Mono not Chasing Tail Lights | Hacker News Reader

Mono not Chasing Tail Lights

jeffreystedfast.blogspot.com

52 pointsoomkiller22 comments

I'm not convinced interoperability isn't key. If all they wanted was a Java clone that handled generics and boxing better, compatible with nothing but themselves, they could have had that years before C#.

Doesn't Vala handle that? Not to mention it runs faster and with a lower memory footprint.

Don't get me wrong, i love mono and vala and use them both regularly. But vala doesn't have stack protection, etc which is quite useful for certain situations.

He assumes a MS patent claim won't hurt a mono-using company. He asserts they'll just remove the code and happy days again.

But that's not the worry here: MS could sell a mono-related patent to someone who's willing to attempt to bankrupt a mono-using company.

And so, Mono puts a company at more risk of dying in the court-room, gives MS more "proof" Linux is not safe, and gives MS more leverage in negotiations via patent-threats.

"MS could sell a mono-related patent to someone who's willing to attempt to bankrupt a mono-using company"

Pardon my ignorance, but does this really work this way? I was under the impression that the patent usage has to do with who owns the patent at the time it's used. What prevents companies from reassigning patents for nothing more than the intention to destroy competition?

You think Microsoft is going to sell the patents to it's core development technology to someone else so that they can bankrupt a company using Mono? That's a pretty far fetched scenario.

Beyond that, as the article points out, you can use Mono and not use Microsoft's patented technology. It just requires a little more work (again, C# is an International Standard which is not covered by any patent)

You think Microsoft is going to sell the patents to it's core development technology to someone else so that they can bankrupt a company using Mono? That's a pretty far fetched scenario.

Considering they funded SCO to try and kill Linux and called open source a cancer, yes, yes I do think that.

Well, and as fun as taking quotes and events out of context is, that really doesn't tell the story.

Point #1 (SCO): Microsoft paid SCO a relatively small sum ($16 million in a year where they made over $10 billion in profits) which they claim was to protect themselves from any legal action. Open Source advocates have always claimed the money (and the later introduction to Baystar which led to more funding) was to destroy Linux but if Microsoft had wanted to do that they could have easily bought out SCO and just asserted the patents themselves.

Point #2 (Linux is a Cancer): Ballmer made this quote in 2001 when referring to the GNU GPL. Put in context it isn't the smoking gun Open Software advocates represent it as. From the Register's article regarding the quote (http://www.theregister.co.uk/2001/06/02/ballmer_linux_is_a_c...)...

Microsoft CEO and incontinent over-stater of facts Steve Ballmer said that "Linux is a cancer that attaches itself in an intellectual property sense to everything it touches," during a commercial spot masquerading as a media interview with the Chicago Sun-Times Friday.

Ballmer was trying to articulate his concern, whether real or imagined, that limited recourse to the GNU GPL requires that all software be made open source.

"The way the license is written, if you use any open-source software, you have to make the rest of your software open source," Ballmer explained to an excessively credulous, un-named Sun-Times reporter who, predictably, neglected to question this bold assertion.

http://press.redhat.com/2009/09/09/microsoft-and-patent-trol...

The Open Invention Network (OIN) learned recently that Microsoft was planning to auction off some of its software patents, which we understand it marketed to trolls and some other non-practicing entities. It also used marketing materials that highlighted offensive uses of the patents against open source software, including a number of the most popular open source packages.

This isn’t some fanboy’s blog, this is a Red Hat press release.

I can't seem to find any basis for this other than "entity x recently learned from anonymous entity Y that Microsoft might do something bad"

That said, the point here is that the patents in question in regards to Mono are the cornerstones of Microsoft's own development efforts. So they aren't going to sell off their ADO.NET related patents because it's the core of data access in their entire development environment.

For the scenario in which a patent affecting Mono was sold to happen Microsoft would have to auction off pieces of the .NET Framework itself and that's just not going to happen.

Besides, doing so would a PR debacle of epic proportions. Microsoft wants developers to use their technologies (duh). Does anyone seriously believe that they are going to stab those developers in the back after publicly promising to do nothing of the sort?

Anything that has the potential to make desktop software better on Linux is good with me. In case anyone hasn't noticed it's in an abysmal state. There are many examples of Linux being a decade or more behind comparable commercial solutions on other platforms. If Mono happens to interest some Windows developers or promote interoperability it would improve the situation greatly in my opinion. As far as patent concerns go with Microsoft -- I say force their hand. Let them litigate if they want to. I think the US DOJ and EU would take a special interest in the case. OSS can't cower in the shadows afraid of big bad Microsoft forever.

The one point I'd like to make is that Microsoft has never threatened to sue anyone over Mono. They made an official agreement with Novell to appease Novell's corporate customers but that doesn't mean they'll sue anyone else. In fact, just the opposite. Scott Guthrie (VP of Development at Microsoft) regularly gives time on his popular blog to Mono projects (see one example: http://weblogs.asp.net/scottgu/archive/2009/02/11/moonlight-...)

The whole "Microsoft Patent" issue is based on the Free Software Foundation stating a concern that Microsoft could someday sue. I don't dispute that concern but understand Mono is very much in Microsoft's best interest because it competes with Apache on Linux.

Finally, as the article points out, even if Microsoft asserts it's patents they'll only be taking away certain support functions like ADO.NET. The heart of Mono is the C# compiler which is an ECMA and ISO standard.

Mono happens to interest quite a few non-Linux developers. The Sims 3 (running on Windows and OS X) heavily uses Mono for scripting in-game events.

Considering tools like MoMA are exclusive to Mono and don't have an equivalent in the .Net world (http://www.mono-project.com/MoMA), I'd state that in some cases Mono is a little ahead of its Windows counterpart.

I'm curious how "Java is actually already way behind Mono". What can the one do that the other struggles with?

To clarify for other readers: I believe he means the JVM, not the Java language.

I haven't been keeping up with Mono's development, but if they have C# 3.0, it blows away Java right there. C# 3.0 is a functional programming gateway drug, but encourages high readability even for non-functional developers. However, that's not a fair comparison, as there are other great JVM languages such as Scala and Clojure.

As for Mono or the CLR vs the JVM, there has been a lot about this written on the net. You can find discussions about garbage collection, module systems, JIT performance, security, library support, etc.

Here's one example: MSIL is more rich than Java Byte Code. This makes writing compilers easier and can lead to better performance via unsigned types, objects on the stack, unsafe pointer use, etc. See: http://stackoverflow.com/questions/95163/differences-between...

I'm not an expert on the CLR by any means, but I do quite a bit of C# development on Windows. I greatly prefer C# to Java, but couldn't really care less about CLR/Mono vs JVM. I'm really more interested in LLVM as a compile target for interop and optimization of open source languages.

The "Debian shootout" suggests the JVM is consistently faster, but more of a memory hog.

http://shootout.alioth.debian.org/u32q/benchmark.php?test=al...

quad core, or not quad core - that is the question:

http://shootout.alioth.debian.org/u32/benchmark.php?test=all...

The answer to the question seems to be "doesn't matter much", and the same for 64-bit. Java is always faster, nearly always fatter.

More of the Java programs have been re-written for quad core, which is partly why those programs faster and fatter.

A plug for Software Engineering Radio http://se-radio.net/podcast/2009-07/second-life-and-mono-jim...

I've personally never used Mono, but if people are putting their time and sweat into the project then it obviously has value for them. Nobody is forcing anyone to use it. Free as in beer.

I wrote five paragraphs after this, but then realized that nothing else really matters. So what we are witnessing here is a circle jerk. Moving on.

Reply on news.ycombinator.com