Submitted by Open-source pioneer and Novell Vice President Miguel de Icaza Thursday for the first time publicly
slammed his company’s cross-patent licensing agreement with Microsoft as he defended himself against lack of patent protection for third parties that distribute his company’s Moonlight project, which ports Microsoft’s Silverlight technology to Linux.
Novell’s de Icaza Criticizes Microsoft Patent Deal
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSAlert.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
That’s an interesting development because DeIcaza is usually pro Microsoft.
He is not “pro” anything. He always tries to be objective, so he sees things from afar, rather than getting all psyched with his own position, or OSS’ position, or Microsoft’s position. This is why I like Miguel.
I really really respect Miguel. Although he is an idealist he is also practical. He wants good tech just as much as he wants ideology(freedom).
Futhermore he never seems to have allready choses sides on anything. He always considers and he sees good and bad on both sides. *As is it (obviously) really is*
Agreed, but it seems to me that Miguel is always on the pro-Microsoft side of any camp that he respectfully belongs to.
Ditto. He’s a pragmatic and objective guy.
Nope.
IMHO you have to be pro-MS to say OOXML is a superb standard ( Cause the 6000 page blob riden monster obviously isnt! )
http://groups.google.com/group/tiraniaorg-blog-comments/browse_thre…
Well, he has implemented a spreadsheet (Gnumeric) before, so unless you have better credentials, I think I’ll take his word over yours on that.
Miguel seems to have a peculiar bug in his firmware which might be expressed as:
while True:
>>if provider == microsoft:
>>>>technology.category = “cool”
>>>>implement_for_linux(technology)
>>>>claim(NO_LEGAL_PROBLEMS, technology)
>>>>set_warnings(off)
>>>>ignore(signals, ALL)
(‘twould be nice if we had [code][/code] tags.)
Edited 2008-03-08 00:02 UTC
He is free to like Microsoft technologies and implement them. Novell is free to throw money at the projects. The developers are free to work on them. And you, the distros, and the other projects are free not to include their code in your Linuxes fearing Microsoft patent claims.
In the end it is beliefs vs practical needs. RMS and Theo the Raadt don’t like the binary blobs you use for practical reasons(eg. you want 3d games). You don’t like Microsoft technology, but some people will want to write word documents in Linux and compile .NET apps.
As the SCO case so clearly demonstrates, even an unskilled opponent can get years worth of PR mileage and FUD value out of a case which has no merit at all. SCO had no case 5 years ago, but are still making headlines, even though they didn’t really own anything and we didn’t copy any of what they were claiming anyway. Just imagine what a more skilled opponent, with deeper pockets, could do with Mono, where we cannot even deny that we are reimplementing Microsoft technology.
People can (and do) argue the finer points of patent law all day long. But at the *end* of the day, the battle is not a legal one. The real battle is fought on the field of PR, and has little to nothing to do with patent law, and everything to do with the public perception of uncertainty and doubt.
To a great degree, I feel the same about Mono as I do about Samba. I wish to the gods that we didn’t need it but I thank them that we have it.
But… we’d be crazy to start basing any of our infrastructure on Mono. Unfortunately, with Tomboy, Beagle, and or F-spot shipping by default in some distros, it appears that the creeping Mono-ism has already begun.
And besides, there are other good reasons to avoid Mono. Yesterday, I blew away Tomboy from my largest XDMCP server because it is insane to devote 16MB per user, or a total of over a gigabyte of system ram, to a freaking sticky-note note app!
Why isn’t most of that 16MB shared? If mono is intelligently built, all of the core libraries and the runtime should be AOT-compiled images which are shareable. The libraries of Tomboy should also be compiled this way, so that the only private memory would be the GC segments. In steady-state operation (no one writing tons of notes and deleting them), no GCs will be happening, and most of the GC segments would be paged out.
Is 16 MB really the reference set of Tomboy?
33 MB resident with 18 MB shared as I check right now. So about 33MB + 15MB per additional user. That’s on a 32bit box. Slightly higher on a 64 bit box. With swap on, RES – SHR decreases to more like 10MB. Which still makes it a pig of a post-it note app… just as Beagle is a pig of an indexer. I have never even considered allowing F-Spot, so I don’t know how bad that is.
There is certainly a lot of room for optimization there, then!
I have already performed that optimization:
rpm -e tomboy f-spot libbeagle mono-core-1.2.5.1-3.fc8 mono-data-sqlite-1.2.5.1-3.fc8 mono-data-1.2.5.1-3.fc8 mono-winforms-1.2.5.1-3.fc8 mono-web-1.2.5.1-3.fc8
It achieved 100% memory savings at essentially zero cost to productivity.
Edited 2008-03-09 00:42 UTC
Rereading your post made me think of another question…
Why would you run your system with swap off? When you say swap off, do you mean that the OS has no place to page to besides the original executable images for memory that’s not in use?
I mean that swap muddies the waters when reporting memory usage numbers. I always wonder what people mean when they say “Application X uses Y MB of memory”. (Windows users, for some reason, seem particularly smug in their confidence in the absolute numbers they report.) Shared memory and swap make it very difficult to report absolute numbers in a meaningful way.
As it happens, I *was* running with swap turned off throughout my whole Epiphany session, by accident. I had it turned off for a completely different experimental reason, and then forgot to turn it back on again. I don’t believe that swap is bad and leave it turned off as some people apparently do. As Andrew Morton said, you really don’t want megabytes and megabytes of bloatyapp floating around in your ram.
With swap turned off, “resident – shared” actually means something. If there is stuff in swap, you really can’t know what “resident” truly means. On the other hand, one could make an argument that having swap turned on, and measuring when the system is under memory pressure, might give a more meaningful representation of the true cost of the app under real life conditions. If it has a lot of stuff that the OS can just swap out once and forget about, that lowers its impact.
At any rate, I always feel conspicuous when I report memory usage numbers, because the real answer lies out there in the gestalt somewhere.
Edit: Oops! I was responding as if you had replied to this post I just made a few minutes ago:
http://www.osnews.com/thread?304014
But aside from the Epiphany reference, I think it is still appropriate.
Edited 2008-03-09 22:35 UTC
That’s why I used the phrase ‘reference set.’ Apps can take tons of memory, but only really touch a little bit of it. “Working Set” is probably the NT equivalent of what you called “Resident Memory.” You’re right about how to measure the true memory cost of a program… force the OS to drop all of its pages and see what gets pulled back in by the program.
This is really what people should be concerned about when they complain about the bloat of a program. It might have tons of code (to support many features), but if it is well-designed and modular and compiled with a good profile-guided toolchain, the actual cost of all those features is not too bad unless someone is using everything at once.
Once upon a time there lived a beautiful princess in a lovely castle, and the sky above the castle was always blue, except for when wispy clouds would float overhead and draw pictures for her. And the VM on her computer always knew the right things to put into swap, and the right time to bring them back into ram, and when it didn’t, the application programmers and kernel coders knew how to fix that. And they all lived happily ever after.
Edited 2008-03-10 01:30 UTC
if you read his full reasoning on it here http://tirania.org/blog/archive/2007/Jan-30.html, you will see that his opinion on OOXML is well reasoned, and coming from his position as the creator of the most used spreadsheet in the free software world, not as the VP of a company which has partnered with MS.
Is it?
1. He tries to defend the 6000 pages of OOXML by trying to somehow argue that this gives us more detail than we’ve ever had before from Microsoft. Alas, those 6000 pages are largely a dump of the huge number of quirks of Microsoft Office, and gives little away in terms of how to actually implement them.
2. He tries to argue that with the information available on formulas in ODF that formulas simply cannot be implemented. However, given that Lotus, Open Office, KOffice, Google and Corel are all implementing ODF formulas, and there is feedback into improving and changing ODF in successive versions, his evidence for this is thin on the ground. I don’t see anyone that has managed to implement the functions and formulas system as specified in OOXML, apart from that which has already been reverse engineered from Excel, which ironically makes the job easier. That doesn’t make the spec any better, however.
3. He tries to criticise Groklaw for keeping track of problems and inconsistencies with OOXML. He tries to point out that OOXML references seven ISO standards where ODF only references three. While I can’t verify that off-hand, that isn’t the point. The fact is that OOXML comes up with, and references a lot of Windows-only implemented technology at the moment, where existing ISO or other standards could have been logically used to better benefit. He completely ignores the W3C standards that ODF uses as well.
4. He tries to claim that the information for Windows Metafiles is publicly available. WMF is a one-to-one mapping of Windows API calls. He would vote to add such information to the specification, but of course, Microsoft hasn’t and won’t do this. The complete lack of any amendments to OOXML after comments have been submitted shows this up. ODF, on the other hand, has successive versions.
5. He tries to give a weak justification as to why SVG shouldn’t be used. Basically, he argues that it’s too much work and would pull in too many other W3C specs. Incredible. However, other developers are using SVG now, or are at least using a subset of it, and if Microsoft actually had a web engine that adhered to many W3C standards properly then they wouldn’t have such a problem.
6. He tries to argue that it is within reach to bring XAML and WPF to non-Windows platforms, but this just shows up where Miguel conceptually just doesn’t get it. You can get 20%, 40%, 60% or even 80% of what Microsoft has implemented, but you can never have a situation where you have a 100% drop-in replacement. As Microsoft will also be first with any new implementation, you can never, ever be on a par with what they’re doing. That’s not what standards are about.
7. Like people like Rick Jelliffe, he brings up the response to OOXML’s 6000 pages that seems to be doing the rounds – OOXML uses 1.5 line spacing versus ODF’s single spacing! If this isn’t straw grasping, I don’t know what is.
I wasn’t aware that he wrote Open Office Calc.
There’s no evidence for that I’m afraid.
1 – What he says is that from the spreadsheet point of view, you need that kind of size to actually have a usable spec. ODF does not give you what you need to write a usable spreadsheet, OOXML does. I do not claim to be an expert on any of this, but I am willing to bet he knows what he is talking about when it comes to spreadsheets.
2 – Yes, you have proven that it is possible to come up with your own formulas. Just because you can, doesn’t mean that it isn’t easier for someone to give them to you.
3 – I don’t have a hard time buying that facts coming out of groklaw are only slightly less distorted and biased then those coming out of slashdot comments, but that is a secondary point. He is not saying that MS is not referencing their own stuff, what he is saying is that critising them for not referencing ISO standards is nit-picking, due to that they due in many cases, and some of the cases that people pull out to critise MS are strawmen. Just because it is an ISO standard doesn’t mean it is used, or that it is even a good idea.
4 – That is very weak, are you saying that there are not equivilents to those APIs on other systems? If you have the spec for the file itself, it would be trivial to map it to other APIs.
5 – Just because it is a standard does not make it good for all purposes. SVG is a good intermediary vector format, but it was designed for compatibility, not for performance or size. Completely ignoring the valid points he brings up about SVG being a monster of a spec, for non trivial images, SVG is a good choice for only very specific circumstances due to it being XML based. Unless what your goal is to have the most wordy, cpu intensive format possible. Its like making an xml based music format, sure, it would be the most compatible format on earth, but that would be the only thing going for it.
6 – WPF is kind of beyond the scope of this specific discussion, but since .net is one of my favorite things to talk about, I’ll address it anyways.
WPF should be alot easier to implement then winforms was, the problem is that WPF is so massive it isn’t feasible right now due to the rediculus amount of code that would need to be written. When it comes to the vectory/animation stuff, there are plenty of libraries in the linux world that could be easily leveraged though.
You are right that they will never be on par, but you are wrong about drop in replacements. 3.5 just came out last christmas, and mono is close to 100% 2.0, and about 80% 3.0 compliant. Since most shops are conservative about new technology, and 3.0 didn’t really have all that much to offer, the state of things in the .net world is still very 2.0.
As for WPF, it is actually the least adopted of the 3.5 technologies (either WF or WCF would be the most). People in the community are speculating that when Silverlight 2 hits, it will drive WPF adoption as people will have serious incentive to start learning it. As it stands now, the tools for both designers and developers are immature, and scalable, glitzy UIs are not enough incentive for everyone to learn a completely different technology (other then me of course, as a web guy, WPF is the first UI technology I have ever really enjoyed learning.)
As a side note, the reason they are able to get such good binary compatibility in the first place is that MS shares their test cases with them.
7 – His argument is that size is a good thing, so pointing out that it is actually 4500 pages is actually a negative. Are you a developer? If so, you will understand what he is talking about, its not like you can really have too detailed a spec, so the whole size issue is kinda dumb. I would take 6000 pages over a cocktail napkin any day.
Either you only started using linux in the last few years, or you are being pedantic for no good reason. He wrote a full featured, widely used spreadsheet program. Based on his obvious credentials, I am willing to take his word about issues to do with spreadsheets.
I honestly find it sad at how rabid the linux community has gotten. Miguel is really one of the rockstars in that world, it is hard to think of people who have contributed more to making the linux desktop experience what it is then him. I started out as a mac developer, then became a linux developer. I am now a .net guy, and while windows is by far my least favorite of the three operating systems, the developer community is equally my favorite, mostly due to the lack of religion.
If you seriously can’t see that he was giving his opinion as the guy who wrote Gnumeric in that post, you have gone beyond any point of objectivity.
Open Office, Lotus, KOffice, Google and Corel have ODF formulas implemented now. Ironically, people have been implementing Excel’s formulas and functions functionality with no documentation at all. The formulas specification in ODF needs more detail, and that’s happening, but the notion that nobody can implement formulas with ODF is false. Miguel doesn’t even say what is missing.
*Shrugs shoulders*. Can’t argue with that logic. Claiming that Miguel is an expert on this stuff so we can take his word at face value doesn’t address what he’s saying.
Different applications using ODF are doing that now. I don’t see half a dozen applications using OOXML that have been able to implement formulas and exchange documents with them in. I haven’t seen any applications able to open Excel 2007 documents with formulas in them.
I’m afraid that’s a worthless statement unless you can go through the points made on Groklaw and say why they’re wrong. Miguel hasn’t been able to do it and people banging away on their MSDN blogs avoid a point-by-point rebuttal of Groklaw’s objections like the plague.
Microsoft are free to use what they like, but when you are trying to get something agreed on as an ISO standard there is a reason why you use existing ISO and other standards and specification. It’s so that the standard is reproduceable and reliable and communication between standards groups increases to solve any problems. I read that in a manual somewhere.
With such impeccable logic, why bother with standards at all then?
To WMF? No. WMF is an insecure format from another era that maps 1-to-1 on to the Windows API. To implement WMF and to understand it when it appears you’ll have to understand and have at least a partial implementation of the Windows API. It doesn’t explain why it is used in place of CGM.
No it isn’t. Just ask the WINE people.
SVG is being implemented by many software developers today, or at least a subset of it. Moaning about how verbose it is does not justify telling us that we should all just reimplement XAML and WPF, as Miguel seems to be doing.
I’m not sure how reimplementing XAML or VML is any better.
You’re missing the point. Why are you and Miguel simply assuming that people should just implement WPF at all whilst railing on existing standards that people are actually implementing such as SVG?
There you have it.
You’re right there. There’s also lots of legacy code.
If a specification rumbles on for a whole chapter about a feature, and then tells me at the end that I need to reimplement the features of WordPerfect, Word 95 and Word 97, and it repeats this umpteen times, what does that tell me?
I don’t like to be pedantic about details, but you said that he created the most used spreadsheet application the the open source world. Gnumeric was, at best, an Excel wannabe that got nowhere near to being Excel.
I’m not. I’d rather read what he says and make a judgement based on that.
I find it sad when people take what someone says at face value, based on what they think his reputation is, without reading what he’s written and dismissing Groklaw’s objections without saying why. It’s pretty old and tired now.
While he started Gnome, others have done far more to carry it forwards than he has. He found it more productive to start an incomplete clone of Excel, and a perpetually incomplete clone of .Netwhilst regurgitating Microsoft’s marketing material.
Objectivity is looking at what people say, write and do and making a judgement based on that. The opposite is to come up with a preconceived idea of someone’s reputation and saying “Well, he wrote this so I’m willing to just believe him”.
What Microsoft really need is to push their new products.
Usually new Microsoft applications come with new incompatible formats. I.e. you can’t open VS2005 project in VS2003 even if they’re almost identical XMLs. It is not a technical problem, but marketing trick. Obviously you need to force incompatibility issues, because a typical XML document allows easy forward compatibility.
Probably anyone who stands for Microsoft “standards” was bought by Microsoft or listening too much of marketing blahblah.
Edited 2008-03-09 12:31 UTC
Yeah and have to be pro[something] to think [insert opinion].
I know this is always true. I can see their thoughts!
Eugenia, I agree with you Miguel is not pro anything. Possibly dealing with Microsoft and expecting the relationship to be beneficial to each party is usually a mistake. Microsoft deals with others when it has the upper hand.
Edited 2008-03-08 00:24 UTC
I’m not entirely sure who you’ve got him mixed up with. For good or for bad, and people can come up with all the conspiracy theories they like, he has always had a puzzling and burning admiration for what Microsoft comes up with. Whether it be Gnumeric and Excel, Mono and .Net or support for OOXML, he is anything but objective. His bizarre support for OOXML was widely discredited, and around the time Mono was created he even started regurgitating a lot of Microsoft marketing material on .Net, such as ASP.Net being used to write 80% less code than anything else.
It just appears that Miguel is now backtracking somewhat, and if you read the debate:
http://www.linuxworld.com/news/2008/030608-mix-novells-de-icaza-cri…
A lot of it looks quite painful and difficult to reconcile. He even comes up with this gem:
Owning end users? That’s not really a logical response to……….anything.
“I’m not entirely sure who you’ve got him mixed up with.”
Miguel and Eugenia know each other personally and they’re having the same stance which might originate in their natural naivety but is a part of their economical model now.
Edited 2008-03-10 10:34 UTC
Either the guy is bought by Microsoft or incredibly naive.
You are confusing “does not have a rabid, irrational hatred for” with “pro”.
“That’s an interesting development because DeIcaza is usually pro Microsoft.”
It could also be a PR stunt; they were quiet for some time.
Hardly “slammed” his company, he should know that working for company like Novell something like that would happen.
Hate to say it but I hope Microsoft dont pull the carpet from under his feet, if Microsoft are being more open just who is pulling them strings. It’s not so long ago they were threatening patents and IP infringements, I guess he has to just sit on the side line then really.
What a paradox!
I don’t know what to believe. Normally, history has shown that everything Miguel thinks is great from Microsoft is totally wrong….so if he does not like something does that make it actually good?
So…confusing.
No. It just makes it so horrendously and obviously bad that even Miguel can’t credibly rationalize it.
Someone opened a can of worms
HAVE FUN JUST LOOKING AT IT…DOLT.
That statement must have been painful to make, it having been obvious for a long time that De Icaza really likes MS tech.
As far as I can see, the anti MS sentiments displayed by De Icaza are actually quite obvious statements to make, I just don’t understand why it has taken him so long to actually see it.
It must be true what they say, you lie down with dogs, your gonna get fleas.
As for his statement about the Desktop becoming irrelevant, at least as far as making money goes, I think he has his head up his rear end. Sure, the mobile sector, and to an extent Google with it’s content delivery and advertisement business models, are taking the IT industry by storm, but to think that it’s going to sweep away the current business models is absurd and displays his usual lack of thought on this, and many other issues.
It’s been a long time since I had any respect for Miguel De Icaza, and these statements haven’t changes my mind about him one bit.
miguel de Ichsa is an Idiot
Edited 2008-03-08 19:56 UTC