Dear James,

Just a stupid question: did you contact the authors and asked them whether you could have a copy of the software?

While I agree that one cannot expect from authors/funding agencies to support any software forever, in my opinion they should provide the version described in the paper as is, either for a licence fee or free, so others can reproduce/check their results.

 

Best,

Herman

 

 

 

 

Sounds rather extreme considering what's actually annoying you is you weren't told up-front that's it's not worth your time trying to install whatever you were installing.  You should be complaining to whomever misled you into wasting your week. 

The other thing you seem to be railing against is the universe, for not magically funding all software developers in perpetuity.  That's one for the funding bodies... though to be fair, can they be expected to support every brain-fart that made it into code and hence a paper?

(You asked for a flame! :)
phx

I hereby call on the broadest community of academics and researchers, including scientists, historians, economists, sociologists, psychologists, and whoever else has ever published a paper or read from the literature thereof, to reject any and all papers that describe new software that itself is not released under an open source model.

I further declare that this post is designed to ruffle feathers and incite incendiary conversation, to provoke all-caps and evoke multiple exclamation marks with interposed “1”s where anger prevents one from properly holding the shift key.

My rationale for this post: I have just spent a week installing software for structural biology (not crystallography) only to find that some of the key utilities needed were described in a recent publication but were not OSS. The authors have decided to stop supporting the software but have not retracted their paper, which is completely irrelevant without the availability of the software package they describe.

Let’s hammer this one out and come to the rational conclusion that non-OSS software should not be awarded publications.

James