[plug] Opensource groupware STILL SUCKS!

Mr Shayne shayne at guild.murdoch.edu.au
Wed Nov 9 16:41:31 WST 2005


>> Another option is to get a group of clients together to offer some
>> sort of incentive to programmers / developers with clear goals,
>> deadlines and requirements spelt out to meet your need. 
>
> Yes, that's how things normally work. You want a product or a piece of 
> software, you pay someone to make it for you. Then you sell it to recover 
> your money.

The problem as I see it, is that all the "little pieces" of the puzzle are 
held by multiple different companies, and dev groups, all with there own 
little agendas. Now, if you look at the "biggie", thats the wholl kolab 
thing, wierd stuff emerges.

kolab was funded by the german govt to build a real exchange killer. To a 
degree they succeeded , granted Im not convinced the "lots of little 
parts" aproach is particularly coherent, but whatever. Now, the protocols 
where based to talk via IMAP, and have specially formatted messages to 
cope with calendars, apointments and the various fruit. hackish, but it 
worked. Along comes the old warhorse of FOSS groupware, citadel, and says 
"Yep. That'll do." and implements the protocol. A "groupdav" committee is 
also put together, and soon groupdav etc looks to be a new standard , with 
the opengroupware and others coming on board. Now citadel got there first, 
put in support for the kolab protocol, and you now have *2* products that 
will talk tango, plus some upcomers. Add to that Toltek etc with the 
connectors, its all looking good.

Sudenly the kolab people release version 2 and come up with a NEW nutty 
protocol, and all the plugin makers change to support the new one, leaving 
the rest of the market in the lurch. Huge shitfights emerge and the whole 
thing falls in a heap. Kolab worry me, my experience with the developers 
left me deeply concerned. Insisting distributions use there own custom 
ultra-nutty packaging system (infact when I said that surely they should 
support rpm and deb, they said "No. there crap because they dont support 
our packaging system". alarm bells ring at that point.)

So we are back at square one. kolab , which has to be one of the most 
invasive packages I've seen, may be one of the few that'll work, if you 
are prepared to pony $50+ for the connectors, and btw, these connectors do 
*NOT* provide a full mapi backend, just a sync facility. Citadel is back 
on the border despite being probably the most competant server out there 
but clientless (except for aethera). 
But Aethera is still stuck in the stone age. Evolution is still strictly 
linux/bsd. chandler is still a joke. the connectors all suck. Sun seems to 
of given up on ever getting glow to work (The web page at sun still says 
0.2 alpha will be out reaaal soon now late 2003 or something , but 
probably not that) and ah man. what a mess!

Meanwhile, its still really just groupwise or the ibm thingee thats 
available, and if your a tiny little student org with precisely zero 
budget and a real grim future cos of vsu, well.... tough shit.

>
>
> Alex
> _______________________________________________
> PLUG discussion list: plug at plug.org.au
> http://www.plug.org.au/mailman/listinfo/plug
> Committee e-mail: committee at plug.linux.org.au
>



More information about the plug mailing list