On Friday 10 March 2006 14:57, Paolo Molaro wrote:
> Well, either you have an open source app you can share or you have a
> proprietary app and say: "We have this app, but we don't want it to work
> on mono because we spent so much time making it work on pnet already".
> If it's a policy decision to not try to have your app working on mono
> but try to make it work on pnet, it's fine: it's your app and your
> policy. It's important though that you made it explicit that it is
> your policy and not something that depends on mono.

(Note that I'm not a Trumpf developer working in a company that was contracted 
by Trumpf to fix bugs in dotGNU)

Back when the decision between Mono or dotGNU was made dotGNU was the only 
choice because Mono was still using WINE and Trumpf wanted the target 
plattform to be PowerPC.

Today the application runs fine on dotGNU and I guess there is just no 
interest in having it run on Mono as they already have a working solution. 
It's probably not policy but other issues: reporting/finding/fixing bugs in 
Mono takes developer time and thus costs money, but there wouldn't be any 
immediate gain for them to do this...

> I looked up a couple of threads you initiated: they are in areas of the
> code (System.Data, not related to the point of SWF) that I don't deal
> with. The replies were mostly from Mainsoft or from the Novell devs in
> Bangalore and at least in one case they suggested you how to correctly
> fix one issue and there was no followup. I guess they could have been
> more encouraging; however, in Mono we always request test cases for the
> test suite, because the codebase is so large and often external
> contributors just post workaround patches instead of real fixes.
> Consider the test cases as a way to filter the good contributions: sadly
> this results sometimes in good contributions being dropped as it may
> have been in your case. Since pnet just takes a copy of the System.Data
> implementation from Mono, you might want to push fixes you may have
> anyway, so pnet will benefit, too.

That's why I've resubmitted one patch three times, it was critical :-) There 
were only two or three bugs we've found in Mono's System.Data stuff, I think 
I reported them all...

I completely understand why you want Unit tests, but if my employer says "no 
time for that, you've got more urgent things to do"... well, let's just say 
I'm normally not keen on continuing developing stuff from work in my free 
time ;-) Got other hobbies...

> > Yes, last time I checked (about half a year ago or so) I saw that their
> > SWF implementation has changed but wasn't sure off the top of my head
> > whether there still is any WINE dependency.
>
> There isn't so yur assesment is based on assumed but incorrect info.

Yes, I'm sorry.

> > See above: I gave up on it. I remember reporting to Mono how they can fix
> > a bug in their layouting code so it behaves correctly but that wasn't
> > taken seriously. Again, that may just be bad luck on my side.
>
> Do you have a reference for this layout issue? Anyway, since you made
> the list of Mono contributors in at least two mono releases, I guess
> your changes were accepted and credited.

I couldn't find an e-mail about the layout issue... I'm pretty sure I reported 
it since I remember clearly that Mono and dotGNU both did that particular 
layouting thing wrong, but I think it was in IRC.

It was about this diff:
http://cvs.savannah.gnu.org/viewcvs/dotgnu-pnet/pnetlib/System.Windows.Forms/Control.cs?r1=1.85&r2=1.86

In short: for anchoring to work like on MS.NET you may not save the 
positions/sizes but have to save the *distances* from the edges so when you 
make an application smaller than the layouting intended and then back to 
normal size everything is restored correctly. See my comment in the diff for 
an explanation.

> I'm perfectly fine if you say: "I like pnet SWF code, I worked on it a
> lot". It becomes mud throwing when you say "Mono's SWF code is not so
> good", especially since you made it clear you haven't recently seen it.
> Many people will ask about mono and pnet comparisons: I just suggest
> that you explain the strenghts of pnet instead of trying to put mono in
> bad light, especially with arguments that are easily rebutted: it will
> give more value to your arguments:)

Aye, you're right :-)

> lupus

C'ya,
        Marc

-- 
Marc Haisenko
http://darkdust.net
Today is Prickle-Prickle, the 69th day of Chaos in the YOLD 3172

Attachment: pgpREU6BVVXM9.pgp
Description: PGP signature

_______________________________________________
Developers mailing list
[email protected]
http://dotgnu.org/mailman/listinfo/developers

Reply via email to