[Foss-cafe] Novell's new desktop

[dep]

quoth Fred Smith:

| Well, it allows it via an accessable API, but I don't believe it's
| any sort of standard.  Kmail isn't just going to work right out of
| the box unless someone writes an EDS kioslave or a module for EDS
| that publishes the data as an IMAP-style service.

*any* email app is compatible with any other so long as the same api is 
used. as you put it, it would be just as easy -- actually, easier, in 
that kmail existed before evolution did -- to say that kmail produces 
the standard, so long as everybody else writes to the kmail format.

| KMail *could* just access the standard mbox files in ~/evolution (or
| ~/.evolution in 2.0), but that's not the best way to do things as
| mbox files are flat files and have problems when multiple
| applications try to write to them.

which is my entire point from my first post in this thread: that the 
locus of the standard needs to be in mail storage and index files, so 
that *any* mail app can go there and read and write with the same 
effect as if it had handled all the mail there. 

| Well, there *are* standards for mailboxes with
| subdirectories(MailDir), address books(VCF), and much of what
| constitutes a mail client configuration (mailcap).  If Kmail is
| breaking mailboxes or address books between versions, then Kmail
| likely isn't following standards yet again.

not really. kmail is apparently breaking its own configuration files, 
nothing else (though to me that's quite enough). there is no *good* 
reason why i cannot fire up kmail, xfmail, evolution, sylpheed, mutt, 
pine, or any other mail client and have it compatible in both 
directions with every other mail client on the machine without its 
creating *any* client-specific things beyond the mailer identifier line 
in the headers of sent mail. fact is, there is no reason why this 
shouldn't even extend to read/replied flags. if novell were to work 
with, say, the free desktop group to create standards, and would then 
employ those standards in their desktop of the future or whatever they 
call it, it could be very useful, even as the corel people actually did 
good by bringing a certain set of standards to kde apps. (and i am 
first to confess that i was tremendously skeptical of corel getting 
involved. as it turned out, i was wrong.)
-- 
dep

the mind earns by doing. the heart earns by trying.