W3C home > Mailing lists > Public > public-html@w3.org > September 2010

Re: Zero-edit Change Proposal for ISSUE-41 (distributed extensibility)

From: Jonas Sicking <jonas@sicking.cc>
Date: Mon, 20 Sep 2010 15:18:54 -0700
Message-ID: <AANLkTik2qinXTaO4UztjJxvpnPuZfm5m7iS7WAAg=vXD@mail.gmail.com>
To: "Edward O'Connor" <hober0@gmail.com>
Cc: HTML WG <public-html@w3.org>
On Mon, Sep 20, 2010 at 2:43 PM, Edward O'Connor <hober0@gmail.com> wrote:
> Hi,
>
> I believe our null Change Proposal for ISSUE-41 is ready for
> consideration by the chairs. It lives on the WG wiki:
>
> http://www.w3.org/html/wg/wiki/User:Eoconnor/ISSUE-41
>
> It's not yet the 22nd, so I'd very much appreciate any improvements or
> comments other WG members would care to offer between now and then.

I would expand more on the positive impacts section. The main concern
that I keep hearing from web developers that they do not want to deal
with namespaces. Even the fact that SVG and HTML is in different
namespaces is a source of complexity. On top of this the DOM APIs
aren't particularly well suited for namespaces. For example the *NS
methods in DOM-core requires that two parameters instead of one is
passed around everywhere. And in other APIs, like querySelector and
DOM-XPath it is even more painful.

It has been suggested that this could be fixed by making modifications
to the DOM APIs. However so far no such proposals have been put
forward as part of the other change proposals for ISSUE-41, and so at
this time this is a purely hypothetical argument.

The result is that any time that there are localName collisions it's a
pain for developers. Even if the elements are in different namespaces.
Thus namespaces aren't buying us much in terms of avoiding naming
collisions, while adding complexity for authors.

Similarly, if other languages are developed which have localName
collisions with HTML, SVG or MathML, then it will make it harder and
more error prone to include those languages into the HTML parsing
algorithm. Even if such languages are integrated in the parsing
algorithm it is more prone that a small author error will result in
the wrong element being closed and a resulting in bigger errors in the
resulting DOM.

/ Jonas
Received on Monday, 20 September 2010 22:19:47 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:39:19 UTC