Stress – This news is from my previous group at Microsoft, and this is speculation about my current group. Both groups are excellent, but both are very different. In an internal, product-oriented group, you have to argue ferociously with everyone to make sure the right thing happens. At least, this is what I do now, so if that is incorrect I will be unemployed soon. I have been arguing for a week straight now, at the same time as I am suffering from a replica of my daughter’s first cold. I believe it will be worth it, when things finally get done the right way. Jack Handy always said, “When you die, if you get a choice between going to regular heaven or
pie heaven, choose pie heaven. It might be a trick, but if it’s not,
Hailstorm? – Dave Winer slams hailstorm today, and Meta Group has a fairly level-headed discussion of challenges to hailstorm as well. These are good comments, and I hope people at Microsoft incorporate this advice when we actually get around to deciding what hailstorm really is. (joke, for the humor-impaired)
Hierarchy of Trust – Dan Gillmor bemoans the fact that the Internet has lots of anonymous people who spread misinformation. But maybe he is judging too harshly, though. For example, I have suspected Dan of spreading misinformation at times, and I do not blame the Internet or even him for that. The more people we communicate with, the more we have to be fuzzy about how much we are willing to trust. This is no different than being in a library and overhearing someone tell his friend about “this great investment opportunity”. The Internet simply lets us communicate with more people, so we have more trust decisions to make. But by the same token, the Internet makes it more difficult for us to have deep communications with everyone we meet, and online relationships tend to be more transient. In fact, as our lives increasingly are besieged by interruptions to our time, hundreds of e-mails a day, instant messages, and so on — we automatically erect attentional barriers that greatly reduce the impact these communications. At the same time, the base of opinions for those willing to seek conflicting information (as opposed to hanging out in online communities that amplify their own biases) is much greater. Dan makes a point about sites that allow comments to be posted, but ironically fails to include a link to allow readers to post comments to his article. Clearly, sites that allow dissenting opinions to be posted will be somewhat more trustworthy, but like any community, an open mike can serve to amplify the biases of that community as well. On the Internet, as in life, the only way to trust anything is to take personal responsibility for seeking out the most compelling conflicting views that you can find. And even then, keep it fuzzy. In the paper, “Making a Semantic Web,” I talk about some ways that people can universally add commentary or trust recommendations to other people’s published works. It is quite easy to imagine how you could use a generic data-mining “clustering” algorithm to have the system intelligently show you only opinions from those people who had similar views to yours. Even more interesting, however, would be to periodically have the system show you the world through the eyes of the people who were clustered very differently from you. Why doesn’t someone do this? Interestingly, Novell’s former CEO is taking over CEO position at Google, so maybe something will happen there. Novell understood metadata (at least, Craig Burton did), and Google is currently the dominant metadata supplier. And despite the fact that Sergey Brin and Larry Page seem to be fairly anti-Microsoft, they are not stupid, and I would love to see them succeed at expanding the uses of metadata on the Internet (somebody needs to do it, and I am too busy).
Rationalize – One of the oft-cited reasons for working at Microsoft is that “you get to work with lots of smart people.” From my experience, it is true: for an organization of this size, the density of hyper-creative and smart people is unmatched. If you ever get tired of always having to be the smartest one in your office, Microsoft is the cure. However, this privilege of being surrounded by a bunch of other ruthlessly intelligent and motivated people has its drawbacks. One of them is that people here tend to be really good at rationalizing whatever they want to do. You’ve heard the saying, “people make decisions emotionally first, then they make up reasons to justify their emotional decision.” If you’ve ever seen how insanely clever the average cigarette smoker can be when rationalizing his smoking, just imagine a bunch of smokers with 160 IQs in the same room all trying to rationalize different things. Some of the designs people come up with are so insanely complex that only a person from Microsoft could dream them up. And when asked why they didn’t use a simpler design, they can invent reasons why this particular problem was SO unique that it deserved such drastic measures. We hire people specifically for creativity, so it’s no surprise that people are so creative in defending their designs. This is probably why the culture at Microsoft is so confrontational. There is just no way to get to the bottom of a disagreement in time to ship unless you lock everyone in a room and “discuss” for a few hours. Of course, if you let the guy write his entire data loader in Perl instead of running through IRowsetFastload, he’s going to be more motivated to prove himself right, and maybe that is a better outcome than having him use the most appropriate technique. And no developer wants to typecast himself by only doing easy, boring stuff. In fact, trade-offs like this happen all the time. On the whole, I would say that Microsoft is orders of magnitude more pragmatic about doing things the simplest way possible than the recent Open/FSF “movements” (I had an OSS advocate recently tell me that TeX was a suitable document processing software for most consumers.) On the other hand, Microsoft tends to do things the hard way a bit more than most of our dot-com customers. And most Microsoft stuff is way more complicated than the work of Postel, Kantor, and the others who I feel best represent the true roots of any collaborative development movement.
India Sneezes – Madhu Menon discusses the rather cold way that CNET India fired their entire staff this week. This was the same day that I remarked on the strange way that CNET USA ran a sort of emotional dirge for employees laid-off at San Jose Mercury News. Maybe Jai Singh ran this article to divert attention from the firings taking place within his own organization? One interesting note about both the SJ Mercury and CNET is the apparent high premium that both organizations placed on being a “special place to work”, and “a family; a community”. What ever happened to being the best and shaming your rivals with the accuracy of your reporting? But despite talk in India of H1-B visas drying up, the demand for good tech people will stay very high, and the sort of people who feel threatened by smart foreign workers are not the sort who hold much sway in development organizations. News is that Rediff will be taking over another portal company, so things are still moving.
Granny Cams – CNN is running an article about a movement to put surveillance cameras in nursing homes to protect patients from abuse by staff. Some nursing homes are protesting, saying that it will be harder to hire employees. This begs the question: why would the sort of employees who work at nursing homes (allegedly) not want to have their actions monitored? And are nursing home staff really that scarce that *anyone* can get a job? Our demographic in the U.S. is quickly shifting toward a population that is top-heavy with elderly. The elderly will be demanding increasingly more services at the same time that the working population shrinks. We’re going to have to do more work with less people, and that means software will become even more fundamental. The problem is, we’ve already got severe shortages of people that can do software, and this industry is not doing the best job of creating software that doesn’t need people.
Glass Houses – At Microsoft we have a bunch of servers from which we can watch on-demand recordings of various conferences, presentations, and so on. This makes it convenient to see what things a visiting researcher said, share lessons-learned from major projects, and so on. For example, the Office team has given some talks about running software development projects that are still useful years later. One topic that has appeared in recent years is running a software service. Just like everyone else in the industry, Microsoft has been learning how to run online “mega services”. Some people will recognize that the telephone companies and cable television companies have already got a lot of experience providing services that are based on software. But even telephone and cable service is fairly unreliable. The idea that people see damage and “route around it” is about as Internet as it gets. “My DSL line doesn’t work? Oh well, I’ll just get some food and it will be working when I get back.” But the really disturbing question is why there is so much damage to route around in the first place? Why do our DSL, Cable, Telephone, eBay, MSN Messenger, and other services break so often? We all know at least one crufty old-timer who will make the claim that his operating system or tool is the only way to build a service that runs. But software is just one small part of the puzzle, and no software is free from association with high-profile outages. Hardware is another small part of the equation. The rest comes down to planning, discipline, and organization. So why do we suck so bad at this? And what can companies do to improve?
Robert Scoble – Speaking of Hailstorm and his Train Simulator Fan Site. “HailStorm is cool (our Train Simulator Fan Site, when it’s up, is very cool and is free) but there are some horrendous risks with a centralized system. Today I’m living in a HailStorm nightmare. It’s a free nightmare, but a nightmare nonetheless. If Microsoft charges for HailStorm services, will they be any more reliable than MSN Communities is right now? We’ll see, but the track record isn’t good.”
But still signifyin’ – Now the opinions about Hailstorm are flying. The whole discussion of “secret APIs” is particularly funny. I remember about two years ago, my brother and I decided we wanted to use the MSN Messenger protocol for P2P communications. It took us about 40 minutes to look through the type library and figure out how to do it (with no inside “special” knowledge of Messenger required). I wondered for two years why nobody was using this as a platform. Now I know why — apparently people have been convinced that the only software tools that are customizable must be C++ source code that compiles under gcc. The sheer simplicity of a language-neutral type library throws people into cognitive arrest.
No Money in Signifyin’ – Today the Mercury News announced that they would be laying off people, and the Editor resigned in emotional protest. Said some employees of the layoffs, “They will wreck the newsroom and ruin the paper’s ability to cover the community”. Perhaps to find company in misery, the Mercury is today running a story about all of the other journals that are hurting from the loss of dot-com revenue. But I am having trouble how these people got themselves confused with journalists. Miguel de Icaza wrote me regarding my recent piece on gratuitous linkage, thinking that I was ranting against Gnome. Actually, the post is mostly about fluffy filler that poses as content that comes from people posing as journalists. I can save a more reasoned analysis of Gnome for another day. It seems clear to me that most of these sites were about entertainment, not news. The public wanted to be titillated with stories of instant riches and flamboyant lifestyles, and these journalists provided the stories, with the flimsiest basis in “news”. As long as the journalists kept dazzling the minds of baby-boomers intent on making up lost time in their retirement investing, the baby-boomer retirement dollars kept flowing to the dot-coms. Since none of these companies were actually making any money, they needed to sell the dream of revenues. As long as the storytellers could keep selling the dreams, the dot-coms were smart to invest in the storytellers. And this notion of news.com running a story about another Editor at a different tech paper — has the smog from Hollywood gone to Silicon Valley’s head? I’m having visions of awards ceremonies put on by celebrities to award the celebrities who have done the best job of running awards ceremonies for celebrities … or something.