You are here: SE » ResearchHome » WhereWePublish

Where to publish

Some considerations for choosing the place where to send our potential publications, be it a conference (or other meeting) or a journal or magazine.

Base rules

As a rule of thumb, we try to publish in first-class places: One of the best available conferences or journals in the respective specialty (or in software engineering overall).

However, it may make sense to publish incomplete, preliminary, or lower quality initial results first and that will of course be done in lower-quality places such as workshops or national conferences. We do not, however, publish at PseudoConferences or in predatory Open Access journals.

It can also be useful to go to venues that are more heavily visited by practitioners. The scientific quality of practitioner-oriented journals and meetings is clearly lower, but there is something in return: You may be able to get the attention of practitioners, the people who should eventually use our results, and you may be able to start an industrial cooperation in this way, which can greatly increase the relevance of the research.

If you want to pursue a career, keep this in mind:
  • to be counted as a scientific publication, your contribution has to undergo peer review,
  • to get you much formal reputation it has to appear in a high-impact journal or well-known top conference,
  • to get you much actual reputation is has to be good, has to be read, and has to be cited by others,
  • to get feedback and make contacts, choose places with a large readership (typically magazines or large conferences, not journals).
  • If your story is very complicated or unusual, chances of success are better at a journal than at a conference (because of the two-phase reviewing process).
  • If your work is preliminary, a workshop (or perhaps a minor conference) is the right place. In particular, consider the workshops at ICSE (especially CHASE) and perhaps PPIG.

Which conferences?

  • Prefer conferences sponsored by ACM or the IEEE Computer Society. (See their webpages for lists.) The top place by far is ICSE, followed by its little cousin ESEC/FSE. Of the smaller, more specialized conferences, ESEM is our most likely choice.
  • EASE is also a possiblity.
  • For national meetings, the conferences and workshops organized by the GI and its parts (Fachbereiche, Fachgruppen) are often a sensible choice. See for instance GI Fachbereich Softwaretechnik: It has plenty of different topic groups (Fachgruppen, FG) and working groups (Arbeitskreise, AG). Most of the former have a workshop at more or less regular intervals. The Fachbereich has a yearly conference, the 'SE' (Software Engineering), and the GI overall also has a yearly conference, the GI-Jahrestagung.
  • For practitioner-oriented conferences and magazines, see for instance iSQI/ASQF, SIGS Datacom, Heise, and some others. (These would be for presenting, not publishing.)
  • An interesting international research/practitioner hybrid is XP/Agile.

Which journals?

  • IEEE Transactions on Software Engineering (short: TSE) is the most highly regarded scientific journal on software engineering. It is not easy to get something published there, but far from impossible either. Try it.
  • Empirical Software Engineering (short: EMSE) is the best-regarded journal that is specialized to empirical work and hence sometimes more open in this area than is TSE.
  • Information and Software Technology (IST) is strong (despite its name) on empirical SE work since a number of years. For our work approaches, the reviewing experience at IST is often better than at EMSE.
  • Journal of Systems and Software (short: JSS) is published commercially by Elsevier and hence not quite as easy to access for everybody, but has a good reputation to publish relevant, practical scientific work.
  • Communications of the ACM (short: CACM) is the central publication of the ACM. Every member automatically receives it. Most issues have a central topic and outside of that topic it is difficult to get full length submissions in, but if you manage to, it has very high reputation and a lot of readers. CACM is considered a magazine, not a journal, which means it is scientifically not as rigorous (but has color graphics, side bars and other nice features for easier reading).
  • IEEE Software another magazine, is a highly interesting place for a publication, because many practioners read it, but like all magazines tends to expect shorter submissions. It has plenty of different kinds of contributions, go and see.
  • ACM Computing Surveys (short: CS) publishes overview articles for all areas of informatics. If you manage to get one article in there in your lifetime that is a major feat (and a major piece of work before).
  • Journal of Universal Computer Science (short: J.UCS) is a European alternative for anything that does not appear to quite fit in anywhere else.
  • Other SE journals such as Software Quality Journal, Software Process: Improvement and Practice, Journal of Software Maintenance and Evolution, and others are also possible, but clearly lower in reputation, readership, and impact.
  • ACM SIGSOFT Software Engineering Notes (short: SEN) is not a journal but a kind of newsletter (of the ACM Special Interest Group (SIG) on Software Engineering, called SIGSOFT). It does not perform peer reviewing but can be an appropriate place for informal kinds of contributions.
  • and of course: if your contribution is not (or not purely) software engineering, many other places can be relevant, too.

By the way: Journals are often ranked according to the number of citations that the articles appearing in these journals get in subsequent years ("impact factor"). One such citation database and corresponding ranking information available at FUB is the ISI JCR: This statistics is problematic in many respects, but since so many researchers and universities pay a lot of attention to it, indeed gives a rough indication how valuable a publication in a given journal is likely going to be. Go to JCR, select category "Computer science, software engineering" and sort the results by impact factor.