forward the dbus session?

[Ben Lindstrom]

On Feb 14, 2009, at 11:19 AM, Peter Stuge wrote:

> Brian J. Murrell wrote:
>>> I guess the objection is simply that dbus is too fluffy for anyone
>>> to jump on it,
>>
>> Objection, or just that nobody has done it yet?
>
> Possible objection or so..
>

To clarify what Peter is may be getting at...

Next the KDE people will want their version of DBUS to have a special  
hook into OpenSSH.  Maybe Enlightenment will invent their own protocol  
and want it in.

The objection would be to implement something that isn't a universal  
standard.  If all the desktops used one DBUS-like protocol it may be a  
different story.  Also, what security risks are opened up by this?    
Do we have issues of trust and security like we do with X forward?


>>> but if you can provide a patch which turns out to be very
>>> small and if you can demonstrate/explain how the feature is a great
>>> benefit for users then it has a better chance of being included  
>>> anyway.
>>
>> Well, the benefit is clear to anyone who understands what DBUS is and
>> what it's good for.
>
> My point is that we probably do not. I certainly don't. I really
> dislike DBUS because it adds thousands (well, almost) of idiotic
> dependencies and I can not understand wtf it is trying to accomplish.

My issue is this is being presented without a list of changes (or  
patch, or list of current hackery need to make DBUS work) that would  
need to be made.

e.g.

Variable FOO needs to be set to <magic>:<number>:<set>  so that  
<insert reason>
OpenSSH needs to grok <concept A> so that <insert reason>
OpenSSH needs to grok <concept b> os that <insert reason>
..etc..

I was hoping my first question to Brian Murrel would product such a  
list besides.  But I guess I wasn't clear enough.


>> It would be nice if a desktop app started on a remote machine but
>> displaying to my local desktop could interact with my desktop as if
>> I ran it locally.
>
> Sure, but for the purpose of a great many it might already do that.
> I wouldn't know, I don't want a desktop. But please explain with a
> good example and I'm sure you'll win me over.
>

I'd argue that was the idea behind X11 myself, and DBUS  decided to do  
an end-run around the that group and implement their own.  And now  
we're stuck with yet another protocol to tunnel.

>> Given that there have not been any strong rejections of the idea
>> (yet) maybe I will spend an hour or two and hack up a patch.
>
> Please do - and again - please explain, concretely, why it is useful,
> in order to improve it's standing.
>
> Is there DBUS on OpenBSD as well? In that case I guess the change
> should go upstream rather than here.

I agree.  Until I know the full extend of the hackery needed to  
implement DBUS support I'd be inclined to ignore the idea. =)


- Ben