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

Comments on BP2

From: Sean Owen <srowen@google.com>
Date: Thu, 13 Mar 2008 15:56:06 -0400
Message-ID: <e920a71c0803131256x114288c9k1e8cf634ff36a142@mail.gmail.com>
To: "Mobile Web Best Practices Working Group WG" <public-bpwg@w3.org>

We were asked on the call today to again review the BP2 draft at this
time, and comment.

Hats off to the editor, naturally. What a job it is to condense
miscellaneous thinking into a coherent recommendation.


I suppose I still take issue with the types of things that have been
contributed by the group into this document so far. The majority of
these BPs are not ones I personally would have included. Some that I
definitely would include have not been. Let me call up a few
representative examples that illustrate some issues I see:


5.2.1 Protect Personal Information Used in Transactions
This is not mobile-specific, it seems to me. Nothing about the
recommendation does not apply to the web at large. I don't disagree
with the sentiment a bit, but think our charter in the *M*WI means we
are to confine ourselves to discussing what is new, or different, in
the mobile context. This is one practice I would simply remove.


5.3.2 Inform the User About Device Memory Impact for Application
Installation and Use
This seems to be veering into application-land, out of web-land. I
think this is categorically out of scope for the M*W*I.


5.4.2 Minimize Redirects in Server-Server API's
What does this add over BP1?


5.9.4 Provide Alternatives to AJAX
Other BPs here seem to tacitly assume we're talking about a device
with "AJAX-ish" scripting capability. This BP says, but if that's not
so, do something else. Yes... but this seems like writing a best
practice like "provide alternatives to mobile", instructing mobile
applications to provide a desktop experience to desktop browsers. That
is, I suppose it goes without saying. I would think it more natural to
say "we're talking about devices that do AJAX here in BP2" rather than
finesse it with a content-free suggestion like this.


On the other hand I find BPs like this quite good:

5.9.2 Use Reliable Methods for Determining Script Support
This offers a concrete, clear suggestion that is relevant to mobile
insofar as the question of script support is far from a foregone
conclusion in mobile, is relevant to web technologies, and is not a
restatement of BP1.


I see in the Seoul meeting notes that my suggestion to look at Apple's
developer guidelines was more or less shelved. I suppose I had seen
that as precisely the sort of thing we should be writing. In
particular the "viewport" trick seems like by far the #1 thing I'd
mention to someone wondering what's new in the world of the high-end
mobile browsers. Comparing that to what is coming into the document, I
suppose I'm left wondering whether this is on the right track?


To be concrete: I'd remove the BPs I name above to start. I would like
to at least reconsider a viewport meta tag BP. And so on from there.
Received on Thursday, 13 March 2008 19:56:47 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 13 March 2008 19:56:48 GMT