[PATCH] Multi-Protocol IO-Slave

[Kevin Krammer]

On Sunday 13 January 2008, nf2 wrote:
> Aaron J. Seigo wrote:
> > On Sunday 13 January 2008, nf2 wrote:
> >> Oh well - unfortunately there is a really unlucky tradition on both
> >> sides not to touch, nor even look at the other projects code. Feels a
> >> bit like cold-war: Xdg standards were the maximum to get for
> >> interoperability. Arms control agreements ;-). Common code? - no way.
> >
> > this is complete and utter, excuse my language here, bullshit.
>
> No - it's not. How come that i'm the only person who at least tries to
> dive into VFS code of both desktops?

I guess the root of this disagreement between Aaron and yourself is based on 
different definitions.
Aaron and many other KDE developers including myself are quite sick hearing 
that we do not particpate in or adopt things developed elsewhere while we 
definitely are.

Shared implementation as opposed to shared specification (which quite some 
have their origins within the KDE project) are just way more difficult to get 
right and often still need some formal agreement (i.e. a specification).

So yes, shared specifications have been more successful than shared 
implementations so far, IMHO also because until lately shared implementations 
meant linking to the same thing.
I expect that the common IPC system we have since the widespread adoption of 
D-Bus will make shared service implementations a lot more common, having 
shared client implementations as a bonus but not a requirement, thus not 
blocking the adoption of the service just because of disagreements on API 
design.

> How come that desktop developers 
> don't even read the mailing lists of the other projects?

I guess some of both sides do, but since those lists tend to have quite some 
traffic due to their importance for each project, such "lurkers" will likely 
skip all thread which are not totally obviously interesting to them as 
oberservers.

> Why didn't GVFS 
> developers communicate that they started something new which might be of
> relevance for  KDE. Probably you wouldn't even know that GIO/GVFS
> exists, neither have the slightest idea of it's design, without me
> posting here. That fact just surprises me. That's what i wanted to
> express - sorry for being a bit provocative.

They probably didn't want to risk any negative feedback.
There has been a similar proposal years ago on the XDG lists and it recieved 
quite positive remarks from GNOME and KDE developers, but some developers not 
associated with either project reacted so hostile that the leading developer 
gave up.

Additionally they probably anticipated that any formal proposal would lead to 
the request of a formal specification and they quite likely wanted to get 
some in-the-wild testing before detailing something they might regret later 
on.

Cheers,
Kevin

-- 
Kevin Krammer, KDE developer, xdg-utils developer
KDE user support, developer mentoring
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.kde.org/pipermail/kde-core-devel/attachments/20080114/5fceb990/attachment.sig>