[whatwg] <meter> feedback

[Ian Hickson]

Summary: no changes made. If anyone has a better idea for a name for the 
element, let me know. <meter> is a little confusing to people, but 
<gauge>, which would be better, is just too hard to type for people.


On Thu, 8 Feb 2007, David Latapie wrote:
>
> On Thu, 08 Feb 2007 20:39:54 +0100, Anne van Kesteren wrote:
> > Have you read what <meter> is about? Because it seems to me like you 
> > didn't...
> 
> meter has some meaning attached to that the proposed <gauge> word does 
> not have. I would tend to favour this one.

<gauge> isn't an option due to spelling and typing issues.


On Tue, 13 Feb 2007, Karl Dubost wrote:
> 
> I agree that meter will lead to a lot of confusion. I fear many 
> questions or misunderstanding from people. Just because it has two 
> possible meanings. 
> http://www.whatwg.org/specs/web-apps/current-work/#meter0
> 
> possible words are:
> * tools
>   - gauge
>   - indicator
>   - scale
> * semantics of content
>   - magnitude
>   - quantity
>   - amount
> 
> another possibility is to use
> the latine name: metrum
> or the greek name: metron
> or japanese: 『計器』 pronunciation: keiki (sure to not be used elsewhere)

Maybe <indicator> would be ok, what do people think? Should we rename 
<meter> to <indicator>?


On Tue, 15 May 2007, Elliotte Harold wrote:
>
> This may have already been asked (sounds familair) but I don't see it in 
> the archives. What if the min or max values for a meter is unbounded? 
> For example, the 7th member of the Fibonacci series? How does one 
> indicate that?

You can't; the point of <meter> is a bounded range, lke a gauge.


On Thu, 12 Jul 2007, Smylers wrote:
>
> Under the current spec these <meter> elements both have a value of 7:
> 
>   <meter>7/10</meter>
> 
>   <meter max="10">7</meter>
> 
> but this one has a value of zero (while still having a max of 10):
> 
>   <meter max="10">7/10</meter>
> 
> While specifying the max value twice like this is obviously redundant, 
> in the cases where both values are the same it's unambiguous and 
> arguably harmless.
> 
> Is there merit in allowing this case?
>
> If not, would some other failure mode be less confusing?  It seems a 
> little awkward that when double-specifying one attribute that 
> attribute's value gets set as specified and it's an entirely different 
> one which has its apparent value ignored.

The idea was to have a bad failure mode so that it would be obvious. The 
problem is if someone does:

   <meter max="50"> 10% </meter>

...you might not notice the problem straight away, and debugging it might 
suck. By making failure very obvious (it'll always be zero and never 
move), we make it much more likely that this kind of thing will get 
caught.

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'