Re: [Slightly OT] How to process lightweight text markup languages?

["Randy Brukardt" <randy@rrsoftware.com>]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 2259 bytes --]

<gautier_niouzes@hotmail.com> wrote in message 
news:a26a1003-dff6-4f07-90e7-b9aa041ad1bd@googlegroups.com...
>Le lundi 19 janvier 2015 21:12:41 UTC+1, Randy Brukardt a écrit :
>
>> No, this is a fallicy. The Internet (and world in general) would be 
>> better
>> off if documents that are sufficiently malformed were simply not 
>> displayed
>> at all. In that case, the document source would be fixed if anyone cares
>> enough.
>
>Alas the forces of competition resulted in the opposite: if a browser 
>doesn't display
>a malformed HTML, users blame the browser and try another one.

You're right today, of course, but it didn't have to be that way. And 
shouldn't have been...

> By developing
>Wasabee I was amazed by the proportion of sites with malformed HTML code.
>I guess it was always close to 100% but over time it is even closer. 
>Chances are
>that only the W3 site is okay...

We checked all of the pages on the previous AdaIC site through the WC3 
tools. (Us Ada people care about Standards, you see.) We didn't intentially 
post anything with errors (although it happened a few times). So there was 
at least two such sites. I'm pretty sure that the various Ada standards and 
other documents produced by our tools are also correct (I've checked them 
periodically).

...
>Only the input is idiot. The tool has to be smart. See for instance the 
>algorithm
>in Wasabee that corrects an ill-formed HTML tree. Same in real life: 
>sometimes
>smart guys have to find solutions for idiots (customers, bosses,...)

It's impossible to guess all of the ways that some "smart" tool will 
interpret malformed input, and indeed those ways can be different for 
different tools. That provides a near-infinite number of ways for bad guys 
to sneak junk past one's defenses.

>> And for what? So that people who can't be bothered to get their documents
>> formatted correctly still get some results (thus reenforcing their bad
>> behavior). Bah humbug!!
>
>You'd like to start a new, parallel Web with only well formatted documents 
>?

Sounds good to me. I pretty much do that for e-mail today, and that 
certainly helps prevent spam from being delivered. I'd do it for the web, 
too, if I could figure out how.

                               Randy.