Hi Nancy,

On 12/12/2011 20:25, Nancy Johnson wrote:
Thanks,

I love the more graphical layout and organization putting critical
issues on top.

Yes, that's a good feature. There's a half-made plan to use the same design for the main validator but it's a big job.

[..]

I have been moving image sizing to the style sheet and not left inline..

NNooooo!!! That's one thing that the mobile checker is definitely good for - stopping this bad practice of using CSS to define the size of an image and, even worse, using CSS to resize the image.

W3C Best Practice: http://www.w3.org/TR/mobile-bp/#IMAGES_SPECIFY_SIZE

My take on it: http://philarcher.org/diary/2011/phpimageadaptation/

HTH

Phil.



On Mon, Dec 12, 2011 at 1:19 PM, Phil Archer<ph...@w3.org>  wrote:

On 12/12/2011 17:28, Patrick H. Lauke wrote:

On 12/12/2011 17:18, Nancy Johnson wrote:

I noticed this validator only checks for xhtml 1.1 basic or mp1.2. Is
it going to checking again html5? http://validator.w3.org/mobile/


Not to my knowledge, no. You could argue that it's aimed at older
generations of phones/browsers.


We (W3C) have been discussing this issue. The mobile checker is an
implementation of the mobileOK Basic Tests [1] which is the machine testable
subset of the Mobile Web Best Practices [2]. As long as that is true we
have:

- a checker rooted firmly in a specification - which is a good thing;
- a checker that is growing old and, as is obvious, increasingly out of date
- which is a bad thing.

If we were to update the checker to, for example, cover HTML5 or any other
technology (CSS3, SVG or whatever) then how would we root that in a spec? It
becomes a dynamic system without a reference point.

Now - since a lot of work went in to the checker (and the specs behind it)
and it's a potentially useful tool, we don't want to lose it. However, we
would need some sort of community effort to determine what the checker would
check. There's also an issue of cost - maintaining the validation suite
means writing new code.

For now, I think we can say that the mobileOK checker is a useful guide. A
lot of the best practices are still entirely valid. Taken with the Mobile
Web Applications Best Practices [3] they form good advice to any mobile
developer. However, it does need some interpretation - which is a pity.

For example, the checker will warn you if you don't use the
application/xhtml+xml MIME type. If you're coding in HTML5 that's simply
wrong and I haven't seen an instance where there's an advantage in using the
XHTML MIME type.

The checker will scream at you if you don't include cache control or image
dimensions - those are very much right!




What about media queries... Is the mobile checker suitable for if
you are creating one set of htmls code and for mulitple devices?


No.


I'd say not yet.  What we need is the mechanism for how to manage change and
how to effect change in the checker. Keep nagging us - that might help us
get it higher on the agenda.

HTH

Phil.



[1] http://www.w3.org/TR/mobileOK-basic10-tests/
[2] http://www.w3.org/TR/mobile-bp/
[3] http://www.w3.org/TR/mwabp/


--

Phil Archer
W3C eGovernment
http://www.w3.org/egov/

http://philarcher.org
@philarcher1



*******************************************************************
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
*******************************************************************



*******************************************************************
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
*******************************************************************



*******************************************************************
List Guidelines: http://webstandardsgroup.org/mail/guidelines.cfm
Unsubscribe: http://webstandardsgroup.org/join/unsubscribe.cfm
Help: memberh...@webstandardsgroup.org
*******************************************************************

Reply via email to