Re: what is appropriate to post to post to the WHATWG list?

Hi Ian, thanks for the reply.

Just to be clear, I am not and have not been (since I specced it)
attempting to persuade you as the editor of the WHATWG HTML spec to add
<main> to it.

I am pursuing it as a HTML5 extension spec, which can be folded back into
HTML5 or HTML.next if and when it is implemented.

The 'plea to hixie' thread was started by zcorpan, I only responded at
times where I thought I had relevant new arguments or data, again not to
persuade you, but in response to statements in emails on the thread.

Any one alone is probably fine, it's just the volume that you are sending
> on what to many people on the list is really uninteresting that is causing
> them to complain.
>

So the volume of my responses was cause for complaint, if so I will think
more carefully  before I respond. I understand that unfortunately features
that improve accessibility may not be interesting to many people, but I
don't think that is a good reason not to discuss them.I must say that much
of  the discussion that is on the list is really uninteresting to me
personally, but I understand that the list is not there for my pleasure,
but for the discussion of html features.

>It's not forbidden, but it's probably more appropriate to post to the list
>for the spec than the WHATWG list.

I have posted to the WHATWG list if I think it is where I will get
(primarily) implementer feedback, which is my motivation for posting to the
list and has so far proved useful.


What's not appropriate for the WHATWG list is repeating material that has
> already been posted, posting opinions without data or logical arguments
> ("I won't implement this", or "this would be hard to implement", or "I
> would prefer to implement this", when sent by user agent implementors,
> count as data, since they affect what gets implemented and thus whether
> the spec is useful), e-mails that just say things like "me too" or "+1",
> and the like.
>

I have not been doing that as far as I can tell,  but please disabuse me if
I am wrong.


Basically, if it won't lead to the editor of a WHATWG spec considering
> changing a spec, then it doesn't belong on the list.
>

This statement appears contrary to above and as previously stated I am have
not been trying to convince you, I have been engaging with implementers and
responding to the emails of others on the list.

in the case of <main>, the arguments for and against have been made in
> detail (and indeed, were made in detail back in 2005), the situation has
> been considered carefully multiple times, and unless there's new data to
> be considered, posting more about it just increases the noise side of the
> signal to noise ratio.
>

Yes, you keep alluding to decisions and discussions made as far back  2005,
I have searched the archives and can't find any associated detailed uses
cases and data, if you can point me to this it would be useful.

I did provide new data, which you disagree with, but which has persuaded
others including implementers to consider main.


Having said that, as far as <main> goes, despite it being a bad idea,
> there does seem to be some support for it amongst implementors. So your
> best move, if you think it's a good idea, would be to convince them to
> implement it. That would be new data which would almost immediately cause
> the spec to have it added, regardless of how good an idea it is.
>

It is not a 'bad idea' you saying it is, does not make it so and none of
the arguments you have put forth as to why it is bad have been convincing.

as maciej stated [1] in the plea to hixie thread:

"I would not fall on my sword to get the <main> element into WebKit but I
would not reject a patch to add it either, assuming a sufficiently good
spec exists for it somewhere."

 I think that  the main spec [2] after useful feedback mostly via the
WHATWG list and IRC contains sufficient detail for it to be implemented and
that is where I will be putting my efforts going forward.

so thanks for putting up with it thus far.


[1]
http://lists.w3.org/Archives/Public/public-whatwg-archive/2012Nov/0104.html
[2]
https://dvcs.w3.org/hg/html-extensions/raw-file/tip/maincontent/index.html

regards
SteveF

On 23 November 2012 08:01, Ian Hickson <ian@hixie.ch> wrote:

> On Tue, 20 Nov 2012, Steve Faulkner wrote:
> >
> > So I can better understand what you consider appropriate to post to the
> > WHATWG list and what may be a cause for complaint from other list
> > members, of the following emails I posted on the topic of the main
> > element this month, which are not acceptable:
>
> Any one alone is probably fine, it's just the volume that you are sending
> on what to many people on the list is really uninteresting that is causing
> them to complain.
>
> (Just so we're clear, it wasn't a lot of complaints. If it had been, I
> would have been much more proactive about killing the relevant threads.)
>
> On Tue, 20 Nov 2012, Steve Faulkner wrote:
> >
> > In particular is it OK to post to the WHATWG list, asking for and
> > responding to feedback on a non WHATWG spec:
>
> It's not forbidden, but it's probably more appropriate to post to the list
> for the spec than the WHATWG list. (e.g. XMLHttpRequest, Selectors, HTML
> Editing APIs, and DOM have all received feedback on the WHATWG list, even
> while not being WHATWG specs.)
>
> What's not appropriate for the WHATWG list is repeating material that has
> already been posted, posting opinions without data or logical arguments
> ("I won't implement this", or "this would be hard to implement", or "I
> would prefer to implement this", when sent by user agent implementors,
> count as data, since they affect what gets implemented and thus whether
> the spec is useful), e-mails that just say things like "me too" or "+1",
> and the like.
>
> Basically, if it won't lead to the editor of a WHATWG spec considering
> changing a spec, then it doesn't belong on the list.
>
> In the case of <main>, the arguments for and against have been made in
> detail (and indeed, were made in detail back in 2005), the situation has
> been considered carefully multiple times, and unless there's new data to
> be considered, posting more about it just increases the noise side of the
> signal to noise ratio.
>
> Having said that, as far as <main> goes, despite it being a bad idea,
> there does seem to be some support for it amongst implementors. So your
> best move, if you think it's a good idea, would be to convince them to
> implement it. That would be new data which would almost immediately cause
> the spec to have it added, regardless of how good an idea it is.
>
> Cheers,
> --
> Ian Hickson               U+1047E                )\._.,--....,'``.    fL
> http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
> Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'
>