Crunch Week

For the last six months, since sometime back in November, the developers at Zango have been working on version 10.0 of our toolbar/desktop advertising client; and we're finally getting ready to release; and I’m not happy. The issue at the moment is that we’re scheduled to release in less than two weeks – and we still haven’t quite hit Zero Bug Bounce. I’ve been watching the count of P1 bugs carefully each day for the last two weeks, and we keep hovering around 10-15. That’s got me worried. We released our internal beta yesterday, and as expected, that’s released a whole new slew of bugs, so maybe I should be grateful that today’s bug P1 count was still at 13 – but I’m not quite there yet. On the surface, I'll confess, this release isn’t a lot to look at: our toolbar (in all its various brands) will look mostly the same, and the ads we serve will look and behave mostly the same way. But there's a lot that's changed under the hood. For starters, we'll now support Vista -- which was a much larger challenge than anyone had thought -- maybe I'll throw out a blog on that one sometime. Perhaps the biggest change is that we can now cross-promote content across our different brands. In other words, it used to be that if you had Hotbar installed, and wanted to access content on Zango.com, you had to install Zango -- with the result that you had two toolbars, which looked almost identical, and pretty silly, stacked on top of each other. In addition, I'll confess, they often interfered with each other, resulting in more system instability than anyone here liked. This release will change all that: if a Hotbar user wants to watch a video on Zango.com (I recommend Daniel and the Bus), they just get access, without any prompts or anything to install. This means that we can promote our videos and games to our Hotbar and SBU users – and Hotbar and SBU to our Zango users – which will hopefully help us improve how long our users stick around. From my perspective, though, the most interesting bit has been how this release came together. Since Zango acquired Hotbar a year ago, we’ve been trying to figure out how to merge our different cultures, mostly with success, but certainly not without some hitches. As a part of the merger, we lost the product managers in Tel Aviv who had been keeping the product moving forward, and it took us a while to get those positions refilled with solid folks. In addition, it’s just hard bringing two teams together, who speak different languages, live thousands of miles apart, have mostly never met each other, and who have maybe eight hours of overlap a week (if the Bellevue folks come in early and the Tel Aviv folks stay late). But we made a special effort on this release to act as one team. Our product managers, whether in Bellevue, Montreal or Tel Aviv, now report up through a single Director of Product Management (and no, you can’t have her). We worked off a set of unified business requirements, and a unified functional spec. We moved all our code to Team Foundation Server (dumping StarTeam in Bellevue and VSS in Tel Aviv), worked off a common bug database, managed the project centrally here in Bellevue, and implemented a common, automated daily build system. The toolbar (Tel Aviv) and the search assistant (Bellevue) are still separate code bases, and don’t share much: but we do finally have a common installer for all of our brands. All of that is good. But I remain worried that we’re behind. I know that folks on both sides of the Atlantic are working overtime to get this release out. It means a lot to us: just getting Vista support should get us another 5% in daily installs. There’s not much more I can do at this point: I’ve never been qualified to write production-level C++, and it would be a mistake of the highest order to let me loose on the code base. So I alternately encourage and harangue and kibbitz, and do my best, on the whole, to let the folks that know what they’re about keep at it. Hopefully I’ll have some good news in a couple weeks.