Hi Jan,
Hi all,

thank you for the summary and thank you for your efforts. Maybe things have 
changed also for PediaPress/the Collection extension? Maybe we can have another 
try to get support from them? 

As Denny Vrandecic is now working for the Wikimedia Foundation Germany and 
PediaPress (the company behind the Collection extension) is stating to have a 
partnership with the Wikimedia Foundation, maybe the time is now better?

Denny, can you maybe talk to the PediaPress people about this?

regards,
Bernhard

----- Ursprüngliche Mail -----
> The way I understand Collection, it grabs the raw wikitext and
> renders
> it itself. It "knows" the standard magic words, transclusion etc, and
> the main extensions such as Cite/Math/ParserFunctions ("main" meaning
> they're used on Wikipedia), but it does not understand anything that
> any
> other extension adds to the wiki unless its functionality is
> implemented
> "within" the Collection render server as well. Also, any information
> that's not directly in the wikitext is not available to the render
> server, even if that render server is on the same machine.
> 
> You can see this for yourself by configuring Collection to use the
> PediaPress render server; *obviously* that server has no access to
> the
> SMW tables, and thus no knowledge of the semantics. Thus, while that
> server doesn't understand #ask and #show, it couldn't access the
> semantic data even if it did.
> 
> October '09 I've dived into Collection, exchanged some mails with the
> mailing list and one Ralf Schmitt, but they were unwilling to support
> SMW, and indeed unwilling to explain how to solve the problem. They
> just
> suggested to send in a patch.
> 
> At the time I believed a solution could involve something like
> extension
> ExpandTemplate on the source server, where Collection would first run
> the raw wikitext through some sort of expandagizmo so that all parser
> functions had done their jobs, and be replaced with pure MW code
> (e.g.
> an {{#ask: <criteria>|format=count}} woud be replaced with <result>).
> This could then be rendered on the Collection render server. However,
> since then I learned about js-stuff and css, so there probably are
> some
> more things to tackle anyways.
> 
> I still think it'd be doable, but the problem should be tackled in
> general, not just for SMW, or else Collection could fall on its face
> again as soon as I install any other extension like Variables,
> ArrayExtension and HeaderTabs (which I now use). This then requires
> cooperation between at the least a Collection application architect
> and
> a MW application architect (if there even are such persons/roles);
> then
> adding SMW application architects to the discussion would likely
> help.
> But the Collection people seemed quite unwilling to act in '09, and I
> couldn't say if that has changed anywhere in the mean time. Back then
> they seemed only interested in good prints from Wikipedia.
> 
> On 20:59, Krabina Bernhard wrote:
> > Hi,
> >
> > I think making the Collection extension&  SMW work together is a
> > very important functionality!
> >
> > Are there SMW developers that would be interested to have an eye on
> > that?
> >
> > Are there implementors/customers who would be willing to provide
> > funding?
> >
> > regards,
> > Bernhard
> 
> --
> Jan "Saruman!" S.
> "I'm a stream of noughts and crosses in your R.A.M."
> 
> 

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure 
contains a definitive record of customers, application performance, 
security threats, fraudulent activity, and more. Splunk takes this 
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
Semediawiki-devel mailing list
Semediawiki-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/semediawiki-devel

Reply via email to