[Top][All Lists]

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

RE: [help-GIFT] Looking for an MRML DTD

From: Phil McConchie
Subject: RE: [help-GIFT] Looking for an MRML DTD
Date: Thu, 5 Aug 2004 18:28:03 +0100

Oh OK, I have read in the specs about one of the primary objectives being
extensibility but thought that there might be an XML DTD floating about that
nailed down the exact element and attribute names for those parts of MRML
that appear from the specs to be concrete.
Such as tags for "get-configuration", "get-server-properties" etc.

Perhaps the chap you mention who has been investigating the practicalities
of MRML implementations would indeed be someone to start with.  Is he on
this list?

I figure that anyone who has done an MRML implementation must have a MRML
DTD and we might as well move towards some kind of standardisation for at
least the tags that will be definitely included in any final version.

I would be very interested in seeing your students' error list.  If you
could forward my address as you suggest it would be much appreciated.

Yes, Josh is my project supervisor.  It's not problem asking project related
questions on this list.  Incidentally, how many subscribers to this list are

I am working on this project fulltime at the moment and hope to have the
coding more or less finished in 2 weeks.

Cheers, Phil.

-----Original Message-----
From: Wolfgang Müller [mailto:address@hidden 
Sent: 05 August 2004 17:32
To: address@hidden
Cc: address@hidden
Subject: Re: [help-GIFT] Looking for an MRML DTD


> I am writing an open source (Java) MRML client/server pair of applications
> to be used for establishing MRML connections to multimedia databases and
> subsequently querying/searching them for 'like' files by supplying an
> example multimedia file (QBE).

> I have perused the website and read version 1.0 and 2.0 MRML
> specifications but can't find a definitive 

Not being final, but being extensible was one idea of MRML. 

> and comprehensive MRML DTD file 
> or document anywhere.
> I obviously need this to ensure I make my applications true MRML compliant
> 'beasts'.


> Can someone point me in the right direction?

I think *the* pointer would be to have a chat with Stephane Marchand-Maillet

who (IMHO) has thought most about the software engineering implications of 
MRML 2.0 so far.

Regarding 1.0 I just had some students write an MRML client as a software 
engineering project (i.e. focus on the "making of" and not on the product), 
and they pointed me to some outdatings, errors and typos of the MRML 1.0
report. How much time do you have? I could put you into contact with one of 
my students who has not communicated me a comprehensive error list yet 
(finals --> no time). As I am not sure if he wants his address on the net, I

will forward your address to him and ask him to contact you.

> Regards,
> Phil McConchie
> (MSc Student - Queen Mary, University of London)

You're with Josh Reiss, right? Greetings. May I ask you more precise
about your project publically on this list? Or would you prefer this 
questions be asked and aswered off-list?

> PS:  The idea is that my end deliverables will be able to be used for any
> multimedia file type, not just images like GIFT.

Oh, that's the idea. An unpardonnable mistake when choosing the names of
tags was to go for image and thumbnail, instead of media-item and 
media-item-summary or the like. However, as what's shipped are really URLs 
and (in the HTTP request we learn the MIME type of the media items) the 
initial needs for true *multi* media are cared for.


Dr. Wolfgang Müller
Universität Bayreuth

reply via email to

[Prev in Thread] Current Thread [Next in Thread]