Microdata support

[Elliott Sprehn]

We have an implementation of Microdata [1], but it's not turned on in Chromium. Firefox 16+, Zombie-Presto and Maxthon apparently all have support and it's in the HTML5 spec [2].

I also see http://google.com includes some random microdata bits though they don't make much sense to me.

Does anyone know why we never turned it on? Can we turn it on now?

If not we should probably delete it since we don't even compile it in.

[1] http://en.wikipedia.org/wiki/Microdata_(HTML) 

[2] http://www.whatwg.org/specs/web-apps/current-work/multipage/microdata.html

- E

[Ojan Vafai]

I don't have strong opinions about Microdata as a feature. It seems useful, so I vaguely support turning it on, especially as Firefox already has.

I have one concern with the feature as specced is that getItems and the various Collection returning properties/methods all return live NodeLists/Collections. Does the current implementation do so? I think if we can at all avoid it, we should try to get that changed if we want to ship this. Live NodeLists/Collections impose a large cost on the rest of the codebase and fundamentally make regular DOM operations slower.

[Adam Barth]

I have some concerns about the quality of the implementation in Blink.  If we do decide to turn it on, we should first review it for quality.

Adam

[Alex Russell]

The current micro data API is...poor.

I think we should write it off and try again. No opinions in what that means for our impl in the meantime, though (other than it shouldn't ship, of course).

I'm happy to put work into a better API if someone will collaborate on impl.

Regards

[Dimitri Glazkov]

I have similar concerns about quality.

:DG<

[Ojan Vafai]

Sounds like there are enough concerns with the current code/api that we are not comfortable shipping it. As we don't have anyone actively owning fixing it and it's not even clear any of this code would be useful for an API we'd be happy with, I think that means we need to delete the code for now.

[Adam Barth]

+1

[Jonathan Dixon]

Worth adding an entry to http://www.chromestatus.com/features to record this?

[Elliott Sprehn]

On Fri, Apr 12, 2013 at 11:03 AM, Ojan Vafai <oj...@chromium.org> wrote:

Sounds like there are enough concerns with the current code/api that we are not comfortable shipping it. As we don't have anyone actively owning fixing it and it's not even clear any of this code would be useful for an API we'd be happy with, I think that means we need to delete the code for now.

Okay I'll delete the code then. We can always roll out the deletion if needed.

- E 

[Max Heinritz]

>Worth adding an entry to http://www.chromestatus.com/features to record this?

Done! ctrl-f for "microdata": https://docs.google.com/a/chromium.org/spreadsheet/ccc?key=0AjGgk26K1Cc-dFdfUlRQRWtjUm5XdjB4cmFGWjhTZmc#gid=0 

[Max Heinritz]

Elliott, did you delete the Microdata code?  Do you know which version of Chrome was the first to ship without it?

[Elliott Sprehn]

I did delete it. We never shipped it, the ifdef it was behind was never turned on.

[Max Heinritz]

Cool, thanks for clarifying.

[Dan Brickley]

On 12 April 2013 10:32, Alex Russell <sligh...@google.com> wrote:
> The current micro data API is...poor.
>
> I think we should write it off and try again. No opinions in what that means
> for our impl in the meantime, though (other than it shouldn't ship, of
> course).
>
> I'm happy to put work into a better API if someone will collaborate on impl.

I've heard this from a few people now, and would be happy to join any
discussion of better APIs.

Is anyone here aware of uses of Microdata for purposes beyond
schema.org markup?

What was the original intent of the API? Did anyone use it?

Dan

[Michael[tm] Smith]

Dan Brickley <dan...@google.com>, 2013-08-08 11:44 +0200:

> On 12 April 2013 10:32, Alex Russell <sligh...@google.com> wrote:
> > The current micro data API is...poor.
> >
> > I think we should write it off and try again. No opinions in what that means
> > for our impl in the meantime, though (other than it shouldn't ship, of
> > course).
> >
> > I'm happy to put work into a better API if someone will collaborate on impl.
>
> I've heard this from a few people now, and would be happy to join any
> discussion of better APIs.
>
> Is anyone here aware of uses of Microdata for purposes beyond
> schema.org markup?

As far as applications that consume content that uses Microdata and that
actually do something useful with it, I don't think there are any
well-known ones other than schema.org.

I know Drupal has some specific support for Microdata but it's not clear to
me that enables anything more than just producing document with Microdata
attributes included (rather than consuming that Microdata content and doing
something special with it).

> What was the original intent of the API? Did anyone use it?

The original intent was just to provide a standard way for the Web platform
to expose the Microdata content of Web documents to scripts.

--Mike

--
Michael[tm] Smith http://people.w3.org/mike

[benjamin...@gmail.com]

I'd also be interested in seeing this move forward with an API that folks are happy with. For reference, here's the commit that removed the old microdata support.

commit 88e7fb9f921306929788b79bdbfca85d727f589b

Date:   Fri Apr 12 23:11:02 2013 +0000

    Remove MicroData implementation

    

    Per the discussion on blink-dev we should remove the MicroData implementation.

    There's questions of quality, a maintainer, and that no port of WebKit ever

    turned it on.

    

    See: https://groups.google.com/a/chromium.org/forum/?fromgroups=#!topic/blink-dev/b54nW_mGSVU

    

    Review URL: https://codereview.chromium.org/14028014

    

    git-svn-id: svn://svn.chromium.org/blink/trunk@148315 bbb929c8-8fbe-4397-9dbb-9b2b20218538