Re: [pre-review feedback] qa-html-language-declarations

Hi Gunnar,

On 25/08/2014 15:09, Gunnar Bittersmann wrote:
> Hi Richard,
>  From some WG protocoll I knew you were working on an update of
> http://www.w3.org/International/questions/qa-html-language-declarations-new
>
> Will you continue?

Well, actually that file ought not to be there. I migrated all the 
changes to 
http://www.w3.org/International/questions/qa-html-language-declarations 
at the end of May. I don't seem to have changed the dates on the proper 
version of the document though - not sure why.

So these replies relate to the version at
http://www.w3.org/International/questions/qa-html-language-declarations

And I'll change the dates and "About this article" section while I'm 
about it.

Now done.

I'll also add warnings to the translations - basically it looks like I 
had a major distraction part way through upgrading to the new version :(

Also done.

Btw, i also removed the namespace attribute in the html tag, and all the 
xml:lang attributes from the article, per the most recent practice (ie. 
pages are no longer polyglot, just html5).

>
> I’ve already incorporated your changes into my translation:
> http://dev.bittersmann.de/International/questions/qa-html-language-declarations.de
>
>
> You guessed it, I’ve found a thing or two: ;-)
>
> »»
> Always use a language attribute on the html tag
> ««
>
> Using ‘element’ instead of ‘tag’ would be more precise:
> Always use a language attribute on the html element

Actually I chose 'tag' here because it was more precise. It means the 
bit of source code <html ....>.  I agree with you in a couple of later 
places, and change to 'element', but I really mean tag here.


>
>
> In some alt attribute:
> »»
> from the document text 'español'
> ««
>
> Uppercase, actually:
> from the document text 'Español'

Actually, I believe that one doesn't uppercase the start of the word 
español in Spanish.  See http://es.wikipedia.org/wiki/Idioma_espa%C3%B1ol

I know I do that at the top right of the page (links to translations) 
but that's just to make it look consistent, and I probably should change 
it for the Spanish pages.  Note that Wikipedia actually does the same 
thing in the left column.

>
>
> In the source code (two times):
> »»
> <span xml:lang="sv" lang="sv">Español</span>
> ««
>
> Fits well under heading ‘What if element content and attribute values
> are in different languages?‘ ;-) Not intended, I guess.
>
> Make it both times:
> <span xml:lang="es" lang="es">Español</span>

Good catch. Fixed. Thanks


>
>
> Sample code:
> »»
> <span title="Spanish"><a lang="es"
> href="qa-html-language-declarations.es">Español</a></span>
> ««
>
> IMHO, the tooltip belongs to the link, i.e. the a element. Hence the
> elements should rather be nested the other way around, span inside a:
>
> <a href="qa-html-language-declarations.es" title="Spanish"><span
> lang="es">Español</span></a>
>
> (This doesn’t necessarily mean to change the article template, but would
> make a better example.)

Hmm. I see what you're saying, but I'm not convinced it needs changing.

>
>
> »»
> Instead, move the attribute containing text in a different language to
> another element, as shown in this example, where the span tag inherits
> the default en setting of the html tag.
> ««
>
> Again, ‘element’ instead of ‘tag’ (both times):
> Instead, move the attribute containing text in a different language to
> another element, as shown in this example, where the span element
> inherits the default en setting of the html element.

Changed.


>
>
> Under ‘Specifying metadata about the audience language’:
> »»
> Here is an example of an HTTP header that declares the resource to be a
> mixture of English, Hindi and Punjabi:”
> ««
>
> I wonder if “resource” wouldn’t refer to the text (instead of the target
> audience), i.e. to what Content-Language is *not* about?

A resource is something pointed to by a URL - a file of some kind, 
mostly, so it is appropriate here. (That's why I chose the word 'resource')


>
>
> »»
> HTTP or meta Content-Language information
> ««
>
> Both ‘meta’ and ‘Content-Language’ are code keywords:
> HTTP or <code class="kw">meta</code> <code
> class="kw">Content-Language</code> information

fixed


>
>
> »»
> on the html tag
> ««
>
> Dito:
> on the <code class="kw">html</code> tag
>
> But should be better:
> on the <code class="kw">html</code> element

fixed


>
> Same in the following sentence:
> If you have used the language attribute on the html element, as you
> always should

fixed


>
> »»
> Secondly, the <code class="kw">doctype</code> that should start any
> <del>HTML or</del> XHTML file
> ««
>
> ‘DOCTYPE’ must be in capitals for XHTML/polyglot HTML5:
> Secondly, the <code class="kw">DOCTYPE</code> that should start any
> <del>HTML or</del> XHTML file
>
> »»
> The <code class="kw">doctype</code> in the example below
> ««
>
> Dito:
> The <code class="kw">DOCTYPE</code> in the example below

fixed both


>
>
> »»
> Latin 1 (iso-8859-1)
> ««
>
> Use uppercase (as elsewhere):
> Latin 1 (ISO-8859-1)
>
>
> »»
> All this, however, is nowadays moot,
> ««
>
> “Moot” refers to point ‘Thirdly’, not to all above. Remove ‘All‘. Don’t
> make it a separate paragraph, rather append the sentence to the previous
> paragraph.

Changed to "All these encoding examples, however, are nowadays moot"


>
>
> »»
> For example, Azerbaijani can be written using both right-to-left and
> left-to-right scripts
> ««
>
> This would be clearer with the addition of the actual scripts:
> For example, Azerbaijani can be written using both right-to-left
> (Arabic) and left-to-right scripts (Latin, Cyrillic)
>
> Or without parentheses:
> For example, Azerbaijani can be written using both right-to-left Arabic
> script and left-to-right Latin and Cyrillic scripts

good idea. Used:
right-to-left (Arabic) and left-to-right (Latin or Cyrillic) scripts


Thanks,
RI


>
> Cheers,
> Gunnar
>
>

Received on Wednesday, 27 August 2014 17:04:10 UTC