Re: Preparing a document for publication

I've created

for info and links related to WG process.

has links to Lee's minutes email and Birte's document prpearation email.


On 08/10/10 06:50, Axel Polleres wrote:
> Thanks for this write-up, Birte!
> One more thing, which is a necessary last step...
>   Check whether the "This Version:" *and* "Previous Version:" links are correct... checked and changed this now for all of the docs.
> cheers,
> Axel
> On 6 Oct 2010, at 19:02, Birte Glimm wrote:
>> Hi all,
>> I finally managed to write down what I do to get the entailment
>> regimes doc ready for publication. I forget it myself each time I have
>> to do it and waste my time on trying to produce an html version with
>> xmlproc instead of xsltproc, etc, so I hope this collection of steps
>> will at least help me.
>> I use a Mac, so the xsltproc step (3) might be different for others.
>> 1) adjust:
>>      <!ENTITY draft.year "2010">
>>      <!ENTITY "October">
>>      <!ENTITY draft.month "10">
>>      <!ENTITY "14">
>> to the new pub date and possibly
>>      <!ENTITY maturity.level "WD">
>> in you .xml file (for me that is xmlspec.xml in the entailment folder)
>> 2) Open shell and cd to the folder of my xmlspec.xml
>> 3) xsltproc -o gen.html xmlspec.xml
>> This will generate the html version of my file which can be checked by
>> the pub rule tools
>> 4) commit the generated gen.html into CVS
>> 5) figure out the complete URL of gen.html
>> I go to, click on my
>> editor's draft link:
>> and change xmlspec.xml to gen.html
>> Now do the checking. When the checking shows errors, I try to fix all
>> the errors in the xmlspec.xml and start again from the top. This way I
>> won't have to change the same errors in each publication round.
>> 6) Go to and enter the complete URL
>> for gen.html into the checker and run the checker
>> or enter,pubrules,
>> i.e., yourURL,pubrules
>> This should give one error at least:
>> (1 occurrence)
>>      ->  404 (Not Found)
>> Since the doc that we are going to publish on 14 Oct 2010 is not yet
>> there, we are just preparing the publication.
>> Other errors have to be explained I guess (e.g., we have the
>> rif:imports URI which is a placeholder and does not yet exist)
>> 7) On the pub rules result page there is a menu on the right hand side
>> with a Tools section. From there I select link checker at
>> This should give an error such as
>> Line: 42
>>      Status: 404 Not Found
>> because the document is not yet published, but already mentioned in
>> the boilerplate stuff.
>> Other errors should be fixed.
>> 8) From the pub rules page there is also a link to the markup validator at
>> or enter,validate
>> This should not give errors.
>> 9) Finally, I go to the CSS validator from the pub rules Tools menu at:
>> and enter my gen.html URL
>> This should not give errors.
>> 10) W3C requires conformance with Web Content Accessibility Guidelines
>> 2.0, Level AA. This can be checked with
>> I never did that before and now that I did I found that xmlspec.xsl
>> does not produce accessible html files. I fixed that now (b tags for
>> bold should be strong tags, i should be em, tables from ednotes should
>> not have a summary) and my document now validates.
>> Copy gen.html and if you have local.css (or whatever is needed) into
>> the appropriate subfolder of pub and commit into CVS.
>> All done, up to Lee, Axel and the webmasters now I hope.
>> Birte
>> --
>> Dr. Birte Glimm, Room 309
>> Computing Laboratory
>> Parks Road
>> Oxford
>> OX1 3QD
>> United Kingdom
>> +44 (0)1865 283520

Received on Friday, 8 October 2010 08:10:39 UTC