Re: Can accessing the device microphone and camera be added to HTML5?

@Ian:

On 3 May 2010 23:38, Ian Hickson <ian@hixie.ch> wrote:

>  I'm arguing that HTML (not HTML5, just HTML) should be in a state of

constant improvement


I can't imagine anyone is arguing against that. It's just that this
statement:

On 2 May 2010 22:11, Ian Hickson <ian@hixie.ch> wrote:

>  I think a better solution would be to have a continuous process of adding

features and fixing bugs, with no frozen versions.


...seemed to suggest some kind of ever-continuing process without
milestones. IMHO, not having milestones is a great way to prevent a project
from succeeding. Very nearly the first task of a project manager (whether on
a software project or a skyscraper construction project) is to identify the
milestones for the project. Milestones help keep the team focussed. In the
case of specification, it seems to me -- and like Jock I'm looking at this
from the outside (although time permitting I'd like to change that) -- that
team is a herd of cats, so the more one can do to focus them on a goal, the
better.

On 3 May 2010 02:35, Ian Hickson <ian@hixie.ch> wrote:

> Browsers don't generally claim to support a particular version, unless

they're lying.


That's an interesting statement. But I'm only slightly interested in what
vendors say (ex: http://samples.msdn.microsoft.com/ietestcenter/); I'm much
more interested in what external testing reveals (ex:
http://www.codedread.com/svg-support.php), and having milestones (like HTML
5.0, HTML 5.1, CSS3 Selectors, SVG 1.1, etc.) provides an invaluable means
of communicating that information usefully.

On 2 May 2010 22:11, Ian Hickson <ian@hixie.ch> wrote:

> > and then start work on HTML 5.1, or HTML 6, or HTML Codename Betty and

> Veronica, or whatever it should be called next.


> That's already started.


Are you referring to the WHATWG spec? (http://www.whatwg.org/specs/) Seems
like it, from the statement on that page that "... the WHATWG HTML
specification is the next generation of the language, while HTML5 is a more
limited subset with a narrower scope."

@Jock:

On 2 May 2010 21:10, Jock Murphy <jockm@stufflabs.com> wrote:

> The idea of a sandboxed file store, and the ability to work with files (I
> would love to be able to know the modification date of the file I may be
> uploading, or working with locally, as an example)


Check out the File API: http://www.w3.org/TR/file-upload/
http://www.w3.org/TR/file-writer-api/ Very exciting stuff. The read stuff
(at least) is supported in Firefox 3.6 and I believe it's coming soon to
Chrome; don't know about the others.

On 2 May 2010 22:13, Jock Murphy <jockm@stufflabs.com> wrote:

> Sadly as I look at the timeline shown on http://www.w3.org/2009/dap/ it
> looks like we are talking next year at the earliest that we might see
> adoption of these -- desperately needed -- enhancements.


Adoption is racing ahead of spec completion. Opera, Safari, Chrome, Firefox,
and IE (yes, really) are all moving forward with HTML5 features, despite the
specification still being a work in progress. I'm not saying spec completion
isn't important, just that adoption isn't being prevented.

@All:

Briefly: I'd like to send a huge "thank you" to everyone who's gotten HTML5
to this point. It's such a massive leap forward, both in terms of what's
specified and in terms of the quality of the specification. It feels to me
like a "reboot" of the language, and it needed rebooting.
--
T.J. Crowder
Independent Software Consultant
tj / crowder software / com
www.crowdersoftware.com


On 3 May 2010 23:38, Ian Hickson <ian@hixie.ch> wrote:

> On Mon, 3 May 2010, Nathan wrote:
> > Ian Hickson wrote:
> > > On Thu, 29 Apr 2010, Fenton Travers wrote:
> > >> Get into a 12-18 month cycle for spec upgrades for a while, until we
> > >> run out of major areas of improvement, then things can slow down
> > >> again.
> > >
> > > I think a better solution would be to have a continuous process of
> > > adding features and fixing bugs, with no frozen versions. What's the
> > > point of a cycle? It doesn't match any of the browser vendors, it
> > > doesn't match the authoring community, so why bother? It's just
> > > artificial.
> >
> > Could you expand on this a little - it's left me somewhat confused.
> >
> > Is the takeaway that HTML5 will constantly be in flux and never reach
> > Recommendation / be a Web Standard?
>
> I'm arguing that HTML (not HTML5, just HTML) should be in a state of
> constant improvement, with the stability of each part being tracked
> separate from the whole. This is effectively what's going on now at the
> WHATWG:
>
>   http://whatwg.org/html
>
>
> > Will there be (or is there) some form of core HTML5 specification that
> > vendors must implement and support, and then optional specifications
> > which vendors can choose to implement or not - or at least a structure
> > that would allow vendors to effectively say we "support a,b,c but not d
> > & e"?
>
> Browser vendors are never forced to do anything. It's our job as spec
> writers to make sure that everything in the spec is what the browser
> vendors are willing to implement, and to remove what they don't want to
> implement. This is the case with the current process as well as with the
> process I'm suggesting we follow going forward.
>
>
> > The short is, when I'm authoring a document I need to know what's
> > supported globally, or at least what should be supported globally.
>
> The only way you can know that is to test it.
>
>
> > aside: is there a fixed way of determining at runtime what features are
> > supported before attempting to use them (particularly with js related
> > features) or is it up to userland to figure out how?
>
> It depends on the feature.
>
> --
> Ian Hickson               U+1047E                )\._.,--....,'``.    fL
> http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
> Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
>
>