Posted by Ed Jensen on 11/09/07 16:13

dorayme <doraymeRidThis@optusnet.com.au> wrote:
> Now one more question. What would you imagine about the
> appropriateness and quality of the tools if IE could be taken out
> of the picture?

I'm not really qualified to answer that question for several reasons.

First, I'm not really very familiar with any of the tools on the
market. The little HTML/CSS/JavaScript that I've written was created
entirely by hand.

Second, I'm not a professional web developer, so I'm not necessarily
familiar with best practices (i.e., the "right way" to do things).

That's not to say I'm entirely unfamiliar with web development. My
wife runs a small business, and I maintain a small web site for her.

I tried doing things the Ivory Tower way (i.e., Separate content and
layout!, Tables are for tabular data only!, etc.), but I found the
experience time consuming and frustrating beyond any measure of good
sense.

And I say that only in the most pragmatic sense. It's a small web
site for a small business! I need to be realistic about how much time
should be invested in it.

Yes, I committed the unforgivable sin of using tables for layout, but
at least it renders correctly in IE6/7, Firefox, Opera, Safari, and
handles text resizing correctly. It even passes W3C validation (both
HTML and CSS). As an added bonus, it even renders correctly in lynx
and links!

Any Ivory Tower types who care to do so may now inform me of my gross
incompetence and how CSS based layouts are easy and I'm a stupid dummy
and shouldn't maintain a small web site because I suck.

> In fact, just to keep it simple and isolate the
> tools business, imagine all browsers of any one type (say, visual
> browsers, screen readers, being essentially the same in respect
> to their standards and renderings). Would you guess that
> professional web authors would *still* be getting it "wrong"?

Yes, I still think they'd get it wrong, because it's so *very* much
easier to do it wrong than it is to do it right.

> Unless you have some idea of this, you might be confusing the
> quality of tools with the difficulties of coping with browser
> variation and especially IE. (There may very well be no tools
> that could ever be made to cope with browser variability).
>
> OK, now suppose you came up with a rough idea that they would
> still be getting it *too wrong* even though *less wrong*. But
> there is yet more work to be done before you can simply complain
> about the tools.

Please keep in mind I'm not complaining about web design programs.

The "tool" that I think should probably be considered broken is CSS
(for layout).

Good tools should make it as easy (as is realistic) to do the "right
thing" and hard to do the "wrong thing".

This is becoming increasingly important as more and more low quality
developers (including web developers) enter the field.

Most modern day companies have made it crystal clear that keeping
developer costs down trumps every other consideration. It would be
naive, in my opinion, to think the problem can be solved with
education, experience, or training. This is how modern day companies
think:

Educated people cost more money. Can't do that.

Experienced people cost more money. Can't do that.

Training costs more money. Can't do that.

Cheap developers. That's the ticket!

I think we can continue to expect to see the field flooded with cheap
developers. The only thing that can really fix the problem is to fix
the tool so that it's easy for the new breed of developers to do the
right thing, and hard for them to do the wrong thing.

That being said, I'm pragmatic enough to realize the problem will, in
fact, NOT be solved. There's too much inertia to overcome now, and
too little desire to fix it.

Some day, a new technology will take the world by storm and replace
HTML/CSS/JavaScript, and then (and only then) will the problem be
solved, in my opinion.

> Consider this idea of the "professional" website author. If some
> of these folks are scoring jobs on any basis other than a
> knowledge of the good use of the available tools and a good
> understanding of important website building criteria, is it the
> tools themselves that are to blame?
>
> Perhaps you might argue that if a proper accreditation system was
> implemented, there would not be enough good developers to go
> around because the tools are too tricky to get to grips with and
> few would graduate.
>
> But why? I think you have conceded that some sites are well made,
> so the tools do work in the right hands. People get paid very
> handsomely. It is an attractive profession for young people to go
> into? Perhaps the tools are not harder than many tools in many
> other professions. It is not a breeze to walk into engineering
> and to be able to design and troubleshoot control systems in a
> manufacturing plant. Not anyone can do it just like that. Nor by
> merely reading a book or two and 'having a go'. There needs to be
> a serious study of it. The tools themselves are the maths, the
> electronics, the mechanical or chemical theories and whatever is
> appropriate.
>
> You would get onto stronger grounds and be making more
> substantial insights about the inadequacy of the tools if you
> could show that they were too hard even for a sufficient number
> of educated developers to be turned out.

[Back to original message]