[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[school-discuss] Eduml Dream

Hello and  thank you for a particularly stimulating mailing list
archive this month!

My name is Bruno Vernier, I coined the term "Eduml" around 1998-9 and
co-authored the proposed EDUML (Education Markup Language) with various
members of the seul-edu mailing list and some participants from a Linux in
Education in France mailing list.

I had discovered the importance of XML and was keen to see its potential
realized in our field where we have to  manage an increasingly huge messy ecosystem
of educational software (both pedagogical and administrative) in our spare
time.  More so with commercial software, but very real  in open source /
GNU software: hardly anything seemed to communicate seemlessly with anything

Our Dream with EDUML was to start an esperanto of educational data format
by providing a standard for open source programmers to follow from now on,
but more realistically, to provide a standard for us, spare time system integrators,
to write export and import routines for the stuff actually in our LANs and WANs
with a view to slowly but surely eliminate redundancy and easily integrate
new applications side by side with old ones.

Several month later (1999) , I discovered both SIF and IMS
(http://www.imsproject.org) and several fellow travellers.  Reading the
mailing list archive for this month made me relive my journey back then:

- I was both excited by the possibility that large numbers of commitees
would do all the detail work for ironing out a universally accepted standard
(the best part being that they seemed to have a chance at convincing the
world given their combined P.R. strength)  ...

- I was put off by the purely commercial goals of the "non-profit"
organization running the show:  $1000 membership or no say whatsoever in the
design, and in any case, since their progress was excruciatingly slow, I
felt paralysed: should I continue to work on the Eduml specs and begin my
own "P.R." or should I wait and see what would come out of IMS and SIF?

- I waited.  Eventually, I became convinced that IMS was becoming a very
serious international player, uniting various European groups and the US
army's formidable distance learning branch, etc ...  I was less than
impressed with SIF because of its very US-K-12 centred (notice the canadian
spelling of that last word) selection of tag names. The SIF organizers
clearly stated that they aimed only for the North American market.

- Canada (where I am) has begun to focus on CANCORE which is a slight
variation on IMS and definitely part of the IMS family.

Meanwhile, I got very involved with the zope web application server which is
very XML centric, and now the CMF (Content Management Framework), a module
on top of zope which stores all objects using DUBLIN CORE (which is the
grandmother of all Metadata XML projects, it seems)

I use a combination of SQL (postgresql, specifically), xml/xsl, to convert
data from commercial programs like remark, as400 mainframe, windsor into
something useful for TTW (Through the web) report card writing for teachers,
or library circulation, or mixed online course management.  Now that XML/XSL
and their cousins are readily available, I have renewed interest in the same
thing you are talking about here:

a method, a toolset, an attitude, hub server a la ZIS, that allow us to
share learning objects, import/export routines, and in some very specific
cases even actual students and courses among ourselves

My gut feeling these days is that Microsoft has put all its effort on .NET
and passport which either compete with or extend past the ZIS ideas, just as
the original IMS plans completely revolved around CORBA , only to be totally
scraped in favor of XML ... we live in an era where programming fashions
change so fast ... although I hope and count on XML being popular for a few

Several of the programming - oriented people here have proposed various
technical solutions (and I talked about my own as well).  Let's keep in mind
that it is an utter waste of time to even attempt to agree on a common
platform (web-based or not, perl or java or python or C or C++), platforms
(J2EE, zope) ; all that is over in the XML, RPC (.NET) era; what matters is 
that we agree to have our programs talk to each other seamlessly.  

The only debate should be over which XML (or maybe even non-XML) format is
likely going to be adopted naturally by everyone  (so we don't waste our
time too much programming for tiny niches) ... I claim that if we must
choose one, it ought to be within the IMS XML standard which is truly
international and backed by the biggest players in the online learning
field. Like Canada, we can fork off a little bit and recycle the name EDUML
for the open source variant which would  be sufficiently compatible with the
official commercial norm as to be usefully interoperational but not too
undignifiedly compliant (pardon the pun)

The beauty (and efficiency) of this idea is that our starting point would be
the IMS XML schemas and as we apply them in our real day to day situations,
we would feel free to adapt them to our (open source sharing) way of 
thinking and working. 

P.S. although I have (for the sake of oratory) opposed SIF with IMS, in
practice, they do not compete but rather profess to be compatible one with
the other, with SIF being North American and for traditional LAN/WAN applications
and IMS more oriented towards online (internet-based) applications