[uf-discuss] ISO Dates and Durations
Martin McEvoy
martin at weborganics.co.uk
Thu Sep 25 02:16:19 PDT 2008
Hello Philip
Philip Tellis wrote:
> 2008/9/24 Martin McEvoy <martin at weborganics.co.uk>:
>
>
>> people do not have any problems of putting machine data in the head of a
>> document, for example service discovery links and meta details such as
>> keywords and descriptions. its worth a little thought a think?
>>
>
> >From a performance point of view, dumping too much user invisible data
> into the HEAD section of the document is going to eat up bytes that
> are of no use to most users.
Agreed...
> Personally, I'd leave the HEAD for data
> that the browser needs up front in order to correctly render the page
> (eg: CSS, favicon, content-type, link-rel),
Machine Data also like service discovery links, alternate formats such
as RDF Atom and RSS, a vast amount of websites also use meta tags for
verification such as microid an Google Analytics also for Descriptions
and keywords, how many websites in the web2.0 world have you seen using
external Javascript? this has to be loaded into the browser too, so
where performance is concerned pushing a few extra bits of data up into
the head is really a non-issue to most. Pushing data up into the head
(for machines) would seem like the right place to put ISO durations and
timestamps to separate content from data.
> and push everything else
> lower down to when it's actually needed.
>
> However, keep in mind that one size doesn't fit all. Going too far on
> the performance scale can sometimes mean a loss of validity,
> accessibility, and semantic value,
....
> so it's up to individual site
> owners to make the call. We should provide a couple of options that
> touch different points on this scale.
>
Agreed
Thank you
Martin McEvoy
> _______________________________________________
> microformats-discuss mailing list
> microformats-discuss at microformats.org
> http://microformats.org/mailman/listinfo/microformats-discuss
>
More information about the microformats-discuss
mailing list