[whatwg] Web Forms attributes: minlength / autocomplete

[Anne van Kesteren]

On Wed, 06 Jun 2007 14:15:02 +0200, Sander <html5 at zoid.nl> wrote:
>> Why can't you use pattern= for this?
>
> Maybe I could use that, but as a front-end developer I'm not aware of  
> all the possibilities and grammar of regular expressions (as goes for a  
> lot of front-end developers I guess, even proffessionals).
> I guess pattern can be used instead of the maxlength and required  
> attributes as well and it even makes the input types email and url  
> obsolete (in a technical manner).

No it does not. It does not serve the same purpose as required and does  
not allow the user agent to recognize email or url input controls.


> But I thought HTML5 was to make things better and easier. For most  
> people minlength will be a whole lot easier to use than pattern as  
> regular expressions are not really intuitive to write. And as there is  
> already a maxlength attribute (which will probably not be deprecated) I  
> think it's more logical to have a minlength as well.

What are the use cases?


> [...]
>>> --- autocomplete ---
>>> The autocomplete attribute acts as a boolean, but its values are not
>>> consistent to other booleans, being either attributes or variables.  
>>> <...>
>>
>> This attribute is defined in a way that is compatible with existing  
>> implementations. It was introduced long ago but never formally  
>> specified. So we can't really change it.
>
> I see. But we can add true/false as possible (and prefered) values and  
> keep on/off for backward compatibility.

That would not be backwards compatible. It's better that authors create  
content that is backwards compatible.


> Then again, the autocomplete values are not a real big deal as its  
> functionality doesn't change. But if we want to stay in tune with all  
> existing implementations, then maybe we should at least ask the browser  
> vendors to wait with implementing HTML5 features untill the specs are  
> final. An element like <canvas> is new to HTML5 but also already  
> implemented in a couple of browsers (well, I don't have to tell you that  
> of course). Following your argument a of of its specs may not be changed  
> anymore, even though the HTML5 specs are still in working draft.

Correct. The specification is driven by implementation and vice versa.


-- 
Anne van Kesteren
<http://annevankesteren.nl/>
<http://www.opera.com/>