Slashdot Mirror
Mono Unimplementable?
Prior Restraint writes "According to this ZDNet article, Tony Goodhew, a Microsoft program manager, implies that MS will license C# in such a way that Ximian won't be able to implement the ECMA standard." This comes on the heels of Ximian's announcement of working on .Net aka Mono [?] .
No, Hemos, he doesn't imply that. You infer that.
Most standards bodies will not accept anything patented as a standard unless the holder of the patent agrees to license for a small fee all their patents for this purpose.
It is extreamly common to use something patented in a standard, the owner of the patent just has a standard contract that anyone can agree to: for $x (often $10,000 so it is out of most open source pockets, but cheap for a company) your company can use the foloowing patents in any product as nessicary to impliment y. For instance you would license 8b10b encoding from IBM for fibre channel, but you would not have the right to use 8b10b on your own bus, only in your interface to fibre channel.
There are a couple RFCs which are legal agreement that anyone can as no charge use some patent for IP, so long as you follow some restrictions.
Any standards body that would let Microsoft keep open source out with patent games isn't a real standards body. Most members of technical standards bodies I know of (Members themselves, not the company they represent) are open source aware and pay some attention to these issues.
Why does the summary mention C#, when the article doesn't mention it or any other programming language? Is the submitter interpolating with no basis, or did he read a different article than the one linked to?
--
The next Cmdr Taco duplicate will be ready soon, but subscribers can beat the rush and see it early!
The setting: Early 1999, a documents convention in Atlanta (all about printers and scanners and signature verification technology, etc...). On the stage, left to right, are Linus, a senior Microsoft marketroid, Maddog, and a Wall Street analyst. The Microsoft guy was running off at the mouth about all the huge labs they have in Redmond, where they can replicate any problem known to man. Linus jumps into the fray with a case involving the U.S. Post Office. (Those little barcodes you get on the bottom of all your envelopes? Those barcodes are put there by printers powered by Linux. Only thing that would run that reliably and had good vendor support.) It seems USPS was having a performance problem. They called Red Hat. Red Hat looked at it. They scratched their heads, and forwarded it to Linus. Linus looked at it, scratched his head, vi'ed a couple files, had an "aha" moment (discovered a race condition), tappity, tappity, tappity, compile, init 6, diff-pipe-mail, problem solved. Total turnaround time for USPS, 48 hours.
Then he delivered the zinger.
There was dead silence in the room, and I've never seen anyone look quite so uncomfortable as that Microsoft marketroid sitting there in the spotlight between the two most vocal penguinheads on the planet....I must admit, though, it's fun to see other folks catching on to Linus' PR methods...
--
I'd rather listen to [Sir Isaac] Newton than to [Microsoft's VP Craig] Mundie. He may have been dead for almost three hundred years, but despite that he stinks up the room less.
-- Linus
Microsft is saying that if Ximian uses any of Microsoft's code (ie .NET runtime, libraries) then they are going to have licensing issues.
Ximian is saying that they aren't going to use any of this code so there won't be issues.
ZDNet is making the whole thing muddy so they get all the Slashdot effect and sell more advertising.
Note that both Microsoft and Ximian are corret here.
Fear: When you see B8 00 4C CD 21 and know what it means
There is nothing to stop Microsoft from deviating from the ECMA standard they submitted. Just like there was nothing that stopped them from deviating from the ECMA standard from Javascript that they passed. MARK MY WORDS: We will implement it, and they will change it. We can't extend them if they have more users than we do. We'll just end up writing a bunch of software that works on the "official" implementations and not end up being able to use stuff from them. (They will eat us alive this way -- and easily).
.NET. I think that all of the grumbling they are doing now is a bluff and is not serious.
They will LOVE that Linux developers are doing their work for them in pushing
Also, why not work with a standard that doesn't change? Java is here today, and is VERY open about its future direction. Do you think Java is slow? Well. . . how fast do you think IL interpreters will be on Linux for the next 3 years? (not as fast as Java during that time, I guarantee). Java isn't open source? That may be true, but Sun has a reputation for creating open standards if not open source software(NFS, NIS, their hardware spec, etc.) Microsoft doesn't have this reputation and have already screwed around with developers by "standardizing" ECMAscript and then turning around and breaking it. We know Microsoft isn't trustworthy, so we are going to help them and then try to trick them out of their own game???? What????
If we continue to fall for this, we are too stupid to compete with Microsoft. (So we shouldn't even play with them. . . competing with them wasn't the point anyway - the point was serving our own needs with great software).
Err, not really. See:
In an article I read, the MS rep. said that MONO would "probably" have to use MS IP. And, in the situation where MS IP was used, MS would have to look at how they license their IP because the GPL is so scary.
.NET implementation will also be scary.
This is total FUD, because MONO is only implementing the EMCA standard and would never use MS IP to do it. After all, how could you make quality software if you filled it with/based it on crap?
MS is just creating FUD again by saying that the GPL is scary, so thus this Open/Free
ZDNet reports this propaganda like it's fact.
A great many people think they are thinking when they are merely rearranging their prejudices. -- William James
You're confusing two different aspects of .NET: The application environment and the ASP tools.
.NET object model. Despite the source of these components, I am looking forward to being able to use Mono. This half of the .NET project is actually something I am looking forward to.
The Mono project will be implementing the Common Language Environment, C# compiler, and the
The cool part is that I will be able to use the graphical form designers to create the visual parts of my programs and then implement those in any language I choose: since I don't like C# or (ugh) VB, I'll use Python.NET. If necessary though, I can write some parts in C and subclass them from within python; no more worrying about tools like SIP and SWIG which are by nature imperfect since they must be reimplemented for every language pair.
Unfortunately, taking advantage of this would in the past have limited me solely to Windows, which IMHO outweighs all my possible gains. With Mono, however, I will be able to do all of the above in a cross-platform manner. Even if the Mono project does not release a graphical form designer, I can do that part in Windows and develop the rest in Linux, in my favorite programming language. The resulting program will then be able to run anywhere with only a recompilation.
The part you're worried about includes the Hailstorm initiative and other Microsoft muscle-flexing. Mono is not interested in this at all. Rather, the project implementing this half is GNU.Net, which is going to directly compete with Microsoft even as Mono complements it. If you are worried about this, GNU.Net is the project to watch for a Free (in all meanings of the word) replacement for it.
--
Friends don't let friends misuse the subjunctive.
> Miguel de Icaza, Mono's founder, said, "The
> consensus is that [Microsoft] could stop someone
> from implementing the specs by using patents.
> [But] nothing in dot-Net is really innovative,
> so it would be simple to use alternative
> non-patented approaches.
No one says "We're gonna screw you, and take over you own architecture, oh, and, you can't innovate either" quite like Miguel.
However, Jan van den Beld, ECMA secretary general, said the licence would cover only Microsoft's own implementation, not the standard itself. "There are no known rights owned by Microsoft that would require a licensing agreement," he said.
Miguel de Icaza, Mono's founder, said, "The consensus is that [Microsoft] could stop someone from implementing the specs by using patents. [But] nothing in dot-Net is really innovative, so it would be simple to use alternative non-patented approaches."
This says nothing about Mono being "unimplementable" as the /. headline questions. It's just Microsoft's implementation of the submitted standard that is in question, which is dealt with in the last sentence of the article quoted above.
AFAIK Microsoft cannot patent a "standard", unless it's one of those unofficial standards that they create through their monopoly.
Of course, that's not to say that Microsoft was stupid by using ECMA. They can claim that C# et al is an open standard while not actually allowing anyone to implement it.
Most likely, Microsoft will withdraw their submission from ECMA. If they do not, people will just implement based on the standards, at which point Microsoft will not be able to sue because the standards are open.
Regardless of what Microsoft does, the open-source folks will go ahead with their implementations of C# et al, even if Microsoft withdraws from the ECMA standards process.
--
Oceania has always been at war with Eastasia.
Think back in history. Think City-State. Think of a heavily fortified city at the cross roads of some very heavily travelled trade routes. .NET is the paving on the roads, while the fortified city is MS-Passport. And to move your goods through Passport, you have to pay a tax. Do you think given that model that billg cares whether he owns the roads or not? I don't think so. If I was in bill's shoes, I'd be all for people building roads to my kingdom.
And so who cares if .NET is proprietary or not? If it's easily available for, or ships with, 90+% of the desktops in the world, what does it matter? All of this Mono hype misses the point: what is really needed is a credible competitor for Passport. Forget .NET. Building a .NET work-alike merely reinforces the MS monopoly. We should really be setting our sites on building a competitor to the Passport services.
Am I the only one who thinks it would be much smarter for a competitor to use their own technology. MS has alway been a tough competitior, why does Ximian think they will succeed with this? They should be pressuring Sun to GPL Java (would make sence on the heels of recent news about MS abandoning it) and build an architecture based on that. I think that our own Free .NET type implementation (that also runs on Windows) would be a better strategy than trying to play catch up to a Huge multibillion dollar company that controls the market to start with.
A freely implementable standard is a successful standard. Examples are TCP/IP and the PC architecture.
.NET is already well on it's way to becoming a failure as a standard.
However, there are many technologies that never became standards. Remember MicroChannel from IBM? It never succeeded because IBM held the licensing rights.
You can't have it both ways. Either it is freely implementable or you have to pay, whether it is in the form of licensing fees or other restrictions.
If Microsoft want to tightly control implementations, then in my opinion
Surely there must be some real development going on in the free software world that could be covered instead. A Linux-powered robot that recognizes human faces or Sun's study of Gnome usability?
Unsettling MOTD at my ISP.
In the article, ECMA's geral secretary refutes the MS claims. It's probably some lame misunderstanding. And if not, they've dug a nice big hole for themselves.
-- Another senseless waste of fine bytes.