It’s easy to forget that HTML, which is an extremely simple programming language, is actually just an exceedingly complex markup language. HTML was one of the original building blocks of the web, and its used by web developers to mark-up (or describe) a page with agreed-upon HTML tags that, when rendered by a browser, spits out a website.
The agreed-upon part here is pretty crucial. Agreement comes in the form of the HTML standard, a large specification that is currently under the watchful eye and guidance of the HTML Working Group, itself a part of the W3C (World Wide Web Consortium). What this means is that a bunch of folks (members of the HTMLWG) from all across the professional spectrum get together from time to time and more or less decide the future of HTML. Since 1997, when the HTMLWG was first founded, that’s just about the way things go.
But it wasn’t always that way. The HTML specification actually once lived under the auspices of the IETF, a multi-national standards body responsible for just about every major Internet specification you could think of. Inside of the IETF was a different HTML working group, mostly open to any engineer that wanted to join up.
This made the IETF group both large, and generally speaking, completely outpaced by a much more rapid parallel development of HTML that was happening in the browser world. The two largest browsers, Microsoft’s Internet Explorer and Netscape’s Navigator, were in the midst of a battle for market dominance. So in 1995, while the IETF was meticulously piecing together the second iteration of HTML which would eventually become HTML 2.0 (with very little process in two years), Netscape and Microsoft sprinted full speed ahead with new features.
As it turned out, one of the easiest ways for browsers to edge out the competition was to add in exclusive HTML tags. Microsoft had proprietary tags that would only work in IE, and Netscape had some that would only work in Navigator. Add enough cool features, browsers reasoned, and users would follow.
So then comes one night, when Netscape engineer Lou Montulli was out with a few co-workers. Before joining Netscape, Montulli helped to create Lynx, a text only browser for Unix-like operating systems. Lynx was about as basic as web browser got; it didn’t even support images. In fact, Montulli joked, the only “effect” it probably could handle was blinking text. A few engineers chuckled, and then seemingly went about their business.
But the next day when Montulli came into work, he found that someone had, in fact, implemented his blinking text feature. One of his colleagues had transformed his idea into a new HTML tag, the
blink tag. When text was added inside of a
blink element, the text itself would flash (or blink) in a loop. On. Off. On. Off.
Before long, the
blink tag made its way into the latest version of Netscape Navigator.
Not to be outdone, Microsoft echoed back a few months later with their own effect-based HTML tag:
marquee. Unlike the
blink tag, when text was placed in a
marquee tag it would scroll along the page. It is not a matter of public record whether this too was originally thought of in a bar, but one could at least hazard a guess.
At the time, there was a new breed of web professionals proudly labeled “webmasters.” Part designer, part developer and part evangelist, webmasters held enough esoteric knowledge about the web to manage, edit and create websites, both large and small. For them, the
marquee tags offered a bit of fun, used for under construction pages or website headlines.
But the tags weren’t exactly thought through. It’s not like there was a demand for this functionality, and each was added in without much research. That made them a bit too easy to abuse. As it turns out, blinking or scrolling text can be a distracting, if not irritating, feature for many users who found their eyes wandering back to a bit of text over and over. But their most egregious offense was what they did to accessibility. Because these features were launched but not tested, they caused issues for those with visual impairments, cognitive disabilities, or in the
blink tag’s case, epilepsy.
blink had a quick rise and fall in the browser world. They became known as over-the-top symptoms of the Browser Wars era. But they also revealed a major inadequacy in the HTML standards process. In all this time, the IETF HTML Working Group had slowly been working toward HTML version 2.0, which added a few new things to the HTML specification. Browsers, on the other hand, were adding new things to HTML every day, most from outside the HTML specification and with very little consensus between them.
HTML clearly needed something new.
So in November of 1995, representatives form Netscape, Microsoft, IBM, Sun and the W3C (among others) all met in Chicago to hash things out. With just enough people to get things done, this group decided that it would be best to move the HTML standards process away from the IETF and into the W3C, were a tighter and more dedicated group could guide the direction of the HTML standard. This group came to be known as the HTML Editorial Review Board.
In February of 1996, this new HTML ERB held their first meeting. The board had representation from all across the map, from browser vendors to software makers to standards advocates. And they each had a different idea of what HTML could do. What they needed was some uniform consensus. A common enemy to bring them together.
Which brings us back to
By 1996, both of these tags had fallen way out of favor. Standards advocates pointed out their accessibility failings. Designers saw them as an unnecessary visual blight. And browser makers didn’t like the inconsistency. Even Montulli, the “creator” of
blink regretted that the feature ever made its way into Netscape. Just about everyone agreed.
marquee were bad for the web.
So Netscape and Microsoft made a deal. The former would pull
blink if the latter pulled
marquee. And neither tag would make its way into the HTML standard. It was the first thing that the HTML ERB could agree upon and it set the tone for the standards process to come. With that out of the way, they were able to get to work on HTML’s more pressing issues.
The HTML Editorial Review Board was eventually tasked with taking a number of disparate and competing standards, like HTML+ and HTML 3, and merging them into a single standard that could meet the demands of browsers and users. There were a few starts and stops (HTML 3 was abandoned for instance), but in early 1997, they released HTML 3.2, the most advanced version of HTML yet. It included a whole lot of new features browsers clamored for, while still adhering to backwards compatibility and accessibility requirements.
Right before the release of HTML 3.2, the Editorial Review Board was officially renamed to the HTML Working Group. The HTMLWG would go on to release HTML 4 at the end of 1997, then reach a crossroads at HTML5 and XHTML. But its possible nothing like that ever would have happened if the standards community didn’t have something they could all rally behind. Something just like