petercon at microsoft.com
Tue Dec 1 18:25:02 CET 2015
Michael, I think you're missing Shawn's points.
If WP only needed their internal processes to work, then a private-use subtag would meet their needs. The only reason for having a registered subtag is if you want to be able to interoperate with other parties. If they want to create spelling and grammer checkers, a registered subtag like wpsimple is only needed for those if other parties need to reference WP's spelling/grammer checkers and do so in a way that differentiates from all other English grammer checkers. Is WP going to be creating a grammer checker that you can install on your Mac or PC and use in Word alongside some other simple English grammer checker, and are you (or whoever) actually going to be installing and using those side-by-side? (It doesn't seem to me like a likely scenario.)
For content retrieval, a user looking for simple language content probably doesn't care whether it's WP's particular version of simple English or somebody else's form of simple English. Browsers will send accept-language headers based on language settings in the browser or the user's device. So, if someone wanted wpsimple content, they'd have to configure language settings in their browser or on their device to en-wpsimple. But then if they want to do that for another site — BBC news, say — and that site has its own registered subtag for simple English, then the user will need to configure language settings for that as well. So, they end up configuring both en-wpsimple and en-bbcsimpl (and so on for other sites with their own form of simple English). Again, the user probably doesn't care about the differences between WP's simple English and BBC's; in the user's mind, there is only one thing they care about, simpler English. So, the scenario would work better if all of those sites responded to en-simple: then the user only needs to configure their system for one thing rather than many.
From: Ietf-languages [mailto:ietf-languages-bounces at alvestrand.no] On Behalf Of Michael Everson
Sent: Tuesday, December 1, 2015 2:38 AM
To: Shawn Steele <Shawn.Steele at microsoft.com>
Cc: ietflang IETF Languages Discussion <ietf-languages at iana.org>; Amir E. Aharoni <amir.aharoni at mail.huji.ac.il>
Subject: Re: Pending requests
> On 30 Nov 2015, at 23:01, Shawn Steele <Shawn.Steele at microsoft.com> wrote:
>> What might spell-checkers or grammar-checkers flag if anyone developed software to support “en-simple"?
> By Wikipedia's own description a spell checker seems precluded, since it allows extending the vocabulary when necessary.
It could still flag any word not in its small dictionary as a guide to the writer.
Plus I said grammar-checker.
> If Wikipedia want's to make a spell checker, then an en-x-wpsimple is sufficient. If Wikipedia wants to interoperate with the readers (rather than the developers), then they need something less explicit so that the reader's computer's have a chance of providing Wikipedia with a useful tag.
Please explain this. en-simple and en-wpsimple are the same except that one of them (the first) could mean anything to anybody while one of them (the second) would point to a coherent specification.
> En-simple could work as a valid http-accept-language on Wikipedia or Voice of America or CNN or wherever someone wanted a simplified reading level.
But if the content of a VOA or CNN simplified version differed from the WP specification or from each other, what good would be served?
> Having the user select "English (United States, Simplified)" from their user settings would make detection of those cases automatic and be far more useful to the community that would benefit from such a declaration. I can't say "en-US-voa-wpsimple-…."
Nor should you. VOA’s specification could be very different from WP’s, could it not?
Michael Everson * http://www.evertype.com/
Ietf-languages mailing list
Ietf-languages at alvestrand.no
More information about the Ietf-languages