Let me prefix this by saying for those who don't know me that I am known for 
being very pro-standards and pro-free-software. I'm an invited expert to the CSS 
working group and I am Mozilla's default Style System QA contact. Please don't 
interpret this as me being in favour of an IE-only web!

Richard Stallman <[EMAIL PROTECTED]> wrote:
> 
> I am told there are many sites that only work with IE.  I am trying to
> think of ways to do more to discourage that, and it occurs to me that
> Mozilla could help organize the effort in a very efficient way.

The reason sites only work with IE is because in the real world, only IE 
matters. IE has about 99% of the market by pessimistic counts, and at least 80% 
by conservative estimates. From a business standpoint, ignorant web site authors 
can make make a strong case to their employers that only IE compatability 
matters, and that standards-compliance and testing on non-IE browsers is not 
worth the expense.

(I'm not saying this is the case. I'm saying that people are able to 
convincingly argue this position to the ignorant money-givers.)


> What would you think of adding an automatic feature to send a
> complaint message when certain features are encountered?

The moment any browser with such a feature gained market acceptance, it would 
result in massive distributed denial of service attacks on offending sites.


> If there are several features that cause such trouble

There are *hundreds* of such "features", and most are extremely subtle 
(ironically many are actually bugs in IE, and were not added out of malice at 
all). In fact, it is often very difficult for me to tell whether a site is 
failing in non-IE browsers due to bugs in those browers, bugs in the web page, 
or use of an IE-specific features. When the latter, it is often hard to tell 
which feature is the cause. The time that would be spent writing heuristics to 
analyse web pages would be better spent making the browser in question more 
appealing to users.


> What do you think?  I think that over a period of a couple of years
> this might generate substantial pressure on sites to stay away from
> nonstandard features.

IMHO by *far*, *far* the best pressure to apply would be to get another browser 
to get enough market acceptance that it has enough market share to register on 
web authors' profit margins.

Incidentally, some people are defeatist and say that IE being bundled means that 
other browsers have no chance. The IE6 numbers disprove that theory. IE6 is only 
bundled with Windows XP, which accounts for 9% of Google's users. Yet IE6 
accounts for over twice as many users as all non-IE browsers put together -- 
around 30% if I am reading the graphs correctly [1]. That implies that of the 
people choosing to download a browser, twice as many download IE as any other 
browser (~20% vs ~10%).

[1] http://www.google.com/press/zeitgeist/zeitgeist-jan02.html

Brendan Eich wrote:
 > rms wrote with this idea, and said it was ok to pass along.  I am
 > arguing with him that spamming webmasters about IE-only features
 > actually *could* make things worse -- it could annoy them into moving
 > from a position of ignorance (I didn't know I was using something that
 > isn't standard -- I'll fix it, thanks) to intention (screw you for
 > spamming me, I'm never gonna change -- IE forever!).

That is another likely scenario, yes. Netscape 6.x (based on Mozilla) has some 5 
million web users. If just five thousand of these (around 0.1%) visited 
www.gnu.org and their browser found that site to be using some offending feature 
(say the deprecated <center> element), that would result in someone receiving 
5000 distinct e-mails. As you can see, this suggestion doesn't scale.

-- 
Ian Hickson
``The inability of a user agent to implement part of this specification due to
the limitations of a particular device (e.g., non interactive user agents will
probably not implement dynamic pseudo-classes because they make no sense
without interactivity) does not imply non-conformance.'' -- Selectors, Sec13

Reply via email to