[DotGNU]DotGNU-Jabber Meeting Logs

[Adam Theo]

discussed:

a GPL Jabber Server
P2P Conference
Identity

[14:05] * theo has changed the topic to: The DotGNU-Jabber Meeting
[14:06] <theo> ok, while mike gets over,
[14:07] <theo> i want to say hello, everyone!
[14:07] <theo> and sorry for accidentally planning this on Thanksgiving, 
for all you fellow americans out there  :)
[14:08] <theo> um,
[14:08] <theo> hello?
[14:08] <jan> hi adam, hi everyone else!
[14:08] <theo> ok, good,
[14:09] <nb> Hi Jan, hi theo, hi karjala
[14:09] <theo> i was starting to worry i was in the twilight zone here  :)
[14:09] <karjala> hi
[14:09] <theo> ok, first item: "a GPL'ed Jabber Server"
[14:10] * theo has changed the topic to: GPL'ed Jabber Server
[14:10] <theo> 
http://subscribe.dotgnu.org/pipermail/developers/2001-November/001430.html
[14:10] <theo> that's a link to a post i made a few hours ago,
[14:10] <theo> about my discovery of how a GPL'ed Jabber Server exists 
already.
[14:11] <nb> This is very good news.
[14:11] <theo> it seems that up until v1.1.3 of the open source server, 
it was GPL'ed.
[14:11] %% mike has joined
[14:11] <theo> then starting with v1.2 it was completely relicensed to 
the new JOSL.
[14:11] <theo> hi, mike!
[14:12] <theo> glad you could make it.
[14:12] <mike> heya
[14:12] <mike> yeah. It seems that the jabber.org <> theoretic.com link 
isn't 100% right now. Or it might be my connection
[14:12] <theo> nb: what are your thoughts on this discovery?
[14:12] <mike> Anyway, one q - why is a GPLed Jabber server necessary?
[14:12] <theo> no, i'm having s2s problems with jabber.org, too.
[14:13] <mike> ok. is it the DNS propagation?
[14:13] <theo> mike: because dotGNU would want to run their own Jabber 
implimentations, as part of dotGNU.
[14:13] <nb> mike: Why we need a GPLed Jabber server...
[14:13] <theo> this could only be done in the GPL.
[14:13] <nb> Three reasons:
[14:14] <theo> mike: nope, DNS is done. it's just seems to be another 
jabber.org glitch.
[14:14] <nb> 1) People want an IM system as part of DotGNU...
[14:14] <nb> this means among other things that we must be able to 
integrate a server for whatever IM system we use into the webservices 
server stuff...
[14:15] <nb> This won't work if our stuff is GPL'd and the server has a 
GPL-incompatible license.
[14:16] <nb> 2) Jabber has also been proposed as a protocol that DotGNU 
could use for other things (IM communication between pieces of software)...
[14:16] <nb> Especially the JAM stuff, and the Jabber p2p ideas seem 
suitable for this purpose...
[14:16] <theo> yep.
[14:17] <theo> i'm particularly interested in the JAM aspect.
[14:17] <nb> HOWEVER we cannot for reasons of priciple adopt a family of 
protocols where we would essentially provide momentum to a 
non-GPL-compatible project.
[14:18] <mike> is that still alive? What's the major difference between 
JOSL and GPL then?
[14:18] <theo> JOSL is mozilla-based.
[14:18] <theo> it is not compatible with the GPL, meaning code cannot be 
forked over to the GPL.
[14:19] <nb> 3) If it's true that Jabber,Inc took a GPL'd technology, 
bought it out by hiring the developers, and changed the licensing so 
that they would have some hooks on it, then what they've don eis just 
plain wrong / unethical, and it wouldn't be right to go along with it.
[14:19] <theo> nb: by "providing momentum to a non-GPL-compatible 
project", you mean just not using the JOSL'ed Jabber server, right?
[14:19] <nb> Yes...
[14:19] <mike> well presumably the developers consented to that 
licensing change, yes?
[14:20] <theo> yes.
[14:20] <theo> they would have to.
[14:20] <nb> mike: But people who contributed with stuff like bug 
reports etc weren't asked.
[14:20] <theo> the JOSL is a very nice license.
[14:21] <theo> i met many people at the p2p conf that were using the 
JOSL license in non-jabber projects.
[14:21] <nb> theo: If DotGNU uses Jabber internally (not just as an IM 
system)...
[14:21] <theo> nb: yes, probably so. it was just assumed that bug 
reporters with patches handed over their copyrights.
[14:21] <theo> although,
[14:22] <theo> to be fair, i don't think there were many, or even any, 
of those in the early jabber server.
[14:22] <theo> from what i've seen, the entire codebase was done by the 
early core team.
[14:22] <nb> then that will add a lot of momentum that to the JAM and 
p2p parts of Jabber.
[14:23] <theo> nb: yep, i agree.it will finally get the JAM started again.
[14:23] <nb> which is a good thing probably...
[14:23] <nb> :-)
[14:23] <theo> nb: would you want to takle the 1.1.3 codebase?
[14:23] <theo> or hold out and try to get the 5 or so copyright holders 
to re-GPL it?
[14:23] <mike> it might be worth reimplementing the server in java if 
you're going to do that
[14:23] <nb> but DotGNU will work only with things that 
"proprietary-thinking" companies don't have hooks on.
[14:24] <mike> And perhaps it's worth considering why they changed the 
license : presumably the GPL wasn't suitable for their needs
[14:24] <theo> nb: how so?
[14:24] <theo> mike: well, yes, for their needs. i don't think we are 
saying they were wrong.
[14:25] <theo> they did what they felt they needed to.
[14:25] <theo> what's important right now is dotGNU needs GPL'ed code, 
and there exists (albeit old) GPL'ed code.
[14:25] <mike> ok
[14:25] <nb> For some reason, they thought (maybe still think) that a 
GPL-compatibel license will not meet their needs.
[14:26] <theo> so nb, think that dotGNU could use the 1.1.3 line?
[14:26] <nb> Yes.
[14:26] <theo> i'll get the post explaining the 1.0 landmark.
[14:26] <theo> http://www.jabber.org/?oid=6
[14:26] <theo> that's announcing server 1.0.
[14:26] <nb> My intuition (which is often right and sometimes wrong:) 
says that they won't truly want to coopertae with us.
[14:27] <theo> yes, quite possibly. unfortunate, but likely.
[14:27] <nb> In this case the way to go is to start from 1.1.3 and fork 
from there.
[14:27] <theo> yep.
[14:27] <theo> ok, so that established.
[14:27] <mike> possibly
[14:28] <mike> bear in mind that that line is UNIX only
[14:28] <mike> is that a problem
[14:28] <theo> a nifty new name for this 1.1.3 fork?
[14:28] <theo> no.
[14:28] <theo> shouldn't be at this early stage.
[14:28] <mike> ok
[14:28] <nb> I propose you informally check with the copyright holder if 
they're interested incooperating with us.
[14:28] <nb> If they're not interested, don't go into long negotiations, 
but fork.
[14:29] <theo> nb suggested "DotJab" as a name for this fork,
[14:29] <theo> yep.
[14:29] <theo> i agree, nb.
[14:29] <nb> mike: Actually it's enough for us if the server-side stuff 
works on GNU systems (GNU/Linux, GNU/HUrd) and they're close enough to Unix.
[14:30] <mike> sorry guys, got to go eat
[14:30] <mike> later
[14:30] <mike> i don't think i can contribute much at the moment anyway
[14:30] %% mike has left
[14:30] <theo> mike: ok, talk to you soon, i hope.
[14:30] <theo> nb: ok.
[14:30] <theo> sounds good.
[14:31] <theo> i'll jabber Jer real quick to ask him,
[14:31] <theo> and try and catch temas and dizzyd soon, too.
[14:31] <nb> Good idea... invite them to come over :-)
[14:31] <theo> oh, none of them are online right now.
[14:32] <theo> prolly all celebrating thanksgiving.
[14:32] <nb> Understandable.
[14:32] <theo> oh, wait.
[14:32] <theo> hold on.
[14:32] <nb> So I suggest we invite them to the next meeting.
[14:32] <nb> If they don't come, we have our answer :-)
[14:33] <karjala> Who is mike?
[14:33] <karjala> I mean, what's his relation to the jabber team?
[14:34] <theo> mike is the main developer of Jabber Identity.
[14:34] <karjala> I didn't catch that from the context of this discussion.
[14:34] <nb> He's the primary developer of the (GPL'd) Jabber auth project.
[14:35] <theo> yep, the two of us are developing a complete Identity 
system for Jabber.
[14:35] <theo> ok, so that's established (GPL).
[14:36] %% mass has joined
[14:36] <theo> and i assume you are interested inJabber for instant xml 
routing, right?
[14:36] <theo> hi, mass.
[14:36] <karjala> theo: who is 'you'?
[14:36] <nb> Me :-)
[14:36] <theo> mass is a jabber developer.
[14:36] <mass> heya :-)
[14:36] <theo> "you" was actually meant broader  :)
[14:37] <theo> you in the 2nd person plural  :)
[14:37] * mass has missed quite a few dotgnu/jabber meetings now
[14:37] <theo> hehe, that's fine.
[14:37] <theo> we just finished talking about GPL'ing the Jabber server.
[14:38] <mass> ahh
[14:38] <theo> mass, did you know the JOSS was once GPL'ed? up till the 
1.1.3 line.
[14:38] <theo> i was shocked when i found out.
[14:38] <mass> yeah, you didn't? :-)
[14:38] <theo> always thought it was JOSL'ed the entire time.
[14:38] <mass> I remember when the license changed
[14:38] <mass> I've been with jabber since.. I think 99
[14:38] <theo> so it looks like dotGNU will be doing a fork of the 1.1.3 
line.
[14:38] <mass> either august of 99 or 98
[14:38] <theo> wow.
[14:39] <theo> from the beginning, pretty much.
[14:39] <mass> yep
[14:39] <nb> Mass: Are you one of the copyright holders of JOSS?
[14:39] <mass> about four months before the 0.7 server came out, which 
switched to pretty much what is the current protocol
[14:39] <mass> nah; I wrote jabberbeans
[14:39] <theo> oh, so the protocol implimented in the 1.1.3 line is the 
same as 1.4?
[14:40] <mass> well
[14:40] <mass> more or less
[14:40] <theo> ah, ok.
[14:40] <theo> minor changes.
[14:40] <theo> gotcha.
[14:40] <mass> probably less :-)
[14:40] <theo> ah, ok.
[14:40] <mass> 1.4 had some IO changes, that was the big thing
[14:40] <theo> IO?
[14:40] <mass> the goal was to make the server a good framework for 
writing components
[14:40] <theo> as in c2s, s2s?
[14:41] <mass> and things like transports and yeah, c2s and s2s need to 
handle lots of sockets
[14:41] <theo> ah, ok.
[14:41] <mass> so the idea was that you can create all these components 
as building blocks
[14:41] <mass> and run them all within one process, or in separate processes
[14:42] <nb> We're building something like this in DotGNU anyway.
[14:42] <nb> So we could just use that stuff...
[14:42] <mass> also everything within the docs.jabber.org 'draft 
protocol' section which requires server code was implemented in 1.4
[14:42] <theo> ah, ok.
[14:42] <mass> I would of course recommend trying to get the right 
people in the right place and see if the server can be dual-licensed, 
rather than (basically) forked
[14:43] <theo> yes, going to try that first.
[14:43] %% mike has joined
[14:43] <mass> that would by far be the best for everyone
[14:43] <theo> yep.
[14:43] <nb> Yes...
[14:43] <theo> hi, mike
[14:43] <theo> ok, thanks mass.
[14:44] <mike> heya
[14:44] <theo> well,
[14:44] <nb> I heard from Bradley Kuhn (VP FSF) that he and RMS have 
already tried talking with them about this...
[14:44] <theo> that concludes the GPL topic, i guess :)
[14:44] <theo> nb: eh?
[14:44] <mass> ahh really?
[14:44] <theo> nb: talking to them about what?
[14:44] <nb> it may take a real threat of a fork to get them to change 
their mind :-)
[14:45] <theo> oh, interesting...
[14:45] <theo> didn't know that.
[14:45] <theo> wow, that's interesting. thanks norbert.
[14:47] <mike> ping
[14:47] * theo has changed the topic to: O'Reilly P2P Conference Report
[14:47] <mike> still here then
[14:47] <mike> ok cool
[14:47] <theo> alrighty.
[14:47] <theo> as many of you know, i attended the P2P conference,
[14:47] <mass> (lucky bastard ;-) )
[14:47] <theo> sponsored by IMissary <www.imissary.com>
[14:48] <theo> IMissary is Heg's new start-up,
[14:48] <theo> to develop and sell jabber products, as well as jabber 
consulting.
[14:48] <theo> he was gracious enough to provide money for room and 
board up there.
[14:48] <theo> there was alot of talk about Jabber,
[14:48] <theo> after i brought the topic up.
[14:49] <theo> and even heard dotGNU come up a couple of times.
[14:49] <nb> Is imissary a Free Software company?
[14:49] <nb> (Their website doesn't seem to work)
[14:50] <theo> on dotGNU, i think there was a panel talking about 
Identity Services and the Liberty Alliance that mentioned it as a viable 
alternative to dotNET.
[14:50] <theo> nb: no, it isn't. it's commercial.
[14:50] <mass> website worked ok here
[14:51] <theo> and i brought it up in my Identity Services speach.
[14:52] <theo> overall, alot of focus went to .NET.
[14:52] <theo> *alot*
[14:52] <mike> no surprises there
[14:52] <theo> from the atmosphere of the conference, alot of people are 
considering .NET as a viable platform to adopt.
[14:52] <mass> *grin* I still haven't quite figured out what microsoft 
thinks .Net is
[14:52] <theo> this is mainly to do with 2 things, IMO:
[14:52] <mike> lol. They have a definition somewhere
[14:53] <theo> the "gloss" that Microsoft has surrounded .NET with.
[14:53] <nb> mass: The difnition is changing all the time... .NET is 
simply about catching everyone in a .NET and locking them in.
[14:53] <theo> all the presentations about .NET portrayed it as a this 
bold, innovative new thing that was *clean and simple*.
[14:54] <theo> that is what seemed to be the theme MS used. *clean and 
simple*.
[14:54] <theo> thought that might be of interest to dotGNU.
[14:54] <mike> and i have to admit - it is
[14:55] <theo> and that dotGNU could do the same by adopting a clean and 
simple [platform, Jabber, for an underlaying protocol.
[14:55] <nb> Were those MS representative speaking, or independant 
companies?
[14:55] <theo> mike: it is? yeah, i havn't had a chance to look into it 
much.
[14:55] <mike> The programming model is
[14:55] <mike> HailStorm is something different
[14:55] <theo> MS reps were everywhere, but those were speakers.
[14:55] <theo> mike: ah, ok, i see.
[14:55] <theo> one thing:
[14:56] <theo> .NET is fully using http to route.
[14:56] <theo> dotGNU could do better in the realm of power, 
flexability, and security by using Jabber.
[14:56] <mike> not true
[14:56] <mike> they have developed something called DIME
[14:56] <theo> just thought i'd plug this again :)
[14:56] <theo> DIME? never heard of it....?
[14:56] <mike> it's like a lightweight Jabber, but without all the s2s 
routing etc.
[14:56] <theo> oh, huh....
[14:57] <theo> can it be found anywhere?
[14:57] <mike> no, it's still pretty new. Direct Internet Message 
Encapsulation i think
[14:57] <theo> oh, ok...
[14:57] <theo> sounds interesting.
[14:57] <theo> i'll try to look into it.
[14:57] <nb> Actually DotGNU wants the s2s routing.
[14:57] <theo> hm...
[14:58] <nb> We want dollars, not just dimes :-)
[14:58] <theo> is it just me, or do others see potential battle lines 
being drawn in technology between these different groups?
[14:58] <mike> hehe. Not really
[14:58] <theo> i mean, the Free and Open world potentiall using Jabber,
[14:58] <theo> .NET & other proprietary companies using DIME,
[14:58] <theo> and AOL will whip something out, i'm sure...
[14:59] <nb> AOL hates MS so much...
[14:59] <theo> ah, well. just me  :)
[14:59] <mike> DIME will probably be an open protocol, or basically open.
[14:59] <theo> nb: yep. they will never use .NET.
[14:59] <nb> that we actually have a chance to get them supporting our 
standards.
[15:00] <theo> nb: maybe... but i think AOL sees itself as the "real" 
internet, and will do their own thing, but ultimately blunder it.
[15:00] <mike> hmmmm, doubt it. maybe though. knowing aol they'd rather 
die than use something that wasn't developed by them
[15:00] <theo> yep.
[15:00] <nb> The trick is that we need to get the banks on our side first.
[15:00] <theo> yes, true.
[15:00] <nb> Everyone wants to be able to process payments :-)
[15:00] <theo> ok, any questions about the p2p conf?
[15:01] <mike> banks?
[15:01] <mike> ah yes, i see
[15:01] <theo> mike: yeah, secure, safe transactions.
[15:01] <theo> i've read that many banks that are tech-savvy are 
skeptical of .NET's security.
[15:01] <nb> The banks hate Microsft, too.  Thatä's our big opportunity.
[15:01] <theo> yep.
[15:02] * theo has changed the topic to: Identity
[15:02] <theo> ok, mike, have any updates on Jabber Identity/Genio?
[15:02] <theo> Genio, btw, is the new name for the system, while Jabber 
Identity is the jabber-specific implimentation.
[15:03] <mike> not really. I've been redesigning it a bit lately, and 
getting the hang of Java, as I'm planning on reimplementing it in Java 
using servlets/APACHE SOAP
[15:03] <theo> ah, ok...
[15:03] <mike> Yeah. Genio is the system. Jenio is the jabber based 
implementation. Or at least, that's current thinking
[15:03] <mike> i'm redesigning that oo
[15:03] <theo> ok, cool.
[15:03] <nb> Do you have a website up yes for Genio?  Released code?
[15:04] <blalor> or a quick overview of what Genio is?
[15:04] <mike> this is all subject to change without notice of course :)
[15:04] <mike> almost
[15:04] <mike> it's coming soon
[15:04] <mike> for now, www.theoretic.com/identity
[15:04] <theo> right now it's just at Theoretic
[15:05] <theo> oh,
[15:05] <theo> and i'm going to try and get into the Liberty Alliance, 
and pitch Genio to them as a solution.
[15:05] <theo> they are looking for something, i hear.
[15:06] <theo> and Genio fits everything they need.
[15:06] <mike> Right. We are hopeful about that
[15:06] <blalor> I'm a little confused; how would something like Genio 
work across messaging platforms?
[15:06] <blalor> jabber s2s is great for a distributed network of 
servers, but what about something that's on a different network?
[15:07] <mike> Genio and Messaging is still being designed. Basically I 
see it one of the web services you can associate with offering platform 
independant access to messaging networks
[15:07] <theo> Genio is not language-specific.
[15:07] <mike> So you can use Jabber as the messaging system if you 
want, or else use a different method of sending messages/notifications etc
[15:07] <theo> it just really defines what components are needed in a 
Genio system,
[15:07] <blalor> I think a big-picture overview with lots of little 
pictures and arrows is much needed
[15:07] <theo> and lets the various programming languages and platforms 
do that how they can.
[15:08] <mike> Or platform specific
[15:08] <theo> yep, we're working on that.
[15:08] <theo> i'll be doing the big picture stuff, probably.
[15:08] <mike> It's coming! Don't worry- problem is right now the big 
picture keeps changing, and the little arrow keep moving
[15:08] <theo> hehe.
[15:08] <nb> :-)
[15:09] <theo> ok, i think that wraps this up.
[15:09] <theo> is there anything else real quick?
[15:10] <theo> ok, that's it.
[15:10] <theo> thank you everyone,
[15:10] <theo> i'll put some stuff together and see if there is need for 
another meeting next week.
--
   /\    -- Adam Theo, Age 22, Tallahassee FL USA --
  //\\   Theoretic Solutions (http://www.theoretic.com)
 /____\    "Software, Internet Services and Advocacy"
/--||--\ Personal Website (http://www.theoretic.com/adamtheo)
   ||    Jabber Open IM (http://www.jabber.org)
   ||    Email & Jabber: address@hidden
   ||    AIM: AdamTheo2000   ICQ: 3617306   Y!: AdamTheo2
 "A free-market socialist computer geek patriotic American buddhist."