<div dir="ltr"><br><div class="gmail_extra"><br><br><div class="gmail_quote">On Sun, Apr 14, 2013 at 11:36 PM, Bogdan Costescu <span dir="ltr"><<a href="mailto:bcostescu@gmail.com" target="_blank">bcostescu@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><div dir="ltr"><div class="im">On Sun, Apr 14, 2013 at 10:54 PM, Mark Abraham <span dir="ltr"><<a href="mailto:mark.j.abraham@gmail.com" target="_blank">mark.j.abraham@gmail.com</a>></span> wrote:<br>
</div><div class="gmail_extra"><div class="gmail_quote"><div class="im">
<blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><br><div dir="ltr">I propose we withdraw support for g87 for GROMACS 5, rather than bother to fix it.<span></span></div>
</blockquote></div><div><br>If it's broken anyway now, why not retiring it already in the next versions of the current stables lines ? If someone would have missed it, we'd have heard complaints already... Both 4.5 and 4.6 versions are scheduled to be released very soon, that would make a perfect occasion.<br>
</div></div></div></div></blockquote><div><br></div><div>If it's been broken for years, then there's no rush to withdraw the claim that we have working support for reading .g87, either. :-) There may be aspects of the support that do work that I failed to identify.</div>
<div><br></div><div>In theory, our point releases are supposed to fix bugs and flesh out features that were not fully implemented in the major release. We haven't really talked about a feature-deprecation or feature-removal policy (though I have a number of candidates for deprecation for 5.0). Generally one would expect features to remain in a point release, but if they're already broken then the issue is fairly moot. If there's not time to fix something that would lead to non-obvious wrong results, then adding some kind of hack to save some users from disaster would be the way to go - but that's not applicable in the present case.</div>
<div style><br></div><div style>We currently have no tests for trajectory-reading functionality, so any 4.x code change more invasive than "You want to read .g87? Sorry you can't have that. <fatal error>" runs a slight risk of breaking something else. One advantage of waiting for 5.0 is that we will have a proper test suite that would allow for making this kind of feature change with fair confidence.<br>
</div><div style><br></div><div style>In theory, we're also giving someone the chance to say they really want .g87 - not that I imagine that person exists!</div><div style><br></div><div style>Mark</div></div></div></div>