Care about standards? No, not exactly…

August 29, 2007

Topics: Web standards.

Recently, it seems like I’m writing a lot about web standards from an unusual perspective (for me). Rather than evangelizing techniques such as CSS and Web Standards, I’m trying to give every argument a fair discussion. I want to evangelize thinking. This is the result of a notable shift in thought: I’ve been moving in a slightly different direction in my way of approaching web development principles. I believe it’s a more sophisticated way of thinking, but I could be wrong….

On that subject, Jeff Croft came out with an interesting opinion piece today on differentiating between “standards” and “web standards”. It’s interesting, and definitely worth reading. There’s one particular comment in it, however, which particularly got me thinking:

[…] I’ll say this emphatically: I don’t care about standards. Not at all. Not even a little bit.


My first thought, of course, was less than entirely positive, and I’m not going to share it here. My second thought, however, was a bit more virtuous.

Of course Jeff doesn’t care about standards. Standards are the wrong thing to care about. He continues to state that his priorities are these:

  1. Create solutions that solve the client’s stated problems.
  2. Find ways to achieve the client’s stated goals.
  3. Find ways to solve problems the client doesn’t even know they have.
  4. Use tools that are elegant and efficient, and help me produce quality work within the client’s budget and timeframe.

And these make perfect sense. What do I care about? Is it the details of standards, as laid down by the W3C? Not exactly. My priorities are more along the lines of:

  • Accessibility.
  • Usability.
  • My client’s business success.
  • The successes of visitors accessing the site.
  • Effectively solving existing problems.
  • Anticipating future problems and preparing for them.

I use web standards as a principal tool for achieving these goals: but the means is not the first priority. There are known instances where support for standards is too weak to make them the best choice. Flash is a perfect example: <embed> is well-supported; <object> is the web standard. Sometimes standards just don’t work.

Now, I’m probably more dedicated to the principle of following standards than Jeff Croft is. He says that 95% of the time he considers standards to be the best decision; I’d probably go 99%, at least. But that probably has more to do with the projects we work on than anything else.

Regardless, the important point is made: it’s not the tool you should care about — it’s the product. I’ve written numerous articles about web standards. One common thread I know I’ve mentioned is the idea that web standards are in no way a path to making a good web site. You can follow standards slavishly and make an absolutely horrendous site. Web standards are a tool, and like any tool they can easily be misused.

It’s important to always consider the end-result of your labors to be the primary goal: not the method you’ve used to reach it. Web standards and thoughtful consideration of semantics may get you there 99% of the time, but you still need to be willing to stretch your way of thinking to encompass the methods in that remaining 1%. Best practices mean making a choice.

Have something to contribute?




« Read my Comment Policy

16 Comments to “Care about standards? No, not exactly…”

  1. I have posted about this cake metaphor over on my blog which will be up tom (AU perth time).

    http://germworks.net/blog/2007/09/18/how-to-make-the-perfect-web-cake/

    comments etc most welcome

  2. But the basics are only a small part of a cake, ultimately. There’s a vast difference between a sponge cake and a flourless torte (arguably, a flourless torte is not a cake, of course.)

    Do note, though, that I’m only arguing against a quasi-religious dedication to web standards without any consideration for other fundamental development needs: the kind of site which might provide a rigidly standards-based development by using deeply nested absolutely positioned div tags, br elements for positioning, and a complete lack of semantic elements.

    It’s important to note that web standards are ultimately only a tiny part of what makes a website functionally exceptional.

    Basically, if there is a sound and testable justification for not violating web standards which provides benefit to the user, it is an appropriate activity. Otherwise, no — it’s not acceptable.

  3. ha ha ha
    I am sure many people will not be surprised that I will be remembered on the web for my love of food 😆

    Seriously if the cake metaphor helps people understand the different areas of the web that is the main thing. What we need to do now is figer which cake ingredients are the web ingredients…

    there will alway be the basic ingredients when it come to creating a cake and they cannot vary too much else the thing will end up being something other than a cake.

    Agree totally and I think most people do

  4. I don’t think there is a perfect cake recipe (either metaphorically or actually) — instead, there are great varieties in the tastes and needs of those who will be consuming the cake.

    I agree about the varieties, but the basics must remain basically the same, regardless of tastes, consumer needs, or recipe add-ons. To again use the well-known Jermayn Parker cake metaphor (© 2007, J. Parker), there will alway be the basic ingredients when it come to creating a cake and they cannot vary too much else the thing will end up being something other than a cake.

  5. Heh. 😉

    Authorship, as it happens, is all that is legally required in order to claim copyright. Registration of copyright is only helpful in terms of being able to defend your copyright more easily in cases of legal infringement. The biggest challenge with unregistered copyright is in proving your original work!

  6. Maybe I should copyright the cake metaphor 🙂

  7. No question that web standards are a necessity: but when it comes to passion, I think that the end product should trump the means. It’s critically important to be aware of web standards; to be intimately familiar with them; and, crucially, to be cognizant of their current limitations.

    I definitely agree that if what you do is swap the salt and sugar, you’re going to end up with a lousy cake. But adding an extra pinch of salt might solve a problem.

    I don’t think there is a perfect cake recipe (either metaphorically or actually) — instead, there are great varieties in the tastes and needs of those who will be consuming the cake. You need to create a cake which is best for your consumers, not an abstractly idealized pastry.

    I’m enjoying this metaphor, by the way. Getting to use the phrase “abstractly idealized pastry” isn’t an every day occasion. 😉

  8. In my opinion, web standards are absolutely necessary to the stable future of the internet. Without standards every developer and every browser maker does his or her own thing. That does not for a good internet make. I can’t think of any industry — manufacturing, IT, you name it — that has been around for awhile and has a solid future without the development and adherence to their own industry’s standards. If everyone follows standards, we’ll (the internet industry) all be on the same page and moving forward; innovation, and compatibility will be simplified and assured. Without standards it’ll be like the Wild West. It’s like that now, but the industry should mature if given enough time. Without standards it’ll stay like the Wild West and its future will be dubious and challenging — more so than it needs to be or should be.

    The cake metaphor is a good one. The cake standards (the recipe, the protocol) are a pinch of salt to bring out flavor (yes, that’s what the salt is for in a cake), and cup of sugar to make it sweet. Without the standard we may swap those two (and probably provide a damn good argument for doing so). But, really, swap the two and what do you get? A cake that looks good but tastes like shit. I think we should all work on writing the cake recipe, enhancing the cake recipe, then sticking to it once perfected; one and all.

    Regarding too much accessibility. I agree there is such a thing and I’d like to offer this article on the subject: Avoiding Extreme Accessibility.

    Nice article, Joe. Very thought provoking.

  9. That may be true when it comes to revising a website in order to create accessibility; I’d say that when it comes to maintaining an existing level of accessibility for a new site or developing an accessible web site, that 5% is well worth while.

    I can see what you’re saying, of course: you need to spend time making certain that your web site is a quality resource, or all the accessibility in the world just means you’re making sure people can access out of date information. Not so valuable, on the whole, compared with keeping the information up to date.

    There’s a point where you have to compromise; but 5% is too high a price for me to pay.

  10. what’s “too much accessibility?, after all?

    Given that a cake recipe is about proportions, I’d say it means spending too much time looking at accessibility issues, and not enough time writing good quality content or designing an aesthetically attractive and intuitively interface.

    That’s the problem I have with one of my websites – yes, I could improve the accessibility of it further, which would help maybe 5% of the site’s users at the absolute most – but in doing so, something else would have to give, which would probably mean the content would go out of date more, which would hinder a much higher proportion of users.

  11. I do validate – I think it’s a great double check, since I do want to avoid making stupid mistakes…but I don’t worry much about validation buttons. They don’t prove anything to the average site visitor; just take up screen space that could be used to better purpose.

  12. Good stuff Joe, I’ve been thinking along those lines a bit recently too.
    I’m not often bothered about actually validating my code, I’m really not fussed about having one of those ‘valid html’ buttons, and at the end of the day, the sites going to be measured on how well it achieves it’s goals, rather than how good the code is.

    Good code (I use good in place of valid) is just one of the tools used to help make the sites I build better and more effective at what they do.

  13. Indeed – as I said, our different perspectives on standards probably relate closely to our projects. You’re creating different cakes than I am.

    Thanks for your comments, Jeff!

  14. Thanks for writing this, Joe! Glad you saw my point. 🙂

    As to the cake metaphor, I’d add a few things:

    1. Everyone’s recipie for the perfect cake is different.
    2. You may not want the same cake for, say, a wedding, as you world for, say, a three year old’s birthday party. Each cake has different requirements, and therefore may require different ingredients.
    3. Sometimes, if you don’t have a parituclar ingredient, you can get by without it (or find a substitute) and still make a perfectly serviceable cake.

  15. Heh. Actually, I love that metaphor. Tasty. 😉 I’ll take a yummy cake web site any day. Too much of any one ingredient and you could run into trouble as well.

    The metaphor struggles a bit, I guess, when you start to go that direction — what’s “too much accessibility,” after all? I guess, answering my own question, it’s including Pseudo accessibility

  16. Standards, accessibility, semantic xhtml etc are like ingredients to a cake, you leave one out and it wont taste and look as nice BUT with all in with the right amount, you will get a yummy cake. Yinder silly illustration I know, but one of the best ways I see it 🙂