LWN.net Logo

WebKit rising - Perhaps I tend to worry to much ...

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 2, 2008 17:10 UTC (Wed) by TxtEdMacs (subscriber, #5983)
Parent article: WebKit rising

Admission of bias: I prefer Gnome over KDE (by appearance alone), however, I am bothered by
the presence of Mono at Gnome's base code.  Moreover, the recent propensity of some of those
associated with Gnome to adhere too closely to the questionable assertions of MS on standards
worries me.  Hence, despite my bias, I will leave Gnome if this trend turns into their true
future course.  But where do I go?

Another problem I have with Gnome, was their email application used to pop up in preference to
my choice, Thunderbird (that has stopped). Nonetheless, I would like the option to remove
entire default applications from Gnome, e.g. the email program and browser.  My preference is
Firefox, that I download myself and keep current (or nearly so ...) rather than allowing the
distribution to rule (Ubuntu in my case).

Finally, I happen to test for Safari.  Moreover, I am on the webkit mail list, no argument
inappropriately so, since I reported a flaw in Safari version 2.x handling of percentages in
padding*.  It was already fixed by the time of my report in the beta version of 3.0.  Hence,
while I am not an adversary of webkit <i>per se</i> I would appreciate the ability to remove
applications I have no intention of putting to use.

Could that be viewed as a rant?  It was not meant to be.  When Gnome was started, I thought
the argument was it was meant to be fully object oriented and modular.  So why is it a problem
to cut the clutter? 

* Only integer values worked.  If decimal values were used it was rounded down to the lower
integer.


(Log in to post comments)

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 2, 2008 17:32 UTC (Wed) by debacle (subscriber, #7114) [Link]

Are you sure Mono is in GNOMEs "base code"? I'm mainly on XFCE, but use GNOME  occasionally,
and I definitely do not have Mono installed.

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 2, 2008 17:46 UTC (Wed) by TxtEdMacs (subscriber, #5983) [Link]

Certain, no.  However, I thought I read that assertion recently here in comments or content
relating to the danger of Mono being accepted by free software.  That is, it might be
implicitly encumbered by existing MS patents.

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 4, 2008 3:57 UTC (Fri) by drag (subscriber, #31333) [Link]

I think people are easily confused by the .NET vs C# vs Winforms vs etc etc.

Microsoft doesn't make it easy, seeing how they are monolythic about everything, but Mono as
it's used in Gnome is the language that is used to make .NET. 

As long as Gnome avoids the parts of Mono that is not covered by the ECMA language standard
(334 and 335 I believe) then I do not believe that Mono posses any more threat in terms of
software patents (at least not any more scary then non-Sun patents that may cover Java or any
patents that may cover techniques used by QT libs or python or anything else software-related)

Now the stuff that Mono does to extend support to cover things like .NET extensions,
Silverlight, or Winforms then I think that is worth avoiding. Those are the things that are
questionable and I would be scared of if I was developing a application. I don't think that
should be a problem for Gnome since I doubt very much that GTK# resemble Microsoft's
application frameworks in any significant manner. 

http://www.ecma-international.org/memento/guidance.htm
> General Declaration:
> The General Assembly of Ecma shall not approve recommendations of Standards which are
covered by patents when such patents will not be licensed by their owners on a reasonable and
non-discriminatory basis.

If anybody has anything concrete on any sort of software patents that are a danger for Gnome
(beyond any hand waving) that Microsoft may have on the ECMA language standard then they
should get together with some of those free software lawyers and try to get ECMA to renigh on
their standard unless Microsoft comes clean.

I am serious about this. If this is really a problem for Gnome or for any of the dozens and
dozens of Linux distributions that ship Mono or anything else then it needs to come to light.
I am sure that if there is _evidence_ in the form of patents documents or Microsoft's own docs
that mention patents that are not licensed as part of the ECMA standard. Or if there is a
problem with ECMA's definition of 'non-discriminatory' then I would be very surprised if Gnome
devs didn't pull out of using Mono and replace it with something more safer.

I don't think that anybody has anything even remotely strong though. I think it's all just
pure guesswork and hand waving. 

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 4, 2008 19:29 UTC (Fri) by TxtEdMacs (subscriber, #5983) [Link]

I am not certain, however, it is my impression it is hard to get a copy of a submitted patent
application where no office action has been taken.  [Where do you think the term submarine
patent arises?]

Next to show how hard the program you suggest is to implement, consider this threat:
http://www.engadget.com/2007/05/14/microsoft-linux-and-ot...

I remember it as 258 patents. Have any of these been even identified?  I think they hide in
the mist just about as completely as when this attempt at intimidation was made.  It should
have been easy to document with just a bit of cooperation.

OK, let's get back to Mono.  The real issue is we have no certainty, so why take unnecessary
risks, particularly when the major use is implementing the preferred MS model for web
development.  If they are prone to make threats and snarl without provocation, why do them the
favor that supports their strategic goals?  In addition, what reason is there to opening a
project to what might be, under current statutes, valid claims of infringement?

There is another flaw in some Gnome project actions, why support OOXML as an official standard
when, by market clout alone, it could have existed as a de facto implementation?  

Having done research on technological topics, I can attest that facts are very difficult to
ascertain without either close set of knowledgeable contacts or the desire of an involved
participant to disseminate information.  If it is the latter, what you might be fed could be
pure fantasy either due to their self delusion or it could be a simple marketing ploy.
Disinformation can arise under other situations* too.  When you are lucky and/or well connected
you get facts.  Much of the information I gathered was patently false.  [You decide if I am
punning, that way you get a taste of market research.]

* In one case, I was speaking to someone high up in an organization of some repute, that
simply rejected my inquiries and further berated my efforts.  Several years later that
organization came under attack for their lax treatment afforded the hazard.  As a result,
reputations lost luster, positions were lost and the problem even then was not remedied in a
timely fashion.  That is, self interest can be delusional and contagious.

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 30, 2008 8:08 UTC (Wed) by jonasj (subscriber, #44344) [Link]

"There is another flaw in some Gnome project actions, why support OOXML as an official
standard when, by market clout alone, it could have existed as a de facto implementation?"

What are you talking about? Gnome does not support MSOOXML.

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 30, 2008 8:10 UTC (Wed) by jonasj (subscriber, #44344) [Link]

(I mean support as in supporting its standardization, not as in supporting the file formats in
the software)

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 12, 2008 17:24 UTC (Sat) by Duncan (guest, #6647) [Link]

Re: your ECMA quote.  

You are aware that "Reasonable and non-discriminatory" (RAND) licensing, 
as traditionally formulated, is not anything of the sort when it comes to 
Free Software, right?  The problem is that RAND allows per-copy royalties, 
as long as they are offered at a standard "reasonable" rate to everyone.  
Of course, per-copy royalties do a number on Free Software, since that 
means it's not freely redistributable without by definition making the 
distributor liable for infinite royalties!

So while the term /sounds/ good, those concerned about their Free Software 
rights have learned that RAND, by itself, ensures nothing.  In fact, it's 
a negative, since it's likely to effectively ban Free Software 
use /at/ /all/, unless there are additional clauses that specifically nail 
down those royalties to zero or at minimum, a one-time fee that covers all 
use downstream and is specifically /not/ purpose restricted, for some form 
of Free Software license.  (If the licencer wishes to maintain some 
restriction, they may limit it to copy-left licenses or to say the GPL 
specifically, if not, they may include MIT/BSD as well.)  Or, a legally 
binding pledge to the same effect can also be acceptable.

Other than that, I've nothing specific to point to; I'm simply pointing 
out that your quote of the RAND licensing conditions doesn't mean what it 
might intuitively look like it does, because in fact, RAND *IS* 
discriminatory, as the term was developed in a commercial context without 
reference to Free Software and due to terms, discriminates against it.  As 
such, it's a favorite of MS, since they can claim to be non-discriminatory 
while in fact being nothing of the sort.

Duncan

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 2, 2008 19:32 UTC (Wed) by salimma (subscriber, #34460) [Link]

Not in the base GNOME libraries, but some programs that are released as part of the GNOME
Desktop are written using C#/Mono (e.g. Tomboy, F-Spot. Not sure if Beagle is officially part
of GNOME or not)

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 2, 2008 21:37 UTC (Wed) by Frej (subscriber, #4165) [Link]

Beagle isn't, neither is rhythmbox. Same stuff rhythmbox (in c) being the older project.

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 13, 2008 2:40 UTC (Sun) by shirish (guest, #35547) [Link]

Hi all, I'm on Ubuntu Hardy and these following packages use mono on my system namely
autopano-sift,f-spot,gbrainy,tomboy,hugin 

WebKit rising - Perhaps I tend to worry to much ...

Posted Apr 7, 2008 12:01 UTC (Mon) by ceplm (subscriber, #41334) [Link]

> I am bothered by the presence of Mono at Gnome's base code. ... But where
> do I go?

yum erase \*mono\* \*sharp\*

works pretty well for me :-)

Mat&#283;j

Copyright © 2008, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds