W3C home > Mailing lists > Public > public-html@w3.org > November 2011

noted 3 issues re: time/data (was Re: minutes for HTML WG f2f, 2011-11-04, part 1)

From: Marat Tanalin | tanalin.com <mtanalin@yandex.ru>
Date: Thu, 17 Nov 2011 21:57:21 +0400
To: Tab Atkins Jr. <jackalmage@gmail.com>
Cc: (wrong string) Çelik <tantek@cs.stanford.edu>, Sam Ruby <rubys@intertwingly.net>, public-html@w3.org
Message-Id: <219311321552641@web151.yandex.ru>

14.11.2011, 21:33, "Tab Atkins Jr." <jackalmage@gmail.com>:

> 2011/11/14 Marat Tanalin | tanalin.com <mtanalin@yandex.ru>:
>> 14.11.2011, 19:38, "Tab Atkins Jr." <jackalmage@gmail.com>:
>>>> If to consider syntax with 'itempropvalue' attribute (as I've mentioned earlier and in the bug 14679 comment 3):
>>>>
>>>> <span itemprop="foo" itempropvalue="bar" itemscope>
>>>>
>>>> or CSS-like one I've proposed later:
>>>>
>>>> <span itemprop="foo: bar" itemscope>
>>>>
>>>> then 'bar' is obviously value of this span as property.
>>> That doesn't seem obvious to me.
>> What exactly is unobvious for you? (What exact of the two syntaxes, what exactly in each of them.)
> In the former example, that @itempropvalue wins over @itemscope in
> determining the value of the <span>. There's no clear reason that one
> should win over the other, so it will end up confusing people.

How @itempropvalue could interfere with @itemscope if they are related to _different_ levels/itemscopes according to your example with 'review'/'location'/'geo'/'lat'/'long'?

We have property (@itemprop) 'foo' with value 'bar' (@itempropvalue). Also we have nested itemscope with its own @itemprops/@itepropvalues exposed via _child_ elements (not shown in our examples) of itemscope element. Where is the confusion here?

> I'm ignoring the latter example for the time being. It's somewhat
> clearer, but it has its own problems, such as defining a second syntax
> for @itemprop. Remember, @itemprop currently takes a space-separated
> list of properties.

It's questionable how common in real world are usecases where different (space-separated) microdata properties share same value.

>  Your suggestion would probably require *also*
> defining a comma-separated syntax for your colon-separated pairs.

To be clear just in case: @itemprop in my CSS-like-syntax proposal is not intended to store more than one name/value pair.

>  How
>  whitespace is treated before and after the value is unclear as well.
>  (CSS gets around this by making whitespace insignificant.  You can't
>  do that with Microdata.)

Quotes around a value would probably be enough (CSS-like way as well):

<span itemprop="foo: 'lorem ipsum'">

> A property's value may, itself, be another Microdata item. For
> example, the 'location' property of a 'review' item may be a 'geo'
> item with 'lat' and 'long' properties. That's indicated by putting
> @itemscope on the element with the @itemprop.

Thanks, so it's just about nested itemscopes. This does not make @itempropvalue attribute confusing at all.

Only extra thing that probably should be changed in the spec if @itempropvalue attribute will be accepted is to rename @itemprop to @itempropname. Then we would have completely clear/transparent name/value pair via [itemprop]name / [itemprop]value attributes, respectively.

Dedicated DATA element ("an element intended just to store a value") still looks unneeded and littering.
Received on Thursday, 17 November 2011 17:58:01 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:17:41 GMT