In a recent interview with the online edition of the Austrian daily Newspaper Der Standard, Mono project-lead Miguel de Icaza pleads for a cooperation between Mono and Microsoft’s .Net: “I think that the deal should include a technical Mono/.NET collaboration, and even go as far as Microsoft recommending Mono for all of their developers looking at migration.”
I wouldn’t hold my breath on such a deal; I think de Icaza will sooner be hit by a lawsuit than a business deal.
He must not have heard about the story of “The Scorpion and the Frog”. This is business and Microsoft did what was necessary to protect its competitive advantage (whether present or future). What they should do is continue to innovate and maintain that competitive advantage. There are so many instances where the focus has been on keeping up where it doesn’t quite matter and not leveraging/marketing the advantages enough. I like that they have developed a tool to aid with migrations but as Daniel Robbins found out(leaving due to frustration over not being able to work at his “full level of technical ability”), when entering into scenarios such as this, competitors are only going to assist you as far as it benefits them, no further. You still have to work on what makes you better. So when he describes doing WPF better, I applaud that direction (while still maintaining compatibility of course)
Edited 2007-03-25 19:02
Actually don’t be so sure about that. Microsoft is well aware of the existence of Mono, and them supporting Mono only boasts .NET over Java. However their support isn’t active with money or sharing of code, because they want to make sure Mono and .NET forge ahead on two different paths. But many of the Microsoft developers also work on the Mono project.
Also you have to understand that Miguel is shooting for Mars in hopes of reaching the Moon.
Actually don’t be so sure about that. Microsoft is well aware of the existence of Mono, and them supporting Mono only boasts .NET over Java. However their support isn’t active with money or sharing of code, because they want to make sure Mono and .NET forge ahead on two different paths. But many of the Microsoft developers also work on the Mono project.*
Yeah, right. Microsoft acquiesced to pirated copies of their business software and OSes, too.
Until the day they had achieved dominance.
Now, copyright infringement is clearly illegal while the status of Mono is questionable. But if you’re a Mono developer, you should pray that .NET will never ever win against Java, because the next day MS will start collecting royalties and sue everyone who doesn’t comply.
And the way the system works it will take tens of millions of dollars to win those lawsuits even if Mono is perfectly, 100% legal.
*: Is the preview broken? Because if I follow the bold blue statement and use the q tag it just shows the tags. But blockquote seems to work. Strange.
EDIT: But not in the posted version. Is the new osnews usable? And is its preview less screwed up?
REEDIT: I can now answer my last two questions:
Seems to be. No.
Edited 2007-03-26 17:45 UTC
I don’t think you really understand the benefits of .NET working on other platforms. Or at least you don’t see it from Microsoft’s perspective. Who do you think is going to benefit from a wealth of software being developed on .NET? Especially all the new compilers that compile against the CLR. There is Python, PHP, COBOL, FORTRAN, some supported by organizations some supported by guys that just program on the weekends. So what good does it do for Microsoft to shut down a very useful compiler, like PHP.NET, that still compiles .NET code and runs on Windows perfectly. There is none.
And Microsoft understand that not all these guys that are going to be providing very useful tools for its Windows developer that adds to the value of running Windows are going to come from the Windows side of the world. I think you are still stuck in the 90’s with your mentality about Microsoft.
I think you are still stuck in the 90’s with your mentality about Microsoft.
I think you are still stuck in the reality-distortion field with the idea that Microsoft has changed. They haven’t.
And neither have you obviously, so why expect anything from Microsoft?
Obviously I don’t believe what I just said, but it is rather self centered to believe you have all the answers.
Should they step further into that Microsoft shit – sorry, but that’s what it is – I’ll switch to KDE. De Icaza hasn’t listened carefully to what the communitry has said about the Novell-MS deal. Tying GNOME even closer to .Net is even worse, because the day will come when Microsoft goes to court with that. I wouldn’t be surprised if it was Mono they’re after (well, to start with).
Ballmer has said it on and off that OSS is violating MS patents. So it’s just a matter of time when he’ll be the one to open the flood gates.
Should they step further into that Microsoft shit – sorry, but that’s what it is – I’ll switch to KDE
The only connection between Gnome and mono is the GTK# bindings. mono is a cross language runtime and particular bindings for a GUI api are just a side show. You are only tied to Microsoft via mono if you use a Microsoft api such as Windows Forms or ASP.NET. If you use GTK#, Cocoa# or KDE’s Qyoto/Kimono bindings (that I work on), you are only tied to the C# language and other CLR languages. The CLR is an ECMA standard, whereas Windows Forms isn’t, so please separate issues about using Microsoft specific GUI’s from using mono as a basis for implementing interesting non-Microsoft apis via bindings.
Yes, I know that.
But the point is that the lawyers in Redmond are already evaluating things. MS _will_ eventually stand up and sue OSS for patent violation. Ballmer has to act quite soon if he does not want to appear a total idiot to the Micrposoft shareholders. I bet he’ll start with mono – well, ok, just a not so wild guess.
Sorry to say this, but mono was a waste of time from the start. It’s too close to the devil, and Novell has seen what happens when you embrace the devil.
Now it’s time to talk about Java. It’s not too late – thank God …
It’s too close to the devil
Isn’t there a version of Godwin’s law about comparing Microsoft to Satan?
…mono was a waste of time from the start.
Considering C# is an ISO standards based language, I don’t think you can say mono was a waste of time. It brings C# to the playing field for platforms other than Windows. If anything, it has allowed people to bring their ASP.NET apps written in C# to Linux based web servers. Yes, we’ve had Java for a long time and it’s run things cross platform, but this is another option for the masses. Besides, there have been plenty of really good apps for Linux written in C# using Mono.
Considering C# is an ISO standards based language, I don’t think you can say mono was a waste of time.
That’s a good point. It’s not like Mono is completely walking on thin ice, from a legal perspective. (You always have to be wary of Microsoft, of course)
What’s wrong with you people 🙂 C# is a good language. Java, since 5.0, is also good enough (since it copied over certain C# features). Personally, I wouldn’t mind *either* of these languages prevailing the Linux desktop (both GTK and QT based). How long do you think a non-paid developer can go with C or C++ nowadays? I’d trade more features, due to easier development, to the skeleton apps written in C that want to pass as Gnome apps.
Conclusion: higher-level languages are developer friendly and grass roots OSS friendly. Embrace them!
Why not do it properly then, and use even nicer languages? Several are available. Free.
I personally love Python, but since C# is an ECMA and ISO approved standard, I see nothing wrong with it.
I’m sure Java infringes on some MS patents, that’s the nature of software patents, everybody who makes software is infringing on someone’s patents.
But Java has Sun behind it, sun has money and other patents. So there is a good chance microsoft wouldn’t attempt to sue over such patents.
Mono is currently being well supported by Novell, who have a deal with microsoft to licence microsoft patents.
Microsoft suing the Mono project would be very bad for Novell and hardly an advantage for Microsoft as Mono is less of a threat than say, OpenOffice or Firefox.
There are two sides of this issue.
Sure, if you don’t use these bindings, you aren’t affected at all.
The GNOME people seem to get fed up with C. So it is even evaluated to switch to Mono/GTK# as main development platform. This wouldn’t be something done today or tomorrow. But GNOME already ships applications using GTK#.
Mate, don’t worry about it; GNOME hasn’t included anything relating to Mono within the standard distribution – what you see in distributions are merely the result of its users asking for a said application, which happen to be written using mono.
The debate is over which framework should be used as the default for future development if GNOME developers choose to go down the road of having a managed environment – Mono, as much as De Lcaza argued that the technological arguments were compelling vs. the obvious licencing issues back then in regards to Java, the elephant in the room can no longer be ignored; the underlying motivation for the creation was the restrictive and closed source nature of Java.
Flash forward to 2007; Java has been open sourced under GPL; Sun is now an active opensource development company, with Solaris Enterprise System (the old iPlanet stack) will be opensource soon as well – which brings forward the question; which one will be chosen? for me, the choice is obvious now – a couple of years ago, I would have said mono after weighing up the patent, licencing and technological issues, but today with Java being opensource, more mature technologically, a big community of developers overall, and the fact that the issue of patents don’t hover over the neck of Java like a guillotine over the neck of a falsely accused person, as with the case of Mono.
With that being said, I do think, however, that GNOME need to think long and carefully about using Java – use it where it works well, but don’t use it for the sake of simply being able to say, “look at us! we use Java!”.
Edited 2007-03-25 23:44
If I’m not mistaken, Tomboy is in 2.18. Tomboy uses Mono and GTK#. You can take it out, but it is now part of mainline GNOME
Yes, you are correct; it is part of the ‘package list’ but at the same time, however, Mono isn’t playing a key roll in the core of the desktop environment; if it sits on the side as an option I have no worries – people should have the right to choose over what VM based language they wish to use, but at the same time, I would concerned if it started to become a ‘must install’ component of GNOME.
The debate is over which framework should be used as the default for future development if GNOME developers choose to go down the road of having a managed environment
You’re quite clueless regarding the antipathy towards Java, whether it’s open source or not. You make the same bad logical analysis that developers are going to flock to Java just because it’s almost open source.
The debate isn’t over, because Gnome won’t have an official runtime. Java is hated for various reasons, Mono is hated because it’s MS, and other developers would rather stay away from both of them and use Python or Ruby. Gnome is a C platform – plain and simple, with various bindings that developers use for various applications that might or might not make it into the official Gnome applications releases.
With that being said, I do think, however, that GNOME need to think long and carefully about using Java – use it where it works well, but don’t use it for the sake of simply being able to say, “look at us! we use Java!”
That’s something you don’t have to worry about. Java will be ignored only a little less than it is now.
You’re quite clueless regarding the antipathy towards Java, whether it’s open source or not. You make the same bad logical analysis that developers are going to flock to Java just because it’s almost open source.
You talk about Java as if no one uses it. Sounds to me like you personally don’t use it, so you maintain this illusion to justify your decision to yourself.
Java is hated for various reasons
In other words, you hate it, for reasons you don’t care to disclose.
Java will be ignored only a little less than it is now.
You’re right, it’s just being ignored so much.
http://www.eclipse.org
http://www.limewire.org
http://azureus.sourceforge.net
http://www.neooffice.org
http://www.trolltech.com/developer/downloads/qt/qtjambi-techpreview
And no one’s interested in developing anything new for Java on the desktop.
http://community.java.net/projects/community/javadesktop
http://java.sun.com/developer/technicalArticles/tools/nb_guibuilder…
Sorry, you seem to be blinded by your own prejudices. Please, go back to your hole.
Please try to pay attention. Eclipse, limewire, or any of the others in your list have nothing to do with Gnome.
You have a point, but my point is, Java already has quite a bit of traction on the desktop, and Gnome is in a position to join in. If they don’t, then Gtk-based apps may just end up getting left behind by applications that are not Gtk-based, but instead SWT or Swing-based, yet still integrate nearly seamlessly with the Gtk look-and-feel (not to mention Windows and Mac L&Fs). A lot of projects, particularly those not confined to the Linux community, are interested in the idea of compile-once, run-everywhere. Gtk+ by itself does not fulfill that goal; Mono only goes halfway there.
P.S. I forgot to mention that the mathematics software Maple and the modelling tool/IDE Visual Paradigm are also written in Java. Again, not Gtk-based, but pretty visible and important apps that run across Windows, Linux and Mac. Yes, they are commercial and not open-souce, so you may count them out of the “Linux desktop” equation if you feel like it; I’m just pointing out the potential that has already proven to exist for implementing large, complex Java desktop apps.
Edited 2007-03-29 06:28
You have a point, but my point is, Java already has quite a bit of traction on the desktop, and Gnome is in a position to join in
It’s the other way around. Gnome already has quite a bit of traction on the desktop, and “Java” is in a position to join in. Well, not really. Sun has never cared about the Unix desktop as witnessed by their complete ignorance over everything not Swing-based.
You can stop pointing out Swing apps as they are completely pointless to anything that would ever be accepted in a DE distribution.
I’m just pointing out the potential that has already proven to exist for implementing large, complex Java desktop apps.
Yes, everything has “potential”, but desktop Java is always on the verge of being the comeback kid.
Should they step further into that Microsoft shit – sorry, but that’s what it is – I’ll switch to KDE. De Icaza hasn’t listened carefully to what the communitry has said about the Novell-MS deal. Tying GNOME even closer to .Net is even worse, because the day will come when Microsoft goes to court with that. I wouldn’t be surprised if it was Mono they’re after (well, to start with).
Ballmer has said it on and off that OSS is violating MS patents. So it’s just a matter of time when he’ll be the one to open the flood gates.
I don’t really see how Miguel has anything to worry aouut. Mono was a Ximian project. Ximian was bought by Novell. Novell has a patent agreement with Microsoft. Novell has the copyright to Mono and many of the Mono projects which it licneses under the GPL. The only company that Microsoft could sue is Novell which according to their agreement they cannot do.
Sure, if all the Mono projects are tied only to Novell-created distros. But as soon as another distro, or the Gnome project itself, decides to include a Mono application, then they are in risky territory.
Sure, if all the Mono projects are tied only to Novell-created distros. But as soon as another distro, or the Gnome project itself, decides to include a Mono application, then they are in risky territory.
Not true at all. They could only sue if the application itself infringed. The language itself is a standard. The only issue could be the libraries themselves, much like java classes. No one got sued for making GPL java applications and there is a reason why. If Mono is indeed infringing and Microsoft sues the only one they can sue is Novell because the code is copyrighted and licensed by Novell. Microsoft’s deal with Novell prevents that.
wrong thread again.
Edited 2007-03-28 05:24
Man, if Miguel is telling the truth there may be more to the deal then at first realized. Check this out –
It’s also funny how people justify they’re actions to themselves. Read the following for a prime example –
As for wanting MS to tell developers to use Mono for cross platform development, the guy must be hallucinating. MS does not want cross platform developers, they want to tie you to they’re platform. End of story.
really, this guy is on crack
Miguel can publish what he wants, I’ll never use mono. I don’t need it and I don’t want it, this little child of the Novell-Microsoft cooperation.
Even in OpenSuse 10.2 it is still possible to live without mono. Although it is part of the default install, it can be uninstalled without causing any problems. As long as this is the case, I have no problem to use or test an OpenSuse distro.
MS does not want cross platform developers, they want to tie you to they’re platform
Very true.
However, one thing we have learned from the Microsoft/Novell deal is that Microsoft knowns that some of their customers are at least considering Linux for certain areas.
So in order to avoid having them consider other development frameworks like Java, which are totally not under their control, they might “recommend” a framework they are indirekt control over, e.g. they control the direction of .Net, thus controlling what mono will have to do as well.
An application runable under mono will of course not be as tied to Microsoft as a native application would be, but on the other hand they can more easily convince customers to switch back if the application’s performance on Windows can be boosted above the performance on Linux+Mono due to optimiziations in their runtime.
Nicely said and I certainly didn’t see it that way. But what if a day comes where Mono is more feature rich, better supported and commands a larger number of developers than .Net (however unlikely)? Do you really see MS lay down and give up? Or, much more likely IMO, start a patent and/or IP infringement lawsuit?
Do you really see MS lay down and give up? Or, much more likely IMO, start a patent and/or IP infringement lawsuit?
Of course they won’t give up, but as you said such a situation is highly unlikely, there is IMHO no point in speculating what they would do then.
In order not to get into this situation, they will keep on extending .Net in a way that is hard (or even impossible) for others to implement.
This is something they have huge expertise in.
If that day comes, Microsift has a problem:
Should they sue Mono, thereby hurting their own development environment (because at that point in time the feature “platform independent” will be a killer-feature), they will find themselves hurting their own business.
So they will most probaly do what they are doing now: FUD-ing around but not really sueing the free projects.
As for wanting MS to tell developers to use Mono for cross platform development, the guy must be hallucinating. MS does not want cross platform developers, they want to tie you to they’re platform. End of story.
That is definitely 100% true – just wait till Microsoft starts prancing through the Mono developer community demanding royalties and payments for those not running their software on the Novell Linux platform.
The licence agreement between Novell and Microsoft, is just that, a licence agreement between them, and only them, it does not cover the whole Linux or opensource community – which is what Miguel is some how trying to lay claim to.
Mono is still the same licence and patent riddled mess it was last year and the year before, the only difference is that only Novell customers are going to be spared the wrath of Microsoft – for all those who aren’t Novell customers, I would think long and hard about whether the risk of getting sued into an early grave is worth the risk in the long run for so-called perfect portability which doesn’t exist.
I for one am hoping that with Java going GPL that great things will be happening in the Linux world with it.
+1000
I have no more to say.
MONO shit, I don’t know how they started that project. Miguel de Icaza really seems a Novell mercenary that only makes the OSS shit that Novell directives say him.
Seriously, Icaza has lost his way and the sad part is this is affecting GNOME project too. One of the reasons is being mostly owned by Novell.
I think Gnome needs an business-independent fork and avoid that Icaza madness.
A decent and up-to-date implementation of .NET without a host of potential legal issues would be a really cool thing. It would make it much cheaper to switch between platforms. Choice is good.
Unfortunately, the greatest advantage would be when migrating from Windows to Linux. As such, Microsoft would never accept such a deal.
Such a thing would only happen, if Microsoft lost its monopoly power, and would have to fight on equal terms (f.ex. if Microsoft had been split up in the first place).
The fact that Icaza can even suggest such a thing, suggests that he is living in a fantasy land.
the more you use mono, the more Miguel sees his team’s resources (and probably his own income) increase.
derStandard.at: Rumour has it, that Novell is substantially staffing up it’s Mono-team, any chance you’ll put this into numbers?
Miguel de Icaza: We are doing a big push for Mono.
derStandard.at: Where will this increased resources go?
Miguel de Icaza: We will try to staff up all the areas in Mono that need better support, it is still something that we are discussing internally.
That’s pretty logic if you think it .. but I can’t avoid wondering to what extent is Miguel really interested on anything else but that …
just a thought ..
Edited 2007-03-25 19:10
Oh, his income is increasing. Holy mother of god, how dare he!
sorry, maybe I was to subtle for you …
it won’t happen again.
No, subtlety implies intelligence.
too subtle again .. sorry, I just can’t help it =(
I haven’t agreed with the reasons behind starting Mono over the years, particularly some of the regurgitated Microsoft marketing about .Net. However, if they could have looked at some of the ways they thought .Net was better, as well as Java and up and coming stuff like Ruby, then I think they could really have had something. As it is, they wanted to create a .Net compatible clone which will never get them anywhere. Additionally, Windows.Forms is stillborn even in the Windows world because Avalon is supposedly the next big thing.
If he thinks he’s going to get collaboration from Microsoft for Mono to be anything other than an interesting little educational project, like Rota, he really should check himself in somewhere. There is ample historical evidence that this will simply not happen, and the deal was an ill advised sham.
Additionally, Windows.Forms is stillborn even in the Windows world because Avalon is supposedly the next big thing.
Wrong. Microsoft isn’t Sun. These days, Microsoft is pushing out tech faster than adoption. As was stated in the interview, Avalon(WPF) is years away from having a substantial presence. Business tends to be conservative, and doesn’t just rewrite stuff because Microsoft pushes it out.
“”I think that the deal should include a technical Mono/.NET collaboration, and even go as far as Microsoft recommending Mono for all of their developers looking at migration.”
Microsoft will never recommend anyone “migrate” away from the Microsoft platform, and as such Icaza is looking at this the wrong way. In the realm of software, there are always business constraints as to why a particular software package needs to be developed on a specific platform (Windows, Unix/Linux, etc). What they need to do is make Mono a compelling reason to write software on a platform other than Windows, i.e. RAD, a good IDE, truly portable bindings, etc.
Once software is written for a specific platform, there is almost zero justification for porting it to another platform (in the realm of IT at least). New development is an entirely different story though. Give us a reason to use Mono on other platforms, and we’ll choose it.
Again, migration: Not an option. New development: huge opportunities so long as there are compelling enough reasons.
FWIW, there are a ton of Microsoft bloggers who regularly post about Mono, and do indeed highly recommend it for non-Windows development. MS can’t take an official stance on it because it doesn’t make much sense…it would be like them recommending someone use Oracle.
Microsoft will never recommend anyone “migrate” away from the Microsoft platform, and as such Icaza is looking at this the wrong way.
Migration can easily be in the opposite direction. Of course Microsoft would never recommend that Windows developers start using Mono in order to migrate to non-windows platforms. That’s just poor business sense. However, The opposite cwould make perfect sense. Microsoft would be well off to encourage existing Linux / Unix software developers to migrate to Mono in order for their applications to be ‘Windows Ready’.
It’s really not a 2 way street in the development community. Windows developers could usually care less if their applications run on other platforms. Linux, Unix, and most OSS developers, on the other hand, are usually concerned with their applications running on ALL platforms, often even windows. Mono provides such an oportunity.
Now granted Java can provide such an oportunity as well, but that’s for another topic.
Now that Microsoft and Novell have a deal, Microsofts money seems to be funding the increase in developer resources for Mono.
Mono seemed to have been slipping further and further behinds Microsofts .Net. With the increase of paid developers maybe Novell can get v2.0 and winForms finished which could make Mono attractive again – it seems to have lost a lot of its lustre and support. Novell probably realises that if Microsoft did go after companies without agreemenets like the Novell one – after Mono became entrenched in the GNU/Linux ecosystem – then it would, at least in the short term, be a huge win for them.
Fortunately Java, on the desktop in particular, is on the rise and may stand a chance of becoming the standard for cross platform development on Windows, Unix, Linux and MacOS. KDE may have other ideas but its hard to see them beating Java.
We’ll see.
Miguel is like Linus – TECHNOLOGY DRIVEN! I love them for that.
Mono rocks, there is no question about it. Java on linux – well, since it is GPL now, there is still hope, but there is a huge hill to climb for java to be any good for the linux desktop. This far, it has only been a detriment, a big time-sink. (kaffe, gcj, ugly homemade mismatched gtk-themes for java) (Note that I am not mentioning java server apps. Java rocks on linux for that.)
But Mono+C#+Gtk# is a mix of GPL, LGPL, MIT Licence, Ecma standardization..it can’t get any better than that. Meanwhile everyone just poured their love over Java, a severely locked in language before the GPL. Because Sun is somehow “good” for “us”. (“Us” being the collective unix/linux-geekdom.)
The politics and business deals around OSS platforms are important, and Microsoft is a company that is much more selfish than many other. (They all are, Sun, Google, MS, Redhat.. MS is just so much better at it, making them scary and untrustworthy).
But any patent that MS can threaten Mono with, they could just as well threaten ANY other runtime with, it being parrot, kaffe, the jvm.
And in any case, if you still think Mono+C# is a bad bet, why ALL THE F*CKI’N hate? Grow up people! The long and short of it is that Miguel rocks, and there is a big chance that you don’t.
(For the record, I rock too.)
That rant was a notch too trollish, so let me clarify some bits. It looks like I poured crap on some really great tech there. I didn’t mean it like that.
1) Gcj – Yup, really cool. But not really taking the OSS un*x desktop anywhere. IMHO. But it rocks.
2) Gtk themes for java – Any work in this direction is great! Sorely needed, hugely wanted. Still not good enough. But it rocks. Or will rock.
3) I wrote that Kaffe represented a time-sink. Wrong. Edit it out. Kaffe is beyond awesome. Maybe just not a great hope for a desktop-level OSS runtime.
4) Java in general – No other language/runtime in the history of man have such great development environments as Java has today, free or for money. Not smalltalk, eiffel, .Net, C++. Do I have to mention it rocks? Rocks hard even? But not in the area of desktop apps. Nope.
(note – I don’t believe in the edit function)
Edited 2007-03-25 20:48
Quote:
(note – I don’t believe in the edit function)
Edited 2007-03-25 20:48
/quote
Great sense of humour.
I’m just wondering, how many other things “Rock”? or “Rock hard” even? You sound like an expert in the area so maybe you could enlighten us. 😉
Oh and for the most part I agree the projects you mentioned are all worthwhile.
Miguel is like Linus – TECHNOLOGY DRIVEN!
And here I was thinking de Icaza set out to start Gnome because of the evil Qt license, that is to say, because of politics. It’s funny how things are evolved…
And here I was thinking de Icaza set out to start Gnome because of the evil Qt license, that is to say, because of politics. It’s funny how things are evolved…
Not “evil”, just not free and not open source. Just like the original .NET implementation that Miguel is working to replace with a free one.
Yep,
Miguel seems to have become a love boy for all technologies that Microsoft produces. Maybe that is a way for him to compensate the fact of not being able to work there.
… if he expects (or even thinks) that MS will “cooperate”.
This is ***Microsoft*** we’re talking about here!
One gets the impression that more than a few commenters here simply do not like the idea of “Mono as an improved developer platform for a fairly substantial segment of the developer market, and also as a tool to assist Windows developers make their software run on Linux, MacOS X, BSD and Solaris”. Hence the rather childish name calling against Miguel — one of the biggest figures in the free software world.
I can see where your coming from but as I am not a developer, I can’t really comment on the validity and/or usefulness of Mono. However, I do understand the general feeling of betrayal many people in the Linux community express in relation towards Miguel De Icaza.
Sure, calling the guy names is a really childish thing to do and I think you are very right to point out that fact but keep in mind that it is Miguel’s (former) standing in the FLOSS world that makes the betrayal (perceived or not) all the more hurtful and hard to take.
No wonder people are getting upset about it! 😉
Just my €0.2
Just read the book.
If Mono isn’t stupidity I don’t know what is.
If you get the chance read the Microsoft/IBM interoperability agreement story between Win32 API and OS/2.
Yikes!
Gad. Stay away from that Ximian Mono Monkey Sheeeet.
-Hack
If you get the chance read the Microsoft/IBM interoperability agreement story between Win32 API and OS/2.
Indeed. People just never want to learn from the past.
Most posters here seem to have missed the relationship between .NET and Windows.
It seems clear that .NET is the new application-level API for Windows. I see Mono as a way to extend that API – that is, extend the Windows API – to other operating systems. From that perspective, the notion that Microsoft would support Mono and recommend it to partners is rational.
Windows is more than an operating system, a fact that can be see in the myriad computing devices on which it appears. It is a +software platform+ grounded on interoperability between its various forms. The mobility of .NET is a natural consequence.
From that perspective, the notion that Microsoft would support Mono and recommend it to partners is rational.
It’s not rational for any company to make decisions that actively damage themselves.
“It seems clear that .NET is the new application-level API for Windows. I see Mono as a way to extend that API – that is, extend the Windows API – to other operating systems. From that perspective, the notion that Microsoft would support Mono and recommend it to partners is rational.”
Microsoft is not interested in other operating systems – the only way they are interested in them is to search and destroy them.
It is by no means not different with Novell. MS knows that some of its partners want to run a Linux OS. So they’re cooperating with a strong distributor. In the future that may drastically change. Who knows? MS could buy Novell, and run SUSE 12.0 in a virtualized invironment inside Vista X (or whatever the name for that new fantastic OS is going to be).
The moment you start to trust Microsoft, you’ve lost already. So don’t even _think_ about it.
Microsoft may indeed be interested in other operating systems insofar as they can serve as a substrate for the Windows software platform. It is the same rationale that drove them to build Internet Explorer for Mac.
Look at Java – it is likewise a software platform that runs on many operating systems. Microsoft may want Windows to do the same via .NET and previously HTML. Think “obiquity”.
I remember when Icaza first made the news, he was perhaps one of the most hyped person in open source. However, his constant pushing for Mono/Net when noone in open source really wanted it, has pushed both him and the Mono project completely on the sideline. When I see his name nowadays, I think what a joke and sell out. Now that Java is GPL’d, Mono is now even more useless.
His Hyped was well earned. Because he didn’t just talked like most, he did a lot of good work.
Gnumeric, Evolution, Mono, etc… He’s one of OSS’ few good assets.
Java’s slow and bloated. Still doesn’t have a desktop application that could help it gain some popularity. Most Java applets found inside web pages have been replaced by Flash. Apart from embedded and server, Mono has nothing to fear from Java on Linux. Even the most misguided OSS fundamentalists/idealists won’t be able to stop Mono. It’s already included in GNOME. ^_^
You obviously are talking about Java from years ago. Java6 is fast and Java7 will probably be more so. I am hoping that Java being GPL that there will be a lot of push technology wise to make it even better on the desktop.
Java’s slow and bloated. Still doesn’t have a desktop application that could help it gain some popularity
Azureus
LimeWire/FrostWire
Eclipse
Lotus Notes 8 client
RSSowl
Jedit
DrawSWF
Freemind
Ganttproject
Arachnophilia
Jabref
Bioclipse
Jdiskreport
Marvinsketch (essential for me and the best program in its field)
Marvinspace
Jmol
JUploadr
Jgnash
Jajuk
Jchempaint
Jago
Batik Squiggle
and more…
Good list but you didn’t even mention all the countless enterprise management tools that use Java. Java is not slow and bloated. It goes from powering the smallest cell phones to the largest super computers. .Net is dead on everything but Microsoft.
Java’s slow and bloated. Still doesn’t have a desktop application that could help it gain some popularity
Azureus
It’s funny that when people want to pimp Java on the desktop, they mention Azureus as one of the killer apps. The reason why this is funny is because most people (at least in the Windows world) who try it usually end up switching to something else because it is such a f**king strain on memory/resources and crashes every 15 minutes. So you guys can keep the whole ‘Java on the desktop’ shit over on Linux where it belongs
Edited 2007-03-26 04:54
I use Azureus on Linux, and it works just fine. It does use some memory, but that’s not a problem with an OS with decent memory management. All of the complaints I’ve seen about Azureus have been from Windows users, which suggests where the real problem is.
Another desktop Java application that I use is ProjectX. It does a good job converting my MPEG2 DVB captures into the peculiar version of MPEG2 required for DVD players.
When I searched for special-purpose utility programs, I used to always find them written in Perl. Then there was a shift to Python, and now Java seems to be coming on strong. My sample size is too small to project a trend, but I wouldn’t write off Java, either.
If Azureus crashed every 15 minutes on my computer, I would change my computer 🙂
On the other hand, Mono, basically, has four apps:
1. Beagle
2. F-Spot
3. Tomboy
4. HelloWorld
The first three of these apps use more memory, in my testing, than Firefox. That’s right. A simple sticky note app is more resource intensive than a full featured web browser. (I haven’t tested HelloWorld.)
Could this have anything to do with… Mono?
The first three of these apps use more memory, in my testing, than Firefox. That’s right. A simple sticky note app is more resource intensive than a full featured web browser. (I haven’t tested HelloWorld.)
Just a quick couple of plugs. I have replaced Tomboy by Zim which is PERL-GTK2 and a far superior and more capable program.
As for Beagle, I use Tracker instead which uses a very small daemon written in C and therefore is a much lighter and very capable substitute with a good GTK2 client, Nautilus integration and can even be used from the command line.
“””
As for Beagle, I use Tracker instead which uses a very small daemon written in C and therefore is a much lighter and very capable substitute with a good GTK2 client
“””
Yes, tracker is good. Its engine is far faster; There really is no comparison.
Its only disadvantage at this time is that it does not have as many data sources from which it can pull. But they are making rapid progress on that.
In fact, its small development team has come from way behind and has almost caught up with Beagle, which is written in Mono and has full backing from Novell.
Doesn’t that pretty much discredit whatever claim that Mono might have had left regarding rapid application development?
And with that discredited, what else does Mono have left to recommend it?
Edited 2007-03-26 16:44
Doesn’t that pretty much discredit whatever claim that Mono might have had left regarding rapid application development?
And with that discredited, what else does Mono have left to recommend it?
One thing, it makes it easier to port .NET apps from Windows to Linux and other *nix.
This is what makes MONO a double edged sword for MS. On the one hand it legitimates .NET as a cross platform technology and hopefully for them, draw developers away from Java. On the other it provides an escape root from MS lock-in. There is the potential for a lot of ISV to start porting their stuff to Linux etc., and it could turn into a floodgate.
So I too am in two minds about MONO.
Edited 2007-03-26 17:20
“””
One thing, it makes it easier to port .NET apps from Windows to Linux and other *nix.
“””
I agree with this. Mono is good to have for that segment of the developer community which would be 100% Microsoft-centric otherwise.
It is using Mono as a foundation upon which to build *our* infrastructure that I oppose.
A friend of mine from long ago was fond of saying: “Better to have it and not need it than to need it and not have it.”
That’s how I feel about these MS compatibility projects like Wine, Mono, and Samba. (Though with the Unix extensions, Samba is really coming into its own as a good solution for Unix<->Unix connectivity.)
As the author of tracker, it is quite wrong to compare tracker and beagle as is.
Tracker is a extensible metadata daemon with a high performance RDF style triple store database + a high speed indexer. Beagle in comparison is just an indexer.
What that means is you can use tracker to store tons of metadata, do desktop tagging, cross reference and cross query all your metadata and provide a next generation integrated desktop. Beagle can not do any of that as its not designed for the task. Tracker has the capability to do almost anything the nepomuk guts are doing for KDE4 as well. So Tracker is much much more then Beagle as things stand atm.
Of course we are also fast catching up with beagle on the indexing side of things and we expect to have similar level of support within the next two months.
As for development speed, for non-OO apps like a daemon you will find C to be as fast to develop in as C# (C# and java force you to use classes and OO style which can result in typing more code). In areas that are more OO based like using widgets and xml, C# wins hands down.
Also a lot of the donkey work in tracker is done in sql which is of course orders of magnitude more productive than either of them + it also benefits from having a KISS design.
“””
for non-OO apps like a daemon you will find C to be as fast to develop in as C# (C# and java force you to use classes and OO style which can result in typing more code). In areas that are more OO based like using widgets and xml, C# wins hands down.
“””
No doubt true. But in the case of handling widgets, python beats both of them hands down.
And where execution speed *is* an important issue, a tiny amount of C here and there at critical points often works very well, indeed, since Python makes it so easy to mix and match as necessary.
Eclipse – I have to wait 10s for their version of IntelliSense (not even comparable to MS) to upadate on Pentium D 2.8GHz with 1GB of RAM.
If java is fast, then what is so god damn slow ?
Edited 2007-03-26 16:26
Eclipse != Java. The fact that Eclipse is slow does not automatically mean that Java is slow. Your grasp of logic fails.
On my Macbook and my P4 1.8 Ghz Linux machine, I get no delays in IntelliSense. Perhaps you should check your preferences to make sure that there isnt a 10 second delay in showing the tooltips?
Ever heard of IKVM? Now that Java is GPL’d, Mono gains in power. If I can use all of my Mono *and* Java libraries, I am better off than if I have only Java alone. And for all the Microsoft hatred spewed around here, there is little recognition that with .Net and C#, they finally did something that has a great deal of technical merit over Java. Anyone who would argue otherwise would probably also deny that Java has been trying to follow C#’s lead for years now.
As for Microsoft not wanting to promote Mono, why would this be the case? For the same reason they decided not to port Office to the Macintosh? Having greater portability means that you can reach a larger audience with your products, and I would think that this would be a big win in any corporation’s eyes.
Dude, they ported Office to Windows. Office was originally released for the Mac (1990), then released for Windows (1992).
I find it hard to take anybody seriously that does not at least try and look up what they’re talking about. I’m sure many other people feel the same.
there is little recognition that with .Net and C#, they finally did something that has a great deal of technical merit over Java.
…like the original Windows API had advantages over the original Mac API: it was based on years of experience developing for Mac’s, with no legacy issues to make changes hard (not that MS didn’t make some real boneheaded mistakes of their own, of course.) Sun, like Apple, had no similar preexisting system to base an improved design on.
As for Microsoft not wanting to promote Mono, why would this be the case? For the same reason they decided not to port Office to the Macintosh?
Mac Office makes a lot of money for MS, but it’s probably more valuable as a bargaining tool since it’s so important to Apple (how often has MS threatened to drop it?)
Of course, they have to studiously ignore the embarrassing fact that Mac Office, while better than Windows Office, is not a good enough product of its kind to be competitive except that it is MS Office.
Mono is in the same position as GNUStep: crossplatform, but the “official” version has many more features and stays well in the lead. Mono, like GNUStep, makes it much easier to port to the “real” version than away, and may encourage people to try .NET on Windows who’d otherwise have no interest in *anything* from MS. It also serves as an emergency trojan horse to hurt anyone they don’t like who uses it, but that applies to most MS products.
I’ve said it before and I’ll say it again: all the time, energy and political will invested in getting Mono to the point where it’s a mere shadow of .Net could have brought Parrot to maturity a year or three ago. We have plenty of free and open language implementations to choose from, all of which are investigating byte-compiled interoperabilty via a common bytecode and interpreter. The work had begun long before the Mono crowd started pimping it as the best idea since toilet paper.
Instead, we’ve got a half-assed implementation of a possibly-encumbered runtime that can only actually execute the most basic command-line apps.
Way to go.
Miguel & co., the community thanks you. We didn’t want Ruby, Python, Perl, etc. (you know, existing, useful, cross-platform languages with appealing and different syntaxes and applications) to run faster on a common runtime…
I’ve got this great idea for an invention. It’s round. I’m thinking of calling it the “wheel”. I wonder if Novell will fund my work.
You can write stable Gtk applications in Mono. Gnome+Mono is actually a pretty nice development suite.
And as immature as Mono is, it’s pretty fast.
Then maybe you should stop saying it and do a little homework. It seems you feel quite comfortable to write a message bashing Mono and pimping (your word) Parrot. What you fail to do is justify things like the following:
Do you really have so little knowledge of the current technology, or are you deliberately trying to spread FUD?
Yep. Poor us. Now we have the ability to do C#, Java, VB.Net, F#, Python, Ruby, Ada, Smalltalk, Boo, Pascal, etc, etc, etc with a common runtime.
Oh but we can’t be happy about that, now can we? We can’t recognize the significance that that brings us. To face this debate with such honesty would be rather off putting for those of us looking to advance a personal agenda.
Good luck with your bird.
I’d like to turn that question around… Do you even know what Parrot is?
Here, read up: http://en.wikipedia.org/wiki/Parrot_virtual_machine
And yes, I’ve done a number of in-house projects using Mono/.Net, but without access to a cross-platform GUI framework, it’s not really much of a cross-platform tool, is it? At least, not for the kind of jobs they advertise it for.
There is plenty of language support in Parrot already. Look at the list! More than on Mono/.Net, and the Parrot folks didn’t even have to make up half of those.
My point was that such a wasteful allocation of resources hurt the community and served only Novell. Your post did nothing to contradict those assertions. Way to go, fanboy.
It would be great if they did. I’m only being taught in Visual Basic and Java, so if VB.NET apps could be compiled to work in other operating systems with little trouble, we’d have both of them being cross-platform. 🙂
Microsoft is Microsoft however.
We evaluated writing a free Qt replacement, but reimplementing an API would most likely result in less efficient software and would have taken too long to implement. GNUstep, Wine and LessTif were other projects that had attempted to reimplement a proprietary API and just had a limited success after a long development history.
http://primates.ximian.com/~miguel/gnome-history.html
So at one point Miguel agreed that chasing someone else’s API with a free implementation was ultimately doomed to fail. What changed his mind?
So at one point Miguel agreed that chasing someone else’s API with a free implementation was ultimately doomed to fail. What changed his mind?
Chasing after someone else’s API is only a small part of Mono. Even if that were to fail, there’s still the free stack (the CLI virtual machine, C# compiler, Gtk#, a truckload of free Mono assemblies, Monodevelop, etc.) which is a nice and modern free software development platform.
…it’s all been said before, but let me just add my voice:
Proof, (if proof were needed – which it patently isn’t) that Miguel de Icaza is either an MS mole or just a plain nutter.
Reason enough to ignore Mono? Methinks it is.
The debate about which language GNOME should use (aside from C) has been ongoing for an extended period of time. People have brought up pro`s and contra’s of Java vs. mono and in so doing created an ideological battle. But the reality is while people were speaking out for java and against mono code was being written in mono and now we have a dearth for good desktop programs being written in mono-not java. During the entire debate not one line of desktop java application was written. Of course during this debate a tremendous amount of work was poured into getting a free java implementation to enable future developers to write in java.
The only linux distro which is pushing java properly is redhat- and redhat is simply not pushing desktop linux. Additionally it was redhat pushing python for administration tools that led to python being acceptable for desktop tools-and now Ubuntu has fully embraced python-I expect to see more python-gui from ubuntu on the desktop.
If you look at the hackers on planet.gnome.org the only one who routinely works on java is Havoc Pennington- and he is not working on the GNOME desktop anymore. There are at least 5-10 regulars on planet.gnome.org that are currently working on mono projects.
Now that Java is going GPL one would think we would see a big push for java on the desktop-but GNOME doesn’t have any desktop app hackers who prefer to work in Java. This may of course change in time- but given the momentum that mono now has new recruits to GNOME development are far more likely to embrace mono than java.
Personally i have nothing against java. Java is an incredible platform. Many look down on Java as bad for the desktop- how can one judge java by looking at desktop apps ?- the areas where java really shines is *not* the desktop-it is the server. The development of java desktop apps is simply irrelevant to the vast majority of java developers and java deployments-with the exception of in-house java-apps it is unlikely that there will ever be much in the way of desktop java applications. In java space server is #1 and java development tools are #2 and java desktop apps a distant #3.
Mono is different. For mono the server usage and desktop application development both vie for position #1. And the desktop development taking place using mono is primarily or the GNOME desktop, for Linux.
I hope mono becomes the common vm for the GNOME desktop-and that instead of writing panel applets in python people would use Iron-Python or boa. Mono still needs a lot of work- a lot of optimization- each mono app should share a common vm and thus allow for each app to profit from good garbage collection. (note I am not a developer- I may be talking out of my ass-but it seems to me the memory inefficiency of mono apps is due to them not sharing resources sufficiently).
I wish people would stop spouting the “Java just shines on the server” meme. SWT and the work gone into Java 5 and 6 show that this is not the case. Java on the desktop was on the back burner in the days when Sun “was the dot in the dot com”, but those days are long over. Spewing this nonsense is almost as bad as saying Java is slow because it’s interpreted.
As for the “dearth for good desktop programs”, where are they? See sbergman27’s post above.
I don’t know much about Mono desktop applications, but I recently sold one pure Java (Swing) desktop application
to a customer. They run it on Windows, and it has native look and feel, so they have no problem with it. It could run on any other OS as well.
DG
Hello folks,
The size of Tomboy might look large to people, but that is mostly shared libraries that are shared with other processes. In fact, every Gnome app has large VSIZE numbers precisely because they are sharing a lot of code with other applications.
The overhead from Mono for Tomboy is:
Native code size: 422927
Created object count: 82262
Total memory allocated (total, does not include reused data): 6517 KB
Tomboy consumes 4 megs more of memory than an empty GEdit buffer. Looking at the pmap output a large difference between GEdit and Tomboy is that they link to different set of libraries, and most of the memory consumption.
If I load the same data into GEdit, the difference between Tomboy and GEdit goes down to 3.2 megs.
So it is doing quite well, you can bring down that difference further down by two megs by using precompilation on your assemblies (run mono –aot on all of your libraries).
Miguel.
Is this guy for real? I say no integration. Next thing you know they’ll be saying Mono stole/copied their code.VB code can already run in Mono so we should leave it at that and no more.
http://www.mono-project.com/VisualBasic.NET_support
Edited 2007-03-26 15:44
MS will never even hint at recognizing Mono, much less ink a deal: that would just make it harder to sue when they finally decide to… which may well be the day that substantial parts of Gnome move to Mono.
Forget the patented parts: the CLR spec standardization explicitly allows a “reasonable” license fee. What if MS demanded one cent per application using the Mono runtime? No one, especially anyone rich enough to take on MS in court, could challenge that without looking like a cheapskate trying to dodge a tiny and, face it, justifiable royalty. It would be the equivalent of a supervolcano going off under GPL’ed Mono code, though: even if a big distro absorbed the cost, all the purists would loudly abandon it and Gnome in general. Years worth of enterprise respectability would be gone in the scandal, everyone would be mad at everyone else, and we’d all be stuck with KDE.
The way MS tossed .NET out and forgot it the first few years, they seemed to gain nothing more than locking in a few servers and enterprise shops. Another purpose was quite possibly to infiltrate and sabotage OSS while appearing to innovate. Now they seem interested again: Vista on new boxes could make .NET desktop apps actually viable in a few more years. Have they ignored Mono long enough to let it get going and now they’re ready to start acting like they just noticed?
You are correct and I apologize. I think that the application that I was recalling was IE. In any case, Microsoft garners the money for continuing to release new versions of Office on the Mac.
And I find it hard to take anybody seriously who is unwilling to look at the crux of the assertion that I was trying to make. Perhaps you can explain why, if you were Microsoft, would it not be in your best interest to promote your applications as broadly as possible?
Office was reborn on OSX as a result of Microsoft’s settlement with Apple a few years back. Office on Mac is incremental revenue for Microsoft that helps extend their control of office document management without threatening their enterprise domination. If hell froze over tonight and OSX started gaining serious traction in enterprises, you’d see Office disappear very very quickly.
What application? .Net is a framework, MS doesn’t derive a direct revenue benefit from it. It was created in an attempt to mitigate java *and* tie enterprise customers to Windows for development. Mono ultimately builds off of Microsoft’s R&D without providing a tangible return to MS.
MS will ignore mono for now, simply because it’s adoption will help muddy the waters for Java and it’s not directly impacting Windows position in the enterprise. But if there was ever a move for companies to start adopting mono for cross-platform commercial development, the way IBM is doing with Java now by moving their enterprise software applications, then you will see Microsoft pull the rug out.
In fact, IBM’s move should be a concern for both Microsoft and the mono camp. Lotus Notes, for example, has a significant footprint in enterprises and the latest version will be built on a java-application stack to run on Windows/OSX/linux with application parity. That’s the kind of thing that keeps Microsoft up at night, and that’s the kind of thing that will help validate java as an enterprise-credible application platform pushing mono more in the direction of interesting community project, but nothing more.
So let me get this straight. If we distill your argument down, are you basically making the case that developing an application in ‘C’ is more rapid than in Mono?
I suggest you look into programming in both C and C#. Go on, have a nice long look. As someone who has programmed in C, C++, Eiffel, Java, C#, Python, Ruby, etc, I have some understanding of what I am telling you when I say that when it comes to rapid application development, I can run circles around C (RAD-wise) in just about any language you care to name. I have a soft spot in my heart for the language, truly I do, but when you consider language level, C is not far removed from Assembly. Which makes perfect sense, considering its goals.
Writing in C is tremendously useful when you need spectacular performance and small size. But developer cycles factor in as well, and so there are some commonly accepted development approaches:
1) Where there is no benefit to writing an app in a high level language (ie; it has to be as tight and small as possible), languages such as C# and Java have to be ruled out entirely in favor of C and or C++.
2) Where only certain areas of an app need the characteristics of C, the common approach is to code these in C/C++ and address the rest from a higher level language.
3) Where applications don’t have to be either tight or particularly small, there is no need to even give a low level language a second thought.
What Tracker may tell us is that had Beagle been written in C, it would have been tighter and faster. What it absolutely does not tell us is whether a Mono/.Net language can be better at RAD. It may point to a greater relevance in some situations, but to say that it makes Mono/.Net completely ‘discredited’ is the same as saying it makes <insert your favorite high level language here> discredited.
I think that Mono will never catch up with .Net. It is not in Microsoft’s interest to promote the multiplatform tools, because their OS dominates the market. There is simply no reason for them to cooperate with Mono team in equal terms. You can’t play with someone who makes the rules and is not interested to let you play along.
There are many excelent languages on Java platform, like Java itself, Jython, Groovy, Ruby, and even a PHP interpreter. There is even a talk about making Ruby a second official language on Java platform. Python, PHP and Perl work well on any platform.
Java is very well suited to business web and desktop applications, and lets developer to access databases and directory services with no effort. Mixing Java and native code via JNI is not hard, too. Java is deployed easily, and application can be extended by adding new JAR files to CLASSPATH, instead of linking them like libraries.
I was considering Mono, but decided not to invest time in it. I can do anything I need with Java, Python or PHP, even on Microsoft OS platforms.
DG
I think that Mono will never catch up with .Net. It is not in Microsoft’s interest to promote the multiplatform tools, because their OS dominates the market. There is simply no reason for them to cooperate with Mono team in equal terms. You can’t play with someone who makes the rules and is not interested to let you play along.
Indeed. Unfortunately, in talking about more collaboration with Microsoft, and bizarrely getting Microsoft to actually recommend Mono to developers(!!??), Miguel shows a bit of naivety in where the balance of power actually is in that Microsoft/Novell deal.
Novell are losing customers to Microsoft. Plain and simple. Even their CEO has admitted that, and those in the know in the Netware/eDiecrtory world have known the situation for quite some time. Bizarrely, the half-move to Linux in OES seems to have accelerated this. The phone call to Microsoft to hash out an interoperability deal was as desperate a measure as one can take under the circumstances.
As far as I can tell, the extent of interoperability within the deal relates to some federated directory things, which are not of any use to anyone. There is no list of what Microsoft will work on, Microsoft is not, and will not, get involved with Windows running under Xen, will never reveal the underlying Windows technology to make OpenXML work (or make MS Office produced OpenXML files open reliably under anything else) and eDirectory will never be able to replace Active Directory. Even if there was such a list, the emphasis is entirely on Microsoft to work with Novell because they have the high ground technology that matters. They control Windows, they control Active Directory and they control .Net. It’s that simple. Microsoft have even been willing to incur billions in fines over this.
It’s also so stupid it isn’t funny. Microsoft are Novell’s competitor, they are taking customers away from Novell and making money out of it………and Novell want them to cooperate to stop that from happening?! Errrrrrrr………
Jamie is correct, Beagle and Tracker can not be compared, specifically because tracker can only do a tiny fraction of the work that Beagle can do.
When Tracker catches up to what Beagle can do, a comparison would make sense.
Miguel.
Is it just me, or is your statement
Jamie is correct, Beagle and Tracker can not be compared, specifically because tracker can only do a tiny fraction of the work that Beagle can do.
logically inconsistent with the previous post from jamie
Tracker is a extensible metadata daemon with a high performance RDF style triple store database + a high speed indexer. Beagle in comparison is just an indexer.
What that means is you can use tracker to store tons of metadata, do desktop tagging, cross reference and cross query all your metadata and provide a next generation integrated desktop. Beagle can not do any of that as its not designed for the task
(emphasis in both quotations mine)
EDIT: it is not the GP post, it’s a post from a different thread
Edited 2007-03-26 20:12
Is it just me, or is your statement
…Beagle in comparison is just an indexer.
logically inconsistent with the previous post from jamie
…
Indeed. Not only is Beagle just an indexer, it is simply a rewrite of Lucene in C# (at least the parts that actually do anything), so I’m not entirely sure what Beagle is actually doing and in what ways Tracker or anything else has to catch up.
It’s not as if there is some wonderous stuff going on by virtue of it having been written in C#.
When Tracker catches up to what Beagle can do, a comparison would make sense.
Interesting, I wondered what Jamie’s comment about RDF triple stores meant “Tracker is a extensible metadata daemon with a high performance RDF style triple store database”. So does it use RDF, or it uses something like RDF? What features does Beagle have that are beyond free text indexing? Do Beagle have a triple store that you can query via SPARQL?
Interesting, I wondered what Jamie’s comment about RDF triple stores meant “Tracker is a extensible metadata daemon with a high performance RDF style triple store database”. So does it use RDF, or it uses something like RDF? What features does Beagle have that are beyond free text indexing? Do Beagle have a triple store that you can query via SPARQL?
Hi,
Tracker implements a lightweight triple store using a subset of RDF (no reification). It includes Subclassing of metadata so it should be mostly compliant with a semantic web system.
The triple store is optimised for desktop usage and uses a custom implementation with sqlite.
For Query we use RDFQ/XML rather than Sparql although I have not ruled out writing a parser for that although one of the issues is how do you combine Sparql with the indexer language to provide a unified search language.
I will be working with the Nepomuk guys to sort out a common language here.
Beagle AFAIK is just an indexer and implements no extensible metadata or triple store atm.
is a daemon frontend to lucene…
so go figure what it does!