Today was my first day at Google, and a full day too. Orientation started at 8am, and I just got home (it's a little past 8pm) because after orientation, I was in a couple of video conferences with a team in Japan.

I've joined what's called the Japanese Product Focus Team, which is a group of engineers based in Mountain View who work on various products to add features or address issues specific to the Japanese market (as the name would imply). My first specific task relates to geo/maps/search, and sounds quite challenging (ok, very challenging), but also has a pretty significant impact, so I'm pretty excited.

I can't say much after just a day, but it's very clear to me that Google is a very different company to the last one I worked for, and mostly in good ways. Suffice to say that I actually came out more excited to be there than when I walked in.

Tomorrow's my first day on my new job at Google, but before that, I'd like to share some lessons I learned during my 3 years at Yahoo!. The most important lessons I learned, in my opinion, weren't about specific products or features, or even about programming or technology, per se, but more about people and how they go about their business in an organization.

In the interest of time and brevity, I'll just put it in list form, even though this post has, at various times, been much more verbose. Also, standard disclaimer: my own personal opinion, I do not represent anyone else, etc.

• Primacy of great teams. Great products are built by great teams, comprised of great individuals, bonded by excellent teamwork and collaboration. You've probably heard that before, but let me phrase it a little differently: Without great individuals and excellent teamwork, you can't build a great product. I once pointed out to a manager how a couple of sub-par programmers on a team were dragging the rest of the team and the product down with them, but was told "nah, they can do it, they'll just take longer." It turned out they did take longer, which meant it took that much longer for people to realize that what they were building was useless and had to be rebuilt. I learned that talent isn't a sliding scale, but actually more binary; good people can do what bad people can't, and great people can do what good people can't. Teamwork, on the other hand, is less rigid. Sure, you occasionally have people who are asses or difficult to work with, but as long as you have good cultural norms, management, processes, guidelines, communication, etc, it's possible to create good collaboration (some of the later points will related to this).
• Be critical of structure. We live in a competitive world, and most of us want to win. Even if you don't want to win, your investors, shareholders, or bosses will demand that you do better than before, and do better by larger margins. To do better (either than yourself or others), you usually have to do things differently. Of course, we can't do everything differently all the time because that'd be too chaotic, so we impose structure everywhere: named software development processes, organizational structures, guidelines, standard interview questions, where you go for drinks. But structure leads to sameness, so to be different and do better, examine your structures. Do we really need a meeting every morning? Do we really need six product managers? Do we really need 3 layers of VPs? Be critical, and change things until things get better.
• Risk aversion is death. This is closely related to my previous point, so I won't say too much. I said you have to change to do better, but change also incurs risk. People usually are afraid of risk because they fear they might end up worse off. But risk aversion leads to stagnation, or inability to react, change, do better, and you end up doing worse. Ironic isn't it? In a world where staying still isn't an option, you can't be afraid of taking risks.
• Cheerleading isn't leading. And I'm not talking about George W Bush. I've seen plenty of leaders who applauded and congratulated unimpressive feats -- like shipping a crappy product two months late. What are they saying? Are they saying that we've accomplished our task if we ship something, even if it sucks? Are they saying it's good we're only two months behind schedule instead of nine? Maybe they're afraid of demoralizing employees, but all the demoralized employees I saw weren't demoralized because they weren't getting enough pats on their backs from execs. They were demoralized because they weren't building products they could be proud of, or because they spent too much time in meetings, or because there was too much bureaucracy to get anything done, or because execs changed their minds and prematurely shut down the product... again. Maybe it's because I grew up to parents who made me the best person I can be by making me feel like dirt for the first 25 years of my life, but I like it when my leaders say "this is bullshit, we need to do better." I believe good leaders lead people to a better place, and you can't go to a better place if you don't recognize you're in a bad place.

I've been working on some significant changes to my personal homepage for the first time since, oh, 2003(?). You can see the new version of the top page at this secret URL.

The new page reflects the realities of the Web 2.0 era, where (unlike when I last worked on my homepage) most of my content is hosted elsewhere, and my homepage now simply aggregates it all in one place. Think of it as a portal in reverse: instead of me going out to the world, it's where the world looks into me.

Now, this isn't radical or new by any stretch of the imagination, but ironically enough, it's something I spent 3 years at Yahoo! trying to do for everybody while my own site stagnated in neglect. I first did a hack back in the fall of 2005 to make it easier to pull in external content into a Yahoo! 360 profile. Unfortunate, the PMs didn't see the value in that then. I got another chance at it last year while working on Mash, and came pretty close to realizing the idea, but Mash was put on hold a few weeks after the beta launch. The underlying module architecture did live on, however, and was used to power SearchMonkey, and was also subsumed into (and subsequently largely removed from) the Yahoo! Application Platform (which has yet to launch).

So, I finally decided it was time to give up trying to do the world a favor, and just do it for myself. Except, like they say, the third time is the charm; I managed to do it better than my previous attempts. With my homepage,I found an approach that provides both vitality and efficiency: the page aggregates content from 5 sources (for now) and updates should show up within 15 minutes, but the page also generates in 300ms (on 4 year old low-end hardware, PHP 4.3 without APC, and file-based object caching instead of memcache).

I just finished reading Better by Atul Gawande, a book examining improvements in medicine, and the stories he tells are fascinating. Military doctors who lowered death rates among injured soldiers from 25-30% in the Korean and Vietnam Wars to 10% in the current war in Iraq. Doctors and volunteers in India responsible for administering polio vaccines to millions of kids in rural areas in 3 days. Doctors who incessantly worked towards lowering in-hospital infections, simply by finding ways to get medical workers to wash their hands. A doctor who almost single handedly increased the life expectancy of children with cystic fibrosis from 3 to 40 and beyond. He shows how fields advance because certain people make a conscious effort to do better.

While I'd encourage anyone to read the book, he finishes with a list of 5 things he came up with to answer the question "How do I really matter?", which I thought were interesting and worth discussing. They are:

• Ask an unscripted question -- For Gawande, this is a way of adding a human touch to his relationships with coworkers and patients, by asking questions that go beyond the bare necessity. "This is not a forty-six-year-old male with a right inguinal hernia. This is a forty-six-year-old former mortician who hated the funeral business with a right inguinal hernia." But more generally, I think this is also a good idea because it encourages us to break the routine, try something different, and snap out of group-think.


• Don't complain -- Gawande argues that complaining merely fosters negativity, without offering any solutions. That's certainly true, and complaining should be avoided for that purpose alone, but it's not easy to just "not complain". I worked in an environment where people complained a lot. It then occurred to me that people complain when there's no avenue for remedy. So if people are complaining, it's not just an indication that things are broken, but things are so broken that people don't even know how to fix it. That's double bad.


• Count something -- He asserts that all doctors should be scientists, and therefore should count something, i.e. do research. Earlier in the book, he also talks about the Apgar score, which for the firs time quantified the state of new born infants, which then lead to quantifiable improvements in their care and thus survival rate. It's a common theme throughout the book. The first step towards improvement is to know how you're doing, then measure whether or not things are getting better or worse. Metrics are vital for that purpose.


• Write something -- Writing, in short, is a way to contribute to a larger corpus of knowledge, which the medical (and any other field) is based on. Even small ideas and data, cumulatively, could lead to large changes. In my chosen profession (software engineering), an equivalent alternative would be to contribute to Open Source projects, or release code. Writing is also a great way to organize thoughts, or look at things from a different perspective. Even as a developer, I spend a decent amount of time writing documentation, which often gives me new ideas, or allows me to find inconsistencies and unnecessary complexity in my code or design.


• Change -- In short, he says be an early adopter. Or at least be open to change. A necessary condition to adopting new ideas or trying new things, of course, is the ability to take risks. In some ways, creating an environment or mentality where risk is not only assumed, but tolerated, is the hard part. It's way too easy to be crippled by unnecessary fear of risk.

In general, it was interesting to see how criteria for improvement and success in medicine carried over to other fields. For instance, the 5 points above, I think, would be helpful in making improvements in pretty much any field.

When you take a job that can be performed equally well by either sexes and give it (or not give it) to a person based solely on their sex, that's sexism. McCain's choice of Sarah Palin as his running mate is sexist, pure and simple. She wasn't chosen because of her qualifications. She wasn't chosen because she has knowledge or experiences that complement McCain's. She was chosen because she's a woman.

Some early "reviews" of handsets based on Google's Android have been popping up. For example, this one concludes:

Android, while extremely powerful, has a less-elegant, less-user-friendly interface than the iPhone.

You know, the IBM PC and Windows 3.0 combo wasn't nearly as elegant or user friendly as Apple Macintosh computers of the day. But who won that war? I'm as big an Apple fan boy as the next, but I've been around long enough to know that the sleekist, most elegant, shiny and user-loving devices don't always win, especially when going up against commoditized, open platforms and systems.

As if to spite me (or prove my last blog entry incorrect), there's an article in the NYTimes titled Habits May be Good For You, about someone who's trying to use advertising techniques to bestow sanitary habits upon people in Africa. While that sounds all noble, here's a passage that caught my attention:

"For most of our history, we’ve sold newer and better products for habits that already existed," said Dr. Berning, the P.& G. psychologist. "But about a decade ago, we realized we needed to create new products. So we began thinking about how to create habits for products that had never existed before."

The article lists products that we didn't used to have, but have now thanks to advertising. They include wonderful life-saving (and life-enhancing, of course!) products such as sprays to make your couches smell all froofy, things to make your laundry smell like unicorns, chemicals to make your teeth glow in the dark... Clearly, advertising has done wonders for these multi-billion dollar companies. Advertising allows them to create demand. It's kind of like printing money.

This is an aspect of advertising I hadn't considered in my last post. For existing demand, I still believe advertising is unnecessary. If consumers know that they want something, they have access to tools to discover exactly which brand of that item best meets their criteria. Advertising is not necessary there.

But what about this other role advertising plays? How else would companies create new markets? In other words, how would companies make you buy crap you didn't know you needed? Well, my answer is, they shouldn't. Now, this is where you ask: "But wouldn't that stifle innovation? We wouldn't have dish washers, PCs or the internet if companies hadn't created demand for those technologies!" I'm not saying innovation should stop. Companies should continue to innovate and create new products. But instead of depending on advertising to create demand, they should let the market decide. And the market will decide based on the merits of the product, and worthy products will spread virally through social channels. We're not there yet, but I think we'll get there, sooner than advertisers think.

Nonetheless, I do want to get back to the point about companies selling shit people don't need. I think that's a problem. I think that problem exists because public corporations are funded by share holders who expect exponential growth. If demand doesn't grow exponentially (or generally fast enough), the only way to achieve exponential growth is by entering new markets. One way to enter new markets is to create them.

Now, I don't know anything about economics, so I probably have that all wrong. But let's say I'm right. The problem with a system which requires exponential growth, is that infinite exponential growth is impossible, if any dependent factor is finite. That is, you can't create infinite things out of finite things. Our economy depends on things that are finite, or grow less rapidly, such as energy and certain natural resources. I recently heard that we consume more energy per time unit than the amount of solar energy that falls on this planet. That's akin to spending more money than you make, in a world where nobody's there to lend you money. (That's not entirely true, since renewable energy sources do exist... but it won't matter if we don't tap into those sources, and unless those sources can keep up with demand.)

Anyway, I'm sure I got this all wrong. I'm sure our economy will keep growing. I'm sure smart people have figured it all out. I guess I should just be a good boy and go whiten my teeth now.

The internet is great at eliminating the middlemen. One only needs to look at the music industry to see one of the fiercest battles for survival being waged by fat lazy middlemen anywhere. The middlemen in that industry, of course, are the record companies. In an age where musicians (and their fans) can effectively market, distribute and monetize their goods (music) using online tools (many of them free), there really is no need to have record companies. Record companies used to play a role. They are now obsolete.

Advertising used to play a role. Before the internet, the only effective way to reach your potential customers was through advertising. If you launched a new product, how else would anyone find out about it? Word of mouth existed, but was ineffective, with very very low viral coefficients.

But the internet changes that. Search engines are very effective at matching demand with supply. I recently needed to buy pectin. All I had to do was type in "pectin" in Google, and one of the top results directed me straight to a manufacturer's website, where I found that the local Whole Foods stocked the item. No advertising needed. When a hot new product launches, I hear about it in blogs, and my friends talk about it, share items, post tweets, IM links, email articles. Again, no advertising needed.

Of course, there's the inconvenient truth. Advertising can't die, because advertising still makes money. The odd thing is, advertising doesn't make money because consumers demand it. I don't know of anyone who likes ads. No, advertising lives because of inertia. These companies have paid billions of dollars in advertising for decades. Why would they stop? Most of them don't know, for example, that improving your search ranking and paying top dollars for AdWords yield the same results, except the latter costs money. Surely, if it costs money it must be better (you know, the same way free Open Source software couldn't possibly be as good as multi-million dollar enterprise software). Sooner or later, the beast will die. Question is, what comes next? *shrug* I just wanted to rant against ads.

It's been a while since I've posted on this blog. Since I feel like it, here are some disorganized thoughts.

As is being widely reported, Rojo got bought by Six Apart. It seems like SA isn't interested in Rojo as a feed aggregation service, but as a platform. My immediate thought is: It's starting... I need to get FeedMuncher out there NOW!.

It's only a matter of time before someone (else) realizes that feed aggregation today is like email circa 1994, and that we all need standards that can be implemented by open platforms. I created FeedMuncher sort of as a proof of concept, to show that you can have a feed aggregation platform that provides a known set of APIs to interoperate with various clients.

I still personally use FeedMuncher on a daily basis (couldn't live without it), but between a full-time job and a full-time side project (more on that soon), I'm hardly finding enough time to sleep, much less work on another project. Maybe some day...

(As a fun side-note, I visited Rojo a little over a year ago and showed Kevin Burton what I had of FeedMuncher. He seemed to like it, although I never really talked to him much about feed aggregation as a platform...)

On a completely unrelated note, the stop motion animation project I worked on at the UofC is now on YouTube.

• Trailer 1
  
• Trailer 2
  
• Ducky at the Reg (final version)
  

Several days ago a "powerful bomb" demolished a Shiite shrine in Iraq, and I couldn't stop myself from coming up with a conspiracy theory. I'm no expert, but pictures show thick cables sticking out, which, to me, means reinforced concrete. If that's the case, we're not talking about a stick of dynamite, or even a couple of artillery shells duct taped together; we're talking hundreds of pounds of explosives. Also, the dome seems to have been levelled completely flat at the top, which seems to imply equally distributed forces, caused by a centrally located explosive device.

If my assessment is correct (likely not, but stick with me here), this raises some questions. Who would be able to carry out such an attack? And who gains from such an attack? Unfortunately, the answer to both questions would be: the US.

The former case is obvious; we have the most capable military in the world, there's not much we couldn't destroy (well, fixed targets, at least). The second case requires a bit more of a stretch. The last few days have proven the incident to possess sufficient volatility to incite significant sectarian conflict. If the US is unable (or unwilling) to quelch such conflict, it is likely to lead to a messy, bloody, full blown civil war. This would definitely be a loss to US intentions, yet it would give us an exit strategy. If the administration were to say "We tried to bring them democracy, instead, they decided to fight themselves" the US public (and I'm talking about the same public that voted this current administration into office), would likely accept that as reasonable grounds for withdrawal (besides, Vietnam was a civil war --I think people still remember that). Alternatively, and perhaps more likely, a civil war in Iraq would compel the UN and the rest of the world to become involved in Iraq, allowing the US to relinquish control and responsibility. With sufficient international cooperation, there is also some chance of peace in Iraq, thus fulfilling a US objective that is clearly currently unattainable. Either way, the US gets an out.

This past week, Yahoo! caught some flak for cooperating with the DoJ's request for search data, while Google got some good press. From what I've read, the story's been given quite a spin, so here are some thoughts (disclaimer: I work for Yahoo! but not in Search and I don't know any more than what's been reported in the media. I do, however, have not-yet-vested interest in YHOO stock prices):

• Spin 1: Yahoo! (and others) simply rolled on their backs and cooperated with Big Bro. If the official statements from Yahoo! and others are to be believed (and I think they can be believed), this was not the case. They did cooperate, yes, but only to a limited degree, and no personally identifiable data was released. I think that's an important distinction to make.
• Spin 2: It's a slippery slope from here... Um, no. In between giving the government data with no personally identifiable information, and giving them data with identifiable information lies not so much a slope, but a big cliff. If we're at the point where companies like Yahoo! and Google hand over personally identifiable data on millions of private citizens to the government, we've gone over the edge and it's time to stop talking and start packing. But unlike slipping down a slope, you know when you've fallen off a cliff.
• Spin 3: Giving the government data sets a dangerous precedence Perhaps so. But Google went all adolescent on us, and they got themselves sued by the DoJ. What happens if, on the off chance, the DoJ wins? It sets a legal precedence, which is far far more dangerous than this "Oh, but Yahoo! gave us this data once..." type precedence.
• Spin 4: Google resisted the government's request out of ideological righteousness, because Google can't do evil Granted, that was a brilliant PR move, but anyone who thinks they're doing this to be "good" might be under the influence of the Google Reality Distortion Field. Their decision was likely based on technical and business reasons (1. they have tons of data, 2. they're very secretive), at least as much as on moral reasons.

In principle, I support Yahoo's decision. By giving the DoJ limited data with no personal information, Yahoo! (and MS and AOL) managed to balance the need for government compliance with customer privacy. Seeing how Yahoo! got hammered by Wall Street partially for high operating costs early last week, a fight with the government on this one wouldn't have benefitted anyone.

Having said that, this turned out to be a huge PR loss for Yahoo! and others. I've seen a number of people comment on how they are less likely to use Yahoo! over Google now (if they ever did use Yahoo!, of course). Yes, such people are misguided in their reasoning, but ignorance happens, and Yahoo! may suffer for it in terms of decreased search traffic -> decreased ad revenues and market share -> lower stock prices.

Then again, this may not have been a clear victory for Google either. After being heroically defiant, they got sued by the DoJ, which may have caused the sharp drop in GOOG prices. And depending on how the court case ends up, they may ultimately end up giving up data after all, or possibly set a worse legal precedence. So the jury, quite literally, is still out on this one. I guess we'll be hearing more. [Tagged: yahoo search doj google ]

In Jakob Nielsen's comments about the Internet userbase exceeding one billion, he says:

The Mac, for example, already matters less than you think. Although it has a prominent role in the U.S., it's hard to refer to a company with single-digit market share as "dominant." In Asia, the Mac is practically nonexistent.

I disagree with his assessment. Apple has a noticeable presence and market share in Japan, and at one point, something like 55% of Apple's international sales were from Japan. The only reason why Apple has practically no marketshare in the rest of Asia is because most of Asia (population-wise, that means China and India) can't afford Macs. But with economies modernizing in those countries and as disposable incomes increase, I think there's a very good chance Apple will see significant growth in those markets. [Tagged: macs asia jakob+nielsen ]

A week or so after I bloged about my dream search box, Yahoo! internally released a new search feature called Open Shortcuts that was functionally equivalent to my hack (the timing, AFAIK, was completely coincidental). Well, as of yesterday, this feature has apparently become publicly available. Read more over at the Yahoo! Search blog.

Despite working at Yahoo! (though not in the search division), I admittedly still use Google as my default search engine. For me, the main reasons that I haven't been able to switch (I've tried, I swear!) are:

• Google's result are marginally better (more relevant, less spam) than Yahoo
  
• I like the weather forecast display better on G
  
• I like the stock ticker display better on G
  
• Google's calculator is better (try: "0xef to binary","e^4", "14!" --none of these work on Yahoo!)
  

Open Shortcuts is kind of cool because it does lower the barrier for conversion. I can set my default search engine to Yahoo! and selectively send queries to Google with 3 extra key strokes. Or I can send my stock symbol lookups to Yahoo! Finance (which, at the end, is usually what comes up as the #1 result on Google anyway). But then, 3 extra key strokes to lookup a 4 letter ticker symbol adds 60% more work... and that's a lot.

The other cool thing about Open Shortcuts is that it adds stickiness to search. Since the shortcuts are completely customizable, once you've set things up the way you like them and get used to it, you're likely to keep using the product. Whereas with generic non-customized search, the cost of switching is practically zero (assuming equivalent functionality and quality). [Tagged: yahoo search open+shortcuts ]

Lots of talk about the death of IE5 for Mac (well, more like 2nd death, the 1st death was when MS halted development), and the general tone seems celebratory. Now I'm anti-MS, anti-IE, yadda yadda yadda as the next, but I think people are losing context here.

*dreamy transition sequence*

It's a long time ago. Year 2000 or there abouts. MacOS X PR1 comes out to revolutionize the platform, to bring the Mac OS out of the stone ages. Guess what browser it comes with? Yup, MSIE 5. And guess what? It worked. Safari wasn't out yet, neither was Firefox. Lots of people were still using Netscape 4.7, and Netscape 6 was one slow buggy hunk of software. Frankly, for a while, MSIE 5 was the best browser on OS X.

*dreamy transition sequence*

Sure, by today's standard, IE5 is a crappy browser no matter how you look at it. But at it's moment of death, let's at least acknowledge it for what it was. At the very least, the developers who fought an uphill battle to get it out deserve it.

As has been widely reported on the interweb, the Fench Department of Culture is apparently trying to ban the publication of Free/Open Source (and presumably, open source) software source code, and I'm rather baffled. If they had decided to ban all foreign books, for example, I'd at least understand the rationale; I wouldn't agree with it, but I can atleast comprehend how foreign books can affect culture. But source code? Programming languages are a highly constrained form of expression, barely capable of conveying complex ideas, muich less ones of any cultural significance. Hell, I'd say software (in executable form) has much more impact on culture than code, but they don't seem to take issue with that... Baffling. Absolutely baffling. [Tagged: open+source french+department+of+culture ]