W3C home > Mailing lists > Public > public-bpwg-ct@w3.org > November 2008

RE: Mandating respect of some heuristics?

From: Rotan Hanrahan <rotan.hanrahan@mobileaware.com>
Date: Thu, 27 Nov 2008 08:44:39 -0000
Message-ID: <D5306DC72D165F488F56A9E43F2045D301D0A5FF@FTO.mobileaware.com>
To: "Dominique Hazael-Massieux" <dom@w3.org>, <public-bpwg-ct@w3.org>

As suggested, inferring an intention from a so-called URI pattern is a very risky heuristic, as any of the following examples show:

- http://www.gifts-cabin.co.uk/acatalog/mobiles.html
- http://www.babygizmo.com/c/mobiles.html

(These are actually sites for hanging mobile decorations as used in children's nurseries.)

- http://www.cote.azur.fr/portable.htm

(Despite this being a French site, and having a path suggesting "téléphone portable", it is in fact a desktop site offering ringtones.)

- http://validator.w3.org/mobile/

(The W3C's mobileOK checker, which has /mobile at the start of its path, but which is in fact a desktop page and is certainly not mobileOK itself.)

However, regarding the four "heuristics" that were listed, I'm not sure if they should be considered as heuristics. These are very strong indicators of intent. There is no sense of "guesswork" here. If any of the four hold true, it is necessary to record that the retrieved (or referenced) content is intentionally for mobile, unless there is some bug/error. So barring bugs/errors, the four are really *rules*, not heuristics, surely?

---Rotan.


-----Original Message-----
From: public-bpwg-ct-request@w3.org [mailto:public-bpwg-ct-request@w3.org] On Behalf Of Dominique Hazael-Massieux
Sent: 26 November 2008 14:10
To: public-bpwg-ct@w3.org
Subject: Mandating respect of some heuristics?


Hello dear CT TF,

I'm wondering if the CT guidelines shouldn't mandate some of the (for
now qualified as) "heuristics" as meaning that a CT proxy shouldn't
apply any non-trivial transformation (with an explicit list of these;
e.g. only compressing/removing extraneous white space would be allowed).

The heuristics I have in mind in particular:
 * mobile doctypes (XHTML MP and Basic, WML)
 * <link rel="alternate" media="handheld" href=""/>
 * (possibly Content-Type: application/vnd.wap.xhtml+xml)
 * (possibly a mobileOK claim if that gets defined in time for this
document)

The reasons for mandating these ones in particular would be that they
express quite explicitly the intent that the author made his site for
mobile devices (while e.g. a uri pattern cannot be said to have
associated semantics per the Web architecture).

I think the high-level question behind this practical proposal is
whether a mobile CT proxy should be allowed to transform content that
was developed with mobile in mind or not. 

I tend to think that given the mission of our WG (which I understand is
to encourage the development of mobile-friendly content), I think we
should make sure that someone that invested in making his site
mobile friendly won't get his work transformed by a CT.

(I think some CT vendors argue that some mobile-designed web sites are
still not as optimized as they could be, but my take would be that this
is up to the users/the market to decide, not to the operators/CT)

I know some of these questions have already been heavily discussed by
the group and the task force, but I think the approach of favoring
mobile-content creators by default would be a better one for the mobile
web ecosystem as a whole.

Let me know what you think,

Dom
Received on Thursday, 27 November 2008 08:45:21 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 27 November 2008 08:45:23 GMT