Re: LYNX-DEV Java/JavaScript

[Mark H. Wood]

On Fri, 20 Mar 1998, Dave Eaton wrote:

> On Fri, 20 Mar 1998, Philip Webb wrote:
> 
> > it would seem very simple for Lynx at least to alert the user,
> > since the JavaScript is enclosed in  <SCRIPT ... > ... </SCRIPT> :
> > i picture a suitable formatted `Message from Lynx' inserted in the page;
> > it could perhaps be done for image-heavy HTML too;
> > any problem about adding this for 2-8-1 ?
> 
> Yes, it is counter to the specifications.
> 
> HTML 3.2 (http://www.w3.org/TR/REC-html32.html) stated:
>  "User agents should hide the contents of these [SCRIPT] elements."

Does W3C use "should" in the same way as IETF does?  That is, does it 
mean "we really strongly recommend that you do this, but your code can be 
standard-conforming even if it does something else"?  If so, we could (if 
we choose) decline the suggestion with thanks and a note in the doco to 
the effect that we have done so (and why).

> What authors could do at this point was provide script comment lines
> outside the HTML comment which hid the script. Though perhaps odd looking,
> these comments could warn a user they were missing something.
> 
> HTML 4.0 (http://www.w3.org/TR/REC-html40/):
> HTML 4 has added the NOSCRIPT element to permit authors to provide
> content if they wish. From the spec:
> 
>  "User agents that do not support client-side scripts must render this
>   element's [NOSCRIPT] contents."

This ignores the observation that 99.44% of all authors are also 
browser-patriots (which is not necessarily evil) and assume that, if the 
reader doesn't use the same browser the author does, he will be happy to 
switch (which assumption *is* evil).  Authors tend to custer in three groups:
(1) using some automated tool and entirely unaware of whether it provides 
NOSCRIPT, ALT, and so forth; (2) assuming that everybody uses the same 
browser, so they don't care whether they use these tags; and (3) the 69 
or so members of the any-browser movement.  In summary:  yes, authors 
could do the Right Thing but they often don't.

I realize that *doing* something about all the incomplete code out there 
goes against the principle that one should trust the author and accept 
what he gives you, but surely in this case (since there are so many 
authors proven untrustworthy) one could accept a non-default option not to 
trust?

> If Lynx starts warning on SCRIPT tags without scanning to see if there is
> also an associated NOSCRIPT tag, it will cause a properly composed HTML 4
> page to be displayed with erroneous warnings.

True.  And since the bulk of the HTML out there seems to be improperly 
composed, this could be a significant annoyance.

-- 
Mark H. Wood, Lead System Programmer   address@hidden
One more time:  a (level-2) switch is a bridge.  A "level-3 switch" is
a router.  Deal with it.