<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div>To some extent, these issues can be alleviated by better and easier to find documentation of platform capabilities and best practices. What can we do to better ensure that the right things get documented?</div><div><br></div><div>A few things come immediately to mind:</div><div><br></div><div>1. Don't forget to put the dev-doc-needed keyword on any bugs that affect docs. This includes both bugs about new features and bugs fixing documented misbehavior, as well as bugs adding new capabilities to existing code. You can add the keyword before the patch lands, even, which makes it easier to remember to do and helps us schedule the work.</div><div><br></div><div>2. File documentation request bugs (<span style="font-size: 12pt; font-family: Helvetica;"><a href="https://bugzilla.mozilla.org/form.doc">https://bugzilla.mozilla.org/form.doc</a>) for new capabilities, but also anytime you notice something that's not documented but should be, or if you're unable to find docs that help you out of a jam. We will either make new docs happen or find the existing docs (if any) and make them easier to find.</span></div><div><font face="Helvetica"><span style="font-size: 16px;"><br></span></font></div><div><font face="Helvetica"><span style="font-size: 16px;">3. You're always welcome to update or add to the docs! See our new "Getting started on MDN" guide (</span></font><span style="font-size: 12pt; font-family: Helvetica;"><a href="https://developer.mozilla.org/en-US/docs/MDN/Getting_started">https://developer.mozilla.org/en-US/docs/MDN/Getting_started</a>) to learn how.</span></div><div><span style="font-size: 12pt; font-family: Helvetica;"><br></span></div><div><span style="font-size: 12pt; font-family: Helvetica;">4. If you learn something through conversations in email or on IRC that you couldn't uncover on MDN, please, please, *please* share! Use one of these approaches (or in a pinch, just email me) to share what you've learned. If everyone that got help in IRC or email with understanding something that MDN failed at were to share their discovery, MDN would become an infinitely more amazing resource!</span></div><div><span style="font-size: 12pt; font-family: Helvetica;"><br></span></div><div><span style="font-size: 12pt; font-family: Helvetica;">As always, I remind: you don't have to write well to contribute to MDN. We have writers to tweak your words to be pretty. But you developers are our best resource, and we urge you to share what you know!</span></div><div><br>Eric Shepherd<div>Sent from my iPhone</div></div><div><br>On May 18, 2014, at 4:36 AM, "Matthew N." <<a href="mailto:MattN+firefox-dev@mozilla.com">MattN+firefox-dev@mozilla.com</a>> wrote:<br><br></div><blockquote type="cite"><div><span>----- Original Message -----</span><br><blockquote type="cite"><span>From: "L. David Baron" <<a href="mailto:dbaron@dbaron.org">dbaron@dbaron.org</a>></span><br></blockquote><blockquote type="cite"><span>To: "Matthew N." <<a href="mailto:MattN+firefox-dev@mozilla.com">MattN+firefox-dev@mozilla.com</a>></span><br></blockquote><blockquote type="cite"><span>Cc: "Gavin Sharp" <<a href="mailto:gavin@gavinsharp.com">gavin@gavinsharp.com</a>>, "Firefox Dev" <<a href="mailto:firefox-dev@mozilla.org">firefox-dev@mozilla.org</a>></span><br></blockquote><blockquote type="cite"><span>Sent: Sunday, May 18, 2014 12:38:38 AM</span><br></blockquote><blockquote type="cite"><span>Subject: Re: the firefox-backlog flag</span><br></blockquote><blockquote type="cite"><span></span><br></blockquote><blockquote type="cite"><span>Taking a step back, though, I think part of an engineer's job is to</span><br></blockquote><blockquote type="cite"><span>understand the performance characteristics of the features that</span><br></blockquote><blockquote type="cite"><span>you're planning to depend on. It's far better to talk to platform</span><br></blockquote><blockquote type="cite"><span>folks about how a feature performs (and perhaps also measure its</span><br></blockquote><blockquote type="cite"><span>performance in an appropriate way) before building a lot of code</span><br></blockquote><blockquote type="cite"><span>that depends on it, and probably even before committing to UI</span><br></blockquote><blockquote type="cite"><span>*designs* that depend on it.</span><br></blockquote><span></span><br><span>Agreed. In the case of Australis, the tab curve was a hard requirement and there was a lot of complexity around all of the various configurations[1][2] we have to support so implementing the curves without clipping would have been harder and had other performance issues.</span><br><span></span><br><blockquote type="cite"><span>Some features are inherently slower</span><br></blockquote><blockquote type="cite"><span>than others; you're asking the platform to do more work. Other</span><br></blockquote><blockquote type="cite"><span>features could be optimized better than they are if needed, but the</span><br></blockquote><blockquote type="cite"><span>content on the Web doesn't justify that investment. We can't</span><br></blockquote><blockquote type="cite"><span>implement everything everyone wants and make it fast enough that</span><br></blockquote><blockquote type="cite"><span>they don't notice it; like everyone else, we have to prioritize</span><br></blockquote><blockquote type="cite"><span>work.</span><br></blockquote><span></span><br><span>Right, I'm definitely not saying that dependencies for Firefox should become automatic priorities for the platform, more that I don't think other teams are even aware of some of the bugs that we are blocked on so that they can consider them.</span><br><span></span><br><blockquote type="cite"><span>Don't assume that just because something is in the platform that</span><br></blockquote><blockquote type="cite"><span>it's a good way to do what you want; often things are in the</span><br></blockquote><blockquote type="cite"><span>platform for other reasons. Please ask.</span><br></blockquote><span></span><br><span>In this case, we already knew that SVG performance wasn't optimal but it was acceptable on our benchmarks on the hardware we tested months before the Tab Animation Regression Test (TART) appeared. TART came along and set layout.frame_rate to 0 (something we didn't know about) which then showed that there could be performance issues on lower-end hardware and then it became a rush to fix the performance issue as it was considered a blocker to merging to m-c. Given all of the requirements, it was the best choice at the time and I still think it was the right one. I agree with you in general though.</span><br><span></span><br><span>[1] <a href="http://matthew.noorenberghe.com/blog/2012/11/australis-tabs-where-are-you">http://matthew.noorenberghe.com/blog/2012/11/australis-tabs-where-are-you</a></span><br><span>[2] <a href="http://www.gijsk.com/blog/2014/04/why-doing-visual-refreshes-of-firefox-is-hard/">http://www.gijsk.com/blog/2014/04/why-doing-visual-refreshes-of-firefox-is-hard/</a></span><br><span>_______________________________________________</span><br><span>firefox-dev mailing list</span><br><span><a href="mailto:firefox-dev@mozilla.org">firefox-dev@mozilla.org</a></span><br><span><a href="https://mail.mozilla.org/listinfo/firefox-dev">https://mail.mozilla.org/listinfo/firefox-dev</a></span><br></div></blockquote></body></html>