HTML is

Hack your OS, customize your GUI, announce great software, and configure, configure, configure...

HTML is

Postby SOD » 11/04/06, 4:01 am

Not unforgiving.

If you are one of the millions who write
crappy markup or have your own opinion on whats right you can thank your lucky stars that HTML parsers are forgiving. They ignore many faults. For a third grader
that's great for someone who has been writing code for a while that's pathetic.
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Postby SOD » 11/04/06, 8:42 am

Not an end all to everything. If you write say PHP or JS or any other language you have comply with well formed
and valid code. Why would the 'experts' assert that one of the best things about html is that it does not have to be valid to work? Pretty easy to remain an expert is to lower the standards not raise the bar.
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Postby SOD » 11/04/06, 8:58 am

Something that does not have to be 'reinvented' as announced by T.B.L. this week. All browsers SUCK. When oh when are we going to get reference parser that encompasses the new CSS and the latest HTML spec and when can we expect proper support for XHTML?
At that point all other versions of HTML should invalid and browsers should raise errors to reflect this.
If I can teach a fifth grade class to write valid XHTML 1.1 Strict then adults should have no problem.
A browser should be more than a personal preference
it should be a language reference, after all isn't a browser
the interpreter for HTML?
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Postby SOD » 11/04/06, 9:12 am

Is not a place to write bulky, hidden data then to prove your concept extract it from a HTML document with a language that was intended to change the behavior of html output in a browser not deal with discrete data parsing. I'm talking about microformats. They blow.
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Postby Gerry » 11/04/06, 4:34 pm

I'm talking about microformats.
Heh. No shit.
I answer rhetorical questions for my own enjoyment.
Gerry
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5727
Joined: 12/04/01, 12:00 am
Location: Perth, Western Australia

Postby bob » 11/06/06, 12:40 am

"Why would the 'experts' assert that one of the best things about html is that it does not have to be valid to work?"

Reliability
WYSIWTF
bob
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 7565
Joined: 12/03/01, 12:00 am
Location: St. Louis

Postby kenray » 11/07/06, 5:51 pm

the sweet little miracle that turned ARPAnet into the "wheel" of future and present technology.

Just as our entire electric world exists because Nikola Tesla was able to look at Edison's DC, and see the potential of Alternating Current, our online web world exists because of HTML.

complain or not.
kenray
Senior Member I Get Free Beer
Senior Member I Get Free Beer
 
Posts: 264
Joined: 01/17/02, 12:00 am

Postby SOD » 11/07/06, 10:41 pm

Bob: Isn't HTML only as reliable as the parser? Although
anecdotal evidence, research of HTML in RSS files shows a large amount of markup that is that is not well formed. Its nice that the current parsers forgive much of that but a document that is not well formed is not future proof. Change the current vision of what a parser is and it is no more fault tolerant than XML or any doc with malformed markup.

The one benefit to using XHTML is that your document
is then checked by any XML parser in any browser for
well formed markup. XML as you know only allows well formed markup. It's like spell check for markup.

Kenray: I'm not bitching about HTML although it is a wagon wheel. Imagine concurrent processing of documents that not only parse the venerable HTML but parse the other parts of a document with different parsers for different needs. Not that I can do that :roll:
Last edited by SOD on 11/08/06, 9:54 am, edited 1 time in total.
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Postby SOD » 11/08/06, 9:53 am

kenray wrote:our online web world exists because of HTML.


Don't confuse HTTP with HTML Kenray. I can build a parser that is not HTML yet transmit and parse the
format over HTTP no problem. HTML is a start to the future not the destination.

Ya know if you go to the edge of HTML you will not fall off there are many things beyond it. heh kinda like Columbus 'discovering' America he was told the same thing by 'experts'.
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Re: HTML is

Postby Pogrelz » 11/10/06, 12:20 pm

Making HTML parsers unforgiving really wouldn't solve much -- you can still write terrible code with PHP or JS, or any other language. I've been making sure my pages validate ever since... well, for just about as long as I can remember, and most of them have still sucked really bad in one way or another. My validating web page probably has hundreds of faults in the eyes of someone like Jeffrey Zeldman.

HTML parsers try to make things more intuitive. Sure, putting that unordered list right there is against the rules, but come on, I just want the damn list to display in that spot. When learning about the box model, IE's messed up one is a lot more intuitive. Well, at least in my opinion. And I think that intuitive things are good.

Don't get me wrong, SOD, I hear what you're saying and I agree. It's too bad that "Web 2.0" couldn't mean starting over from scratch.
....
Pogrelz
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 1093
Joined: 03/03/02, 12:00 am

Postby SOD » 11/12/06, 2:24 am

I don't think parsers should be intuitive Pogs.
An HTML parser is not about expression. It just transcribes how the data looks. It is the medium of transmission.

Data should be intuitive, for that to happen I guess it is more important that markup is well formed. HTML is not a
good vehicle for this, XML works well and can keep HTML well formed. Yeah, I don't care where the list is on your HTML page just make it well formed.
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Postby Gerry » 11/12/06, 9:43 am

The more pages you create, the more you try to do advanced stuff and the more you understand why it doesn't work in all bowsers when the code is valid, the more you will want browsers to follow the standards.

Currently there are 4 major versions of the web, because of "intuitive parsers", instead of just the one.

The web in Internet Explorer.
The web in Firefox.
The web in Opera.
The web in Safari.

And really the display problems are the least of our worries, standards on how the page opperates are even worse and much harder to work around.
I answer rhetorical questions for my own enjoyment.
Gerry
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5727
Joined: 12/04/01, 12:00 am
Location: Perth, Western Australia

Postby SOD » 11/12/06, 3:35 pm

I have developed a fault tolerant approach when developing for the web. Without a reference parser you have to develop a 'middle ground' of markup that works
in all browsers with little difference.

Gerry: You are right about the parsers. They look like they are pretty much the same but under the bonnet they are radically different. They are built by imperatives an HTML
document is declarative.

We are at the crossroads of Highway 69. Do we begin to understand information structure as declarative or do we continue to build gadgets that do not respect this understanding? Even in HTML it is all about the data model.
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Postby Pogrelz » 11/12/06, 9:23 pm

Gerry wrote:Currently there are 4 major versions of the web, because of "intuitive parsers", instead of just the one.

[...]

And really the display problems are the least of our worries, standards on how the page opperates are even worse and much harder to work around.

You really can't place the blame for there being multiple "versions" of the web on parsers being forgiving. That has more to do with an incomplete/incorrect implementation of the HTML specification.

And, as web pages are a very visually-focused medium, I'd rather we get the display problems worked out before we move on to making sure we can have neat spinning bars and boxes that jump out of the page at you.

So, getting back to the discussion, I'm glad there's a certain level of forgiveness when I write a page. If I forget to close my image tag when I write a quick blog post, I'd rather the page show up fine than completely "crashing." If some guy messes up when he writes a link in my comment box, I don't want that to "crash" my page either. If the browser can figure out what he was trying to link to, then it's even better. I could go on forever about these little things, but my opinion is the best way to deal with them is being a little intelligent about what the user probably wanted.

The web is a place that changes very quickly, very often and I think that uber-strict parsers would hinder that. Maybe I'm pathetic. Maybe I'm a 3rd grader. Or... both at the same time?! :)

disclaimer: IE often takes this too far.
....
Pogrelz
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 1093
Joined: 03/03/02, 12:00 am

Postby SOD » 11/13/06, 9:41 am

Actually when you forget to close a tag that tag is not closed by the parser. It is open until it reaches the next block element ending tag. Being that an img element has the inline property by default.

That is not intelligent.

So the only thing saving you is the block element. Which brings up one of the values of XML. It removes the concept of an inline element and replaces it with child elements. With XML all elements have parity. This is a primary difference.

Again, XML does not allow malformed markup. Data has to be accountable for itself. It IS the source, the beginning of the application.

The big software guys have failed to provide a user metaphor that meets these needs. But that takes
development dollars and who wants to do that when
browsers are FREE!!!! (as in online porn)
It is better to be here than there - SOD
SOD
BIG GIANT HEAD I Get Free Beer
BIG GIANT HEAD I Get Free Beer
 
Posts: 5284
Joined: 12/06/01, 12:00 am
Location: here and there

Next

Return to Own The Machine

Who is online

Users browsing this forum: Majestic-12 [Bot] and 3 guests

cron