If web design is a lemon market, I guess that makes us all sour
| SitePoint Sponsor |
If web design is a lemon market, I guess that makes us all sour
Yeah, stupid, right? The sole and entire purpose of licensing is to restrict entry, and thus competition.Originally Posted by Stormrider
First, read what I wrote. People go to doctors and hospitals, and something done there or done wrongly, not their complaint, kills them.As for the doctors/hospitals thing, what a ridiculous comparison. Of course they have a high death rate in hospitals, they are hospitals, full of injured/ill/dying people!
Second, there was no comparison made. This was evidence that licensing of doctors, nurse, technicians, or hospitals offers no protection from shoddy work.
To what purpose? There is no way that any qualifying body will be anything other than a formalized means of restricting competition. So the guy in the next cubical, a talented amateur web designer/developer cannot take an offered job for pay without jumping through whatever "qualification" hoops that the board sets up? You can bet the hoops will get smaller and be set higher, the longer the board is in operation.The trouble is, most companies or whatever buying someone's web services has no idea what the difference between a well coded and badly coded site is, and who can offer each. With a haircut, you know if you've been ripped off and can just not go there again.
I'm not saying totalitarian rule or anything, nothing should stop hobbyists publishing sites, but I think there should be some kind of qualification for those wanting to charge for web development services, and I think accessibility/standards should be more strictly enforced for public service websites.
The next part of your comment is more desirable. It does not depend on who does the work or how the work was done; only that the work, itself, meets open, published accessibility standards.
cheers,
gary
Anyone can build a usable website. It takes a graphic
designer to make it slow, confusing, and painful to use.
Simple minded html & css demos and tutorials
It still has an effect. The problem is, some mistakes in that environment can be the difference between life and death, not so in other areas, so I understand what you are trying to say, but the bias is still there because of the type of environment it is.
The only type of competition this would restrict is the poor quality competition! Licensing / qualifications is only one of several ways to achieve this, and properly managed, it needn't be a barrier to entry in the web market. As you agreed, stricter enforcement of standards is another way to achieve all this.
That is all we were proposing, but how do you deal with people offering paid services who clearly refuse to meet open standards if there is no regulatory body? I suppose you could create a blacklist of known "serial offenders" but rather than burning peoples reputation I would prefer to have something you can join that only accepts people who meets open standards (otherwise they loose their membership), it would be recognised that people in the "club" would offer a high level of service, businesses would direct automatically in that direction (they love acredited brands) and it would encourage people to get their code up to scratch.The next part of your comment is more desirable. It does not depend on who does the work or how the work was done; only that the work, itself, meets open, published accessibility standards.
I would be a proponent of HTML acting similar to a programming language where invalid code would result in an error. That wouldn't be subjective at all. However, it would require that upon every request or if the request isn't cached that the HTML be validated. Which may be minimal processing, but it would still take time and thus result in speed decrease. Whether it would be noticeable may be dependent on a variety of thing though. However, validation of semantics wouldn't be practical because computers can't think… yet. So any attempt to implement a system where the system makes a decision needs to be driven by some form of concrete logic. However, semantics can be highly subjective and it would be almost impossible to program a system at this point in time that actual takes into account every scenario. The system would need to be capable of abstract decision making to do this.
XHTML follows this trend (to some extent) that malformed code will result in a YSOD (yellow screen of death) exclaiming an error in the document. And many people on here (*glance at tommy*) were opposed to the very nature of validation in such a method (though I come from a programmer background, I like prerender debugging). But the problem with your idea is that accessibility cannot be tested through code alone, much of accessibility is visual or in such a way that no validator (unless it had some terminator / skynet style AI) would be able to determine if the code met WCAG, WAI, PAS78, 508 or the other accessiblity recommendations and laws.
The numbers I quoted are from a peer reviewed journal, Journal of the American Medical Association. Every effort was made to eliminate just such bias. Any bias comes from being in a venue that can kill you. Try actually reading my post.
No, it will also eliminate competition from those who are unwilling or unable to go through the cost of certification. Properly managed? Every new bureaucracy shouts from the rooftops that they won't make the same errors every other bureaucracy throughout history has made. But they will. You can take that to the bank.The only type of competition this would restrict is the poor quality competition! Licensing / qualifications is only one of several ways to achieve this, and properly managed, it needn't be a barrier to entry in the web market. As you agreed, stricter enforcement of standards is another way to achieve all this.
If there is to be regulation, regulate only the result, not the who or how. For example: Consider a man who knows how to build sewer lines. He knows and can do all that is required to make a drain system that meets all building codes. He also knows diddley-squat about any other area of plumbing. He can't get the sewer line job because he's not a licensed plumber, and he can't be a licensed plumber because he knows only unpressurized systems. The plumber's guild has ensured there is no competition from that guy.
The trouble with end result regulation is what we would call scope creep. Every bureaucrat seeks to justify his existence by micro-managing his little piece of the action; or more likely, it's to increase his power in the hierarchy.
gary
Anyone can build a usable website. It takes a graphic
designer to make it slow, confusing, and painful to use.
Simple minded html & css demos and tutorials
oddz ++ that's my dream. No, that's my fantasy, since we know it'll never happen. Tommy would say, "with good reason"... but me, I'll take it.
Gary, it's not that you don't have a point, but I think there's value in at least certification (utopian certification of course, lawlz). I mean, if just all these amateurs just knew some basic code. Man, that would fix so much. I went through the whole certification and licensing for taking X-rays (there's a test for CT but it's not required to run the machine, depends on the hospital it seems), and yeah, there were good techs and bad techs but every one of them knew at least the basics of health care and the basics of the machinery and the basics of what's an acceptable image. You're more than useless without knowing those things. I would never want to work with someone who didn't know those things.
What got stupid was indeed the bureaucracy, the things one supposedly needed to know but as we were told by our instructors, "you only need to know it for the test". That right there is an obvious sign of something wrong. Me, I went into the field because the x-ray physics were cool and interesting. But you don't need to know x-ray physics to be a good tech. In fact, it makes absolutely no difference. And yeah, I always thought I could teach a smart 8-year-old how to take decent pictures, at least the static ones, but that doesn't mean I want him to do it (unless the zombies attack and there's a shortage of techs).
A certification or a license should exist not for the purpose of restricting competition, but to show that someone actually bothered to learn the very stupid basics of how to do something. I want whoever does my hair (if I actually ever had my hair done lawlz) to know what happens if she mixes two chemicals together on my head. I don't need her to be a chemist but she needs to prove, with a test or whatever, that she knows what's safe and what isn't and why.
I don't want the burns on my head to be the proof that she didn't learn the basics. Someone who hasn't learned the basics shouldn't be able to get that certification-- if she can, then the certification is sh*t and worthless.
Yeah, the last thing we need is a worthless certification, but the problem here is, businesses and clients need something, something impartial or 3rd party, that can tell them what the quality is of something (a car, a hairdresser, a web site). So they don't have to rely on just looks, like a non-mechanic does with a used car.
Now, this doesn't have to be certification... if we're willing to let there be some victims in the beginning, a consumentenbond (group of reviewers to say how good or bad a product is based on how well that product worked mixed in with technical knowledge-based reviews) can also do the trick. Other than word-of-mouth and advertising, I've never seen anyone do a real rating of average Joe websites. There's the wcag and people rate for accessibility like over at accesssites, but something a little broader might be nice.
This way, any Joe schmuck can have a go at it, and his customers can choose the Fiat if they can't buy the Mazda, but at least they'll know that they're getting a Fiat. (excuses to any Fiat fans out there)
I agree, yet again, with Oddz, Alex Dawson and Stomme Poes, and I'm assuming they, like I, come from a programming background.
Funny I should say that, because I actually learned HTML and CSS first. Then I learned PHP, Java and C# before returning to learning how to write HTML the right way - so technically my REAL knowledge of markup came after the programming.
And it's definitely the programming which helped. When writing an application, at first you always make newbish mistakes because you ARE one. There are two things which bring you upto speed - learning new tricks and refining the ones you know. The main cause of refining is error debugging - and when programming you KNOW about it because of the descriptive error message displayed.
That's one thing we not only learn from, but depend on initially. So when coming back into the world markup where so many people are doing things wrong because they don't know any better - it's just like coming from crimeless (yet secure) paradise to a crime-filled city with no regulation.
Of course, strict compliance may not be for most web designers but at least that might give an insight to those who disagree with some views posted here.
Jake Arkinstall
"Sometimes you don't need to reinvent the wheel;
Sometimes its enough to make that wheel more rounded"-Molona
One of my largest concerns is competent print designers who get into web without thinking they have a obligation to learn at least very basic HTML and CSS. I took one web class in college and the professor barely knew HTML relying on Dreamweaver almost any step of the way. I'm sure its like this in many web design courses in graphic design departments across the nation. The focus becomes the design and only the design. That needs to stop in my opinion. Design is important, but the technical aspects are just as important. A college level design course would require that a logo be vector so why isn't it encouraged that a web site should be technically correct – because its "coding"? Give me a break… if you don't like "coding" you shouldn't be practicing web design period.
I have a programming background, too. Worked as a professional programmer for 14 years. That's what makes me realise that programming is something entirely different from content publishing.
Birnam wood is come to Dunsinane
That's very true. Content publishing and programming are very different - though it wouldn't be wrong to argue that content and markup are different too.
I think wysiwyg editors that are standards compliant are great - for content. I can't and won't take any designer seriously if they use a wysiwyg editor though. If you want to design a site, learn the standards - if you want to publish content, learn to write content.
But doing them both, you should know how to do them both correctly or hire someone who can.
Jake Arkinstall
"Sometimes you don't need to reinvent the wheel;
Sometimes its enough to make that wheel more rounded"-Molona
The HTML5 spec allows this behavior for parse errors.
http://www.whatwg.org/specs/web-apps...ml#parse-error
Simon Pieters
Now that I agree with
Can we expect all/most browsers to actually pay attention to this?
Jake Arkinstall
"Sometimes you don't need to reinvent the wheel;
Sometimes its enough to make that wheel more rounded"-Molona
Pay attention? Yes. Can we expect any browser to abort parsing upon a syntax error by default? No. Doing so would literally make more than 95% of the Web not render and would thus nullify the browser's market share.
Can we expect other user agents to do this? Yes. Validator.nu does this for certain errors (in order to have streaming output without buffering). See e.g. http://validator.nu/?doc=data%3Atext...53Ctable%253Ex
Simon Pieters
I come from a programming background as well (though I too got into programming from web development, which sorta makes sense I guess!) but this post has me rather confused. What does design have to do with the technical side? Of course they're both equally important, but they're also very distinct with not a lot of overlap.
The focus of the design should of course be the design and only the design. The designer should have some knowledge of the web to know what's possible and whats not, to account for things like resizing browsers, etc. but that's about where the designers role ends. It's then up to the developer to make the design a reality... or am I mixed up in my definitions again?
(it wouldn't be the first time)
Actually, markup and design are different concepts for me.
I keep my HTML as basic and readable as possible, without thinking about design, e.g. a typical body for me is:
No design incorporated into it at all, just plain markup. The design is added in the CSS - which, again, isn't HTML.PHP Code:<body>
<h1>Site Title</h1>
<p class="slogan">slogan</p>
<ul class="menu">
<li><a href="#">#</a></li> <!-- etc -->
</ul>
<div class="main area">
<h2>Content Title</h2>
<p>Content info...</p>
</div>
</body>
So, design may be design but markup is markup - if it isn't marked-up correctly, how can one call it markup?
Jake Arkinstall
"Sometimes you don't need to reinvent the wheel;
Sometimes its enough to make that wheel more rounded"-Molona
@Karpie
Design determines how the page will look, thus determining how the html is written as well as what 'functionality' will be carried out backend.
*Note Usually with strict 'wire frames' this gives the designer little to no room for error
*Even using the content-out approach poor graphic placement can cause many headaches for the front-ender
So when a designer has experience of what can or can't be done and knows a little of the tech. side, then the entire team will benefit.
@arkinstall
It looks like you follow the content-out approach which is nice.
Absolute positioning will be used more with this approach.
Why something as ghastly as absolute positioning?
I always follow the margin/padding technique; Absolute positioning has way too many flaws.
Jake Arkinstall
"Sometimes you don't need to reinvent the wheel;
Sometimes its enough to make that wheel more rounded"-Molona
Ghastly, not quite
If you are truly marking up content using the content-out approach then position absolute will come into play, e.g:
*NOTE - These can be placed wherever you like on the page.Code HTML4Strict:<div id="navigation"> <ul id="primary-navigation"> <li><a href="#">...</a></li> <li><a href="#">...</a></li> <li><a href="#">...</a></li> </ul> <ul id="secondary-navigation"> <li><a href="#">...</a></li> <li><a href="#">...</a></li> <li><a href="#">...</a></li> </ul> <ul id="tertiary-navigation"> <li><a href="#">...</a></li> <li><a href="#">...</a></li> <li><a href="#">...</a></li> </ul> </div>
I feel floating elements on the whole is the reason for 'unnecessary' markup.
This way you are using divisions how they are supposed to be used and not for presentational purposes. You are encapsulating related content into a container, thus adding more semantics.
Also for accessibility you will find yourself using absolute positioning
Last edited by cooper.semantics; Jun 3, 2009 at 19:50.
I'm curious, which language(s)?
I don't come from a programming background— I'm working towards one, lawlz. But several people on several forums have said, coming from programming to markup, that many of the rules of programming can and should be applied to markup and CSS.
Chicken and egg? The reason 95% of the web wouldn't render is because they are rendering now with errors. The browser who wants to do the Right Thing and Die With Errors can't, because it's in the fisherman's dilemma— everyone wants to fish as many fish out of the sea as possible, but everyone knows the more fish they haul out of the sea, the less fish there will be due to overfishing. It's in every fisherman's interest to fish LESS but the one guy who does it simply becomes the sole loser. It only works if everyone stops fishing or fishes less. That is in the community interest, and against (in the short term) the individual (browser's) interest, though in the long term in everyone's interest, community and individual.
I don't believe Die With Errors can or should stop Joe from blogging. Joe shouldn't have anything to do with the code. Joe should never touch it. It should Die With Errors and Joe can do what we do with our cars— take them in to a shop. We don't have to be automechanics to drive, and Joe shouldn't be touching his bloggity blog-blog-code to write his blog.
I don't understand why Silent Errors should exist "because authors don't know how to write code." I don't speak French, what's the point there? Authors shouldn't know how to write code. Coders should. The web should be able to be free, free as in beer, speech, and errors. Error-free.
How does this stop web democracy again? Writing is writing and coding is coding and they are as different as repairing/building (a vehicle) and driving. Cars shouldn't be broken and driving around with Silent Errors just because "drivers aren't mechanics". Someone may have a better analogy but this argument that the web must remain crap because of "authors" is not one I can (morally) accept.
Of course, like any old prude I will accept it based on "that's life" just like grandma has to accept girls walking around wearing Daisy Dukes. Lawlz.
Yes. But on the other hand, if everyone were Draconian from the start, HTML probably wouldn't be as successful as it was.
Right.
I don't think it's the Right Thing to do.
If all browser vendors lined up their release dates and released new versions of their browsers that Died Upon Error, then users would just simply refuse to upgrade since their old browser Just Works and the new ones don't. I don't think that the fishing story is a working analogy here.
Simon Pieters
Yes, mostly I think because of the unmaintained/unfixable sites out there. So it couldn't just be vendors-- the desired effect is to get web maintainers to fix the errors in the first place. But we all know plenty of orphaned pages out there with useful knowledge. And then all the Yahoos out there with piles of errors it would take a team of intrepid spelunkers months to crawl through all that spaghetti to find and fix the errors....then users would just simply refuse to upgrade since their old browser Just Works and the new ones don't.
A dream and a fantasy, I already know. I also want that car that gets 100 miles to a gallon. Was it built? Quite a few times apparently, but various other reasons (and at least one conspiracy) have kept them out of the market.
While I've got my wishlist out, I'd also like a pony and a CSS parent selector, and Perl6 by Christmas. : )
Off Topic:
Not intended, though I saw it. Sole is correct, and I considered "single" but wth. I'll run with it. : )
Posting Permissions
Bookmarks