(This is chapter 35 of my ongoing memoir of the Internet industry.)

By the start of the year 2000, I had won a new chance to prove myself at iVillage. My letter-writing battle against my former boss was never resolved, but it quickly became old news. We were both too busy, luckily for me, to fight it out to the end.

I found a new role for myself helping to improve our rickety message boards, which were sputtering along on an outdated and inefficient Perl platform called Hypernews. I played middleman between the Community and Tech departments as we reconfigured our software and beefed up our hardware. By the time the new boards launched I was calling myself the Director of Community Services.

It was funny how well the title stuck, since I had just invented it for myself and no official decision was ever made as to where I stood on the organization chart. I didn't report to Michael Rose anymore. I wasn't sure who I reported to, but in a company as crazy as iVillage this was somehow an acceptable state of affairs, and nobody really cared who reported to who. It was likely to change next week anyway.

I liked working for the Community department, because they had real work to do. I'd learned at Time Warner that the best place to be in any company is where people are busiest. Some employees like to work on easy projects so they can avoid stress, but in fact I'm pretty sure the least urgent projects tend to be the most stressful. If you're not doing anything important, you can't prove your value, and your managers can treat you any way they like. If you are working on critical tasks, you can earn respect and win the right to be left alone.

The iVillage Community team was definitely busy. Their message boards brought in 30 million page views a month, about a third of total site traffic, but with a leaner and more innovative structure than the Editorial department, which was packed with entry-level writers and editors churning out banal articles filled with celebrity gossip, diet quizzes and relationship tips day after day. I found this editorial content embarrassingly bad, but the boards were intense, freewheeling, heartbreaking, funny. This was partly due to the hard work of a small off-site staff managing nearly two thousand off-site/volunteer community leaders, all of it reporting in to the fortunately capable Susan Hahn, who dealt with community crises on an hourly basis and was grateful for my help.

I hadn't thought much about online community since my early days on rec.music.dylan and BEAT-L. Literary Kicks had no community interaction at this time, as it was made up of flat (and mostly neglected) HTML pages with no comments. I now began thinking about community in a new way. IVillage was a women's site but men also posted on the boards, and I experimented myself one day by posting to a parenting board, asking a question about a problem Elizabeth was having in high school. I created the username "asheresque" to write this post -- mysterious enough, I thought -- and by the end of the day got back about five useful and caring responses. This ecosystem really worked! After several years of nonsense projects at iVillage and Time Warner, it was a thrill to be working on something I really believed in.

Surprisingly, I was also still expected to lead the product integration of FamilyPoint, the company we'd bought for $30 million last summer, even though I'd screwed this up so badly in the fall. I would have expected Michael Rose to have taken the responsibility from me when I left his team, but he didn't seem to want it, and neither did anybody else. Our management team seemed to have soured on the whole FamilyPoint concept, which was probably the only reason I was still entrusted to lead the effort. I began to hear suggestions like "why don't we kill the whole thing?" and "let's strip it for parts".

The fact that iVillage's top managers could be so fickle -- buying a company for $30 million, then deciding it was a dud and moving on after a single stumbling start -- was a real indication that something was wrong with this management team. In fact, it was the entire dot-com industry that was suffering from attention deficit disorder in early 2000. There were too many opportunities, too many good ideas, too little follow-through. But everybody seemed to think the system was working -- after all, our stock was still trading at over $100 a share.

I made up a series of proposals about the individual components of the FamilyPoint service that we could integrate into our community package: clubs, email invitations, photo galleries, member profiles. Initially, photo galleries were the idea I liked best. During my brief stay in the Marketing department I'd learned that a good product proposal should provide evidence to prove that the product will be popular. So, after consulting with our Research department I was given some traffic metrics from other sites that had launched online photo galleries. But these metrics were not encouraging, and suggested that we'd lose money on the product. (It would be years before a company called Flickr.com found a way to make photo galleries work, and the time wasn't right in 2000). In this case, the exercise of trying to prove that the product would be popular proved that it wouldn't be, so I didn't spend much time on the Photo Gallery proposal.

Elizabeth gave me a better idea one weekend when she showed me a Wicca site she liked. As she walked me through the message boards, I noticed that she kept clicking on the personalized member profiles that contained question-answer information entered by each user. Our boards at iVillage had no member profiles, I knew our members had been asking for them, and this was one of the components of the FamilyPoint software available for our use.

This seemed like a great potential project, but I had to run the numbers to prove it made sense and get the management go-ahead. I went back to the Research department and asked for industry data on websites that ran member profiles. This time, the data was encouraging. I selected a single example and formula as the centerpiece of my proposal to the management team:

Yahoo.com's message boards got about 100 million page views a month. The member profiles linked on these boards got about 3 million page views a month, or 3% of message board traffic.

Our boards got 30 million page views a month, so if we linked profiles to our boards we could reasonably expect to get them to get a million page views a month.

This simple formula, it turned out, was good enough to win an okay. It was a Eureka moment for me. So this was what product managers did! Empowered by the management decision to move forward, I got a graphic treatment from our designer Christina Frederick, enlisted techie Murad Ahmed to customize the code, and began planning the rollout with Susan Hahn.

I was grateful to lose myself in hard work during the early months of 2000, because my state of mind since the divorce was still a wreck. I was still avoiding all my friends in Manhattan, still pretending Literary Kicks didn't exist, still locking myself up in my apartment every night to sit on my windowsill and listen to music and brood. My musical tastes lurched with one extreme to another, mirroring my moods. I was now taking a break from 1930s Broadway showtunes and listening heavily to Bruce Springsteen's Darkness on the Edge of Town, an album that every newly divorced husband and father should keep handy for those moments when he's sitting around his sorry bachelor pad feeling down. It's surprising that Bruce hadn't been divorced yet himself when he made this record; he sure got the anger and the alienation and the loneliness and the frustration right.

The divorce settlement process itself had become horrible, ugly. I had optimistically believed that Meg and I would work with a mediator to find a financial settlement that worked for both sides. Most of last year's IPO money was still in the bank, and I thought our recent good fortune would help the settlement go smoothly. Instead, it had the opposite effect. With large amounts of money involved, the process quickly became suspicious and hostile -- though, thank God, Meg and I never let the hostility get in the way of our co-parenting relationship. I felt angry at times at the choices Meg and her lawyer made, but I also knew that Meg was not a greedy or materialistic person, nor a vengeful one. I wish some of her friends and relatives hadn't given her advice to "go after me", though. I guess they were running the numbers too.

Our busy schedules with the kids helped keep us focused on what really mattered. They stayed with me in Manhattan most weekends, or sometimes I'd take them out to visit my father and stepmother or mother and stepfather on Long Island, and my siblings and various step-siblings would come by with their kids and we'd all have a great time. Back in the city, I started getting the hang of my single-father routine. Some people doubted that I'd be able to run a household by myself -- shopping, cooking, cleaning, laundry -- and I enjoyed proving them wrong.

Dealing with my kids was much easier than dealing with the crazy population of Silicon Alley in the boom months of early 2000. The first wave of the Internet industry was reaching a peak of ridiculous hype at this time. Newer and dumber dot-coms were being launched every day. I began to dread what dumb idea for a major website I'd see advertised on a giant billboard or TV commercial next. Watching the Super Bowl was surreal; it seemed every commercial was for a new website. There was a major new online fashion store called Boo.com that promised to be the Amazon.com of clothing (I was skeptical, because I'd watched iVillage's "iBaby.com" baby-products store flop). Pets.com had a sock puppet. Whoopi Goldberg was selling some new kind of digital currency called "Flooz". Gazoontite.com was for allergy sufferers (I suffered from hay fever myself, but I'd be damned before I'd buy my Allarest at a website called Gazoontite.com). Kozmo.com was a web-based urban food delivery service. Newscaster Lou Dobbs left CNN to found a science website called Space.com. Everybody seemed to be losing their minds.

I was quite surprised to hear, one day in early 2000, that Time Warner had agreed to be acquired by -- wait for it -- America Online. I wish I could have been in my old office on the day this was announced, that I could have seen the faces. Five years earlier, when we kicked off Pathfinder.com, we thought of AOL as among the weakest and most vulnerable competitors in our field. Many in the media industry were shocked by the realization that AOL had acquired Time Warner, rather than the other way around.

It was hard to keep up with all the exciting business/technology news. I was now spending more time every morning with the New York Times business section than with the arts section. I laughed when AOL bought Time Warner, but I cheered -- loudly -- when the Justice Department came down with an antitrust judgment against Microsoft. As a Microsoft-hater from the old school, I was very familiar with the dirty tricks this company played to battle competitors like Netscape, Apache and Linux, and it was very satisfying to see Bill Gates humbled and defanged. (Years later, I would finally come to respect Bill Gates after he donated most of his wealth to worthy global charities. I wonder if the Department of Justice's stunning lawsuit might have been the first step towards his eventual humanization.)

On Friday, March 10, 2000, the NASDAQ composite index reached 5132.32. There was no way to know this at the time, of course, but this was the day the Internet stock market reached its peak.

Many of us had convinced ourselves by now that this would never happen, that the average stock prices of all these Boo.com's and Gazoontite.com's and iVillage.com's would stay in the hundreds forever.

I have no idea how I spent that day, though I'm pretty sure I would have been at my desk working with Susan and the Community team testing the beta version of our member profiles.

Member profile links began showing up on our message boards in early April, and the new service was an immediate hit with our audience. We weren't sure if the community members would take to questions like "Favorite movie" and "Favorite song" and "My heroes", but the feedback was remarkably positive and in just a few days more than a third of the members of our whole gigantic community had filled out their forms.

One particular sweet spot came for me a month later when we got the April traffic reports. My proposal in March had predicted that member profiles would get a million page views a month, and they showed up as slightly over a million in the first month. I think even my co-workers who hated me were impressed.

I had somehow found the best department in the company to work with, and as a team we'd just done something right. It was a nice way for us to spend the final season before the dot-com crash began.