Tom Hume wrote:
> On 29 Oct 2008, at 17:16, Luca Passani wrote:
>
>
>> "explicitly" is a way too weak defense against abuse. They will abuse
>> and start arguing that those who complain have a different
>> perception of
>> "explicitly". It's as simple as that.
>>
>>
> Do you have a suggestion for stronger wording of that sentence -
> retaining the meaning but removing the loophole you see there (which
> I'm not sure I see btw)?
>

yes. Remove the clause that enables UA spoofing when the user requests
it. One possibility (which is also contemplated by the Manifesto) is to
issue a request with unaltered HTTP headers. If the response is not
"mobile-friendly" for any definition of mobile friendly, then you can
re-issue with spoofed headers.

BTW, I know that 4.1.5.1 says that "contest tasting" (i.e. duplicate
requests) are not OK, but the reason for it ("The theoretical
idempotency of GET requests") is bogus, so the problem could be solved
there.



>>> If you can point me to an email or fwd me one (off-list if you like)
>>> I'll make sure it gets raised.
>>>
>> look at the links I posted. There I ask why they do not make CTG only
>> apply to opt-in, and nobody took up the suggestion (and Sean Owen
>> said:
>> "I think that's a reasonable position to take. It's not the one I
>> would
>> take, nor apparently people here")
>>
>>
> What's the difference between requiring a user to explicitly ask for
> transcoding, and making transcoding opt-in? To me, they're the same
> thing: the user has to ask for it, rather than being given it and told
> to opt out.
>

the difference is that most users won't know/understand that a
transcoder has been introduced, let alone that they can opt-out of it.
Look at Novarra/VodaUK, they were claiming that their service was opt-in
simply because users were given a chance to log-out.

>>> I can't see any mention of APNs or handling certain mobile APNs
>>> differently from others in any of the links you've posted. All I can
>>> see are your opinions on the UA string and HTTPS?
>>>
>> look at the links in my previous email. I did not use the term APN,
>> but
>> the point was the same. Devices ship with an APN. If you place your
>> transcoder there, then it's not an opt-in transcoder. People get it by
>> default. So, everything I wrote about making CTG apply to opt-in
>> transcoders was about not placing the transcoder in the middle of
>> connections going through the main/default APN.
>>
>>
> Ah OK. But isn't the definition of "default APN" open to
> interpretation (certainly more so than the word "explicitly" IMHO)?
> After all, different applications or browsers on a device may be
> configured to use different APNs; and what is the default anyway -
> Vodafone WAP or Vodafone Internet, say?
>

right. So we should talk of main APNs, since an operator may have many.
In general, no APN that comes pre-configured with a phone.

> Just to check I've understood your point though: you're suggesting
> that transcoding should be opt-in for the default mobile APN?
>

I am suggesting that transcoding should happen in a way that
knowledgeable users will decide that a certain favorite web site of
theirs should be reached through transcoding, and explicitly access a
transcoding portal where they can type the URL or their
transcoder-managed bookmarks.

Luca


> --
> Future Platforms Ltd
> e: Tom.Hume@...
> t: +44 (0) 1273 819038
> m: +44 (0) 7971 781422
> company: www.futureplatforms.com
> personal: tomhume.org
>
>
>
>
> ------------------------------------
>
> As of July 14 2005, it's much easier to be banned from WMLProgramming!
> Please fail to read http://groups.yahoo.com/group/wmlprogramming/ before you
post.Yahoo! Groups Links
>
>
>
>
>