<div dir="ltr">The way I see it, if you can reasonably run ratt and do the analysis before uploading, that’s great, but it shouldn’t be required.<div><br></div><div>For the compiler, I’d prefer having it available earlier over avoiding breakage.</div><div><br></div><div>For random libraries, I’d prefer the opposite.</div><div><br></div><div>Part of the rationale is that for libraries, there’s often a good chance the library screwed it up, and we should fix the library. For the compiler, I think that’s very rare, so likely we’ll need to fix the downstream packages anyway. Plus, if it’s big projects like docker, the project should have had enough time with the compiler release candidates to ensure compatibility.</div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Oct 15, 2015 at 9:41 PM, Tianon Gravi <span dir="ltr"><<a href="mailto:admwiggin@gmail.com" target="_blank">admwiggin@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Sure! :) On the Docker end, the discussion is in<br>
<a href="https://github.com/docker/docker/pull/15703" rel="noreferrer" target="_blank">https://github.com/docker/docker/pull/15703</a> (and you're correct that<br>
it's Docker doing something "wrong", but isn't it still worth checking<br>
so we don't get random FTBFS? or do we just not worry about it and do<br>
something about them as they come?)<br>
<span class="im HOEnZb"><br>
♥,<br>
- Tianon<br>
4096R / B42F 6819 007F 00F8 8E36 4FD4 036A 9C25 BF35 7DD4<br>
<br>
<br>
</span><div class="HOEnZb"><div class="h5">On 15 October 2015 at 11:08, Michael Stapelberg <<a href="mailto:stapelberg@debian.org">stapelberg@debian.org</a>> wrote:<br>
> Do you have a link with details? I’m curious.<br>
><br>
> On Thu, Oct 15, 2015 at 7:45 PM, Tianon Gravi <<a href="mailto:admwiggin@gmail.com">admwiggin@gmail.com</a>> wrote:<br>
>><br>
>> On 15 October 2015 at 10:40, Michael Stapelberg <<a href="mailto:stapelberg@debian.org">stapelberg@debian.org</a>><br>
>> wrote:<br>
>> > FWIW, I don’t think that’s necessary for the compiler itself. Go adheres<br>
>> > to<br>
>> > its Go 1 stability guarantee, and any package which breaks likely did<br>
>> > something wrong in the first place :). If this is the only thing that’s<br>
>> > holding back the upload, I’d recommend to just go ahead with the upload.<br>
>><br>
>> Right, in theory, that's supposed to be the case. I've seen several<br>
>> examples already where that's not the case, including Docker itself.<br>
>><br>
>> ♥,<br>
>> - Tianon<br>
>> 4096R / B42F 6819 007F 00F8 8E36 4FD4 036A 9C25 BF35 7DD4<br>
><br>
><br>
><br>
><br>
> --<br>
> Best regards,<br>
> Michael<br>
</div></div></blockquote></div><br><br clear="all"><div><br></div>-- <br><div class="gmail_signature">Best regards,<br>Michael</div>
</div>