W3C home > Mailing lists > Public > www-tag@w3.org > December 2012

Re: FYI, tag election links

From: John Kemp <john@jkemp.net>
Date: Thu, 13 Dec 2012 22:56:35 -0500
Message-ID: <50CAA373.2090508@jkemp.net>
To: Brian Kardell <bkardell@gmail.com>
CC: "www-tag@w3.org" <www-tag@w3.org>
Hi Brian,

Comments inline below:

On 12/13/2012 09:24 PM, Brian Kardell wrote:

> On Thu, Dec 13, 2012 at 4:46 PM, John Kemp <john@jkemp.net
> <mailto:john@jkemp.net>> wrote:
>

[...]

>
>
>     For example, I believe that there has been a tremendous shift from
>     "declarative" to "imperative" with the transition to Javascript APIs
>     defining the web platform, and specifications that rely on
>     algorithms rather than defining a set of invariants.
>
>     These are rather historic shifts (in the Web's short history) and
>     have impact far beyond my technical description. And web
>     architecture as documented in AWWW has been proven to scale rather
>     well, and support a particular kind of innovation that has spawned
>     remarkable economic growth.
>
>
> Would you be willing to clarify some on what you are saying above?  I
> might just be missing it...
>

Sure.

HTML is a declarative language. We (as in W3C) went through a phase of 
defining transformations (as Eric pointed out in another thread) via 
XForms and XSLT. More declarative and functional languages. Declarative 
and functional languages are very well proven at scale (see Erlang for a 
relatively modern example) largely because they allow massively parallel 
operations to take place without "shared memory" problems. It is easy to 
define a "client" and a "server" and they have particular, and separate 
responsibilities. Only the interfaces are defined, not implementations.

In the modern "webapp", what is "client" and what is "server" is harder 
to see and define. Other architectural constructs have become exposed 
(DOM for example). We have more algorithmic specifications, which 
provide specific implementation details rather than merely interfaces. 
Javascript itself has particular language properties which impact how 
the Web evolves. That all changes the nature of innovations which are 
possible.

Then we have security. When a cookie is sent to a server, is it the 
client (user agent) on whose behalf the cookie is sent, or the server 
which asked the client to send the cookie? It's not clear today. That 
confusion is what causes CSRF attacks. Without cookies working this way 
though, we wouldn't have a web advertising industry, for example.

And the list goes on.

Architectural constructs on the web have been factors in the development 
of today's economic and social landscape. New companies, new markets, 
new ways of communication, new law. These sort of technical decisions 
make a difference that reaches far beyond their technical audience.

It's not clear, given the spread of the Web, that any one group (even 
all of W3C let alone the TAG) can do more than gently influence the 
course of the Web's development these days.

However, I do think it's still worth the effort for the TAG to consider 
the social impact of the architecture of the Web, and plan accordingly. 
In some cases, though, that may even be in conflict with existing Web 
development practice.

[...]

Regards,

JohnK

> --
> Brian Kardell :: @briankardell :: hitchjs.com <http://hitchjs.com/>
>
Received on Friday, 14 December 2012 03:57:03 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 14 December 2012 03:57:03 GMT