Slashdot Mirror
What is .NET?
CyberBry writes "There's a great technical overview of Microsoft .NET over at arstechnica: "In a remarkable feat of journalistic sleight-of-hand, thousands of column inches in many "reputable" on-line publications have talked at length about .NET whilst remaining largely ignorant of its nature, purpose, and implementation. Ask what .NET is, and you'll receive a wide range of answers, few of them accurate, all of them conflicting. Confusion amongst the press is rampant. The more common claims made of .NET are that it's a Java rip-off, or that it's subscription software. The truth is somewhat different.""
The main bone of contention that people have with .NET is the whole Hailstorm/Passport crap that MS is trying to push on us.
.NET itself is a very cool idea wherein any language can be used to write components that can be used by any other language. It's a means of allowing greater interaction between programs.
Hailstorm/Passport is an ill-devised way of online information management. With the amount of paranoia about this kind of stuff, the idea will be either flounder for a while or will be pushed as hard as possible. I think the former, but that's just me.
.NET is Microsoft's Web Services answer plain and simple, the problem is not too many people understand what Web Services entails.
.NET vs J2EE comparison is the fact that .NET was designed with web services in mind while J2EE has been playing catch up with it various Web Services extensions. I disagree with this view though. I look at it more as J2EE is more mature and has been around longer so that is why web services were not designed into from the begining. Second, I think it shows the power of J2EE that it can be extended in such a manner that works well with the existing platform. J2EE also has a much larger support for other types of protocols with existing enterprise systems. This allows systems to ease into the J2EE platform whereas with .NET you have to make the big leap and move everything into the .NET frame work leaving legacy systems.
One of black marks that J2EE gets in the
Yes but every time I try to see it your way, I get a headache.
When your friends ask, just tell them "It's a language-neutral Java knock-off..."
Why do people try to make it more complicated? Ok,
Others like to confuse the application that can be written by
The Platform != It's Applications
It's Simple: It's a Java rip off!
"Communism is like having one [local] phone company " - Lenny Bruce
Programming Windoze for a living, I couldn't care less about language interoperability and all that hype. :)
It's neat. But that's it.
In my experience, language skills comes a distant second to knowing your OS.
What I really hope for are quirk-free class libraries, and bugfree APIs. I'd have to find a new job then, of course...
No sig to see here. Move along.
Stability before performance, every time.
Or he'd rather be writing, "The JIT produces fast code, but sometimes crashes."? Or, ".NET is vaporware, still three to five years on the horizon."?
The reviewer should recognize and applaud the focus of the developers. Because you know they were sitting around saying, "Wouldn't it be nice if we did this fancy optimization...". Instead, they put first things first.
"Premature optimization is the root of all evil," D.E. Knuth. Learn it. Live it.
Anyone who wants to develop for .NET needs to shell out at least $1,079 for Visual Studio. That's $1,079 more than it takes to develop in Java, and $1,079 too much.
How can Microsoft afford to shut out all the developers that don't have big corporations backing them? Why not at least give away the compiler and class library?
You can't believe it? It's a natural part of their onslaught on the "viral" GPL. And it makes a Mac version just moments away.
Yours Sincerely, Michael.
I'm excited about
And really, Microsoft.com is the only one that could manage to make this a reality. As much as I hate the company, I can't help but feel grateful that I'll finally be able to write apps in a nice high-level type safe garbage collected language and have that be the most well-supported method. (And if others start using high-level languages, maybe my computer will not crash so much, or have so many buffer overflow sercurity holes.)
(As an aside... I fucking hate when people (like the author of this otherwise good article) use the word 'whilst'. Just say 'while'. It's not like we live in Medieval Britain.)
Which goes to show that "language-neutrality" is a myth. The CLR can't even support VB without changing it into a C#-work-alike.
Could be they're being untruthful. Wouldn't be the first time. Although...
Maybe it is portable. Maybe it *IS* real easy to port. Maybe they already have it running on FreeBSD, linux, Solaris, and Plan9.
Doesn't mean they're release it for any platform other than windows...
The author is confused about JIT compilers and virtual machines. A virtual machine is just a piece of software that executes programs in some machine-code-like language. It does not necessarily interpret the instructions. Indeed, almost all virtual machines compile to the local machine's instruction set at some point.
A JIT compiler is a technique used in virtual machine design to speed execution. Technically, a JIT compiler ought to compile code as it reaches it on the execution path for the first time, but thanks to some sly work by Symantec, its become acceptable to call something a JIT compiler even if it actually compiles all the code at load time regardless of whether it is executed or not. Hence the complaints about Java's startup time. Microsoft's efforts in this direction seem faintly bizarre to me. All previous evidence is that keeping compiled code around between runs is not worth it. However, I suppose since they only really support one platform, it won't really cause any problems.
C# is very much not "C++ for rapid application development". It's a completely different language, much more closely related to Java. While C# and Java share C++'s syntax, their underlying semantics are more closely related to Beta or Smalltalk.
As I would expect from someone who obviously doesn't know much about VM or language design, the author also makes far too much of the CL?'s cross-language abilities. While it has good support for implementing functional languages, as far as the much more important OO features are concerned, it is only going to work well for statically typed, single inheritance, single dispatch languages that don't need to do any code generation. Its is my contention that any OO language that can be implemented on the CLR can be implemented equally well on teh JVM.
You can write in "whatever language you want" as long as the language has been designed (or re-designed) to work with the CLR. You don't use VB with the CLR, you use VB.NET. You don't use C++, you use a bastard hybrid Managed-C++. You don't use Smalltalk or Scheme or Perl, you use something that's kind of like Smalltalk or Scheme or Perl. Not that big an advance, if you ask me.
Standard ECMAScript manipulating the DOM in standard ways works pretty much identically (except for implementation bugs) all over the place. This requires version 5.5 of MS's engine or Mozilla (maybe others, don't know, don't care), but it works well.
Browser detection code exists primarily because of people using version 4.x browsers
IMO, this is a demonstration of the standard working. The biggest problem is that it requires people to stop using the fatally flawed implementations (IE 4.x, NS 4.x). If they could be *forced* to upgrade, no-one would have to do that kind of crap any more.
The progress apparent in Java had mostly already been made by 1980, around the same time C++ was being created. The problems with C++ are a result of either ignorance or a deliberate decision to ignore those advances.
Funny, I don't remember the x86 opcode for tail recursion. Or the 68K, SPARC, or PPC one. Would you care to remind me what it is? I guess since there isn't one, it's impossible to run Lisp on Windows, Mac, Sun, or IBM boxes.
Every computer language in existence is Turing-complete; they product code that runs on Turing machines. Some of these Turing machines may be faster or more efficient, but they're all equivalent.
-jon
Remember Amalek.
I was surprised how often the main article preached .NETs language neutrality. .NET is certainly not language neutral. Heck, Microsoft had to neuter their own primary development languages in order to get them into it.
.NET. I mean, you can't be making the wrong decision if it supports all languages equally, right?
I can understand why they would restrict their framework to a single inheritance, single dispatch in order to be easily used from more languages, but they forbid both multiple inheritance and multiple dispatch in the virtual machine.
I suspect we are seeing the "language-neutral" lie pushed so heavily right now in order to convince people to choose
I'm sure a lot of us slashdotters have queried what this dot-net thing was. Anyone with a Hotmail account was aquainted with the chintzy logo in the upper-right corner. eBay is starting to suggest a 'Passport login', which so far has been the most that em-cash has seemed to provide. I've been to trade shows, Microsoft (free hardware swag for false info? that's a steal!)and non-Microsoft pitches and antipitches, and no one has provided an answer.
;)
:)
At least, a good one.
This overview is great from a technical angle (the one me and most slashdotters usually have interest in) and decent from a more mundane perspective (the one you pose as in an em-cash or other sales-derived presentation
It's hazardous to your health as a hacker. It looks like a great way to encapsulate any data in a format which is sufficiently protected under the DMCA. (Yet another reason for that law- and the 99 senators who ayed the vote- to be burned at the stake.)
It's also bad for anyone on a non-MS platform; two of the languages are extremely MS-centric, Visual Basic and C-hash (something that should only be done right before you smoke it).
It's bad - all right, worse - for Java fiends. Bad enough Microsoft feels Java is the worst thing to happen to it since the Wicked Witch of the West was introduced to the business end of a water gone, now they're pulling out all the stops with the theoretically embraced and extended J-hash.
Right now, I just wish there was a way to stop those pricks at Microsoft.. besides a HERF bomb in Redmond, WA
I used to be someone else. Now I'm someone better.
Real life is underrated.
For what its worth, Javascript has been effectively standardized for, oh, I'd say the last five years, even before the ECMA standard. (Note the word 'effectively', it's a critical one to my point.)
.NET?" What use, indeed? The greatest challenge facing those who would implement the CLR is not in writing the class libraries, it's matching them bug-for-bug. (One of the reasons Mozilla and IE still don't work identically is that Mozilla was forced to abandon its plans for bug-for-bug compatibility with IE, due to the feature's excessively high "pie in the sky" factor.)
What you're complaining about (justifiably) is the DOM, or Document Object Model. The DOM was standardized much more recently, and unfortunately contains a few holes large enough to drive a truck through, necessitating the need for non-standard extensions in practice. (One of the most-commonly-used of these is the "innerHTML" attributes, which is *not defined* in the standard, despite the fact that it is wonderfully useful. Mozilla actually explicitly added it many milestones ago because people were screaming for it. The 'standardized' way of doing that was upwards of 10-20 lines of rather difficult-to-read code, involving walking the tree and regenerating the HTML, then nearly-manually parsing the given HTML back into a tree, then swapping the newly-parsed Node tree into the document! Is anyone surprised nobody, even those who understood it, wanted to do that?)
The DOMs are inconsistent, partially because they're hard to get write. But Javascript itself is nearly unchanged since Netscape 3.0. That's not a typo. Yes, a few nice things have been added (for instance, I think an exception mechanism has been added since then), but effectively all of the language anyone uses on a web page script was there in Netscape 3.0. (How many people here have created their own objects in Javascript, or fiddled with the prototypes? IIRC, this feature was in 3.0, and it's still too-advanced to be necessary in most web scripts. Short scripts don't need a lot.)
This works out on topic nicely... because you're very likely looking at the future of Mono. "What use is Mono when the same code doesn't *quite* run on
Yeah, I should have caveated that by saying that it's not legal if you're *NOT* affiliated with a university. If you are student, faculty or staff, though, it's a great deal. I'm not sure what the restrictions are WRT producing commercially released code w/ it, though, so I'd look into that before buying.
I ordered my copy from Genesis. You have to send them proof that you work at or attend a university, but they seem legit.
Sweet jumping savior, how the hell did we get here?
I managed to get two pages into ArsTechnica's explaination of what .NET is, what it isn't, and why it will be used before my brain rebelled: "Great fuck, if people would simply stop schlepping shit around in proprietary binary formats, data could be imported or exported in any damn application that wanted to write the translation".
This would be different if .NET was talking about eliminating the concept of an application (document-centric computing). Hell no, it's a bunch of pointy-head nerds and pointy-haired MBAs adding another goddamn layer of nerd-cruft to everything under creation.
".NET has three packets of information: the IL, the Metadata, and the fuck-this-where-can-i-find-nudie-pics"... please, for the love of Mike (God wouldn't have anything to do with this, it's purely From the Other Side), let's not stop everything and reinvent the wheel. We had the chance to carefully think things through and do it right, back in 1990. We missed the opportunity, we're now stuck with what we've got. Hasn't anybody learned anything from Be? You can't go home again.
If .NET makes any fucking difference before it gets replaced with the Next Big Thing, I'll eat my damned crusty underwear. So far, I could grep for .NET and replace it with "Java", and timewarp my pasty white ass to 1997 when it was going to Save Us from platform-specific languages and Microsoft at the same time. I cut my balls off and drank the poison koolaid, but the fucking UFO hasn't poked it's nose out from behind Hale-Bopp yet, the shy fucking bastard.
Spare me the fucking story. You wanna know what the next great savior is gonna be? Sumbitch, he's already here, and Tim-Berners Lee is his prophet. It's the Church of the Holy Hypertext, and it's vessel is Mozilla. The Web lit up the world because it's simple, it's easy to learn, and it's powerful... and, sin of all sins, it's accessible. Nerds and secretaries are building web pages, because it's easy to do. You think Sally Secretary is gonna benefit from .NET's programming language independance? It's wonky new IDE?
Pfft. Thicker and thicker layers of cruft will not make programming less hard. Thicker and thicker layers of cruft will not change the way we access information. Thicker and thicker layers of cruft will only slow down the spread of knowledge, because everybody's chasing down the next security bug in .fuckingNET instead of sharing what they know.
Potato chips are a by-yourself food.
No, I disagree.
.Net. Back in 1996 when Sun was promoting net development and JITs and the whole concept of a better programming language, Microsoft was still knee-deep in C++ and COM (or was it OLE then?). Six years later, Microsoft looks at the progress Sun has made, copies the most successful parts and bastardizes the rest (again).
Microsoft ripped off Sun the same way it ripped off Apple (who ripped off Xerox) when designing Windows. Yeah, you can say they helped develop the first apps for Mac, etc. etc. but the main idea was still Apple's who pushed the GUI concept and standardized the idea and then Microsoft came along six years later and launched Windows 3.0 based on those concepts developed at Apple.
Now Microsoft is doing the same with
Yes they added some good ideas - just like they did with Windows vs. Mac - but they still ripped off the general concepts and tech gains that were made from Java.
Sorry, a ripoff's a ripoff.
-Russ
Me
One small thing - generally, it's not the developers who pick what they write. It's management, who do it on a basis of what looks popular, what their customers (also upper management) will want, what fits their budget and finally, what fits their strategic partnerships best.
:-(
The bad news is that technology and "what's best" never comes into it.
Sad, but true. C# and VB.NET are so close to isomorphic that choosing between them is mostly a matter of whether you prefer symbolic or "natural language" syntax. Notice the number of long-standing VB developers who are trying to work out the relationship between the tool they've been using until version 6, and the new .NET version.
In fact, from the article itself:
I think the key thing is that .NET really only supports one paradigm properly: single-inheritance OO. C# fits that description, VB.NET has been moulded to match it, managed C++ is forced into it. You get the idea. I know it's theoretically possible to use other programming paradigms from this foundation, but surely the question is how well they are supported and how efficiently they can be done, not what could be done with infinite time and resources available. (Insert obligatory reference to the thread about functional languages on .NET the other day here.)
Consider an obvious example: if .NET is reasonably language-neutral, where is the support for generics? C++ has had templates for years, and they are one of its most powerful features. Java has a proposal that doesn't go as far as C++ templates, but does add parameterised type support to a fair degree. (Anyone know if that made it into 1.4 in the end, BTW?) In ML, functions are implicitly generic unless you specify otherwise. If .NET doesn't support such a fundamental feature, then it's immediately dropped an important aspect for all these languages.
Of course, how important the omissions are depends on your programming style. If you don't use generics, then this particular example is no loss to you. But it should be noted that the current trends in programming-language research are considerably ahead of single-inheritance "pure" OO designs. If .NET can't cope with multiple paradigms and newly developed idioms, it's not going to be leading edge for very long. The power of generic programming, functional programming and other completely separate idioms is being exploited in research already, and has been for some time. I don't think it will be long before they start hitting the maintsream, and then the limitations of .NET's architecture may be its undoing.
If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.