<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Wed, Oct 29, 2014 at 4:33 PM, Chris McMahon <span dir="ltr"><<a href="mailto:cmcmahon@wikimedia.org" target="_blank">cmcmahon@wikimedia.org</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="gmail_extra"><div class="gmail_quote"><div>"...increase unit testing coverage" in our case almost certainly means "rewrite big chunks of Mediawiki Core from scratch". </div></div></div></div>
</blockquote></div><br></div><div class="gmail_extra">Browser tests, on the other hand, are not that hard to write, and don't require any changes in the code under test, or even any understanding of the architecture. Improving our end-to-end test coverage could be a nice Google Code-in task IMO - not as clonable as writing unit tests (too many browser tests is a bad thing as they are slower and require more maintenance) but should still be enough for quite a few students.</div><div class="gmail_extra"><br></div><div class="gmail_extra">Whether writing tests is a good task for attracting smart students is another question. I'd say not, although it might be no worse than the rest of the tasks - "cloneable" means "boring" by definition.</div></div>