-Canonical (not Jo Shields) launches and keeps available a mono-free flavor of Ubuntu

I’m fine with Jo Shields releasing Mono-free ISO images, and I hope it becomes a tradition, and I also hope we do eventually also get the more interesting i386 respin.

not Jo Shields

No pleasing some people

-Release Mono under the GPLv3

-Mono fans stop bashing the FSF and RMS and treating software freedom as something undesirable and different and separate from source openness, and they.stop pretending there is no problem with Microsoft corporation or justifying/dowplayinf their bad deeds and aggresive strategies.

-I would also like the Mono proponents be more vocal against software patents and the extension of software patentability to parts of the world free from this legal threat to FOSS, instead of just claiming that its just the way it is and that we should just accept the statu-quo.

-Novell goes back to the community fully discloses the MSFT-Novell deal’s terms and does not renew them after the agreement has expired -2011- (or better, declares it does not want to continue with it)

-Canonical (not Jo Shields) launches and keeps available a mono-free flavor of Ubuntu (just like it does with kubuntu and xubuntu)

saulgoode :

To be fair, I think accusations of license violations should not have been made without providing conclusive evidence and I concede that you may be right that Microsoft did not violate the GPL. I just feel that IF either assertion is to be made, it has to address deeper issues than the ones presented.
Regardless of whether they were obliged to do so or not, in this instance I am quite pleased with Microsoft’s release of their code licensed under the GPL.

At least we agree on something, then!

Jo Shields :

My problem is people saying “I don’t want that” who simply haven’t researched the issue – they’re just recycling talking points. They’ve heard scary words and, like, zomg!

Why is that a problem? Why does a person have to justify their rejection of ? What if, having been around a while, a person has come to the (rather rational) conclusion that certain words really are rather scary? And what if all the argument they need is to hear one of those scary trigger words? Why do you demand that they sit through your extended apologetics, when they’ve already come to their own conclusions long ago? Who are you to demand that they do so?

I don’t want Microsoft within 100 light years of any machine under my control. It’s just that simple. “Microsoft” is a scary trigger word that sets off alarm bells between my ears. You can belittle me as irrational all you like, but I know better. I was there when Microsoft installed those bells.

Which the GPL/BSD dual-licensed shim using those symbols IS distributed under.

I am extremely skeptical that writing non-compliant works based on GPLed software (e.g., Linux kernel modules) somehow becomes compliant because of an intermediate GPL/BSD shim. Such shim modules may be acceptable if the driver was designed for use with non-Linux OSes (ala NDISwrapper and, perhaps, nVidia). But the issue is not the shim, it is the original target of driver code. If the code was designed to work in Linux kernel space, it should be considered a derivative work of the kernel. If the code was designed to work with Windows (or SunOS, or BSD,…) then it should not be considered a derived work — and the provision of a compliantly licensed shim shouldn’t alter this.

I posted the URL to the original (and still available) download. Take a peek yourself.

I did visit the download page, and found no evidence to indicate the nature of any “proprietary” code (I didn’t download any files to see what was included) — whether it was coded specifically for use in Linux kernel space or was just Windows-targeted code with a shim wrapper.

To be fair, I think accusations of license violations should not have been made without providing conclusive evidence and I concede that you may be right that Microsoft did not violate the GPL. I just feel that IF either assertion is to be made, it has to address deeper issues than the ones presented.

Regardless of whether they were obliged to do so or not, in this instance I am quite pleased with Microsoft’s release of their code licensed under the GPL.

Learn to read, Saul.

directhex@desire:~$ for i in /usr/lib/tomboy/Tomboy.exe /usr/lib/tomboy/addins/*.dll; do echo non-ecma libs in $i:; monodis –assemblyref $i | grep System | grep -v System$ | grep -v System.Xml$; done
:
:
non-ecma libs in /usr/lib/tomboy/addins/Tasque.Addin.dll:
:
:

Tomboy uses Tasque (but does not necessarily depend upon it), which depends upon System.Web, which is not part of the ECMA 335 specification yet is technology developed by Microsoft. It is fine that you hold the opinion that Microsoft does not hold any valid patents on System.Web technologies; but your opinion does not prove the FSF claim incorrect.

I applaud your efforts to provide separation between ECMA and non-ECMA Mono components in your Debian packages — and recognize that you were doing so even before upstream announced their own efforts along those lines. Nonetheless, you have not shown to be false the FSF claim that “there are several libraries that are included with Mono, and commonly used by applications like Tomboy, that are not required by the standard”.

saulgoode :

1) The GPL permits you to distribute GPL-licensed source alongside binary-only proprietary objects,…

Objects designed to link with GPL-licensed code are derived works, and this would be a separate issue than merely aggregating unrelated works in a collection. The GPL does not permit derived works to be distributed unless done so under the GPL.

Which the GPL/BSD dual-licensed shim using those symbols IS distributed under.

2) This topic is NOT a licensing issue – but it IS considered a violation of trust. In the kernel, you can mark a given symbol with EXPORT_GPL_SYMBOL, which basically says “you can only access this piece of code if you are also GPL”.

EXPORT_SYMBOL_GPL does not alter the licensing terms of the kernel, which is GPL2 with the only exception being that “user programs that use kernel services by normal system calls” are not considered derived works. The export flag is a bookkeeping/debugging mechanism that effectively allows the kernel to avoid rejecting non-compliant modules outright. It does NOT mean that non-GPL-compatible modules are thus deemed compliant; merely that the kernel won’t refuse to load them. Just because the kernel is not proactive about enforcing the law, does not mean that the law is not applicable.
As I haven’t investigated the details of what code Microsoft originally distributed (if any) under what licensing terms; there may be other reasons that the MS code was not infringing — however, the specific demurrals presented here by Nobody Reel and Jo Shields are not valid defenses against GPL infringement.

I posted the URL to the original (and still available) download. Take a peek yourself.

saulgoode :

I really doubt there’s anything patentable in System.Web (it’s trivial stuff for mail & web interaction). System.Data is indeed the “scary” ADO.NET, which is .NET speak for “JDBC”, but all concerned are happy to stop using that abstraction once they know what exactly they need to stop using and start using instead. System.Core relates to C# 3.0 (which Do uses heavily), but I SINCERELY doubt that it’s not going to be on the table right now, as the ECMA (currently) debates 334/335 version 5.

You started out by asserting that the FSF’s claims about Tomboy using non-ECMA MS technology can easily be shown to be wrong yet, correct me if I am misunderstanding, your above analysis shows Tomboy does use some System components falling outside of ECMA. Caveats aside as to whether patents exist which cover System.Web; how easily System.Data functionality might be replaced; or plans to have the standard incorporate more components in the future, I fail to see how you’ve shown the FSF claim to be wrong.

Learn to read, Saul.

directhex@desire:~$ for i in /usr/lib/tomboy/Tomboy.exe /usr/lib/tomboy/addins/*.dll; do echo non-ecma libs in $i:; monodis –assemblyref $i | grep System | grep -v System$ | grep -v System.Xml$; done
non-ecma libs in /usr/lib/tomboy/Tomboy.exe:
non-ecma libs in /usr/lib/tomboy/addins/Backlinks.dll:
non-ecma libs in /usr/lib/tomboy/addins/Bugzilla.dll:
non-ecma libs in /usr/lib/tomboy/addins/Evolution.dll:
non-ecma libs in /usr/lib/tomboy/addins/ExportToHtml.dll:
non-ecma libs in /usr/lib/tomboy/addins/FileSystemSyncService.dll:
non-ecma libs in /usr/lib/tomboy/addins/FixedWidth.dll:
non-ecma libs in /usr/lib/tomboy/addins/InsertTimestamp.dll:
non-ecma libs in /usr/lib/tomboy/addins/NoteOfTheDay.dll:
non-ecma libs in /usr/lib/tomboy/addins/PrintNotes.dll:
non-ecma libs in /usr/lib/tomboy/addins/SshSyncServiceAddin.dll:
non-ecma libs in /usr/lib/tomboy/addins/StickyNoteImport.dll:
non-ecma libs in /usr/lib/tomboy/addins/Tasque.Addin.dll:
non-ecma libs in /usr/lib/tomboy/addins/WebDavSyncService.dll: