Posting guide

Revised: 23 August, 2013

Why?

• Good questions attract good responses.

• Formulating a good question can help point to the answer.

• Good questions help avoid embarrassment.

Key points

• Use the latest Bioconductor release (or 'devel') version. Ensure that your packages are up-to-date.

• Provide a minimal, self-contained example (data and code) that someone else can cut-and-paste into a new R session to reproduce your problem.

• Include the output of sessionInfo(). If the example produces an error, provide the output of traceback() after the error occurs.

• Include the maintainer of the package in the email, e.g., packageDescription('GenomicRanges')$Maintainer.

Preparing

• Read the relevant R documentation. Use help.start() to start the HTML search engine. If you are having trouble with function somefunc, try ?somefunc. If you are searching for a function, try help.search("somefunc"). Read the vignette(s) for the package(s) using vignette(package="somepkg")

• Search the FAQ and Bioconductor and R-help archives for similar posts. Try a Google search.

Composing

• Compose a new email message with a new subject line; only reply to an existing email if you are elaborating on or answering a previous question.

• Use an informative subject line. This will help attract responders and helps others when they search the archives. For example, a subject line "time course experiment using limma" is better than just "limma".

• Identify yourself. If you are using a non-professional email account, like gmail or hotmail, or the guest posting facility, then give your full name and professional affiliation.

• If you are asking for advice on how to use a particular function or package, then explain fully what documentation you have already read and why this hasn't yet fully answered your question. This allows a responder to answer your specific question instead of simply referring you to the existing documentation.

• Send email to the appropriate list. Use the Bioconductor mailing list for questions about specific package, or conceptual questions. The R-help mailing list is for questions about the underlying R program. The Bioc-devel and R-devel lists are for discussing code development and other technical issues.

• Use plain text instead of HTML; it is smaller in size and easier to read.

• The following attachment types are accepted: png, pdf, rda/Rdata. Total message size cannot exceed 1MB. If larger attachments are essential, post them to a publicly accessible location and include the link in your email.

• Package developers always appreciate being alerted to possible bugs, but be very sure that you have used the package correctly. In most cases, best practice is (i) to double-check the documentation and then (ii) report the behavior that was unexpected to you.

• Read over your mail. Is it polite and easy to understand? Remember that responses on the Bioconductor mailing list are from volunteers.

Replying

• Respond to everyone in the list, which ensures that your response is archived.

• If possible, write an answer that can be understood by readers with different scientific backgrounds and skill levels / skill profiles.

• When answering a question, consider including some explanation of how you arrived at your solution. This way, you help people not only to solve their problem at hand, but also to help themselves in the future.

• Rudeness and ad hominem comments are not acceptable.

• Brevity is OK. Consider, though, that information that is obvious to you may be very helpful to others.

• Provide URLs to other relevant threads or web sites.

Acknowledgments

This posting guide has been heavily adapted from the R-help posting guide, various comments and suggestions by Bioconductor posters and inspired by Eric Steven Raymond's essay on How To Ask Questions The Smart Way.