<div dir="ltr">Hi,<div><br></div><div>For the analysis tools, I think the do-one-thing-do-it-well approach is far superior than the swiss-army-knife one. Several of the old analysis tools have rather obscure features that confuse people (like the option -ref in g_covar). But the code for those options is often also hacked in, making the base code cluttered and hard to read. As an example, try handing the g_rms code to a student to learn how the RMSD is calculated. If analysis tools are dedicated to a single task, it also helps in making the code clear enough to understand and check what they do. And it makes it easier for those that need some alternative analysis to copy and edit, or write from scratch.</div>
<div><br></div><div>Just my two cents,</div><div><br></div><div>Tsjerk</div></div><div class="gmail_extra"><br><br><div class="gmail_quote">On Mon, Sep 23, 2013 at 7:43 AM, Erik Lindahl <span dir="ltr"><<a href="mailto:erik.lindahl@scilifelab.se" target="_blank">erik.lindahl@scilifelab.se</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>
<div class="im"><br>
On Sep 23, 2013, at 7:21 AM, David van der Spoel <<a href="mailto:spoel@xray.bmc.uu.se">spoel@xray.bmc.uu.se</a>> wrote:<br>
>><br>
> Another (prettier) way rather than hiding the tools would be to introduce an extra level in gmx,<br>
<br>
</div>I think the problem with that option is that it brings back the problem of mixing well-tested code with lots of less tested routines contributed by any developer who wanted to do a quick hack for his own analysis, and it gives the appearance those routines too are part of core Gromacs since they just require another command specifier.<br>
<br>
In particular for new tools I think we should require them to be small, clear and well-documented code-wise, with fewer options and much less fancy analysis built-in (because the number of possible combinations explodes when one tries to do five things at a time), and also make sure we have a responsible maintainer for each tool before considering inclusion in the main distribution. I like the unix idea that each program should only do one thing, but do it well.<br>
<br>
Obviously, we'll have to be a bit more lenient with old tools, but the goal should be the same for those!<br>
<br>
Cheers,<br>
<br>
Erik<br>
<div class="HOEnZb"><div class="h5"><br>
<br>
--<br>
gmx-developers mailing list<br>
<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a><br>
<a href="http://lists.gromacs.org/mailman/listinfo/gmx-developers" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-developers</a><br>
Please don't post (un)subscribe requests to the list. Use the<br>
www interface or send it to <a href="mailto:gmx-developers-request@gromacs.org">gmx-developers-request@gromacs.org</a>.<br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br>Tsjerk A. Wassenaar, Ph.D.<br><br>
</div>