| |||||||||||||
|
| |||||||||||||
Microsoft really has us going...Microsoft really has us going...Posted Jul 7, 2009 17:17 UTC (Tue) by drag (subscriber, #31333)In reply to: Microsoft really has us going... by astrophoenix Parent article: de Icaza: From Microsoft: C# and CLI under the Community Promise
I think it has much more to do with making distributers and users comfortable with Mono's CLI and C# implementation then it has to do with the Microsoft announcement.
Nothing has fundamentally changed with this announcement. Anybody paying attention knew pretty well that C# and CLI were safe to implement.
The questionable portions of Mono, from a legal standpoint, were the parts that implement .NET compatibility.
So Mono working to split off the .NET stuff from the CLI stuff is going to make it easier (not to mention reduce the footprint) for distributions to incorporate C# programs that don't use .NET (like Fspot) without worrying about it.
-----------------------
Personally I am not a huge fan of C# or whatever.. but I hate to see a huge amount of hard work knocked down through nothing more then ignorance and paranoia.
(Log in to post comments)
Microsoft really has us going... Posted Jul 7, 2009 17:38 UTC (Tue) by tzafrir (subscriber, #11501) [Link]
What you write here actually makes me more uncomfortable.
Could you please explain why do you think Mono is useful and should be included in distributions? I wonder how this split affects those arguments as my feeling is that it doesn't.
Microsoft really has us going... Posted Jul 7, 2009 18:18 UTC (Tue) by drag (subscriber, #31333) [Link]
I think that it should be included for the same reasons why they include python or C++ support.
It's a nice language that a lot of people who are new to Linux already know very well and is comfortable with. Schools and universities are full of people that are learning C#. So it's a very nice option to have and is a nice alternative to programming in C for Gnome folks. Python is very high level and is slow, and C is very fast. C# is a nice in-between and it's bindings are very good.
It's just that I feel that distributions should stay away from including the .NET stuff by default. Lots of users need that sort of thing and is a nice option to have, but there are lots of stuff that is not included in distros by default for lots of reasons. Just as long as its packaged and available should be fine.
Microsoft really has us going... Posted Jul 7, 2009 20:06 UTC (Tue) by tzafrir (subscriber, #11501) [Link]
The counter argument to that is that Mono and .Net are not the same thing. "Mono-Safe" seems to only cover basic parts. It also fails to cover later features of C# and the run-time (anything later than 2006. That's the latest thing I see on EMCA's page).
But those students will want features only available in Mono-Risky. And I wonder how we can reduce the risk here.
Microsoft really has us going... Posted Jul 7, 2009 20:24 UTC (Tue) by drag (subscriber, #31333) [Link]
That would be right that Mono isn't .NET. But I think that C# with Gnome would still have positive benefits, even if your dealing with people that are only familar with .NET so far.
Microsoft really has us going... Posted Jul 8, 2009 6:11 UTC (Wed) by jamesh (subscriber, #1159) [Link] Could you please explain why do you think Mono is useful and should be included in distributions? I wonder how this split affects those arguments as my feeling is that it doesn't. The same reason as you'd include any framework in a distribution:
With the proposed split, some distributions might decide that only the core portion of Mono satisfies those goals (e.g. because the apps they want to include use GTK# rather than WinForms), and omit the portions not covered by Microsoft's promise.
Microsoft really has us going... Posted Jul 8, 2009 1:27 UTC (Wed) by ncm (subscriber, #165) [Link] C# and CLI were [already] safe to implement. The questionable portions of Mono, from a legal standpoint, were the parts that implement .NET compatibility.False. MS's US patents appear to cover possibly unavoidable details of implementation of CLI. They may assert these patents against Mono with or without .NET attached. As coriordan noted in 340344, protecting the Mono developers, if in fact this announcement does such a thing, does nothing for Mono distributors, Mono coders, or Mono users.
Microsoft really has us going... Posted Jul 8, 2009 2:08 UTC (Wed) by drag (subscriber, #31333) [Link]
Mono folks worked based off a standard Microsoft set up themselves. They shared code with Microsoft, went to Microsoft conventions, announced products, did demonstrations.
Microsoft even made sure to license them codecs free of charge in order to get silverlight legal to distribute for Novell.
Microsoft has had full knowledge of Mono from _day_one_.
Now Microsoft has publicly announced, under no uncertain terms, that they will absolutely not file claims against anybody who makes, uses, sells, distributes or imports in any manners.
Now you and other people decide that none of this means anything.. that a company can sit on patents for years and help a competitor and then all of a sudden a decade later go after them... even after publicly announcing, in writing on their own website that they won't. But from what I can tell and from what I know that is a whole bunch of flaming horseshit and, in fact, C# is safe to use.
Protestations notwithstanding Posted Jul 8, 2009 2:43 UTC (Wed) by ncm (subscriber, #165) [Link]
As I and others have noted, "going after" users or distributors of Mono, who Microsoft has never addressed, is not the same as "going after" Novell or its employees who MS helped in developing Mono. It's in MS's interest for dependence on Mono to spread to all corners of the earth, just as it's in the interest of a codec or RAM bus developer to get its use standardized and unavoidable. In either case such activity broadens the field of possible licensees. Failing to assert a patent, for any number of years, does not reduce the force of the patent, until of course the patent expires. It's common, though, to stipulate in a license contract that royalty or other demands continue in force after the termination of the patent itself, so there's no hurry. Waiting until a few years before expiration to start extorting is just good business.
Therefore, all your protestations of MS's foreknowledge are irrelevant and, in fact, misleading.
I doesn't makes the rules, I doesn't legally advises about the rules, I only reports the rules.
Microsoft really has us going... Posted Jul 8, 2009 13:45 UTC (Wed) by dgm (subscriber, #49227) [Link]
Never forget that no matter how open the specification, Microsoft still retains full control of the intellectual property.
This "promise" allows you to use their purported property, if and only if you play by their rules (there is, at least, one "kill switch".)
In my opinion, we should be helping people migrate off .NET ASAP, not helping it spread. To that goal, Mono is a disaster. It can barely run non-trivial software written for .NET, and instead does encourage new developments in those patent encumbered technologies.
Microsoft really has us going... Posted Jul 9, 2009 16:44 UTC (Thu) by Quazatron (subscriber, #4368) [Link]
I have nothing against C# -- except that it's yet another runtime bytecode interpreter to have loaded, consuming precious memory, in order to run a simple note taking program.
I don't run Mono apps for the same reason I don't run KDE apps when using a Gnome desktop: memory consumption.
I already have a python instance running on my desktop. I'd like it if Gnome developers kept the language selection to a minimum. If they start including applications made in every language under the sun, soon we'll need Java, PHP, Javascript, Ruby and OCaml runtimes loaded as well.
Microsoft really has us going... Posted Jul 11, 2009 9:52 UTC (Sat) by nix (subscriber, #2304) [Link]
Well, part of this depends on how big the bytecode interpreters are. The
bytecode interpreter is probably small: the downside is the entire heavyweight optimizer, native-code JITter, and enormous library that probably comes with it.
Lua is a pretty good example of an interpreted language so small and fast
|
|
||||||||||||