Fwd: Re: Help from an HTML guru?

Hi everyone,

I'm forwarding Mark's email below to the group - and to Phil!

Mark, thanks a lot for the investigation. This is a huge help, I'd probably never have found myself!

Thanks to Newton for having found a trick that makes the issue less urgent right now. I'm not sure we will want to keep the numbers for the sub-sections (they will probably not be very big, in the end). But now the document looks better without us having to do anything.

And thanks Annette for raising the issue during last call. Without her, I'd have postponed the issue for a while, and it might have been a bad idea given the fact that Mark now hints this is a problem for W3C to solve...


Cheers,

Antoine


-------- Original Message --------
Subject:  Re: Help from an HTML guru?
Date:  Fri, 29 May 2015 02:11:41 +0200
From:  Mark Harrison <mark.harrison@cantab.net>
To:  Antoine Isaac <aisaac@few.vu.nl>



Hi Antoine,

I took a quick look.  The HTML source code looks OK - all of these sections use <h3> headings.  It appears that the dynamic CSS stylesheets generated by https://www.w3.org/Tools/respec/respec-w3c-common are doing something really weird and formatting the first two <h3> headings differently from the remaining <h3> headings - if for example it used CSS selectors such as :nth-of-type(1), :nth-of-type(2)

Taking a further look at https://www.w3.org/Tools/respec/respec-w3c-common - this does appear to be doing some dynamic writing of a stylesheet in Javascript and there are definitely some references to nth selectors within that.

So I did a very simple test using this very minimal HTML file below - which reproduces exactly the same effect.  The error is clearly in the respec-w3c-common shenanigans.

This is an issue that W3C staff need to be alerted to and they need to make necessary adjustments in respec-w3c-common to allow you to use multiple <h3> headings, all consistently formatted the same way.  The problem is almost certainly in the use of 'nth-type' CSS selectors within respec-w3c-common - but it's not very readable because it's minified.   Hopefully somebody has the readable non-minified version and can rework the logic.

I suggest you raise the issue with Phil Archer and he can escalate it and hopefully get it fixed.  Their problem - not yours - so don't waste any further time trying to compensate for their error in respec-w3c-common.

Best wishes,

- Mark



<!DOCTYPE html>
<html>
<head>
<title>Data on the Web Best Practices: Data Quality Vocabulary</title>
<meta charset='utf-8'>
<script src="https://www.w3.org/Tools/respec/respec-w3c-common"
class="remove"></script>
</head>
<body>
<section class="informative">
<h2>Dimensions and metrics hints</h2>
<h3>Availability</h3>
<h3>Processability</h3>
<h3>Accuracy/Consistency/Relevance</h3>
<h3>Completeness</h3>
<h3>Conformance</h3>
<h3>Credibility</h3>
<h3>Timeliness</h3>
</section>
</body>
</html>



On Fri, May 29, 2015 at 12:02 AM, Antoine Isaac <aisaac@few.vu.nl <mailto:aisaac@few.vu.nl>> wrote:

     Dear all,

     Can anyone help with a probably stupid editorial problem I cannot solve for the DQV draft [1]?

     In section "6. Dimensions and metrics hints", all sub-headers are supposed to be at the same level. But for reason only the first two sub-sections (now, 'Availability' and 'Processability') have a heading formatted as a sub-section. The remaining sub-section titles ('Accuracy/Consistency/Relevance' and following) are formatted as sub-sub-sections.

     And this holds, whatever I try. Especially, if I remove the 'Availability' sub-section, then 'Accuracy/Consistency/Relevance' gets the right formatting, but 'Completeness' and the following remain formatted as sub-sub-sections...

     For the record I think I have the same problem with the sub-section "Express that a dataset received an ODI certificate" in section 5...

     Thanks for the help!

     Antoine

     [1] http://w3c.github.io/dwbp/vocab-dqg.html

Received on Friday, 29 May 2015 07:30:04 UTC