[DG: Teaching & Learning] [DG: User Experience] Wiki's and Sakai

[Michael Feldstein]

>> One more note on implementation. I have suggested so far that this is
>> something that could and possibly should be built into the Sakai core. But,
>> as I have brought up several times before on these lists, there is a second
>> way to achieve this aim--one that may require little or even no coding. If
>> every page in the Sakai 3 UI is simply HTML and Javascript, then every page
>> can be managed in a standard web content management system, which would
>> bring page versioning, rollback, and granular access privileges for free,
>> along with some other capabilities such as the ability to require different
>> access privileges for editing different portions of the page, approval
>> workflows, and other good stuff. My current instincts on the best way to go
>> here has evolved somewhat; I now believe that it's a good idea to take the
>> first route (i.e., building the basics into core) but also to make sure that
>> the conventions followed facilitate individual adoptees taking the second
>> route for the extra value added where that makes sense for them.
>>     
>
> On that implementation note, what if it what were deeply rooted in
> Sakai was a 3rd-party framework for providing (scriptable) RESTful
> access to content?  It seems to me that would offer the promise of
> accomplishing the various aims you lay out, doing so in a unified way,
> and not having to write the core functionality ourselves.  "Sakai
> core" itself might be a misnomer where it's really specializations
> (especially in the way of authorization complexities) riding on top of
> something more generic.  I think K2 is heading this direction.
>   
Yes, this is also consistent with suggestions I have made in the past. 
In a world where plenty of decent web content management systems exist, 
I think you have to make a compelling case for building yet another one 
rather than re-using one that already exists. Maybe that case can be 
made, but if so, it should be made as an alternative to one or more 
real, existing WCM systems that somebody in the community has 
investigated for this purpose.

The trick will be to find something that, in addition to having the 
right license and adequate support from its home community, is low-level 
and lightweight enough that it can integrate comfortably into another 
system without bringing a lot of extra baggage. One significant 
challenge will be UI. Whatever WCM system is adopted should either let 
you hide the UI or have a UI that is simple and compatible as an overlay 
to the current "content authoring" capability. That may not be as 
difficult as it sounds. The current style for "content authoring" is 
mostly an editable page underneath a WYSIWYG editor, which is a common 
convention that most WCM systems should be able to handle. Where you may 
run into more dissonance is on site authoring (i.e., adding and managing 
multiple pages) and more complex functionality (e.g., workflow).

Ideally, whether the community chooses a third-party component to handle 
WCM or builds its own capabilities, it would be done in such a way that 
other alternatives could be swapped in (either by an individual adoptee 
or by the community, should the third-party component chosen become less 
desirable down the road) with a manageable amount of work.

- m


-- 


Oracle <http://www.oracle.com>
Michael Feldstein | Principal Product Manager | +1.818.817.2925
Oracle Academic Enterprise Solutions Group
23A Glendale Road, Glendale, MA 01229
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://collab.sakaiproject.org/pipermail/pedagogy/attachments/20090405/0baeab64/attachment.html 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: oracle_sig_logo.gif
Type: image/gif
Size: 658 bytes
Desc: not available
Url : http://collab.sakaiproject.org/pipermail/pedagogy/attachments/20090405/0baeab64/attachment.gif