Re: How to proceed wrt publishing solutions?

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

| What level of effort is perceived as adequate depends on whether it is
| *the* conference in the year where everybody has to publish at any
| price, in order not to perish.
| We must still reach that value perception in the wide audience.

Valid point.

| I wonder if anything more than a solid paper is a suitable precondition
| for accessing the normal (open) June workshop.

OK, lets distinguish between what we require for the June workshop as
paper acceptance prerequisites and what we would like to have in general
and how we deal with the certification and a challenge-only day.

I propose to follow the approach of SIGMOD:
http://www.sigmod08.org/sigmod_research.shtml

Thus, SWSC paper who share their solution code will be tagged by the
organizers with something like "The solution presented in this paper was
verified by the SWSC and is available through the SWSC website."
Other papers will be tagged with something like "The solution presented
in this paper was verified by the SWSC but has not been made available
at the SWSC website."
Certification results at the SWSC-Wiki will be marked in the same way.


Regarding requirements for the uploads:

I would split the other points into a mandatory requirement for
acknowledged uploads (to get the first tag) and further non-mandatory
suggestions.

Requirement:

The content should illustrate the approach and enable the reader to
understand a particular solution on the technical level. With the
solution documentation, a technically experienced reader must be able to
get the solution running and perform the actions that have been
demonstrated during the presentation of the solution at the workshop.

Non-mandatory suggestions:

Try to keep the following issues in mind:
- - Others should be enabled to evaluate the utility of a solution,
perhaps deciding to adapt it and use it in different contexts.
- - Please share content which may be useful for others solving the same
problems. Enable others to reuse your ontologies for instance.
- - Prefer using well known formats and provide pointers to documentation
for any proprietary format you may be using.
- - Try to ease others to get started with your system. Preferably enable
others to investigate your approach gradually without having to install
a large system right in the beginning. Start with the easiest part of
your solution and only then proceed to the more complex parts.
- - The more information is provided, the better.

What do you think?

Best regards,

Uli
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.3 (MingW32)
Comment: GnuPT 2.9.2

iD8DBQFHpvE28VxeCU3I0jARAqtgAJ9vdQWCNQ5tHWnSFm4HCm1RfHIjAgCZAfWe
c1VwtWg/S227mOElYQxDTbI=
=wox2
-----END PGP SIGNATURE-----

Received on Monday, 4 February 2008 11:04:36 UTC