[Members] Is this thing on?

Bart van Bragt jabber at vanbragt.com
Fri Dec 17 04:19:05 CST 2004


johannes.wagener at gmx.net wrote:
> maybe we should not talk about to make a standart to make a standart 
> but rather code and make a standart? just declaring a project as "the 
> new official XMPP lib project" would make it standart.
I don't really see having multiple libraries/APIs as a problem. You'll 
need several APIs anyway, the number of languages that have a 
Jabber/XMPP library available is almost scary :)

If we are going to put development time into something then we should 
put this time into decent servers. There are a LOT of clients out there 
but only a few half decent servers. There still is only one (unfinished) 
pubsub implementation, etc, etc. IMO there is quite a bit of work todo 
before we should start worrying about sub-optimal libraries :) Besides 
that most platforms only have one mostly finished library available.

Regarding the top-down (early adopting companies) vs the bottom-up (home 
users) thing. IMO we shouldn't disregard the homeusers. The needs of the 
companies and the endusers have a LOT of overlap. Both want easy to use, 
fairly feature rich and stable clients. Both need decent servers that 
they can use or deploy. Both need features that set XMPP apart from 'the 
rest'.

Besides that XMPP's share in the corporate world is nice but IMO 
Microsofts LCS is going to be a very tough competitor. Office 
integration, anyone? :D


Anyway. I don't really care that much about the corporate users, I 
became involved in the Jabber world because I'm fed up with the legacy 
protocols that are completely incompatible, start to become more bloated 
by the day and they all have added advertising to their messages. 
Besides that all my traffic gets routed by a large US company (always 
with MSNM and also with the other networks when firewalled), I can't add 
any cool applications to the other clients, etc, etc.

My reason to support Jabber was to make this world a better place :) 
With an open IM protocol that everyone can use, adapt and deploy. IMO 
there should be an XMPP besides other core protocols like HTTP and SMTP. 
IM is starting to become more important by the day but it's still all 
based on the software and servers of some large corporations. IMO that 
sucks :D

So for me the endusers are the main target. I think that's the same with 
most of the people in the JSF and especially in the larger Jabber 
community. Otherwise we would have seen quite a bit more commercial 
clients/servers and less open source projects.

IMO endusers are key, tackle this from the bottom up, IMO the corporate 
battle is very hard to win with all these small commercial XMPP 
companies that are all separately trying to win from companies like MS.

So Jabber should become the Firefox of Instant Messaging :) But before 
that can happen we need decent info for endusers (but 
www.jabbercentral.org is getting in shape. BTW if someone could ask 
Justin Mecham to set the nameservers for jabbercentral.com to the same 
nameservers as jabbercentral.org I'd be VERY grateful. Can't get a hold 
of Justin, have been trying for months now :( ). Then there is a need 
for decent/stable servers and good documentation and registration 
procedures on the server side.

All this to get the early adopters in the enduser world started. Then 
we'll need to get on par with the legacy networks. If I look at the 
people that I know video/voice is starting to become more and more 
important. Lots of people are buying webcams at the moment (ok, mostly 
the 16 yo girls :D). But I'm also using Skype to stay in contact with 
some people when it's too expensive to use the phone. Works perfectly, 
should work perfectly with Jabber too IMO. But implementing this is 
pretty hard so I think just integrating a library/application with the 
already existing clients is the way to go. Client developers have too 
much work to do as it is, adding voice/video would be a very large extra 
burden if they all had to start from scratch.

And then I'm going to stand on my soapbox again and say it once more: 
WEBINTEGRATION. Yes, I know I'm biased because I'm a webdeveloper but 
I'm in the community business (www.phpbb.com mainly) and there are a LOT 
of opportunities here. There are SO many cool things that we could do if 
it was possible to effordlessly integrate a random website with XMPP 
(without having to install servers or bots). I'm planning on adding 
extensive XMPP integration to my main website (www.bokt.nl) which has an 
audience with 99% MSN users (98% is female, most are not exactly 
computer experts). But if I do the integration properly I can probably 
get a very large part of the users to switch or use something like 
Trillian so they can stay in contact with their other contacts through 
MSN (if only Trillian was free :D).

An added advantage (or maybe even the main advantage) of webintegration 
is that it increases the visibility of Jabber a _lot_. Most people out 
there don't even know what Jabber is. If we can get websites to 
integrate Jabber support (because it's cool, because it's fun and 
because it's easy) then millions of people will almost instantly get 
exposed to Jabber technology, they'll see the benefits and if the 
benefits are large enough they'll switch or at least install the Jabber 
plugin in their multi-protocol client...

Ok, I'll get off my soapbox now :D
It's good to see that you are still alive, it has been extremely quiet 
after the RFC numbers got assigned :\

Bart


More information about the Members mailing list