Submitted by “One of the things my team has been working to enable has been the ability for .NET developers to download and browse the source code of the .NET Framework libraries, and to easily enable debugging support in them. Today I’m excited to announce that we’ll be providing this with the .NET 3.5 and VS 2008 release later this year. We’ll begin by offering the source code (with source file comments included) for the .NET Base Class Libraries, ASP.NET, Windows Forms, ADO.NET, XML, and WPF. We’ll then be adding more libraries in the months ahead (including WCF, Workflow, and LINQ). The source code will be released under the Microsoft Reference License.”
MS To Release Source Code to .NET Framework Libraries
121 Comments
-
2007-10-04 1:28 pm
-
2007-10-04 5:48 pm
2007-10-03 8:29 pm
-
2007-10-03 8:54 pm
-
-
2007-10-04 3:42 pm
-
2007-10-04 4:59 pm
-
2007-10-04 5:14 pm
-
2007-10-05 10:36 am
-
-
2007-10-03 8:31 pm
-
2007-10-03 8:51 pm
-
-
2007-10-03 9:22 pm
MollyCI agree with you. Generally I like to think of libraries as a “black box”. And most Windows and .NET programmers do program that way.
But sometimes you’ll run into a problem that isn’t explained by the documentation. “Why the hell am I getting this exception??” Documentation never covers everything.
(Nor would you want it to, as it would unnecessarily bloat the documentation as well as tie it to the particular implementation details of the API, details that may change in future versions of the library.)And even with the source available, one should not write code that depends on the exact source code since the underlying source can change. But programmers can use their own judgement regarding such matters.
-
2007-10-03 10:33 pm
-
MollyC
Actually, stuff like System.Collections, System.IO, etc has had the source available for awhile now through Rotor.
http://www.microsoft.com/downloads/details.aspx?FamilyId=8C09FD61-3…
But this new offering provides the code to a lot of the stuff on top of that lower-level stuff, as well as easily stepping through the .NET code in VS2008. I’m particularly interested in WinForms, WPF, and LINQ.
(I *was* hoping that this would include System.Drawing (it doesn’t appear to), but now that I think of it the System.Drawing code is likely just a bunch of wrapper functions that do little more than call the corresponding GDI+ function, so the source would be of little value there.)
As side note, Microsoft has released code for the purposes of helping developers code and debug before, such as ATL and MFC (and of course, their CRT code).
Edited 2007-10-03 21:26
-
2007-10-03 9:31 pm
2007-10-03 8:59 pm
-
-
-
bobjohnsonmilw
Ok, you’re right that’s portability; but not provided by microsoft is what i meant to say I guess.
Portability in my opinion is of officially supported and released msil interpreters that would support the “write once – run anywhere” paradigm.
So technically, you’re right
But you’re wrong in assuming that small changes are required for true portability, imho. -
-
-
2007-10-03 9:11 pm
2007-10-03 9:27 pm
-
2007-10-03 9:33 pm
-
2007-10-03 9:47 pm
-
-
-
-
-
-
-
PlatformAgnostic
That’s at least graduate-level business. You first have to have the trust before you get into the joyous world of frequently changing rules and power-hungry eurocrats. But this is OT
. -
-
MollyC
“You first have to have the trust before you get into the joyous world of frequently changing rules and power-hungry eurocrats.”
lol @ “power-hungry eurocrats”
I like that.
I hope you don’t mind if I use it in the future. :p -
-
-
-
-
2007-10-04 1:54 pm
-
2007-10-04 8:12 pm
2007-10-03 9:47 pm
-
2007-10-04 7:31 am
2007-10-04 12:29 am
-
-
-
2007-10-04 12:28 pm
-
-
2007-10-04 1:53 pm
-
-
-
-
-
-
-
-
-
-
-
-
-
-
2007-10-04 1:24 pm
2007-10-04 1:37 pm
2007-10-04 2:10 pm
-
MollyC
Your “Another viewpoint” is yet another article with no credibility. The reason I say that is that it starts right off the bat with the insinuation that Microsoft is pretending that this code released under MS-RL is being released under an “open source” license in order trick devs into believing that, when Microsoft has NEVER claimed that this was “open source” code, and has NEVER claimed that MS-RL is an “open source” license.
Second, the article cites Microsoft Hater Steven J. Vaughan-Nichols’ article, so it’s hardly “another viewpoint”, more like one half of “another viewpoint”.
Microsoft is not laying any “trap”. They make no claims that this is “open source”. They make no pretense that this is “open source”. Those that accuse them of such, either explictly or implicitly, have ZERO proof.
Anyone that is working on an alternative implementation of .NET or a .NET-like thing should be leary of looking at this MS-RL code, just as one should be leary of looking at ANY code that is not licensed under terms that allow said code to be used in another (particularly competing) project, but that doesn’t mean that the party that released the code is laying a “trap”. Those that would fall into such a “trap” by using code in an unauthorized manner have only themselves to blame, not those that released the code in the first place.
Note that Mono is not allowing any of its contributors to look at this code, because Microsoft makes it clear that the code is NOT open-source.
Edited 2007-10-04 23:59
-
-
-
-
2007-10-04 3:46 pm
2007-10-05 1:14 pm
2007-10-06 12:30 am
WOW
Microsoft release WHAT???
Make sure you read the fine print on the license.
This is not OSS. It’s a “read-only” license.
One of the more prominent theories circulating around goes something like this:
* MS gives read-only access to .NET source code.
* Developers review it. (Maybe mono developers?)
* Lawsuit time. Evidence shows that you looked at our code. MS kills mono or [insert_app_here].
Yeah. In their dreams, maybe… if you release code, it’s going to be used by someone else.
I predict, in one year’s time, several “competing” versions of .NET will be released, whose sole change will be to occasionally launch advertisements. They will have no technical advantage over vanilla .NET, but they will get broadly advertised, and will be widely adopted by the 50% of the Internet composed of the MySpace generation. These flawed copies will make MS scramble to explain why their cross-platform .NET programs don’t work on Imperfect .NET Implementation XYZ, damaging their reputation, and they’ll only have themselves to blame for releasing the code in the first place.
Edited 2007-10-03 21:24
They will have no technical advantage over vanilla .NET, but they will get broadly advertised, and will be widely adopted by the 50% of the Internet composed of the MySpace generation.
I’m trying to think of a concrete example of what you’re describing that has already happened in the *real world* today…
Considering the amount of open-source code already out there, I would expect this is to be a common problem (where 50% of the internet uses rip-off versions and thinks it is the real thing)…
Edited 2007-10-03 21:32
Which side of their reputation are you talking about?
The side that is known for keeping interoperability information under lock and key so they can continue to monopolies their position as the world’s no1 desktop?
The side that is known to invite partners to develop for their platform only to release a competing product and bundle it for free with their OS?
The side that bullies OEMs into offering systems with only their OS preinstalled?
The side that is known for trying to take over the internet with an inferior product, then providing operating systems without the security mechanisms needed to securely connect to said internet there by leaving the door open for scrip kiddies to run riot causing millions of dollars in downtime and damages?
How about the side that bankrolls other companies to threatens the only thing it can’t buy, steal from or muscle out of the IT industry, i.e. Linux, and when that fails, starts threatening with bogus IP violation charges it is unwilling (read unable) to provide any evidence for?
If MS is damaging it’s reputation, surely that’s a good thing?
How does this differ from any OSS project? Forks happen all the time. That’s the risk of releasing source code.
So what if you look at the code? In most parts of the world the only problem is copyright, and it’s not an issue unless you copy and paste the code. Or, to make absolutely sure, you use “clean room” development, where one developer looks at the original code, tells another what it does and the second developer writes new code from scratch, with the same functionality.
Software patents or reverse engineering are an issue only in some countries.
Yes, but all Microsoft have to do is make the allegations. They don’t have to back them up, or prove them conclusively. Microsoft can claim that such and such app is breaking the licence, and it’s up to the app writers to prove otherwise, not Microsoft. MS fires off some cease and desists to the ISP, and before you know it, Microsoft have spun an allegation into an anti-open source, anti non MS-vendor bit of PR.
There’s only so many ways you can make a simple function return an answer. Making other programmers have to tip toe around Microsoft’s specific implementation will slow down development and increase administration. Just look at the audit that had to be done on ReactOS because of Microsoft allegations.
At least spread correct information please. The audit wasn’t done because of MS allegations, but because of allegations of people from inside the project and from related projects. MS never ever approached ReactOS about it.
Yes they do. If you bring charges of copyright infringement you have to back them up. Of course you do.
I don’t see what ISP’s have to do with this. If you meant hosting companies, they usually check around. They don’t boot their paying customers for allegations alone.
ah, its “shared-source”
This “source release” isn’t anything like “open source” but perhaps it will help the Mono developers. That would be a good thing.
well microsoft coined the term “shared source”, and iirc this kind of look but dont touch was the basic idea for it…
wow, purist trolls have become masters in the art of speculation.
Microsoft’s track record means they are going to have to prove they don’t have those kinds of nefarious goals.
However, I think some good could actually come out of this. If there are any perceptible people in charge over there, they may start to notice better bug reports, maybe even receive a few high quality patches, and some of the other benefits that come from open source, and begin to really understand the advantages.
Again, they are going to have to demonstrate that they are not evil though. There’s just too much history.
I was immediately skeptical about this being some sort of submarine type license, but looks okay..
Looks like your standard “AS-IS” license with a clause of:
if you try to sue us over stuff found in this source code you lose rights to use it.
Nice short license.
was immediately skeptical about this being some sort of submarine type license, but looks okay..
From the FL
2. Grant of Rights
(A) Copyright Grant- Subject to the terms of this license, the Licensor grants you a non-transferable, non-exclusive, worldwide, royalty-free copyright license to reproduce the software for reference use.
“Reference use” means use of the software within your company as a reference, in read only form, for the sole purposes of debugging your products, maintaining your products, or enhancing the interoperability of your products with the software, and specifically excludes the right to distribute the software outside of your company.
Looks like a “submarine type of license” to me.
In what way is it a submarine type of license according to you?
It says what you’re allowed to do and what you’re not allowed to do in quite clear language.
The day Microsoft releases source code to anything of value is the day pigs fl…
*FLAP* *OINK* *FLAP* *OINK*
…well…whatta ya know?!?!
I support the idea behind the viewable source, but I strongly suspect it will be released under some kind of “read only” license, i.e. no compiling possible, etc.
Don’t forget that Microsoft already released parts of the Windows NT and CE to some business partners before.
But IF they would release .NET under some kind of open source license, it could make quite a boom in the Java vs .NET case. But Microsoft is not that smart
java watch out!
this is amazing though. thanks MS! (10$ says they lost a bet and this was the part of the bargin they had to held up) haha
Why? Java will actually be open source.
.Net – it’ll just be there without anyone touching it.
I think you are a bit behind the times, because Java is already free software under the GPL v2
. The source code has been released in the OpenJDK (http://openjdk.java.net) project with some minor binary stubs for those things that cannot be opened up.
The IcedTea project (http://icedtea.classpath.org) attempts to create replacements for these binary stubs with unencumbered free replacements and stubs. It will take some time before it’s complete.
.Net is just a fad that will eventually die out on free software platforms, as there is no reason to come anywhere near it unless you really only strive for Windows compatibility.
Or you could use .NET/Mono because it is likely to be a better runtime since it has the benefits of lessons learned from the Java experience. And having competition allows both platforms to evolve faster.
.Net is not any better than Java except in that it is tied to a single platform (Windows) and can therefore make certain optimizations that are otherwise not available.
It is a lot easier to design something for a single platform/API/processor than it is to try to do what Java does.
.Net has a bit better performance as a result, but it lacks the ability to move stuff between systems – yes, even with Mono which will never be as complete as Microsoft’s .Net implementation is. So your comment is quite a bit off the ball.
I’m not seeing how you get this idea. The .NET class library is tied to Windows, but there is more effort involved in porting the VM runtime between x86, x64, and IA-64 than to other OSes. The VM also runs on ARM.
The singular area in which I could imagine .NET having a performance advantage over Java from being tied to Windows is in exception handling. Java has to be able to deal with Signals from UNIX and SEH in Windows. But SEH isn’t exactly a performance winner and exceptions are… well, exceptional. The underlying OS is not very important to the performance of compiled code except insofar as it calls down to the OS for I/O and gets interrupted by the OS for scheduling and page faults.
Sure, Java has the explicit goal of being portable between OSes. This fact doesn’t prevent Java vendors from writing highly-optimized libraries and runtimes for specific OSes. And they do so. Even Microsoft wrote a JVM that was well-regarded for its performance. .NET is simply an incremental improvement with the lessons of Java in mind.
I’m not seeing how you get this idea. The .NET class library is tied to Windows, but there is more effort involved in porting the VM runtime between x86, x64, and IA-64 than to other OSes. The VM also runs on ARM.
However – all of those are still being written to Windows and the Windows APIs. Where as Java is being written to multiple APIs (POSIX, Windows API, etc.). This can make a big difference in how it performs and the optimizations in the code structure. For example – Signaling (SEH vs. Posix Signals), Error Handling, Threading, Multi-process, Memory models, etc. All of these things change between OS’s, even if in only minor details.
Cross Platform is more than just processor architecture – it is also the Operating System.
Memory models do not change between OSes, AFAIK (I could stand to be corrected if I’m wrong). They’re a facet of the underlying hardware platform. The underlying API might affect the performance of the class library, but the classes that are performance critical are rewritten for each platform anyway, so Java probably uses the exact same mechanisms as .NET on Windows.
As I understand it, Java was designed to be interpreted as well as compiled, while .NET was optimized purely for compilation before execution (and the CLR today doesn’t ever interpret code, AFAIK). The underlying OS shouldn’t make that much of a difference here… and the VMs can be optimized for the target platform anyway. There’s nothing particularly tied into Windows in .NET. In fact, the CLR can be hosted in SQL server which redefines many of the OS primitives to work better for database transactions. In the SQL scenario, .NET does not call directly down into the OS for most important tasks. Instead it calls SQLOS: http://blogs.msdn.com/slavao/articles/441058.aspx.
I have to be honest, though… I do not know if .NET is faster than Java or how much. It may or may not be, but I haven’t done any benchmarking myself, so I’d be hard-pressed to say anything. I’m just trying to argue against your premise that the underlying platform has much of an effect on the speed of executing the Managed Code. I also don’t think a little bit of incremental speed each way makes much of a difference. One would be expected to pick .NET or Java for other reasons than a 10% performance advantage either way. .NET would be chosen for its language independence and Windows support, while Java would be chosen for its maturity and comparative ease of moving applications from one OS to the other (though anyone seriously doing this would need to do significant testing to make sure things really are the same).
I believe that was my point – though I said it that way as Java is not, as you admit, fully there yet. It’s a good start, but still has some way to go.
Fully agree on .Net being just a fad.
#1) Java class library source has been around for at LEAST 10 years, if not since it was first released
#2) The .NET VM is not going to be open, only the API. SUN is just making the move now to open their JVM.
#3) Java needs to watch out more on technical issues, like how .net is language agnostic, C# is java with a bunch of things fixed, and .NET GUI apps don’t suck. MS doesnt need to open their JVM, all they need to do is port it to other platforms (or just get Novell to do it), and that is the final nail in the Java coffin.
The reason .net gui dont suck,
1. unlike Sun, Microsoft have a complete system with nice librairies. Microsoft need to keep its gui “modern” against competition (aqua, gtk, qt)
2. Sun is too political. SWT offer a far better alternative (wx in mono/.net/aqua/python), but because it is IBM’s solution, Sun decide not to support it. Yet they fraud their developpers and supporter telling them that people really like swing and that it’s fast.
> Yet they fraud their developpers and supporter telling them that people really like swing and that it’s fast.
Actually sanctus, Swing is quite fast when done correctly. The greatest problem are Swing developers that don’t correctly deal with the event dispatch thread that blocks repainting. With Sun JDK 1.6 update 5 (currently in early release testing) the *entirety* of Swing is rendered in hardware with DirectX on Windows. Pretty hard to get faster than that (unless you continue to block the Event Dispatch Thread).
If we forget the consumer desktop for a moment and consider the corporate/enterprise/government world, Java applications are very widely deployed. I have seen some .NET projects in the enterprises I deal with, and this restricted opening of the .NET libraries is definitely a help for them, but Java is still the “800 lb” gorilla in this space and will probably be for some time.
Edited 2007-10-04 20:30
When SWT came out, swing was slow (even if done correctly) and looked not only foreign, but unappealing. People were complaining about swing for years and Sun just didn’t care a bit. Then IBM creates an amazing alternative that was faster than swing and the interface look native on every platform out of the box. IBM had listened.
It was too late, like Sun often do, before they start improving their product, they wait until customers were frustrated enough and create an alternative. Swing isn’t alone, the whole jvm get nearly stagnant until .net was unveil.
They invest just more money where competition and development effort weren’t needed.
Sun could have work with IBM to make the whole jvm more appealing for developers, business apps as well as end user in one shot. But they didn’t want IBM to receive any credit. They politically played, against java and their customers.
Still today, swing feel foreign even with the nice themes, especially on Linux where it still look crap. Plus, it keeps burden the developer with extra work.
That is true, but as a consultant, I work for different clients, big enterprise and government. Java is still being use to develop server application, but when it comes to user interface, .net is becoming the standard. I worked for a government entity with more than 20000 employees that switch officially to .net. The look and feel wasn’t the only issue, but it played big time. What people see is the GUI, if the GUI suck, they entire application/framework get tagged. Sadly, a very large amount of people dislike swing. That’s good for consumer desktop and corporate.
I just hope the open source of the jvm will be more “open” and stop this over political attitude.
Edited 2007-10-05 14:39
It was too late, like Sun often do, before they start improving their product, they wait until customers were frustrated enough and create an alternative. Swing isn’t alone, the whole jvm get nearly stagnant until .net was unveil.
This was because for the longest time Sun tried to have their JVM, which had become the de facto, be nothing more than a reference for others to verify against. Unfortunately for them, it became the de facto standard that everyone used, and they were eventually forced to fix stuff so that it actually worked.
[blockquote]
“Reference use” means use of the software within your company as a reference, in read only form, for the sole purposes of debugging your products, maintaining your products, or enhancing the interoperability of your products with the software, and specifically excludes the right to distribute the software outside of your company.
[/blockquote]
So this is nice debugging tool for .net software houses(one of which I work for), but short of some innovative reverse engineering tactics using a third party to review code, useless for mono or dot.GNU
OTOH, since this is “reference” code, some sort of parser could probably be written to extract useful information while shucking any “creative expression”.
Meh, I wil continue to stay away from C# for anything nonWindows
Edited 2007-10-03 20:24
The thing is, with tools like Reflector, you don’t really need the source most of the time.
Where this will be incredibly useful is for debugging, now we will actually be able to step through framework classes, which is quite useful, and one of the few things I miss about my Java development days.
Isn’t it a little early (or late) for April fools.
This is not open source. At all. It just gives you the right download and distribute copies of the source provided you do not modify it or incorporate it into a product.
Basically, it’s useful so that you can look at how the libraries work to make your own interoperable software (eg Mono).
It is not a submarine patent threat, as the license gives patent rights to anyone using the source, so long as the don’t copy/modify the source. That is, Mono could use it to test compatibility or check how a function works, but they can’t copy the code into Mono. So long as it’s clear that they write the code cleanly, it’s in theory safe.
Granted, IANAL…
No one ever claimed it is open source?
Miguel de Icaza coments about it:
http://tirania.org/blog/archive/2007/Oct-03.html
Miguel de Icaza coments about it: