Guidelines for referees of articles for TUGboat

Effective with issue 12, no. 2, technical articles submitted to TUGboat will be examined by an outside referee as well as by the editors. This referee will be selected from the panel of volunteers by the TUGboat editorial staff (Barbara Beeton and others), with a strong effort being made to match the interests and expertise of the referee with the subject matter of the article.

Articles will be sent to referees as soon as it has been determined that they can be processed successfully with the current style files, ltugboat.cls, or tugboat.sty. It is important to attend to the reviewing and return articles promptly, as the schedule allows only four weeks after the editorial deadline (published in the TUGboat Calendar) to complete work on any individual article for inclusion in the current issue. The copy-editing, proofreading, and any special formatting will be handled as part of the issue production.

In areas where there is an active subject editor, a copy of each relevant article will be sent to the appropriate editor at the same time as it is sent to the referee. Communications should thereafter be carried on between referee and subject editor and between subject editor and author until the article is in good (technical) shape to be published.

For the present, we will try to maintain the mutual anonymity of both authors and referees, to avoid any possible personal prejudices. (It should be made clear that we are not aware of any such potential problems, but wish to keep the review process on the highest plane.) Subject editors will be made aware of the identity of both, to permit them to carry on the necessary correspondence; however, they will be asked to adhere to the same practice of not identifying authors and referees to one another. This practice will be reviewed after several issues.

The purpose of the referee's review is not to reject a high proportion of the articles submitted, but to "assist authors in creating articles that are of maximum value to the TUGboat readership." [Victor Eijkhout, TUGboat 11, no. 4, p. 605.] Victor's description of what he saw as the editor's job is, not so surprisingly, similar to what we hope the referees will do, and we suggest that you read it carefully and act on it.

In response to past suggestions and comments, there are a few specific principles that should guide your review.

1. Although similar TeXniques are discovered independently all the time, if the ideas or TeXniques in an article have appeared previously in a TeX "publication" (TUGboat, a proceedings article, TeXhax, TeXmag, UKTeX, etc.) a reference to the earlier work should be given. Related information should be brought to the author's attention even if it is not identical, as it may serve to strengthen the presentation.
2. Major or extensive articles on topics that have already appeared in TUGboat should be examined for new ideas, and if they are absent, documentation should be provided of where the material appeared earlier, to help the editors be specific in their comments when they write to the author. A particularly good survey or tutorial may be exempt from this guideline. Similarly, a short item, particularly one intended for a beginning audience, may be judged less severely if the related material appeared more than two years previous; even so, it is desirable to supply the reference to the earlier item(s).
3. Articles about commercial products are acceptable if they maintain a technical approach. This is particularly true if the author is the creator of the product or is associated with an organization that markets it. Messages that can be construed as advertising should be removed, beyond a name and address to which a reader can write for further information.
4. If a referee finds any conflict of interest in reviewing a particular article, or has any other reason for feeling that the review might be biased, the editors should be informed immediately, and alternate arrangements will be made. It would be useful to let the editors know the nature of the problem, to avoid such assignments in the future.
5. Abstracts of TUGboat articles are now published in several of the newsletters of language-specific TeX user associations. This practice can be supported by encouraging authors to provide informative abstracts that can be translated readily and accurately.
6. There will undoubtedly be occasions when an article is found to be inappropriate for publication in TUGboat, and the decision communicated to the author will be ``No.'' The reasons for such a decision should be provided as clearly as possible to the editors, so that they in turn can explain to the author why the article is unsuitable.

Many TUGboat authors did not learn English as their first language. The editors have tried to edit gently in this area, to make sure that what is being said by an author is clear and understandable to a reader with a reasonable command of English and sufficient knowledge of the subject matter. Nothing is more discouraging to the present Editor than reading a collection of articles by different people all of whom sound the same because of too-heavy editing. TUGboat authors speak with many voices, and this is part of a lively publication. However, if you find anything to be unclear on account of language, please call it to the editors' attention, making suggestions or interpretations, if possible, and we will do our best to get it fixed up.

These guidelines are intended to apply to Proceedings as well as to regular issues of TUGboat. The Proceedings of the TUG annual meeting is published as an issue of TUGboat.

There is one major difference between an article submitted for a regular TUGboat issue and a Proceedings article. All articles submitted to the Proceedings will already have been accepted in principle by the program committee, on the basis of an abstract submitted earlier. It is therefore not practical to reject a Proceedings article; the goal must be to offer constructive suggestions that will assist in raising it to the desired level of clarity and usefulness.

Two other relevant characteristics of the Proceedings are a very tight schedule, and the undesirability of delaying any article to another issue. Thus it is particularly important for the initial review to be very clear and specific about requested changes. Authors will be asked to respond to all comments, either by making changes or by explaining why they are not doing so. Should there be any questions, or should an author respond other than by incorporating suggested changes, the referee may be asked to review the revised article and approve it or make additional comments as appropriate.

Proceedings issues of TUGboat have their own editors, for the TUG annual meeting proceedings often a team of co-editors; both should be named on the TUG web page for the meeting, and should be available via email and fax. Communication regarding the reviewing of Proceedings articles should be carried on directly betwen the Proceedings editor and the referees, and not through the regular TUGboat mailbox. The Proceedings editor is in complete charge of the technical content of the Proceedings issue; the regular TUGboat staff is responsible only for the production of these issues.

There should be a team co-editors for the TUG annual meeting Proceedings. Both should be available via email as well as FAX.

Electronic mail to the editors is best sent to TUGboat@tug.org rather than to an individual. That mailbox is monitored regularly regardless of who happens to be unavailable at any particular time.

We hope that these guidelines will not be found too onerous, and look forward to your help in making TUGboat a more useful tool.