- From: Bruno Racineux <bruno@hexanet.net>
- Date: Mon, 03 Mar 2014 16:25:09 -0800
- To: "Jens O. Meiert" <jens@meiert.com>, Steve Faulkner <faulkner.steve@gmail.com>
- CC: HTMLWG WG <public-html@w3.org>
On 3/2/14 2:30 AM, "Jens O. Meiert" <jens@meiert.com> wrote: >Only responding to this part of your email: Unless youıre suggesting >interest in this to the extent of issuing penalties for non-compliance >on behalf of search engines (which you please provide evidence for) >this is, frankly, nonsense. I am not talking about penalty *at all*. And frankly offended at the notion of you implying I would be making such non-sense assertion. >Furthermore, the delimiter topic itself seems long done, in the sense >of that thereıs rough agreement that ³greater than² characters are >typographically inadequate in the context of breadcrumbs. In addition, >and speaking as an ex-Googler, regarding code that any given company >uses in the assumption that it would even reflect their own views and >standards is maybe a little naive. (Anecdotally, I myself have fished >stuff like <h7> and <h8> elements out of Googleıs code base, and that >code had no significance for company nor industry whatsoever.) By no means was I suggesting a disagreement between SE views and W3C parties as a reason for it. I don't believe it is. And like Steve said: "I don't think google cares what HTML markup is used for breadcrumbs" which I agree with. But if the right angle bracket is bad for AT, that's reason enough to stop placing it in examples on other sites nevertheless. >This issue is about writing accurate documentation and making a sound >recommendation for breadcrumb markup (Steve, please correct me if Iım >confusing things now :). Giving this thought on W3Cıs end makes sense >as the relevant specs have been written or edited on the W3C side, and >as developers may look at the W3C to give sensible recommendations. Only 'interested' or 'thorough' developers look at the W3C specs. Most of the time people just follow the first example they see without questioning it, and I am only trying to help prevent that misinformation from happening. I adopted the right angle bracket symbol based on Google's example not too long ago, even though I am usually keen on extensive research. Now I have to revise it, to be AT friendly. > The world will not end if this documentation makes a recommendation > that not everyone agrees to but pulling in search engines here doesnıt > help anyone. If at all we could ping Ian (Hickson) and former HTML > editors on their take of markup intent and how they think this problem > should be tackled. Perhaps an aside for the delimiter, but accessibility words in a page have practical search engine implications. For example having as <h2>You are here:</h2> has a SE quality impact. It contains no keyword relevance to the overall semantics of the page in the same away a "click here" is not search engine friendly. And it is also becomes quite a meaningless heading in the outline of a page. So a "You are here:" heading is quite a poor choice. David MacDonald's syntax[1] with an aria label on the nav element itself if preferable IMO. i.e. <nav aria-label="You are here, breadcrumb trail"> And if there must be a <h2> heading in the context of using a list, or even for the label, I suggest using "Sitemap trail" as a better cognitive semantic and more appropriate term for the page outline. [1] http://davidmacd.com/test/breadcrumb.html
Received on Tuesday, 4 March 2014 00:25:39 UTC