A draft for a standard for desktop configuration

[Philip Van Hoof]

Hi there,

Sure it's debatable, however . . .

Some people commented (in private) that LaTex is a document format they
don't know about (and that it's holding back their contributing). So to
stimulate contributions, I decided to convert the document to XHTML.

No worries, after using tth I tidy'd it, made some huge manual tweaks,
corrected everything so that it validates as XHTML and created both a
"print" and a "screen" cascading stylesheet for it.

Sooner or later will the location of this document* move to a subversion
service of my company (X-Tend). But no worries, it'll still be available
anonymously and using a web browser (without any authentication so just
like the current viewcvs). For writing to the document, however, we'll
have to create an account for you (which shouldn't be a problem if you
are really planning to work on the document, and we will do that on
request).

--- 

*) http://cvs.pvanhoof.be/cgi-bin/viewcvs.cgi/*checkout*/desktop-config-standard/src/desktop-config-standard.xhtml?rev=HEAD


ps. Sooner or later I will also publish an analysis for an implemen-
tation that on top of implementing the above mentioned specification,
will also attempt to implement remote desktop configuration management.

This (new) analysis of course embeds the desktop-config-standard
specification. It will be the last version of this specification we'd
like to implement (when it's finally accepted and a first version
released).

It also adds some use-cases in a functional analysis. It adds a protocol
specification for remote notification of changes (by pushing them) using
the XMPP protocol and using JEP 0072 (SOAP over XMPP. XMPP is very
interesting cause of "rfc3920.html#rules" which defines that a service
must make sure a message gets delivered, also if the client was
unavailable at the moment of the "sent" event). And finally it adds a
protocol specification for getting key/value pairs using a remote SOAP
service.

ps. So basically if the specification becomes accepted, organisations
and opensource developers interested in the field of desktop config can
(finally) start looking at interesting concepts "like" remote desktop
config management and group policies (yet making sure they are building
something that might actually get used some day). 

This isn't the only interesting concept that could be done. If we just
learn to agree on specifications for the desktop. (this isn't easy)

I.M.H.O. there's an awful big amount of interesting concepts and ideas
that are being blocked or on hold because some critical parts of the
desktop haven't yet been specified (like desktop config management).


-- 
Philip Van Hoof, software developer at x-tend
home: me at pvanhoof dot be
gnome: pvanhoof at gnome dot org
work: vanhoof at x-tend dot be
http://www.pvanhoof.be - http://www.x-tend.be