Question 4 – What is CSS’ ceiling?

The next question from Dallas’ design panel is one from Neill Harmer:

“I keep hearing ‘we can fix that with CSS’. Do you ever see CSS getting to a point like HTML – where we are asking it to do too much? Meaning – what is CSS’ ceiling?”

Good question, and one I’ll probably have a slightly different view on 6 months from now, and again in a year, and so on… ;-)

Answer: In some ways we’re already hitting various ceilings within the current CSS specs, but it centers more to browser differences/limitations being the ‘ceiling’ (at least for me), and of course having the current working proposals in progress at the W3C nailed down so they CAN be implemented.

There are a lot of CSS3 features that I’m really interested in being able to use myself, specifically multicolumn layout and paged media (so you could even look at these being ‘ceilings’ for the time being), but it really depends on when these specs are finalized… and then implemented in the browser. That’s usually the gating factor in my opinion, cross-browser compatibility. If you follow the forward-looking draft specs at the W3C there’s plenty of cool features on the horizon we can’t get to, which is more a case of wanting shiny new toys instead of being happy with where we’re at right now. I guess you could call that a self-imposed ceiling of sorts?

There’s a pretty good list of CSS limitations on wikipedia I like to refer to as well, which sums up some of the current thinking on this subject much better than I’d be able to in a single blog post:
http://en.wikipedia.org/wiki/Cascading_Style_Sheets#CSS_limitations

Honestly, this is a pretty open (and subjective) question that would have really benefitted from a panel discussion – so feel free to pop in your own opinions/thoughts in comments as well. I’d be really interested in seeing the limitations of CSS from others’ eyes myself.

6 Responses to Question 4 – What is CSS’ ceiling?

  1. Woody says:

    Let me guess … Apollo is the answer to these limitations.

  2. Apollo’s HTML/JS runtime engine would of course still be subject to any CSS limitations you may percieve (or your project may have), so no- not really, Woody. I suppose if you wanted to build your Apollo app entirely in Flash/Flex, then sure- but that’s dodging the question, which was what are the percieved limitations/ceiling of CSS. There will still be plenty of need and demand for CSS/XHTML/JS driven *websites* once Apollo is launched.

  3. From an implementation standpoint, the overwhelimng majority of what folks need to make sites and pages (even to push envelopes) is absolutely present in CSS.The lack of uptake for CSS within enterprise applications in particular suggests that –despite great strides– one limitation is a lack of awareness about it in the development community at large. Take a look at markup for something like Oracle Discoverer, or BEA’s off the shelf portal – every portlet within the application is wrapped in a minimum of three tables.Similarly, I believe the web and HTML remain synonymous for most recreational and prosumer site builders. Working with CSS isn’t as challenging as working with XML, but the notion of abstracted presentation seriously lacks mindshare.Another limitation is the cycle time between the introduction of an approach and the support for it within web browsers and enterprise-friendly applications. Take list item drop-down menus. The current level of support for CSS in Dreamweaver is great for writing mark-up but features like Navigation Bar still use tables by default. Fireworks still spits out tag soup.Multicolumn layout in CSS 3 should be great, but workable solutions have been available to designers and developers for years.

  4. Great points all, Mychal. Not sure I’d add anything, other than note that both Dreamweaver and Fireworks will be getting some substantial updates in the deficient areas you noted. This is a critical area to us as well- although DW has spanned essentially two generations of designers there’s a lot of ‘legacy features’ that could do a more noteworthy job of supporting more current best practices.Thanks for your perspective!

  5. I think there’s an unwritten point related to your reply Scott. Each fledgling generation of designers is basically in a position to adopt Dreamweaver markup as a starting point for professional competence – it’s how I made it up some portion of the learning curve anyway.That leaves a sizeable portion of your install base who would essentially have to opt-in to best practices and modify their work habits accordingly. CSS-driven hierarchical menus might be great, but what percentage of users would necessarily want them? Likewise more CSS-support for Fireworks, which folks around here tend to use mostly for HTML emails or Proof of Concept activity.I really only wanted to acknowledge the complexity of knowing when and how to course correct the habits of your users. Cheers.

  6. ADAC says:

    Sense this was written a year ago, I’ve seen a large increase in the number of sites using CSS. I think the only major drawback in the haphazard way many browsers decide to implement it.Unlike HTML which has a set number of tags, and wasn’t really designed to handle the internet as it is now being used. CSS would be easy to add to and has a far more flexible design. I have a hard time picturing CSS having a ceiling anytime in the near future.