I wasn’t at quite the same level as Irving and John, but I was there for the whole process, and my days in the Internet Division were some of my happiest at IBM. My team, the Internet Technology Team, was composed of great people who knew that they were doing important work. We didn’t have to worry (much) about funding; we didn’t have to worry (much) about making the quarter’s numbers; we just had to transform IBM.
My main role during that period was to be IBM’s lead representative to the W3C. This involved a lot of politics, coordination, and tracking, which weren’t my strong suits then (and still aren’t), but I also got to do technical work.
In particular, I served on the HTML Editorial Review Board (later renamed the HTML Working Group) and the Document Object Model WG. This was in the days of the Tag Wars, and although IBM had its own web browser, Web Explorer, with its own proprietary tags, it was clear to me that IBM’s best interests would be served by getting agreement on a real standard for HTML and discouraging vendors (including us) from implementing their own tags. And that’s what happened, though not easily.
I also had the joy of becoming IBM’s porn expert for a time. That didn’t mean that I got to look for porn at work, but rather that I was asked to come up with a strategy to keep the proliferation of porn from slowing the growth of legitimate e-commerce on the Web. (Well, it was that or work on banking security solutions.) So I wrote up a quick proposal for a rating system which could be used by providers or third-party raters, and circulated it around. Eventually, IBM convinced Microsoft and Netscape to work under the auspices of the W3C to create a standard for creating rating systems, and I was heavily involved in the process, most notably as co-author of the PICS Rating Services and Rating Systems Recommendation.
The Internet Division didn’t have a long life, though — we had a perpetual “going out of business” strategy. As projects became successful, they moved into the more traditional development areas, such as Software Group. Eventually, we had colonized enough of the rest of the company that there was no need for a separate Internet Division, and it vanished from the orgcharts.
My team moved with Irving to the Linux strategy area, and I eventually left the team and returned to the Research Division.