Snide comments on run amok

I've gotten a couple of e-mails recently based on my comments about . Now has been getting about 30 distinct hits a day, and in the course of it having been up I've gotten two e-mails pertaining to content (aside from this current brouhaha), so when all of a sudden the mail starts flowing in, obviously there's something going right or wrong.

I don't know whether it's a perverse need to fan the flames, but I figure that since the same questions are getting asked, I ought to put my explanations up for all to see.

The first set of questions are about my design style. Comments like "appallingly ugly" and "We live in a visual culture, and the medium is the message" are the way people are telling me that the visitors to their sites don't have time to read, and need to be pacified by pretty colored blinking lights.

Color Schemes

So, let's take my color scheme first. Black backgrounds aren't popular among graphics designers. Ink costs money. People aren't used to reading light on dark. Even my favorite web site author (Philip Greenspun, proprietor of the various sites under ) advocates black on white. Why then do I persist in my misguided notions?

The short answer: Every time I try to use software with a white background, my eyes end up hurting after 3 or 4 hours.

I think I can justify some reasons for this. When I work in with reflective media, such as paper, the overall lighting of an environment is fairly constant. If I look from one book to another, I'm not forcing my eyes to adapt to heavily different lighting conditions. When I work with an emissive environment, as computer monitors are, looking from a book to a monitor forces all sorts of muscles in my eyes to work, muscles that aren't used to being exercised that often (We have day, and night, two changes in a 24 hour period).

Furthermore, it's only recently that we've started to get high scan rate monitors. If you look at a white background monitor out of the corner of your eye, you can see all the strobing going on. I'm not much on physiology, but that's gotta hurt.

The only light meter I have for doing objective tests responds too quickly to get reasonable readings off the monitor, it jumps and falls as the monitor scans, so I haven't gone through the trouble necessary to get real numbers to justify my opinions. If someone else can point me towards hard data, I'd appreciate it.

This may change as the primary display media of computers changes away from the CRT, black on white on the slightly backlit LCD screen of my laptop is a heck of a lot less annoying than on my Sony Multiscan.

Medium != Message

"Do you think the web would be as popular as it is today if all anyone had was Lynx?"

No. And that would be a good thing. And my opinion on this is why I'm not rich.

There are a bunch of issues to be covered here. Let's start with user interface:

User Interface

When I wrote my first HTML browser in 1993 ("Hyper!", written in TDBS, an xBase variant for the TBBS system), HTML was a handy standard for content exchange. The idea was that we'd all be able to use a common interchange format that people could use their own browsers, customized to their working environments, to read.

My primary browsing environment nowadays happens to be Netscape. My primary environments are Unix and X based, and I haven't yet drunk enough of the Emacs Kool-Aidtm to adopt w3-mode wholeheartedly. In Netscape I assume that links are going to be underlined text, and that when I roll my mouse over a link the destination URL will show up somewhere on that bottom line of the window.

Consistent user interface, it's what made the Mac great. It's what makes many sites on the web completely unusable. Try going to the aforementioned site with JavaScript turned on. What do you get? Links are differentiated from the text by color only (but non-linked text exists in a variety of color), and when moused-over show some portion of redundant text that should really be included in the link itself.

For usability studies about the value of text in the links, check out


I feel kind of guilty about using tags, but I'm vain and don't expect that anyone but me and a few friends actually use my site. And I admit that my personal pages for a while had a table based layouts and colors that'd make the editors of Wired cringe. But I've mended my evil ways.

When Mike, and Debbie, and I, and all of our friends, Robert, Aaron, and all the others, got that first ping over that fractional T-1 line that started Chattanooga On-line , we had dreams about bringing an enlightened community to the world. I can't speak for everyone, but I know that in my naivet I hoped that when we brought free discourse and exchange of ideas to the masses we'd see an enlightened culture.

Laugh if you must, but I was once an idealist.

Of course I was wrong. What we (and so many others in the early days of the Internet Gold Rush) did was destroy the enlightened community with the babble of the masses who didn't care that they were tromping through the flowerbeds, but I still have this vain hope that somehow we can bring the world beyond being a consumer culture, one motivated by the flashing lights and bright colors of feel-good advertising, to one built on the premise of enlightened, educated equals who evaluate the merits of ideas based on the ideas themselves.

So when I see a page that puts a lot of energy into how it looks, but as specific as it gets about actually promoting the goals of the organization is "our goal is to support these core standards and encourage browser makers to do the same", I'm not only skeptical, I'm suspicious that the medium is all of the message. How about some real action plans? BOF sessions at conferences? Petitions? Specific requests? I certainly hope that anyone reading the page isn't going to throw their good name behind an effort to "encourage browser makers" without any real action plan.

Existing Organizations

The only reason that this is even an issue is that so-called "content developers" think it necessary to abuse all of the various features that browser makers have implemented. Off the top of my head, I'd bet that less than 1 in 20 tables is used for tabular data, and of those used for making side-bars (the other use I think might be legitimate) most are used incorrectly.

And tables are something that most of the browsers support well. Do we really need JavaScript onMouseOver compensation for the fact that you couldn't properly describe your links in the main text? Does anyone at all still think that was a good idea?

So rather than crying to the browser makers, how about simply making usable sites? It really isn't that hard.

Anyway, thanks for sitting through this barely coherent rant, I really do value your messages, even the ones that border on the flaming, it makes me reevaluate and better understand my position.

If you'd do me a favor, if you don't mind your message being reproduced in whole or part on my web site, say so? I've used a couple of quotes without attributions in this text in a manner which I hope is fair-use, but I'd feel more comfortable about publicly discussing and responding to messages if I know what you are and aren't comfortable with me using publicly.

Sunday, August 16th, 1998