Last week we talked about whether or not the Debian project would include Mono in its default GNOME installation. This incited some heavy debate on OSNews, but sadly, the Mono debate also lead to some very nasty blog posts in the Debian community. Time for damage control, Debian project leader Steve McIntyre must’ve thought.
iTWire contacted McIntyre about the matter, and the project leader said that in the end, it’s up to the people working in the area in question to make decisions like this. “The GNOME team are free to pick their own selection of applications that they think make the best GNOME desktop that will fit on a single CD, in much the same way the KDE make their own choices (like staying with 3.5.x for Lenny),” McIntyre explains, “Josselin Mouette has updated the dependencies of the gnome2 meta-package, converting Recommends: tomboy into Depends: tomboy | gnote.” We already explained in our previous article the how and why behind this move.
Still, McIntyre also stated that the decision is still up for debate; how exactly that fits into the GNOME team making their own decision – which has already been made by Josselin Mouette – is a bit vague to me, but I’m assuming McIntyre has more insight into the Debian release process than I do.
McIntyre’s own view upon Mono isn’t exactly favourable, but not more or less so than when it comes to C++ or Python. “I’m personally no fan of Mono, but equally I’m not a great lover of C++ or Python either; I’m certainly not about to suggest that we drop any of them right now,” he told iTWire, “There is some fear in the community about possible software patent attacks, but that hasn’t stopped us packaging and shipping other software in the past. If a real problem arises, we’ll deal with at that point. That has been our policy for quite a while now.”
He added that he regrets the tone of the blog posts in this discussion. “It seems that some of our developers are more interested in ranting and raving in their blogs than having a rational, productive discussion on the mailing lists. That’s not helping anyone.”
Util someone shows me how to obtain a royalty free license from .NET i will stay away from Mono.
Please Mr. de Icaza show me the easy steps that is necessary to have my roalty free license
What, you mean something like this?
http://www.waveprotocol.org/patent-license
Insane really. All the Mono people had to do was get Microsoft to make a public statement like this rather than saying they have non-existent letters and referring to cloak and daggers postings on mailing lists.
Edited 2009-06-16 18:33 UTC
You make it sound as though all they had to do was ask Microsoft, and MS would willingly and immediately make such a statement. How do you know someone hasn’t already asked and Microsoft refused? I doubt MS would do something like that unless they had a clear business case for it. But that doesn’t automatically mean there’s a actual patent conflict.
Do you insist on patent licenses from competitors for all software you use? Do MySQL and PostgreSQL have MSSQL patent licences? Does OpenOffice.org have MS Office patent licenses? Does Mozilla have MSIE patent licenses? What makes MONO so special?
its created by someone that has sued linux distributors and has special deals with some of them. Furthermore mono is based on, and implements, code that is created by same company that has at numerous occasions told that linux infringes multiple patents that they own.
basing your future existence (tomboy is just the spearhead of the invasion, ahem) on said platform is naive at best.
yes, I exaggerate to get my point through.
Mono/Novell has sued linux distributors and has special deals with some of them? Oh .. you must mean Microsoft, the people who don’t make Mono. In which case my question still stands. How is Mono different than any other open source product that has a competitor owned and patented by Microsoft?
It’s a stretch to call Mono a “competition” for Microsoft product (.Net stack). Mono is actually boosting and entrenching the .Net environment as viable choice for developers (which some people don’t like).
This matter is much more complicated than it should be. Here is my personal opinion.
No desktop environment should ever depend on Mono.
No distro should ever install Mono by default.
Mono should remain forever completely separated from Linux but it should be available for those who want to use it.
Which is a perfectly valid argument to make, IMO. But it has nothing to do with patents, which is what I was trying to address.
Because .Net and the technology and specifications it is built on is actually Microsoft’s invention. Software like Wine and Samba might have the same interfaces and be moderately compatible but the implementation is not based on an implementable specification from Microsoft.
WINE is an implementation of Win32; Samba is an implementation of the CIFS protocol; both Win32 and CIFS are Microsoft specifications and inventions. I still don’t see the difference.
* development method
* intent
* historical context
* availability of alternatives
and, hey, WINE and SAMBA aren’t usually included in default installs on community-based distros either.
Edited 2009-06-17 06:38 UTC
Which is all well and true, but straying from the original argument. That is, someone asserted that Mono should be avoided entirely unless Microsoft provided patent licenses for .NET. There are plenty of reasons to dislike Mono, or to object to Mono in a default Debian install, but no one has really explained why the patent issue is such an important one.
Samba is. The library component is a soft dependency on nautilus via gvfs-backends via gvfs. Virtually all Debian and Ubuntu default installs have it. I’d wager RH, fedora etc. do too.
you’re right. i don’t know why i said that. errr. my keyboard got disconnected from my brain for a second. i had intended to say that WINE and SAMBA have received their share of scrutiny, as well… but then started thinking about default installs. d’oh!
No. The implementations of Win32 and CIFS in Samba differ wildly to the point where they are completely different pieces of software that happen to be compatible. In the case of Samba and the SMB protocol then there is a ton of history even prior to Microsoft’s additional extending. It’s very difficult to pin anything on them conclusively. It’s the maybe, coulda, mighta scenario that many people try and say applies to Mono.
In the case of .Net and the ECMA specifications then they have an acid test where Microsoft can say “Ahhhh, you’re covered” and this becomes immediately obvious when you read the text of the main patent that they have applied for. There is a ton of descriptive text at the bottom about the technology it applies to as well as various namespaces.
It’s a quagmire that doesn’t need to be gone into where Mono proponents usually start saying “Show me the patents”. Well, as we’ve all seen, there doesn’t need to be any and open source developers just can’t live with that. However, a proper above board grant from Microsoft made public would cover everything as any sensible organisation should do.
Seriously, if you want .Net compatibility then just install Microsoft’s proper .Net framework in Wine and just stop pretending.
Edited 2009-06-17 11:01 UTC
You can not make that claim without looking at the actual patents with a patent lawyer. If you have the actual patent numbers that Samba and Wine infringe and could share those with the public, folks could actually have a discussion.
So far your statements are equivalent to “My dad has a helicopter, and an airplane, and he also has like a million movies, and he is also buddies with Bono from U2”.
You can repeat it as much as you want, but “segedunum said so on osnews.com” will not hold in court.
Mono is wildy different to the point where is completely different software that happens to be compatible.
That sentence works just as well. You also ignore that despite SMB’s history Microsoft still has patents for the protocol itself which makes any implementation subject to lawsuits.
It doesn’t happen to be compatible at all. It is compatible, thanks to a licensed set of specifications.
It doesn’t. You’re not goint to find a set of specifications that describes how to create a base compatible implementation of Windows Networking or the Windows API in the way that has been done with .Net, Mono and Rotor.
No it isn’t. You’re acting like C#+CLR is equivalent to .NET. It isn’t. You can’t just run all .NET software on Mono or the other way around. Both .NET and Mono share some things but they also include system specific libraries. They are very different but happen to overlap in some areas, mainly the language specification.
That’s completely untrue. Microsoft was forced by the EU to turn over SMB protocol docs a couple of years ago. It’s completely irrelevant though. Both SMB and .NET are covered by patents owned by Microsoft.
An Mono is a completely different implementation of ECMA-approved specs that happens to be compatible with Microsoft .NET. There is no other connection between Mono and MS.NET other than the specifications.
And one could say all the same things about almost any other free software project, if one actually sat down and really looked at it. But Mono seems to be the only one that people insist on putting under the lens. I don’t understand why.
Not going to happen, any more than Microsoft is going to make a grant of the patents in MSSQL to MySQL and PostgreSQL developers. But they’re all worthy projects, anyway.
You’re seriously suggesting that distos try to ship Microsoft .NET to run applications like Tomboy, Banshee, F-Spot, etc.? That’s ridiculous. These are applications written to Mono because Mono was determined to be the best application framework available, not because Mono is .NET compatible and the developers are pretending to use Microsoft .NET. Mono is a development platform in its own right.
No. The ECMA specifications describe what is needed for an implementation and those themselves are licensed. That’s what I find clever about it. Whereas many people talk about ‘clean room’ implementations, to get a clean room implementation of .Net then you can’t even look at the specifications.
Sorry, but the “Every other software project is at risk!” angle just doesn’t wash, and it’s really all a lot of people have left.
Since MySQL and Postgres don’t need anything that is in SQL Server in any way, shape or form, and certainly don’t need the specifications to SQL Server then that is a totally non-sensical comparison.
If it’s not going to happen then you walk away.
I thought Mono had nothing to do with .Net?
You generally see all the same well worked arguments from Mono proponents, one of them being this one that that .Net compatibility was ever an issue. However, they never answer the question that that inevitably brings. Why bother with being remotely .Net compatible in the first place? Why was it ever a goal?
They could have simply looked at some ideas they thought were reasonable from .Net, combined them with what was happening in the open source world and created something new and fresh where none of this nonsense would ever have happened.
You’re going to have to give a real reason to back that statement up. You can’t just say it doesn’t “wash” without giving a single reason why it is any different.
It wasn’t ever a main goal of Mono to be compatible. The main goal was to have a modern environment to develop for. Being .NET compatible is a lesser goal that makes complete sense considering both .NET and Mono share the same roots. Mono can survive on its own without compatibility but why forgo compatibility if Mono gets you half way there?
Why? NIH syndrome isn’t considered a good thing. If something works and it works well why re-invent the wheel?
I didn’t say anything about “clean room” implementations; that concept is only relevant if we’re talking about reverse engineering. It has nothing to do with patents or implementing a standard API. In fact, if you violate a software patent, you’re at just the same risk of getting sued whether you’ve made a clean room implementation or not. Totally irrelevant.
Sorry, but the “But there might be a patent in there somewhere” angle just doesn’t wash, either.
And Mono doesn’t need anything that is in Microsoft .NET in any way, shape or form. It is an implementation of a published API with no connection to the Microsoft .NET code. The SQL Server comparison makes sense because MSSQL, MySQL, and PostgreSQL are all implementations of a published, standard, API: the SQL language. What is so hard to understand about that?
It’s closer to a side-effect than a goal.
Edited 2009-06-18 01:25 UTC
WINE started as a reverse engineering project. If reverse engineering is the only way to get one product to work with another, it is legal to do so. If XY Company makes a DVD player and I want to make and sell a universal remote that can be used with it, I am allowed to reverse engineer the IR or radio signals so that my remote will control that DVD player. In the same way, WINE is reverse engineered win32 allowing win32 native software to interact with my non-Windows platform.
If you have a Windows system on your network, CIFS is pretty much a given. (If you know of a working ssh server for Windows and sshfs support; please share the url) If you want to share files and printers across your network with Windows machines then it’s going to be CIFS or some very cumbersom Windows native programs. Thus, reverse engineering CIFS to create compatability with non-Windows platforms is again a legal practice. Also, in this case, the CIFS protocol specs have been released openly thanks to the Samba project team paying off Microsoft to gain specs and permission to release them.
Mono is a direct implementation of .NET outside Windows not a reverse engineering of the framework. It is not reverse engineered to create compatibility and it has no accompanying statement that Microsoft will not consider legal action when times get tougher.
Maybe you’ve heard about that little myth about Linux (the kernel) infringing on 400’ish Microsoft patents. You know, that story they keep advertising to scare customers back too them while continually refusing to actually discuss in detail so that infringements could be addressed? Discussion between patent owner and infringing party to strike a deal or remove infringements being the second step after giving notice and before litigation.
This is one of the most common misconceptions about Mono. It is not a direct implementation of .NET. It is a direct implementation of C# and the CLR. Some .NET compatibility is included but Linux applications like Tomboy, Banshee, and F-Spot use Mono’s own libraries, which are not a part of .NET in any way because Microsoft doesn’t support GTK+ or a *nix platform with .NET.
Cheers. I do like when I learn stuff and programming languages are not my core area of knowledge so it’s good be have the .net/mono relationship clarified.
Samba is an implementation of the SMB protocol (after which Samba is named). SMB is IBM’s invention.
Local LAN network functionality which is implemented on top of SMB has heaps of prior art in Novell Netware. Any patents that apply here are probably Novell patents, donated to the Patent Commons, which would work in defence of Linux rather than against it.
http://www.patentcommons.org/
Wine is largely not an implementation of win32, so much as it a translation between calls made by application to the win32 ABI standard being intercepted by the Wine layer and then translated into the equivalent calls to the Linux kernel and drivers. In this way, WINE works in an entirely different way to Windows. However, some Windows ABI functions do not have a direct close equivalent in the Linux kernel, and hence are implement directly in WINE. However, even here, Microsoft-held patents are NOT likely to apply, because there is any amount of prior art in applications making calls to operating systems.
Caveat: one could probably argue also that there is prior art for .NET (Mono) functionality in the form of Java, and prior art for Silverlight (Moonlight) functionality in the form of Flash, and hence Microsoft can hold no valid patents in these technologies either, but it would be a far more difficult argument to make this case.
Edited 2009-06-18 07:08 UTC
The problem as it stands is that there is no clarity in Microsoft’s position. We’re told by one group of management at Microsoft claiming that the patents are only defensive and that third parties have nothing to worry about. On the other hand we had Steve Ballmer take the stage and literally say that they’ll milk the cash out of those whose software violate Microsoft’s patents.
The difference is this; in the case of OpenOffice.org, it is but one module that can be easily jettisoned at the first sign of problems, wine is a compatibility layer, not a framework – one has no dependency if one develops for Wine given that all wine is, is a win32 implementation. In the case of SQL as mentioned earlier, if Microsoft does anything stupid it would set off a chain reaction that they wished they never had done. If they sue MYSQL/Postresql, then Oracle, IBM, Sybase and every man and his dog start to get involved.
Mono is GPL2: who creates and develops it is, for all practical intents and purposes, inconsequential. If the fact that they have Novel as a patron, in-and-of-itself deeply disturbs many people, some other entity could fork it and solve the problem trivially.
And, as for, “mono replicates a closed-source product, even duplicating some of their interfaces, and they might sue us”… and? This is specific to Mono how?
If you hate Novel that much, I’m guessing you vigorously campaign against OpenSuse. And if you hate replicating closed or proprietary software and interfaces, I’m guessing you’re not using WINE, don’t have Flash either, and don’t have any proprietary media codecs installed, and have not installed any word-processing applications that could read .doc files. In short, I’m guessing your machine falls well short of any reasonable standard of compatibility and functionality.
— Oh, wait, when you say, “created by,” you meant .net being created by Microsoft, not Novell sponsoring Mono (they do, don’t they? I’m not crazy there?). That’s… staggering. In what way does it begin to be significant that Microsoft originated .Net? They did not create Mono in any sense, and pose no greater threat to Mono than any of the innumerable other projects that might step on a patent or re-implement a proprietary interface!
Edited 2009-06-16 20:28 UTC
MS hasn’t sued any linux distributers…yet. Whether they do so in the future is the question.
semantic, shemantic … they have sued at least one company that had hardware with linux onboard, specifically their fat-patent.
now, they chose to sue a “small” company, but there is nothing preventing them to sue everybody else. linux can just pull the support, but its much harder for distributors of linux.
Right, so they’ve proven they have a fair shot at suing any hardware or software vendor they want at their whim. That still doesn’t exactly make Mono a case worthy of special consideration. I sincerely hope the community doesn’t end up running scared over the mere possibility of a MS patent suit; that would absolutely cripple the FOSS movement.
It’s exactly what makes mono special – because mono is something we’re basing future software on. That is, if there are ever any issues with mono, then all the dependent applications will be in trouble too. Much *unlike* wine and samba, which are mere applications.
Edited 2009-06-17 08:54 UTC
From http://www.mono-project.com/Main_Page
It’s not very surprising that there are people who don’t want that to happen (putting all your eggs to a basket that *may be* in a hands of someone that wants those eggs broken)
That still doesn’t make Mono special as opposed to any other application framework, such as Python, Java, GTK, GNOME, or anything else. They all could be subject to patent suits as well. What about Mono’s patent situation is different to any of those?
Thier FAT patent is a different thing all together, and it has nothing to do with mono. THey helped Novell write the thing, I doubt they’ll sue over it.
I am aware that a fat patent doesn’t have anything to do with the patents that are related to mono. That goes without saying. The problem is that by suing *anyone* for patent infringement – and more specifically, within the linux community, they have shown intention to use their patents aggressively. So, if mono ever becomes a burden they can easily stop it – and all the apps based on it. Think about it – what do you think Microsoft would do if Mono *ever* caused people to migrate away from .Net? Would they be clapping their hands?
prob wouldn’t happen, but what if…
yeah, so? – they basically used novell to gain leverage on the whole linux community by issuing a patent covenant. Microsoft bought Novell, and for all intents and purposes I would shun them.
Well, if you doubt it wont happen, I am sure we can all sleep safely…
Bull. It has everything to do with mono. Microsoft put up the full specification for fat on their website and yet sue an organisation who uses an implementation of it. He uses the fat issue as an example of how even when Microsoft is willing to sue when they want to and more importantly willing to sue organisations even for technology that they themselves have provided by way of documentation on their own website which they freely provide to all and sundry.
Whether Microsoft provided the documentation is irrelevant. The issue is software patents. It doesn’t matter who provided the documentation. If there’s a patent, there’s a patent. Microsoft doesn’t need to be the source of the docs if they have a patent they want to beat developers over the head with. Anyway, in the case of Mono, the source of the documentation is actually ECMA.
And talking about ECMA means nothing given that that they only conditions of the portions covered by the ECMA is abiding by RAND. RAND still allows Microsoft to demand royalty payments – so there is no use dancing around naked and pointing to ECMA as if it were the silver bullet to remove all doubt.
I didn’t imply that it removed any doubt re: patents. In fact, I’ve been saying all along that Mono is no different than any other software re: patents, ECMA specs or no. Licensing is a whole different issue.
Microsoft has a database server which uses SQL but they didn’t invent the language nor do they own patents against it as far as I know. They bought and rebranded the SQL database server like most of there other products. Even then, the SQL language is still a separate entity from the database server which happens to interpret it. MySQL and PostgreSQL are two other databases which happen to use SQL language to interact with data tables. Your suggesting that one has ownership over English because they happen to publish a dictionary.
Also, as someone else has pointed out; Microsoft has repeatedly threatened patent litigation and publicly denounced Linux based software as a cancer. To look at this another way, every day on your way to work the same guy jumps out from behind a push and punches you in the face. One day he steps out and offers to shake your hand; are you not going to be watching his other hand to see if it’s making a fist?
Mono is seperate entity from the ECMA specs which it happens to implement. Which makes it two steps removed from Microsoft .NET. Patents within Microsoft .NET are no different that patents within MSSQL. I’m not suggesting there are any patents to the SQL language itself.
Which means *all* Linux software is at risk, not just Mono. That’s what I’ve been trying to say.
Yes!
Just like the royalty free patent grant for compatible implementations included in Sun JDK license.
You might also want to stay away from h.264, mp3, and aac codecs. Don’t use samba either, as the SMB protocol is also patent encumbered. Considering SMB is a part of the Linux kernel I guess that precludes you from using Linux anyway so I guess its time to try a new operating system.
I’ve said it before and I’ll say it again, if people are really that worried about Mono then they should be worried about a slew of other potentially infringing software but they don’t seem to be. The problem is patents, not Mono. If you want to put your head in the sand when it comes to every other piece of software what is making you come up for air when someone mentions Mono?
The discussion is going in circles…
Programming language is a different thing from an application. Even if the whole Linux thing turned out to be completely illegal (unlikely), you could switch over to working on Solaris/whatever in a heartbeat (without eliminating your previous investment). If your programming language turned out to be illegal, you will be in real trouble.
So all of the sudden you wouldn’t be able to use C, C++, Vala, Java, or any other programming language? That doesn’t make any sense. It’s not like Microsoft can force you to uninstall your Mono software before you have an alternative anyway and considering Mono is open source, porting shouldn’t be an issue. Also it’s wishful thinking if you believe that you could just switch to Solaris without a hitch.
No, all of sudden I wouldn’t be able to use Mono. That would be pretty bad, especially if the vision of Novell about Mono having “become the leading choice for development of Linux applications” turned out to be true.
Just to be clear, I do think it’s ok if Debian added Mono to the default Gnome metapackage. This is the original problem we are discussing here. Mono antagonists need to pick their battles, and this isn’t it.
Not true at all. You would be prohibited from distributing Mono but I doubt Microsoft could forcibly stop you from using Mono. Microsoft’s lawsuit against TomTom hasn’t stopped me or anyone else from reading FAT partitions from Linux. In fact prohibiting Linux users from reading FAT partitions would be a much bigger deal than prohibiting people from using Mono because neary every flash device in existance uses FAT and there is no alternative if you want to read camera memory cards or phone memory cards.
well, then you’ve ignored it before and will ignore it again. people ARE really “worried about a slew of other potentially infringing software”. however, this conversation is *specifically about MONO*. right now, there is a concentrated effort to expand mono’s presence in the default installations of several linux distributions, so it’s sort of a hot topic. and it’s not just about patents, either.
Well if people are really worried about patents in general why is their solution to not use Mono instead of fighting against software patents? You can’t stop using EVERYTHING that is potentially encumbered. It isn’t a solution. It doesn’t help anyone and only serves to re-inforce software patents and their validity.
If this isn’t about patents then what is it about? As far as I can tell it has to do with an irrational and blind hatred of absolutely anything that comes from Microsoft.
Fighting against patents is fighting against windmills. How do you plan to fight against software patents? Write some slogans on cardboard and hang around Brussels/wherever? “Not using mono” is a concrete (albeit passive) thing you can actually do.
Probably. I don’t find this anti-Microsoft attitude a “new” or unusual phenomenon in open source circles. This might be a good time to start digging into the (some of the) rationale of that attitude:
http://www.catb.org/~esr/halloween/
(The “Halloween documents”).
How is NOT using Mono doing anything to resolve the problem? It just reinforces the idea that software patents are valid. You can concretely avoid using SMB, and therefore the Linux kernel itself, and also mp3, aac, FAT, and so on. Your argument makes no logical sense unless you avoid pretty much all software.
It’s not new it’s just dumb now. The whole attitude towards open source was what soured me towards Microsoft and caused me to dump Windows entirely in favor of Linux but that was a decade ago! Anti-Linux sentiment from Microsoft peaked years ago. Microsoft isn’t a static company. A lot of things have changed since the “Holloween documents”. If the “Holloween documents” are really what is fueling this then the hatred is irrational.
The SMB protocol is not patent encumbered. The SMB protocol is a free-for-anyone-to-implement invention of IBM’s.
http://en.wikipedia.org/wiki/Server_Message_Block#History
Samba is not part of the Linux kernel.
Samba itself isn’t but a CIFS/SMB VFS layer is in the kernel.
Edited 2009-06-18 14:15 UTC
Microsoft has added a lot of code to the original SMB protocol and hold several patents related to it. The patents are enumerated in the documentation they were forced to give to the samba team so they could try to avoid patent issues but Microsoft may add new patents at any time.
The other thing which really hasn’t been mentioned is that the patents that deal with .NET are not really .NET specific. They have to do with things like XML which could potentially affect hundreds of implementations not related to Mono or .NET in any way. It’s not like .NET itself is patented, or could even be patented, but some .NET libraries have patents pending. If it becomes an issue Mono will have to do what Samba is already attempting to do which is to work around any potentially infringing code.
and let the end user decide if they want to use it or not. For some reason people seem to think that just because it supports a Microsoft technology, it is automatically evil. Mono is licensed under the GPL/LGPL/MIT-X11 licenses. The same licenses in use by a good chunk of free software. I really fail to see the problem. I know MS has a bad track record, but so far they have been improving in leaps and bounds. Some believe MS is totally evil, and the background may very well show that. .NET sprang from the fact that MS was prohibited from distributing Java. At one time that even included Sun’s Java, though I believe that was changed shortly after the court case. Hence was born .NET, so the same functionality could be provided.
and the reason was they did not followed the Java license, they created a “Mono-platform” (pun intended :-P) Java (Windows only), they did not implemented for example the Java 1.1 GUI event model, and still wanted to be called Java. Sun sued, so they threw the towel trying to kill Java that way, so they created a Mono-platform Java clone called .Net, with the advantage of fixing some mistakes with CLR, but still with many hooks to Windows APIs than even Mono is not a full .Net implementation.
They could not control Java they cloned it. On a parallel Universe, MS could have joined to the Java Community Process to express their opinion of what they want for the future of Java, and in that universe too, they could have joined Oasis to help with the OpenDocument standard instead of creating MSOOXML like they did on our universe.
You are absolutely correct. I never said nor implied otherwise. Just stated on why they developed .NET. But then I forget that putting facts in OS News will almost always instigate negative moderation by the folks who disagree with them and live in that parallel universe you mention
True. FYI the intermediate lanuage between Java++/MS Java and C# was codenamed “Cool”. I remember reading something from the developers of Cool they were very surprised no one commented on the fact that it was so “close” to Java (a blatent rip-off in fact at first).
Microsoft lost the court battle because they tried to add extensions so that Java apps would lose their run (nearly) anywhere aspect. They did win by fragmenting the developers to that a choice must be make between developers targeting “Windows-only” (essentially .NET) and those targeting both “Windows *and* everything else” (Java).
This wastes the effort of developers who have to *still* deal with platform-porting issues in this day and age. The developers shouldn’t have to care about the platform (except when embedded-device physical constraints are significant).
…has been lost.
I mean, arguments stem from differing opinions. When one opinion is blatantly wrong an argument will end quickly. When there are some grey areas, arguments will go on longer, but resolution should still be reached. Someone needs to take the “high road” and back off.
Today it seems no one is willing to do that, at least not gracefully. People prefer to beat one another to the ground and still won’t give up.
You know what? It is alright to be wrong. Not only that, it is alright to concede, even if you still think you are right. Try it every once in awhile, people.
I agree, to a point. Taken to an extreme, tho, it can allow loud idiots to rule the world. There’s also a time to call a ridiculous argument a ridiculous argument. (I’m not saying that’s the case here, necessarily, but in general…)
Yeah, there are too many rabid nutjobs on both sides of this argument. If anyone is a voice of reason in this, they’re not getting heard above the noise.
… why not make it an option? For example, in Gentoo there is a “mono” USE flag that controls whether apps that use Mono are pulled in when installing the full GNOME.
EDIT: Though, to be fair, this is an option in Gentoo because everything is an option. However, there are various USE flags having to do with legalese (e.g., “bindist” to sanitize for binary distribution), so this is another good reason to have options. Others include preventing bloat and choosing among alternatives.
Edited 2009-06-16 17:49 UTC
Yes, I think it is the best option, why enforce mono, let it be installed as an option.
I’m pretty certain you’ll have the option to remove it if you don’t want it, even if it is in the default install.
Having recently been happily corrected, I’m working on my workstation build scripts to see about replacing my Mandriva with Lenny (I’d like the same distro on servers and the workstation I admin them from). Basically, I’ll include the parts of KDE I want rather than the KDE metapackage that pulls everything in. Where GNOME my preferred desktop, I’d do the same and include just the parts of it that I want. No need for Tomboy or Mono then don’t use the gnome metapackage and include only what is desired.
Now, I do think it should be a separate option rather than part of the metapackage but there is a way around it at least. Also, if a fundamental part of Gnome like the window framing code became Mono based, this would be a different discussion.
The one thing I do like is that the Debian folk are willing to reconsider the decision where other platform providers would simply say “nope, choice is made, it’s a required component so suck it up.”
I don’t see the issue here. Whether Debian includes Mono default or just as an option in its official repo…either way, the “software patents” argument can be used. To me then, I guess the only choice would be between not including Mono period, or including it in any manner I chose (i.e. installed by default or simply in one of the official repos).
Yeah, I don’t think it matters one way or the other, and seems like a pointless argument. One way or the other, if people want to use Tomboy or any other app that Mono depends on, they’re going to install Mono. So, when/if MS breaks out the big patent stick, whether it was installed by default or not isn’t really going to matter for all the users who depend on it.
So, if you don’t want people using Mono apps, then you need to develop (or find somebody who will) better apps than the ones people are using that are written in Mono. See, at the end of the day, it’s all about the applications. This is what a lot of Mono opponents (and a lot of open source advocates) don’t yet understand. You can take your ideology and shove it up your ass, but if you show me an app that has more functionality and runs better than the one I’m currently using, I’ll definitely make the switch if I can. If it rocks, I don’t give a damn what it’s written in.
Edited 2009-06-17 00:47 UTC
You maybe don’t but I do. I currently refuse to install any java apps on my machine even if there’s some possible application there I possibly could use. I feel the same way about any .NET applications.
You’ll likely find a sizeable number of people like me in the non windows/mac community. We’re already running linux in part because of principle. Whether or not we choose to run apps written with mono, java, etc is just a logical extension of that principle.
What youre really saying here is that Linux and the applications that run on it should effectively be restricted to using technology that predates the widespread issuance of software patents.
I don’t think that is the prevailing attitude in the Linux developer or user community at all, and the idea that we must all subject ourselves to US IP law, or shun technologies that are alleged to infringe patents to ‘remain pure’, is just garbage.
it isn’t just about the possibility of looming patent threats. though, that certainly doesn’t help things.
there is a lot to be said about chasing a moving target. not to mention chasing a target controlled by a company who has a rich history of not playing fair. the mono developers pretty clearly plan on trying to match microsoft’s implementation as closely as they can. (see miguel de icaza’s gusto in keeping up with the evolving platform, for example).
you need to face the facts. mono is a risk. whether a legal risk, a technical risk, a political or strategic risk. or even an ideological risk, for those so inclined.
supporters are keen on saying things like “microsoft probably won’t ever sue, and probably wouldn’t win if they did!”, or “you guys are just blind microsoft haters”, or “if there’s any problem in the future, we can always (fill in the blank)”.
in my opinion, and the opinions of many others, there are just too many “what if”‘s. there is a point at which it just isn’t worth the risk anymore.
* mono is destined to follow a moving .NET-shaped target basically forever
* there are potential legal issues
* for some people, performance is an issue
* there are less controversial alternatives. both for the platform itself and the apps that have been written for it
* lot of people obviously just don’t like it (that in itself is valid)
* having mono apps + dependencies in a default OS installation often takes up more space than alternatives
geez, i am so tired of arguing about this. mono does not belong in a default installation of an OS.
for those of you that argue that the end-user should just uninstall, or those that argue that it’s no different if the app is installed by default or just made available… well, you’re just wrong. if you want to pull in tomboy, a freaking note-taking app, then it should be up to YOU to install it, and all of its dependencies. the difference between ‘installing’ and ‘making available to install’ is huge. pretending like it makes no difference is grossly untrue.
these people who spout “screw ideology; i use what works!!!” need to realize that the FOSS ecosystem takes all kinds, and you simply can’t ignore that ideology is a HUUUUUUUUGE factor for a lot of people.
It’s not necessarily destined to that fate. It’s entirely possible the project will take on a life of it’s one once it reaches a certain state of maturity. It might take a fork to get that to happen if Miguel or Novell are stubborn, but the option is always there.
And there are potential legal issues with every software project (free or proprietary) thanks to the minefield that is software patents. We’ll survive.
Well, those are just the kinds decisions that developers have to make all the time. There are always tradeoffs between speed/footprint/development effort/etc. Mono provides faster development time at the expense of footprint and performance (in theory, anyway). C provides great speed at the expense of development time. There’s middle ground, too. What’s new about that?
That’s for sure.
I don’t feel there’s any one wrong or right, here. I’m sure there’s people that argue say that Evolution shouldn’t be in the base install. Or Epiphany, because it’s based on evil-Apple’s WebKit. In the end, though, it’s the people driving GNOME and Debian development that are making these decisions so their opinion ends up carrying a lot of weight in this debate – no surprise there.
You’re confusing facts with opinions… oh wait, you got it right a bit further down:
But let’s look at your points one by one:
No it isn’t. You’re deliberately confusing the issue by blurring the lines between C#/cil/clr and the Microsoft-specific bits built on top of it (ASP.NET, Windows.Forms, etc)
As has been pointed out ad nauseum: there are potential legal issues in most software. At least Mono has the power of oin behind it.
I’ll gladly concede that this point. Some people have performance issues with Mono and Mono-based applications.
Totally bogus points. To exclude a technology because some vocal minority objects to it is simply absurd. If you want 100% approval for every single decision you’re never going to accomplish anything.
Not so. Considering that the “alternatives” have no compelling apps, including them would be a greater waste of space.
Then STOP arguing about it. It really is as simple as that. You don’t like Mono? Fine; don’t use it! I don’t care! I for one have no strong feelings about it one way or another, but I would certainly not want to be without Tomboy and/or F-spot.
Why should I (or any other user) put up with inferior software just because you have your panties in a bunch. You’re in no position to demand that.
Sigh. Whenever a discussion starts to become a “Point by point” issue you can be fairly certain that it is about to degenerate into an anal-retentive nit-picking contest. Never the less, into the breach I go:
If Mono is not attempting to provide a .NET compatible environment, that’s news to me. Because I’m fairly sure that one of the very early goals (aside from the CLR) was a C# compiler and the C# class libraries, including Windows.Forms.
If Mono is not intended to be a .NET compatible runtime environment then it becomes less useful and the arguments in favour of other technologies such as Java or Parrot become much stronger, with Java at the top of the pile because it is a complete implementation of an established and open cross-platform standard.
I’m not exactly a fan of Java, but I’m fairly certain it has quite a few applications which could be described as “compelling”, at least given the same criteria as applications such as Fspot and Tomboy are being given.
A .NET-compatible environment is not what’s being discussed. Those parts of Mono are packaged separately and are not used by Tomboy, F-spot, Banshee, etc.
Really? Which desktop applications do you have in mind? I’m certainly a fan of both Java and Parrot, but there are simply no compelling desktop linux applications using those frameworks.
I was responding specifically to the claim that Mono is not chasing .NET.
Vuze (Azureus) and a lot of developers are rather fond of Eclipse.
Ok, but the relevant parts (the parts that are used by desktop linux applications) aren’t chasing .NET. Therefor, whether or not Mono is chasing .NET is orthogonal to the question at hand.
I’ve never used it, so I have no first-hand knowledge but judging from screenshots it doesn’t appear to integrate well with the desktop. I have yet to find a Java-app that does.
Gtk (and subsequently Gtk#) doesn’t integrate that will with desktop either, unless the desktop happens to be Gnome.
Which, incidentally, is the desktop that’s being discussed. Tomboy is a native GNOME application, and noone is suggesting that it be included by default in any other desktop environment.
It’s quite enlightening to know the only compelling apps are born by Mono developers.
Well, that’s the great thing about linux and free software, IMO. If you don’t like it you don’t have to use it. Nothing about this news changes any of that. And if it’s an issue that many people take exception to, I’m sure someone will start a new distribution without mono.
However, I still don’t understand where principles come into play here. Mono (the code) is and always has been 100% free software. (The situation is more complicated for Java.) If people object to it out of the distaste for the connection to Microsoft and alleged patent issues, they’re going to have to realize that not everyone is going to have the same opinion, and some decisions are going to be made that don’t fit into that world-view, for better or for worse.
BTW, I’m no great fan of mono myself, but I don’t really take issue if another developer decides it’s the best tool for the job.
Apparently, one should be careful about allowing Mono into your system. Once it enteres your system, it can be very hard to get rid of Mono. Signs and symptoms of Mono infiltration include reduced performance, sluggishness and general system malaise. The treatmenet regimen consists of a small surgical procedure to remove any infected material (Packus Monoectomy). As always, please talk to your local allied disk health providor.
Actually that system slowness is caused by Beagle which happens to be implemented in Mono. 🙂
I once installed an experimental kernel on my laptop that made Mono applications crash, so it killed Beagle. I spent some time wondering why everything felt so much faster. 🙂
(Although other document indexers slow the system down as well. It’s just the price necessary to get full text search of everything.)
Perhaps it is time to just redo Mono. Anything that they do (for apps) we just re-write it in C++ with FLTK or Qt 4.5+. Then name them the opposite… Tomboy becomes Sade’s Notes or Lola. F-stop becomes G-stop, etc.
I would have advocated Gtk for this but I think it has been compromised by MSFT.
People are dissing options without even having fckin clue.
Wine is in the same waters as mono.
– implements features to provide underlaying binary compatibility (checked)
– rewritten without original code (checked)
– original is patent riddled (checked)
– protected by OIN (checked)
– Part of it is in ECMA…? hey, … wine not, .Net yes
Personally I’m using mono because it is the best crossplatform development ever (Gtk + .Net framework) = nice, although I’m waiting for Vala to come to 1.0 to rewrite some of my work with it. At the time of my choosing between Java and Mono,… well mono was free, java wasn’t.
and…
Striking against mono developers would be impossible for MS. They’d only lose. And the fact that patents from MS against OIN are in the same s_hit as cold war was. Everyone had nuclear weapons, and no one had guts (press the button and everyone goes kaboom). It is a stale mate without a winner.
Sorry, the user doesn’t care if C# is the developer’s language of choice (which might’ve chosen after some valid consideration).
They only care ‘omg it uses mono! it must be rewritten in qt/gtk/java/whatever (which the original developer doesn’t care/have knowledge of)’
Edited 2009-06-17 03:14 UTC
Rather than cloning Tomboy in an attempt to keep Mono off the desktops of users, why not write something better?
The real problem here is that apparently there aren’t any talented C/C++/python developers up to writing something as simple as a note-taking application for GNOME that kicks tomboy’s ass feature, performance and functionality-wise.
If that happened, this entire discussion would be redundant.
But apparently, these Mono-using developers are just plain better at writing these apps than the rest of you. And as such they deserve their place in debian.
If you guys want to cry about debian bundling the best apps available, then you need your heads read.
A fork of Tomboy was ported to C++ earlier this year, called Gnote. If you’d go back and read some of the history behind the Mono-in-Debian decision you’ll see it mentioned quite a few times.
You’d think that, but apparently some people in the Debian project are desperate to push Mono into the default install.
I’m aware of GNote, and the fact its a clone of Tomboy with less features.
It’s redundant, obviously designed to be divisive, disrespectful to the Tomboy developers’ creativity and its pretty sad to see so much effort wasted on such a pointless endeavour.
I don’t use either of the apps, and i don’t particularly care for Mono myself, but i know which team of developers i respect more for their creativity and contribution to the community, and it sure isn’t the one behind GNote.
One great achievement with GNote is that it proved Mono apps can be ported “mechanically” to C++. That should alleviate concerns about Mono painting us to a corner – there is always a reasonable exit strategy.
That being said, there is nothing that “special” about Tomboy apart from great UI design. If I had the time (outside normal work & family & other open source activities), I’d make a Qt port – should be a breeze with QTextEdit.
Yeah, but QT/KDE already has a better note taking app than Tomboy: BasKet Note Pads
Then why did you make the claim
if you know there is one?
It’s a port of Tomboy. It also less dependencies. That’s the argument that both sides are currently battling over.
I’m also at a loss to quantify how many features one can put into a note taking application, but maybe I just don’t have enough imagination.
Whilst this wasn’t the point he/she was making, parent did say a “better” note taking app. Gnote is a line-by-line rewrite from C# to C++ and re-implements nearly all of Tomboy’s functionality and existing bugs faithfully. IMHO that isn’t “better” by any stretch…
Rather than cloning Tomboy in an attempt to keep Mono off the desktops of users, why not write something better?
Because average user seems to be uneducated and trolling a_ss like Robert Miller. First completely blatantly copy/pastes Tomboy and then bashes over original.
There is one reason why Gnote will never bee my computer. I won’t support trolling and a_ssing.
The real problem here is that apparently there aren’t any talented C/C++/python developers up to writing something as simple as a note-taking application for GNOME that kicks tomboy’s ass feature, performance and functionality-wise.
Inkscape, scribus…? There is a lot of good applications out there
But apparently, these Mono-using developers are just plain better at writing these apps than the rest of you. And as such they deserve their place in debian.
Hey, I use mono for my development, but even so… no. It should go in main repositories, but keep base small. In my opinion even open office or java doesn’t belong into base, same for gcc and friends.
And even more truthfully? I would be happiest if mono (java and others too) team would provide mini install(as such… who’d care for few kb preinstalled on his computer). Then after, everything comes from central repository. something like using
perl -e ‘use CPAN; install XXX’ in command line. they could extend that to repositories too. dependancies fullfilled and so on
If you guys want to cry about debian bundling the best apps available, then you need your heads read.
Rather than cloning Tomboy in an attempt to keep Mono off the desktops of users, why not write something better?
won’t happen until a_sses like RM exist. reaping other peoples work is so much easier
Robert Miller didn’t write GNote, he’s packaging it IIUC. Hubert Figuiere is the dude that did the porting work.
But the Mono itself is too bloated. Debian couldn’t afford to be more bloated with Mono-related apps and components.
At the moment: Mono = bad