuff...

Duncan Smart
Thursday, January 17, 2002

I get the impression that it also helps having a nebulous "stretch goal" for the company.  Vagueness can be interestingly powerful.

(BTW, that term comes from GE.  This article might be interesting to readers of this forum: )

Their plans are a wee bit ambitious, after all.  Controlling the money that flows through the internet is control over the internet.  I assume that there is no direct series of steps to do this; instead it takes all the creativity from the organization.

Mandy Lindahl
Thursday, January 17, 2002

"I get the impression that it also helps having a nebulous "stretch goal" for the company. Vagueness can be interestingly powerful."

Yep. But the quote I posted was from someone in a technical position, talking to technical people, and attempting to answer a question from a technical person on what .net is. If that was Bill Gates or Steve Ballmer giving that answer, we'd say "Yeah, marketing, strategic objective, ok" but as a "Technical definition for technical people" it just don't cut it.

Robert Moir
Friday, January 18, 2002

If its just an introductory remark its ok, even if it does sound a little like solving world poverty.

I'd expect more detail after that, considerably more argument as to why a new feature rich thingy (cos all the previous ones were), to do much the same as we already do (discounting passing all transactions through a Passport server that is).

But apart from yet another attempt to integrate all the MS development tools into a single family, which has failed anyway as Visual Foxpro couldn't be made to fit, I can't really see the point of it.

Simon Lucy
Friday, January 18, 2002

Well for a start, apparently, it's not just about development environments. I'm a sysadmin not a developer, and I can buy .net stuff too. All my microsoft servers are .net applications servers, or at least will be when they all have the latest stuff installed.

So it's a bunch of web services
It's a bunch of servers
It's a bunch of development environments.

hmmmmm. It's the holy grail isn't it?

Robert Moir
Friday, January 18, 2002

One of my favorite Joel articles is "Microsoft Goes Bonkers"  Joel articulates my thoughts better than I can!

This whole .NET phenomenon amazes me to no end.  Or does it?

At the risk of sounding elistist, I have always considered software developers to be "smarter than the average bear".  In recent years that seems to have changed as I've noticed a staggering increase in the number of developers who seem to champion the Microsoft cause regardless what it may be.

Case in point:  Delphi is superior to VB in all respects, yet VB is vastly more popular in development shops throughout the industry.  Somebody is making these decisions, but technical merit of the actual tool is obviously not one of the considerations.

(NOTE:  I am not interested in starting or participating in a religious war.  For the record I was trained in VB and have logged more hours with VB than any other tool in my arsenal [umm, except maybe Clipper, but that goes back too far].  I have hit the limits of VB more times than I can count and would have to resort to my C++ compiler to get around them.  A friend turned me on to Delphi after complaining to him over a pint.  Actually he issued a challenge to me, something from which I could not back down.  Eight months have passed, I now use Delphi exclusively, have never once had to haul out my C++ compiler, and wonder what the hell I was thinking for all those years.)

Back to my point.

The whole .NET thing seems par for the course these days.  Six years ago I would have been blown away by such mass acceptance of such a vague, intangible specification.  Today it seems the norm.

Imagine GM or Ford releasing an incredibly vague yet all-encompassing description of a gee-whiz-bang new automobile they're building without demonstrating a prototype.  But do the same thing with software, especially Microsoft(TM) software, and watch the industry bend over backwards to become .Net-compliant and obtain their pretty little certifications at $5,000 a pop.

I hate to think this represents a general dumbing-down of the IT industry, but I'm not sure what else it can be.  Thanks for starting the thread Robert, I'm glad to see that I'm not the only one who can't understand why the industry is eating this stuff up.

DB Cooper
Friday, January 18, 2002

> This whole .NET phenomenon amazes me to no end.

Mainly because MSFT has a pretty good success rate on large projects.  And what they're proposing is not exactly vaporware.  It's just not eloquently packaged.

I personally like the idea of the IL.  Not because of MSFT, but because of Mono's efforts.

Remember, the big alternative everyone thinks of is Java.  Java fulfills its design goals nicely, but isn't a great chair to be strapped in.

But you might retort, "But what about the nebulous rest of .NET?"  And my answer is, well, I haven't spent a single cent on .NET, so what do I care?  Of course I'm not entering into contracts where I rewrite huge systems in C#, or purchasing the .NET Defrobulator Sysadmin Special.  I just hope that you can run Lisp code at worst 1 order of magnitude slower under the IL.  Tail-recursion & continuations intact.

Greg Munoz
Friday, January 18, 2002

D B Cooper,

Thanks for pointing me back to the 'Bonkers' article.  I've just read the postscript and insider response for the first time.

Spot on, as usual.

Ged Byrne
Friday, January 18, 2002

Ever hear the story of the blind men and the elephant? Think about that the next time you're tempted to start a sentence with ".NET is..." Microsoft is in the process of applying the .NET moniker to everything they're doing that's new, whether it has anything to do with web services or the CLR or not. They've done this before, notably with ActiveX. From the marketing perspective, .NET is the new branding initiative at Microsoft; from the technical perspective, it will be a year or two before we see where the name sticks.

Mike Gunderloy
Friday, January 18, 2002

I know what .net is.  It's Java.

It's not quite the same technically.  All programmers care about are technical issues, but that's like not seeing the forest for the trees.  The truth is that no one really knew what Java was 6 years ago either, because it was a vague revolutionary movement made up of multiple different parts.

And now we come to .net.  How many people even today know what Java is?  It's a language aimed at average programmers, a huge API, and three platforms (micro, standard, enterprise).  Somewhere along the way it was also supposed to be JINI (to connect the toaster to the alarm clock), and a raft of other things.

Now, .net is supposed to connect stuff up too.  Not exactly cross-platform, but cross-language.  A host of vague technologies looking for consumers to adopt them.  The ones that aren't adopted will die, or at least reincarnated.  Microsoft will hopefully learn from Sun, so it will do documentation a bit earlier, bake security in immediately, and grab all the money it can proactively.

None of it is new.  But Microsoft has never been about innovation, just good business sense.

Maybe this is the age where programmers are the real consuming lusers.  Why else is Ballmer letting loose the marketers on them, unless they have become consumers?

Red Davis
Friday, January 18, 2002

.Net isn't really anything new and is composed of many ideas taken from other languages and architectures, and adds a few new twists.  But you can say that for any other language also, .... including Java.

The thing that makes .Net really useful is that it makes programming with SOAP, COM+, etc. a whole lot easier, since the framework takes care of most of the details for the developer.  The .NET framework is gigantic and offers almost everything under the sun in regards to the regular plumbing that you would have had to do if you were to use SOAP, COM+, MSMQ, etc. via C++, VB, etc.  Programmer productivity is the real selling point with .NET.  This was also the reason Java became very popular, due to its extensive class library which allowed developers to create applications faster by not having everyone write their own hash table class, etc.

If .NET was just another language with a few conviences then it wouldn't have taken off as it did.  Just look at COM.  COM was a great idea (albeit not original).  However, their vision of interoperability between disparate systems didn't succeed due to many of COM's shortcomings; but the main shortcoming was that it is extremely difficult to understand and use.  You could easily spends six months or so in a fog trying to figure it out.

.Net doesn't really offer any advantages in terms of a more user rich experience IMHO, and only benefits the developer in terms of productivity.  The only resulting benefit to end users is that apps get produced much faster.

Pros Chum
Saturday, January 19, 2002

NET is not a language - it is really a number of not always related technologies.
From language point of view .NET is not a Java. This is an evolution of development environment - next generation of IDE.
It is better than Delphi in its visual component framework, it have nice system API and system libraries. The only thing I have problem with is a runtime environment. Hopefully it will not be a problem for us.

Roman Eremin
Saturday, January 19, 2002

Irrespective of .NET' s success or failure, it is going to be everywhere and with a host of features.  This will kill a lot of competion, and Microsoft willl be the "last man standing".

eg: Netscape vs IE for one.

Prakash S
Saturday, January 19, 2002

It may further polarise what is already a polarised arena but I don't think its going to make the competitive market any worse.  Microsoft's current standing with corporate buyers is quite low, conversion rates of Office are reportedly lower than for previous versions and its still too early to say how many will convert to XP.

Given the continuing public failings with its security (which is not to say that any other environment is that much better), Microsoft is going to have a tough case convincing e-retail vendors that the benefits will more than cover the costs to be involved.  It isn't a given either that just because Passport is almost a 100% sign up for every home user with XP that consumers will trust it for purchasing either.

Competitive vendors such as Visa with its virtual credit card (which I have prior art for so they'd better not try and patent it :-)), and other financial institutions managing e-commerce transactions have an easier way to market their solutions to both the consumer and the trader.

Simon Lucy
Sunday, January 20, 2002

".net is yet another productivity tool". I've been around for so many of them that all I need to do is think of an application now, and its completely written before I finish thinking about it. With .net I'm hoping that it will be finished before I even start to think about it.

Monday, January 21, 2002

I have played around with VisualStudio.NET for a while now and I am very impressed for a number of reasons. At first I was impressed by the sheer amount of resources it eats up, but that's a different story. }:)

Seems to me that Microsoft is finally getting it. It has "abandoned" its development tools for about 3 years now. And it hasn't done any serious updates to the Visual Studio product line, apart from service packs. Development tools from Microsoft are beginning to show its age. Both as a development environment and as a set of core technologies and APIs. I think we all know how awful OLE and the MFCs are. And I think we all know that it could have been done a lot better, back in the day they were released.

What impresses me about .NET is not Visual Studio, in fact although packed with neat features, I could live without it. It's the fact that they have revamped windows development, created a "clean" fully object oriented framework around it, seamlessly integrated technologies that previously required developers to go through steep learning curves with separate APIs. I mean, the fact that in order to open up a window using plain win32 you barely need about 70 lines of code, unless you resort to higher level APIs probably gives you a hint of what I am talking about.

.NET isn't Java (the language) because .NET is language agnostic. And .NET is not Java (the JVM) although it borrows some ideas from it, but so did Java from Smalltalk. I think that comparing .NET to J2EE (the platform) is incorrect although not entirely misleading. There's more than meets the eye in .NET and in some ways it is superior to J2EE, specially when doing native Windows development.

.NET is a blessing for people whose core market is native windows development, it simplifies development so much that it would be silly not to adopt it. But it's just "an alternative" when doing web related work, where J2EE still has a few things to say.

I am no Microsoft fanboy, but I think that any developer that doesn't consider to learn .NET is bound to become outdated real quick.

Honestly, although I like .NET I don't see it as a huge leap forward. It basically has very little that can be considered new. They are not inventing a completely new thing. Microsoft has seen the recent trends in software development, and did what it does best. Pick everything up, integrate it nicely so it looks like one technology although is a set of things we are already familiar with and they have made it relatively easy to migrate to. And probably as a side effect, they have simplified windows native development by a large amount.

There's one thing I dislike though. And this thing only can make migration a bit difficult for some. It is the difficulty of making managed code play well together with unmanaged code. I understand that it's a very tough engineering problem to overcome. But I also think it could have been done more intuitive. And in my opinion it could be the source to many of those bugs that are difficult to pin down. I think that many software developers will spend quite some time reading raw IL (generated by managed languages), to figure out things such as parameter passing to unmanaged code. Untill, of course, everyone moves to CLR managed code.

Beka Pantone
Tuesday, January 22, 2002

*  Recent Topics

*  Fog Creek Home