On 5/28/15 2:04 PM, Travis Leithead wrote:
That sounds good to me (working the UI challenges for IME together with
grammar/spell checking). Is this a direction the current IME spec could
take--possibly with a big refactor to consider dropping the ClientRect
exposure--or would it be better to publish as Note the current approach and
start a new spec?
Perhaps it doesn't really matter as the above is a process question, and what
is really needed is someone to start suggesting some concrete proposals
here--I've been pretty much ignoring this spec for the past year, and I don't
see that changing in the near future. It's still something I'd like to see
moved forward, I just don't believe I have the time to move it substantially
forward at the present moment :)
Given this, it seems like the SoTD section should include some type of
large-ish warning/note that include text along the lines of `no one is
actively working on this spec nor its implementation` + `help wanted`.
Also, if the spec's [Issues] and [Bugs] haven't properly captured the
ClientRect exposure issue, perhaps one should be created (and/or the
spec updated to reflect this).
On the other hand, if you propose to formally stop work on the spec "as
it is" now, then it would be appropriate to have a CfC to publish a WG
Note (and if/when there is a firm commitment to do some type of followup
work, a new spec can be created).
WDYT?
-Thanks, ArtB
[Issues] https://github.com/w3c/ime-api/issues
[Bugs]
https://www.w3.org/Bugs/Public/buglist.cgi?component=IME%20API&list_id=57282&product=WebAppsWG&resolution=---
-----Original Message-----
From: Ryosuke Niwa [mailto:rn...@apple.com]
Sent: Wednesday, May 27, 2015 7:00 PM
To: Travis Leithead
Cc: Arthur Barstow; public-webapps
Subject: Re: [ime-api] [blink-dev] Removing IME API code from Blink
On May 27, 2015, at 11:46 AM, Travis Leithead <travis.leith...@microsoft.com>
wrote:
I believed the use-cases for avoiding UI clashes between site-driven
auto-complete lists and IME auto-complete boxes is still a valid use case, and
I think the spec is still valid to try to push to recommendation. However, I'd
also like to follow up on usage of the ms- prefixed API so that I can get an
idea of what its real usage is.
I agree avoiding UI clashes between auto-completions of IME and web page is a great use
case but I'm not convinced that exposing ClientRect for IME is the right API as many Web
developers aren't even aware of UI challenges IME imposes. For example, a similar UI
challenge emerges when dealing with auto-corrections in grammar/spell checking features
as well. It would be ideal if IME and spell/grammar corrections are handled in a similar
manner so that Web apps supporting either feature will "just work" with both
features.
- R. Niwa
