[SCL] SCL draft standard first version here

[Tanel Tammet]

Hi SCL-ers,

Since the SCL interest has clearly warmed up, we have been
moving closer to an understanding on basic issues and
last not least, Harry could perhaps use our SCL papers
if we manage to do these in time, I have taken the courage
to write the "first draft version of SCL spec".

Now, this is "first draft" most definitely in quotes:
Pat and other people had the basic ideas ready
long time ago, Chris wrote several serious specs
on SCL syntax and semantics, I wrote several versions
of SCL->FOL translations, people have been working on
XML syntax etc etc.

My goal was to:

- Write up contents page and the structure
- Write the drafts of the introductory parts

so that it would be easier for everybody to
move on with practical steps: that is,
filling in the concrete chapters and
modifying the structure as necessary.

In particular, I have _not_ yet put
into this draft the semantics chapter
by Chris or the SCL->FOL chapter, not
to speak of XML syntax.

All these need a little bit of work and
can then be included by their authors at
the first possibility.

Please feel free when adding content
to the draft, as well as adding and changing
the structure.

It would be most welcome if anybody comes
up with the improvements or rewrites of the
introductory part as well.

A note about the proposed structure
of SCL into core and full SCL:
observe that the Horrocks problem
and similar unpleasant effects do
not appear at all if we consider SCL
or FOL with no defined predicates
or functions. In this minimal fragment
everything is fine and simple. Hence
it IMHO makes sense to NOT have
equality (or integers, etc) in core
SCL. It would also make it much easier
to write semantics and translation
algorithms: we can first present the
main stuff just about the minimal core,
and then bring the somewhat more complex
stuff in at suitable places, where
we introduce equality, annotations, etc,
all independently of each other.

What I propose to do next for SCL:

- please read the draft through (as said,
   it is just the structure (modelled after
   the RDF spec paper) and intro parts

- let us keep discussions WITHOUT making
   changes to the paper to the bare minimum:
   it is more productive at the current
   stage if ideas are accompanied with
   concrete additions and modifications
   to the draft.

- if you have any ideas of changes, simply
   make the changes and send the modified paper
   to the SCL list along with notes in the email
   about where the changes are

- if you can add some chapters or other parts,
   please do so and send the modified paper to
   SCL list along with the notes in the email.

- I will volunteer to attempt to merge the
   modified and filled versions periodically
   and will send these to the list,
   unless somebody else volunteers to do it
   (would be most welcome :-)

- If nobody comes up with changes and additions
   (which I hope will NOT happen, in particular,
   since people already have large draft parts of the
   SCL draft :-) then I'll simply continue to write
   the other chapters (except standalone semantics).
   We need the SCL spec.


Regards,
      Tanel Tammet

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://philebus.tamu.edu/pipermail/scl/attachments/20031031/77cd140b/scl30oct-0001.html