At the beginning of 2007 the old mash of incompatible web browsers had settled out enough I was willing to bet of developing a complex / rich web application. Before this there were some pretty impressive demos of things you could do in a web browser, but most were massively specific to a particular web browser (and even a particular version) - not suitable for an application that had to work with whatever the customer was using.

Microsoft had pretty much abandoned web browser development. That meant Internet Explorer 6 (IE6) had time to displace practically all the prior versions of IE. IE6 was … not too horrible.

Looking ahead, my guess was that for the small group / niche product I was working on, there was an approaching make-or-break point. We had to come up with a replacement for a critical (once) licensed third-party Windows application. Ideally, the replacement should be a web application, to integrate well with our existing backend and web application.

Please understand, when I say “we”, I mean “me”. (Our development group was small, and shrinking.) I had never done a complex / highly interactive web application before.

In that time, I generated a series of public examples, reflecting bits of my ongoing work.

Also understand that the Javascript frameworks (like jQuery) were at that time still very young, and not useful for my purpose.

In the end, the project took the better part of a year to finish. I wrote new code for the massively-optimal server-side C++ bulk processing application. I wrote a Java web application to interact with the browser-based user interface code. I put together a custom packaging of Apache Tomcat, to be as secure as possible for that application. (I was once a committer to Apache Tomcat, so familiar with the code.) I wrote the HTML / CSS / Javascript for the user interface that ran within the web browser.

In the end, I delivered a highly interactive web application, that was better than the Windows application on which we previously dependent.

When the vendor pulled the plug on Windows application, on which we were previously critically dependent, we could offer a better replacement. Our product survived that crisis.

Also, at the end of the project, I was exhausted. Spanning so many very different domains as an individual developer, is not fun.

After, I have not done any web application / user interface work. My skills in that area are very much out of date, and there has been much progress.

Rather pleased with the emergence of AngularJS. I tried to make a similar argument to the HTML5 working group, but without success. Lost interest, as my work was elsewhere.

Reading about how recently the web-designer community seems to be headed away from the “cascading” nature of style sheets. Bit relieved as I was not ever convinced this was a good approach. (No special insight here. I thought maybe with enough work it might make sense. Then I went off and did other things.)

HTML5 has gone from a hoped-for future, to the living present.

Could spend time coming up to current on the HTML5 / CSS3 / Javascript side, but perhaps not worthwhile. Do not know (yet) what will be my next development chapter.

Slightly amused the (generally) top-right “menu” icon, driven by smart-phone user interface concerns, has become common. Three decades later, top-level user interfaces have wrapped back to a notion very similar to what I put into a user interface style guide in the early 1980’s (when at Burroughs).