jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From St├ęphane Croisier <scroisi...@jahia.com>
Subject Re: On a Slide/DAV merge.
Date Thu, 22 Dec 2005 14:17:10 GMT
At 18:05 21.12.2005, Stefano Mazzocchi wrote:
>Here's my opinion: the value of an apache 
>project is its community, not the code. Code can 
>be fixed *waaay* more easily than any personal 
>issue or personal opinion (that, yes, can be 
>changed by not by means of a svn commit).
>JackRabbit needs to be able to handle all that. 
>WebDAV interoperability is a nightmare and Slide 
>is far ahead of JackRabbit on that front.
>But should we integrate or simply learn from the code?
>I honestly don't care, whatever it's easier and 
>it seems to be that since there is a community 
>of interested people here, probably an 
>independent effort is easy to manage and avoids a lot of political discussions.

I fully agree with Stefano. What's important is 
the community behind a project not the code. 
Other Apache projects had some hard time because 
some committers and other lead developers left 
suddently the project (e.g. Jetspeed 1.x a few 
years ago which restarted again a few months 
later, from scratch, with a new team and 
delivered Jetspeed 2.0). In fact the same can 
also happens tomorrow to Jackrabbit. What if Day 
decided suddently, for some strategical and 
financial reason, to stop their ongoing investment in Jackrabbit?

So let's not focus on the name of the project. 
Whatever it is: Slide 3.0; DAVrabbit;... what is 
important is that the community behind is large 
enough to develop and maintain, on the long run, 
a good DAV library. Lots of time has been spent 
on Slide during the last years. There are 
certainly a certain number of practices, hacks, 
workarounds we may reuse and combine with 
Jackrabbit. But as of today both of the 
communities are quite small. So why not trying to 
plan to make something more thin, reusable, sexy, 
standard, etc... all together rather than trying 
to keep both projects separated? That is the key 
question. Nobody never spoke about reusing the whole Slide code as is....

On our side (=jahia.org) we would be ready to 
help for such a project. We are currently using 
Slide to store our customers binary files (ey! 
Jackrabbit did not exist 3 years ago ;-) ). As 
mentionned by other Slide users, we are also 
encountering a certain number of issues with 
Slide. So we have the choice of trying to 
continue to make some light workarounds and 
patches (in fact what we did for the last 18 
months) or start again with a new DAV layer. I am 
personaly +1 for the latter. In fact, in our 
roadmap, we plan in all the cases to refactor the 
Jahia back-end to support JSR170 (= Jahia 6.0 - 
2007) as soon as we released Jahia 5.0 (Q1 2006) 
and there are a certain number of reasonnable 
chance we will refactor the whole based on 
Jackrabbit rather than trying to make a whole new 
JSR170 implementation ;-). So why not inserting a 
Jahia 5.5 in between by only starting to move our 
current DAV based "document system" to 
Jackrabbit... Briefly speaking we will be ready 
to help and invest ressources in 2006 on such a DAV project!

So will it be in a Slide 3.0 project, in a 
Jackrabbit DAV extension, in a new dav.apache.org 
top level project, I do not really care if we can 
gather a good community of interested people all together...


View raw message