- From: Tobie Langel <tobie@sensors.codespeaks.com>
- Date: Wed, 10 Jun 2015 16:16:52 +0200
- To: Boris Zbarsky <bzbarsky@mit.edu>, public-script-coord@w3.org
- Cc: Mounir Lamouri <mounir@lamouri.fr>, Marcos Caceres <marcos@marcosc.com>
On Wed, Jun 10, 2015, at 16:09, Boris Zbarsky wrote: > On 6/10/15 5:38 AM, Tobie Langel wrote: > > Adding support for partial enums in WebIDL would do the trick. > > This came up before in > https://www.w3.org/Bugs/Public/show_bug.cgi?id=27048 yes? Not hat any > conclusions were reached... Sorry had missed that. Not sure either re conclusions. > > Is that doable? Else what other option would allow for this use case? > > Either using a string or modifying the base specification that defines > the enum. I guess using a string would work. :) > I should note that the concern over fragmented specs is very real. > There are some working groups producing lots of small specs with > intricate and hidden interdependencies that are hellish to implement (or > indeed to understand; you have to read all of them, including ones not > referenced from the one you're reading, to understand how the spec > you're reading works). Care to point to some examples of that, off-list if you prefer, so that I avoid reproducing the same mistakes? > The permissions example is not quite that extreme, I guess... I'm not an implementor and thus not a good judge of that. --tobie
Received on Wednesday, 10 June 2015 14:17:19 UTC