Call for Papers: The Forum for Negative Results (FNR)

The Forum for Negative Results (FNR) is a permanent special section of the Journal of Universal Computer Science (J.UCS) and exclusively publishes negative results, i.e., research that did not have the desired outcome, but still advances knowledge. J.UCS is an electronic journal published by Springer Verlag.

All Computer Science researchers are invited to submit appropriate papers at any time. Please distribute this call for papers to interested colleagues and/or place a link to the FNR homepage http://page.inf.fu-berlin.de/~prechelt/fnr/ in relevant link collections.

Rationale

As most of Computer Science is rather usefulness-oriented, it is currently difficult to publish work that demonstrates a non-progress, or negative result, with respect to usefulness. Therefore today,

• lessons to be learned from negative results are often lost, and
• many works tend to demonstrate neither progress nor non-progress.

Description

FNR is a top-class forum for publishing Computer Science negative results that imply scientific insights. Just like J.UCS, FNR does not restrict contributions to particular topics. However, only papers with the following properties qualify for publication in FNR:

1. The work described had a clear goal, stated in the paper.

2. The starting point or approach of the work was promising and had plausible chances of success. These chances are explained in the paper.

3. Still, the goal was not met. The failure was not foreseeable during the implementation phase of the work; it was apparent only from the evaluation phase. (This rules out most purely theoretical work.)

4. There must be danger of somebody else trying a similar approach again, and failing.

5. Both implementation and evaluation were carried out according to strong scientific standards. These standards are documented in the paper.

6. At least part of the reason for the failure was understood in the evaluation or in subsequent analysis. The explanation is given in the paper. This explanation represents the scientific contribution of the paper.

FNR will be quite selective. Any paper to appear in FNR must be impeccable with respect to points (1) and (5). High standards will also be applied to points (2) and (3). As for point (6), the lesson learned must become clear, but no cure needs to be known.

Articles should be as concise as possible and concentrate on goals, approach, and reasons for failure instead of on technical details of implementation and evaluation. The reviewers of a paper submitted to FNR will apply these criteria when judging the contribution. For further information see the FNR homepage at http://page.inf.fu-berlin.de/~prechelt/fnr/ and also the J.UCS homepage at http://www.jucs.org/, in particular the FNR foundation paper http://www.iicm.edu/jucs_3_9/why_we_need_an

---

Lutz Prechelt, prechelt@inf.fu-berlin.de

Last modified: Wed Mar 29 18:16:52 MET DST 2000