[jdev] Google Summer of Code

[Christopher Zorn]

On Wed, Jun 15, 2005 at 12:19:44PM -0300, Patrick Dalla Bernardina wrote:
> Does anybody know about implementations of workflow using jabber protocol?

I know alittle bit about sluice[1]. I have not been able to work on this project 
in a long while. It is used by MUSC. I want to redo it in python and use idavoll[2] to 
put a pubsub interface to it. But that is when I get some time. 


[1] http://www.itlab.musc.edu/sluice/
[2] http://idavoll.jabberstudio.org/2/


> 
> Richard Dobson wrote:
> 
> >>You're right, but regardless of scaleability and efficiency, it's still
> >>defining more or less the same thing which was already defined in 
> >>JEP-0070.
> >>The title of JEP-0070 pretty much reads as exactly what JEP-0101 does.
> >>It could have been co-authored or something to merge the two ideas so
> >>that it would work either way. :-)
> >
> >
> >It might have but I was working on JEP-0101 long before JEP-0070 
> >actually came out, and I did submit it long before the date it 
> >actually says on it as it took a while and a couple of attempts to get 
> >through to Peter, ive got a feeling it might even have been before 
> >JEP-0070 came out that I started trying to submit it.
> >
> >>Actually, I'm not entirely convinced that JEP-0070 requires a Jabber
> >>server to be embedded in the web server.  The same protocol would work
> >>with only a client embedded in the server, just like existing web sites
> >>don't require an embedded email server to send email confirmations.
> >
> >
> >Sure using the word "server" was probably the wrong term to use, but 
> >it does require some kind of jabber "component" or other to be tied 
> >into the webserver for it to work.
> >
> >>It's a little inefficient, yes.  It's not so conceptually perfect, 
> >>because
> >>it did require the user confirming the message, rather than having 
> >>anything
> >>automated... but I do think the two could have been merged somehow.
> >
> >
> >Maybe so but im not sure how as the author of the other spec seemed to 
> >want to go about it in an entirely different way which was not really 
> >compatible with how I needed it to work (I use a version of this spec 
> >in production and had been before either protocol came out), all I was 
> >doing was documenting a method I use that works very well. After 
> >JEP-0070 came out I considered implementing that instead but it 
> >wouldnt have worked very well in my situation for the reasons already 
> >outlined and since they do go about it in two fundamentally different 
> >ways I didnt see the problem at the time with releasing my spec too, I 
> >thought it might come in useful for someone.
> >
> >Richard
> >
> >
> >_______________________________________________
> >jdev mailing list
> >jdev at jabber.org
> >http://mail.jabber.org/mailman/listinfo/jdev
> >
> _______________________________________________
> jdev mailing list
> jdev at jabber.org
> http://mail.jabber.org/mailman/listinfo/jdev
> 
>