6
$\begingroup$

I often put a name with the URL for the free program(s) I used in writing an answer. Also, adopting a standard practice among theoretical chemists (which I'm unfortunately not), sometimes I drop chunks of code or stdin/stdout listings if it can help to reproduce the results (under ### Appendix, ### Notes or similar section at the end of the post). For example, I try to replace Excel and Matlab by Python and LaTeX; Diamond and CrystalMaker by Olex2 and VESTA; Illustrator and CorelDRAW by Inkscape; ChemDraw by ChemSketch etc., and leave a brief note how I managed to get by using alternatives freely available to everyone (I don't touch piracy here).

By doing so I'm trying to distance as far away as possible from software evangelism, which I think is rather ineffective. The awareness imposed by evangelists easily comes and go. People are lazy and won't break their workflow or invest own time in learning a new toolkit of arguable value and applicability which may or may not return beneficial. For the same reason I don't believe in plain advertisement, even though I put up some ad banners on Community Promotion Ads — 2019 section as a part-experiment.

On the contrary, when people see how a particular problem (question) related to their needs is solved (answer), they actually start to show interest in the tools that could help them get the same result. Not really monkey see, monkey do, but the concept is similar (not in a derogatory way).

One may ask why care to share the info about the programs they use if they are already power users and mastered all the tools they might ever need? Well, let me answer with a quote by L. Ozerov: "Talents should be helped, as mediocrities will shoulder their way through". Paid software is fine on its own as it's backed up financially by the established business models and corporations, and won't go away as long as it's profitable. Free and open-source software development is bumpy and often slower due to the lack of resources and smaller community, which is especially the case for the highly specialized scientific packages. However, it can be self-propagated, partially by its user base:

User achieves a result → informs audience → some people find it useful → acquiring more users → forcing development → getting tools improved → achieving more results → …

Long story short, I believe that unobtrusively demonstrating by own example how people can use free alternative tools to solve their problem is helpful for stable and reproducible research.

Do you think this would really serve increasing awareness and improves quality or just add unnecessary noise and all the tools should solely be listed in Community Wiki?

$\endgroup$
12
  • $\begingroup$ I think you do a service to many by offering to share tools and techniques that are freely available, and you should be commended. However I am not sure I understand what your intention is here. Do you suggest enforcing or encouraging reporting of tools/code/methods? It's hard enough to get people to provide a little info to help answer their question... $\endgroup$
    – Buck Thorn Mod
    Mar 17 '19 at 21:12
  • $\begingroup$ To elaborate: is your emphasis on the question or answer side? $\endgroup$
    – Buck Thorn Mod
    Mar 17 '19 at 21:13
  • $\begingroup$ The question is only about good practices, no enforcement; and it's more relevant for the answers, I think. $\endgroup$
    – andselisk Mod
    Mar 17 '19 at 21:18
  • 1
    $\begingroup$ There's absolutely no harm in posting a list of suggested good practices, but you don't want to increase the red tape on the site too much lest you discourage answers. I do agree that an answer is more useful when there are clear methods reported, as necessary, but then this is often left for the poster to request, say in the comments. Perhaps you might share examples that would clarify where such a code has been violated in your opinion, if you have any clear cases in mind. (Aside: Note that matlab can be substituted (often) with octave). $\endgroup$
    – Buck Thorn Mod
    Mar 17 '19 at 21:33
  • $\begingroup$ As with all of science, I think it's good practice to have sufficient explanation such that what you do is reproducible. Beyond that, though, I'm not sure that we should necessarily adopt a site philosophy as to what type of software is the best. I make no apology for using MATLAB in my latest answer... $\endgroup$
    – orthocresol Mod
    Mar 17 '19 at 21:37
  • $\begingroup$ I must admit I am still a little confused, because you seem to want people to use "free/shareware" more, but don't consider yourself a software evangelist (I am very much for such software, don't get me wrong) that is you want to promote use of such software? $\endgroup$
    – Buck Thorn Mod
    Mar 17 '19 at 21:37
  • 2
    $\begingroup$ No no, I don't insist on transparency: if someone wants to use Excel/Matlab/Gaussian it's absolutely fine, there are many cases when it's the only option available. The point is whether it's appropriate to unobtrusively draw some extra attention (as promotion) to the free software if it's being used in the answer. As for Octave, it's probably a fine package, but I never came along with it for some reasons. Also, the knowledge of Python can be used outside chemistry and science in general, whereas with Octave I pretty much stuck with numerical analysis. $\endgroup$
    – andselisk Mod
    Mar 17 '19 at 21:43
  • $\begingroup$ @NightWriter Showing by example ≠ evangelism $\endgroup$
    – andselisk Mod
    Mar 17 '19 at 21:45
  • 1
    $\begingroup$ Gotcha, but you do see the goal as promotion of free/shareware? Imho promotion is a close relative of evangelism. Anyway, comments risk getting long :) so I'll cede this space to others opinions now... seems the SE webgods must have an opinion on "implicit software promotion" on the site. $\endgroup$
    – Buck Thorn Mod
    Mar 17 '19 at 21:47
  • 1
    $\begingroup$ @NightWriter Well, evangelism in my eyes is similar to cold calling. People won't bother if it doesn't directly affects them. Promotion by an example is like "look ma, no hands" (probably bad example) and it reaches those and only those who are interested and doesn't push. Another example: one can prey all day long how great and secure Linux is and no one would bat an eye because they are happy with Windows or MacOS; show some screenshots from /r/unixporn/ and there are going to be geeks who would want to have that desktop on their computer. Motivate; spread awareness; show, don't tell. $\endgroup$
    – andselisk Mod
    Mar 17 '19 at 21:59
  • 2
    $\begingroup$ I think this is a Q&A site, not a software promotion vehicle, but perhaps those SE webgods have an interest in promoting free/shareware "by example". I don't think the chem SE moderators in particular should use the site to explicitly promote specific software, even as much as I like free/shareware. However if you'd like to write that you used a particular package in every answer you post, that should be entirely up to you. I don't think it's entirely appropriate to edit someone elses answer to show how the same solution is arrived at with another package than that originally/explicitly used $\endgroup$
    – Buck Thorn Mod
    Mar 17 '19 at 22:26
  • $\begingroup$ This "editing someone elses post" part could be argued to violate the "not changing original intent" clause of an edit. But I like the free/shareware ads. Keep them up! $\endgroup$
    – Buck Thorn Mod
    Mar 17 '19 at 22:28

You must log in to answer this question.

Browse other questions tagged .