CDFN: WG organization

[Andreas Neumann]

Hello!

Peter Matthews from the UIS informatics commission asked me to organize
the Working-Group discussing the establishing of a cave and later karst
data exchange format. As the discussion already started and I didn=B4t have
time to set up a website til now (I will do that on the weekend) I just
want to add my ideas concerning this group. Most of the common goals have
already been summarized by Peter Matthews in his mail from Tuesday January
9. I do not have much to add in the short run. I will publish these ideas
on the future website.

My personal background is in fields of cartography. I am currently system
administrator at the Institute of Cartography, ETH Zurich. I did some work
on the use of webstandards for the purpose of representing cartographic
data (SVG, XML, VRML, serverside techniques) - see http://www.carto.net/
and http://www.karto.ethz.ch/neumann/ for some examples. I work with
students of our institute and also closely with Martin Heller from
Toporobot.

I did not want to push myself forward to do this job, but was asked by
Peter if I want to do this. If anyone else feels like (s)he wants to do
this they should please contact Peter or me. Otherwise I would be willing
to do this job, knowing that other people already started to work on this
topic (f.e. Devin Koutes, Martin or Mike Lake who already published their
work on their websites). Of course I want to collaborate with them and
integrate their ideas.

The website to be started on the weekend could be hosted at speleo.ch or
karto.ethz.ch as I do not have access to the official UIS website - Is
that ok for you? or do you have any better alternatives? This page would
cover first general goals and purpose of the group, a general introduction
to XML and the reasons why we want to work with XML and related
techniques. Furthermore we should provide links to related webpages and
tools to work with XML and cavedata as well as this discussion list and
the archive. Finally I would suggest a roadmap on how to go on with the
work.

Given the experience from todays postings I also would like to suggest
that we establish a new mailing-list covering only the problems of the XML
exchange format.

Concerning the name of the standard I would suggest the name "CDX" or
"KDX" for "Cave/Karst Data Exchange Format" with the X representing
"eXtensible/eXchange/XML". "CKX", as Peter proposes, is not so easy to
pronounce but integrates both. If we are going to cover more general data
on karst also, as I personally would like to (depends on the group
activity) - "CKX" is probably best of these three. John is right that
caves may also be found outside of karst, although a minority. "CKX" does
not explicitely exclude caves outside karst. In order not to bloat it and
to get realistic milestones I suggest to start with "CDX" or "CaveXML". We
could even split it up in sections like "CDX-S" (Surveying) "CDX"-Mapping
"CDX-G"-Geology, etc. - but anyway, its primarily a working title ... SML
(Subterrainean Markup Language) elegantly combines caves with mines and
other artifical cavities but is maybe too general as we cannot primarilly
integrate other underground structures such as tectonic, stratigraphic or
geologic underground features in the short run.

I do not want to hurry, although I know that there is quite a demand for
such a standard, because discussion and first results should be
deliberate. Just think of that every standard (such as the ones provided
by the W3C need there time) to gain quality and completeness from draft to
draft. However I want to provide a roadmap and deadlines for discussion.

Concerning the discussion "DTD" vs. "Schema" my opinion is definitely that
we should provide both: a DTD because it can be used by the newest
browsers to provide at least some sort of validation and is easy to
generate. A "Schema" because we can specify specific datatypes and ranges,
although it will take some to time until it really supported by all
involved tools and parsers.

Related techniques such as XSL (both XSLFO and XSLT), XPointer and XLink,
later XQL and the likes should be covered to the extent they are useful
for our purpose. In priority they are queued after the XML/DTD/SCHEMA
topic but people can already work in parallel. In particular interest for
us should also be SVG and VRML/X3D.

What makes CDX/CaveXML more complicated are different philosophies in how
a cave should be structured. Most programs store the data chronologically
and/or hierarchically. Toporobot and CaveRender store morphological
entities - currently galleries - that also better allow topologic
operations and 3D visualization. We have to integrate both approaches in
the data-format ...

Ok, so long for this evening ...

I will shortly post an alternative list to separate discussion from the
more general surveying issues ... although I am still unsure of the name
the list should have - my favourites are "CDX" and "CaveXML".

Andreas Neumann



--
"If you have got the right attitude,
 interesting problems will find you."
                    (Eric S. Raymond)

----------------------------------------------
Andreas Neumann - Department of Cartography
Swiss Federal Institute of Technology (ETH)
ETH Hoenggerberg, CH-8093  Zurich, Switzerland
Phone: ++41-1-633 3037, Fax: ++41-1-633 1153
e-mail: andreas.neumann@karto.baug.ethz.ch
www: http://www.karto.ethz.ch/~an/