- From: Mark S. Miller <erights@google.com>
- Date: Mon, 14 Oct 2013 06:46:05 -0700
- To: Dirk Schulze <dschulze@adobe.com>
- Cc: Simon Pieters <simonp@opera.com>, "Robert O'Callahan" <robert@ocallahan.org>, www-style <www-style@w3.org>, Boris Zbarsky <bzbarsky@mit.edu>, "public-script-coord@w3.org" <public-script-coord@w3.org>
- Message-ID: <CABHxS9jrYvuER2YUKjP-bRKvjiPWWizjPzA4aQq+nHkXiPA0kw@mail.gmail.com>
On Sun, Oct 13, 2013 at 11:05 PM, Dirk Schulze <dschulze@adobe.com> wrote:
>
> On Oct 2, 2013, at 4:22 PM, Simon Pieters <simonp@opera.com> wrote:
>
> > On Wed, 02 Oct 2013 14:44:31 +0200, Robert O'Callahan
> > <robert@ocallahan.org> wrote:
> >
> >> On Wed, Oct 2, 2013 at 3:42 AM, Simon Pieters <simonp@opera.com> wrote:
> >>
> >>> On Tue, 01 Oct 2013 21:15:30 +0200, Robert O'Callahan <
> >>> robert@ocallahan.org> wrote:
> >>>
> >>> Now we have three interfaces where only two are needed. Why shouldn't
> >>>> DOMRect extend DOMRectReadOnly directly?
> >>>>
> >>>
> >>> In
> >>> http://lists.w3.org/Archives/**Public/www-style/2013Sep/0725.**html<
> http://lists.w3.org/Archives/Public/www-style/2013Sep/0725.html>you
> >>> argued that that would be confusing. Did you change your mind, or was
> >>> it only confusing for "immutable"?
> >>>
> >>
> >> As long as it's clear that ReadOnly != Immutable, I think it's OK.
> >>
> >> I think readonly<Foo> is going to be difficult to define and will also
> be
> >> clumsy. An interface might have arbitrary mutation methods on it, e.g.
> >> Foo::reset(), and there will be no easy way to automatically obviate
> such
> >> methods without additional annotations. I don't think we should go that
> >> way.
> >
> > OK.
>
> I would like to know if we all agree on the model: Foo (mutable and
> modifiable); FooReadOnly (just readable, might be mutable or immutable).
"just readable, might be mutable" is self contradictory. I think you mean
"readable, might be mutable". I suggest FooReader. This clearly states that
it provides the ability to read a Foo, and does not state what other access
it does or does not provide.
"Only" should mean only.
> Can we settle on this model? If their are still doubts and requests for
> readonly<Foo> instead, we should resolve them before we introduce APIs that
> need to be deprecated later.
>
> Does the stringifier for Foo look different to FooReadOnly? I would expect
> so, otherwise we would need to introduce an indicator for web developers if
> the current object is readonly.
>
> Greetings,
> Dirk
>
> >
> > https://dvcs.w3.org/hg/csswg/rev/6eba16fd75cb
> >
> > --
> > Simon Pieters
> > Opera Software
> >
>
>
>
--
Cheers,
--MarkM
Received on Monday, 14 October 2013 13:46:33 UTC