Moishe's Blog

Technical Interview Workshop Links

2011-10-06T11:12:00.000-07:00

I'm giving a talk/workshop about how to prepare for the sorts of interviews we conduct here at Google for engineers. Here's a list of links to outside resources that may be valuable if you're thinking of interviewing at Google, Microsoft, Amazon or other tech companies that ask coding questions.

Steve Yegge has a couple of excellent posts, one of which I mostly cribbed to put my presentation together:

MIT has course notes for the course from which I stole the title of my talk.

I wrote a fairly detailed breakdown of one of my favorite questions.

Please add more links in the comments.

Vermont 50

2010-09-29T17:35:00.000-07:00

The first time I participated in the Vermont 50, I rode my bike, and I rode my bike with lots of friends who also rode their bikes. I remember seeing the runners at the start line -- and seeing them pass me during the race -- and thinking, "those people are crazy." Riding my bike was fun but I was emphatically not a runner and the idea of running even a mile seemed ludicrous, let alone 50 at once.

I'd encountered ultra runners before, at the Leadville 100, which I also did on my bicycle, so the idea of ultra running wasn't totally foreign to me. But that idea seemed terrible.

About 5 years later, I started running short distances on trails to see if it would help me lose weight. It did, but I also enjoyed it way more than I thought I would. I started falling asleep with the rhythm of my feet hitting the trail replaying in my head, the way the rhythm of my legs turning pedals did after a long, long ride.

I ran some beautiful trails in Washington State. I ran a marathon. I (briefly) moved to Austin, TX and ran some trails there. In early 2009 I returned to Seattle and ran again in the forest that I'd missed so much, and decided that doing something longer than a marathon seemed like a great idea. So I called up my friend Mark, who I'd ridden with in the Vermont 50 way back when, and asked if he was planning to ride it that fall. He was, so I decided a great intro to ultrarunning would be to run 50K on a course I was somewhat familiar with, in terrain I loved, and also get the chance to hang out with a friend I'd known since 7th grade.

So last year, I ran the 50K. It was fantastic. It rained all day, pretty much non-stop, and the footing was terrible and my shoes weighed about 3 pounds each with mud and the mountain bikers were complaining -- but it was beautiful. The dark grey sky enhanced the colors of the trees; the sound of the rain was soothing and soft; the height of the hills was obscured by the fog. And the best part of the race was that the whole time, I knew I was going to run further than I ever had. I did. I didn't run it fast (in fact I ran it very, very slowly) but I finished and got the medal and was ecstatic. And I'm sure you all know how this goes: the next day I decided that I wanted to run the "real" race, the 50 miler, the following year.

As soon as registration opened for the 2010 Vermont 50, I signed up. Pressing that "submit" button scared the bejesus out of me, and invigorated me with by making concrete a goal that might be too hard.

I used the race as an excuse to go on long solo run/hikes through the wilderness of the Cascades. I got a little tan, and I lost a little weight, and I saw some parts of the Cascades I couldn't believe I hadn't seen after living in the Seattle area for more than 15 years. I ran up to Cutthroat Pass as the sun rose, then ran down the other side to Rainy Pass where I'd stashed some food and drink, and couldn't wait to run back up to be surrounded by the granite peaks at Cutthroat again. I ran near Salmon La Sac and lost the trail in the snow and felt like I was 11 years old again, just out in the woods playing, confident that I would find the trail eventually. I was surrounded by Bighorn Sheep in the Enchantments and had a moment where I wasn't sure what to do if sheep turned unfriendly (do they? What in the world would you do?). I ran miles and miles on my hometown trails, seeing my own footprints from the previous day, continuously finding a better line through each series of rocks. I watched my 4-year-old daughter run her first kid's race and was filled with hope that her grace and speed and joy would stay with her forever.

Then, sooner than seemed fair, summer was over, and I flew out to my parents' house in Massachusetts with my running shoes (three pair, because I still couldn't decide which to wear on race day). And then drove up to Vermont with Mark and Greg, who I'd ridden the 2001 race with, who rode in 2009, and who were riding again in 2010.

Weather on race day was perfect. It wasn't too cold at the start -- I was comfortable in shorts and a t-shirt -- but never got too hot. It was slightly overcast. The trails were in perfect shape, hardened by rain for a few days and dried by steady winds on Saturday.

The first 12 miles passed quickly, and then I ran for a bit with a couple of guys, Bruce and David. Bruce was experienced, and vocal, and David was young and talkative and enthusiastic, and I ran in the wake of their conversation for about 10 miles, through the long 7-mile stretch between aid stations, up & down some challenging hills and through some good singletrack. It was a joy to run with them, but their pace was a little faster than I was comfortable with, so I asked them to go ahead at the 23 mile aid station.

So I kept running, and then ran some more. I must admit that I don't understand how some of you write race reports full of detail and landmarks and specific events -- my mind seems incapable of recording things like this when I'm running. I remember passing the halfway point and feeling excellent. I remember passing the 32-mile mark and getting a huge boost with the thought that every footstep meant one footstep further than I'd ever traveled on foot in one day. I had a couple of low points on uphills, I got frustrated with some downhill singletrack, but mostly I wasn't thinking, I was just running, walking, drinking or eating.

I do remember getting to the 41-mile aid station and realizing I had one hour and 20 minutes to cover 6.5 miles. This seemed within my grasp, if I just kept doing what I was doing.

Unfortunately the next few miles had some really tricky downhill singletrack (at least tricky for me, since I'm not coordinated to start with, and especially not so after running 40 miles). I lost a lot of time here, and the next thing I saw was the "5 miles to go sign" at exactly 5:25 -- giving me one hour to run 5 miles, and, more to the point, 20 minutes to run 2 miles to the next aid station at mile 47.

I realized at this point that my race was over, but something interesting happened: I felt like really running. I'm not saying I was running fast or even not slow -- obviously I wasn't, because I didn't make the cutoff -- but I was running in a way that I never thought I would after so many miles. I was able to maintain my momentum up and down hills, my legs felt great, I was... well, I wasn't flying, not by any stretch of the imagination, but man I felt good, and I felt like I could run forever.

Of course, I got to the aid station after 5:45, and one of the volunteers told me I'd missed the cutoff and to climb into the van for a ride to the finish. And I had to find my friends, who were waiting for me at the finish line, and did not expect to see me walk up behind them, and tell them I'd been pulled. And I ran into Bruce and David, wearing their medals.

But the weird thing was, like running those last 2 miles, it didn't hurt. It felt okay. Of course I wish I'd finished the race, and I can't wait to run another 50 and actually cross the finish line. But the race functioned so perfectly as a catalyst -- a reason to run with friends from work, a reason to immerse myself in the wilderness, a reason to see my friends on the other side of the country, a reason to see my family, a way to force myself into that state of being that only comes with long, intense activity, and lasts for days after -- that those 3 miles just don't seem to matter.

Now I'm counting the days to next year's Vermont 50, not to redeem myself, but because I simply can't wait to do it again.

My talk at Seattle GTUG

2009-11-22T17:48:00.000-08:00

As I mentioned earlier I spoke at the Seattle Google Technology User's Group last week about the gadgets.sharedstate API and AppEngine XMPP support. The video of my talk has been posted. Enjoy!

Tic Tac Toe bot with AppEngine/XMPP and gadgets.sharedstate

2009-11-16T22:16:00.000-08:00

Add 'motactoe@appspot.com' to your roster on http://talkgadget.google.com/talkgadget/sandbox. Then, send it an IM. The bot will reply with a Tic-Tac-Toe gadget (and witty repartée).

Be forewarned: this bot is the worst Tic-Tac-Toe player in the world.

I'll publish the code once I clean it up.

Speaking tomorrow night at StartPad

2009-11-15T18:09:00.001-08:00

Hey all, I'll be giving a brief talk tomorrow night about chat gadgets and the AppEngine XMPP API.

I'll talk about how to write gadgets that are usable inside orkut chat and the Talk Sandbox, how I implemented my "game lobby" for chat gadgets, how to use the AppEngine XMPP API, and probably a little bit of history and other info about the Google Talk team.

I think the talk is full, but the waiting list is short. It'd be cool to see some (more*) familiar faces there.

*Mike Koss, who was the development manager for Outlook when I worked on it in the early/mid-90s, runs StartPad and will be there.

Directory of shared gadgets

2009-11-09T21:48:00.001-08:00

Over the weekend, I used the XMPP AppEngine API to hack together a showcase for gadgets that use the gadgets.sharedstate API that I've talked about before on this blog.

Here's the directory: http://chatgadget-directory.appspot.com/index.jsp

Basically gadget authors can list their gadgets, and users can sign up to "play" the gadget with another user. The app serves as matchmaker and message-relay -- its bot relays the XMPP messages (which are contained in stanzas) between endpoints. Doing so introduces a little more latency than Google Talk alone has, so this is much more suited to turn-based games than to anything "realtime".

You can only "play" these gadgets on the Talk Sandbox (http://talkgadget.google.com/talkgadget/sandbox) or the orkut sandbox right now.

A caveat: as I mention on the text on the site, your email address will be exposed so if you're not comfortable with that, don't use this page, or create a test account.

There's a lot still in progress here, mostly related to detecting when the other player is online (I could be smarter with a queue of players who sign up to play only when they're online), and UI tweaks and so on, but it's mostly functional and I think interesting.

Enjoy, and let me know if you have any questions!

Also: I'll be talking about XMPP/AppEngine, the gadgets.sharedstate API, and other Talk-related stuff at StartPad on Monday, November 16th. Here's the link for more info.

A nerdly confession, and a chance to get a T-Shirt

2009-10-22T08:17:00.000-07:00

I started programming computers when I was in 5th grade, and bought myself a (crappy) computer in 6th grade with money I'd saved from a paper route coupled with "matching funds" from my parents. Believe it or not, this is not the nerdly confession referenced in the title.

When I was in 8th and 9th grade, I got really excited by computer graphics and animation. Join the club, right? If you're a kid who's into computers, there's a high likelihood that you want to make pretty pictures.

Anyway, I wrote all sorts of cool (in my mind, at the time) mathematical pictures, and plotted them on my crappy little Radio Shack plotter, or animated them with some kludgy hack.

So anyway, being in 8th or 9th grade, and a giant socially-awkward nerd, I of course had unrequited crushes on any number of girls. Obviously I was in no position to impress anyone with my athletic prowess or conversational skills so in my mind my only hope was to impress them with my programming acumen. How ludicrous this is must have occurred to me because thankfully I never actually tried this.

However! If you are the 2009 version of me in 1985, and use orkut, I am working on the tools to let you do something awesome! I've been working on making it easy to embed gadgets -- that you can write yourself! -- into IM conversations. So here is my formula for wooing a girl, young nerd:

NOTE: If you're a girl, just s/girl/boy/ below. Or you know, s/girl/boy/ at your discretion, no matter your gender.

Join orkut. (you could do this all on facebook or probably most other social networks, too, but without the real-time awesomeness)
Sign up for the orkut developer sandbox.
Make friends online with a girl. This is way easier than being friends in real life!
Make sure said girl is also on the orkut developer sandbox. Hopefully she's a nerd too.
Write an awesome gadget. See the orkut chat gadgets documentation for more info.
Send this gadget to the girl you're friends with.
Impress her with the excellence of your programming skills.
If you used the gadgets.sharedstate API, interact with the girl in real time!
Figure the rest out on your own, buddy!

If you do this, PLEASE LET ME KNOW (with a link to your gadget). While supplies last, I will send you a Google T-shirt. These are way hotter than varsity jackets, trust me.

What I've been working on

2009-05-28T08:52:00.001-07:00

I've been working on a javascript API to let gadget developers use realtime communication. I'll write in more depth later (I'm in the keynote at Google I/O right now).

Here's the documentation and here's a link to the discussion group.

There are a ton of cool things this will let developers do. For now, this is a developer preview and the feature's only available on the Talk Developer Sandbox, but it will be available soon in orkut and iGoogle. We wanted to get it out to developers as soon as possible for feedback on the API and to let you get a head start writing gadgets before we launch this to the public.

At SXSW

2009-03-14T13:14:00.000-07:00

I'm back in Austin for the weekend at SXSW. I'll be at the Google both (421-423) in the exhibit hall from 4-6 today -- come by and say hi if you're in town!

Also I just added the FriendConnect "followers" gadget to my site -- try it out to follow me in reader and discover friends who also follow this site. FriendConnect's pretty cool and I hope to write more about it soon (when I'm not sitting on the floor at the convention center).

Interested in trying out some early Google tech?

2008-12-16T15:02:00.001-08:00

My first post in a very long time, and I'm shilling :)

The project I'm working on is doing a limited invite-only release, and we're looking for some javascript developers to help test it. If you enjoy writing gadgets and want a chance to try out something fun, email me or leave a comment here!

UPDATE: please leave me an email address if you comment. I won't publish these comments so no worries about spam.

New experimental version of Talk

2008-04-05T08:04:00.000-07:00

We just launched a version of Talk to Labs:

Talk Labs Edition.

Enjoy!

Chatback launched!

2008-02-25T16:51:00.000-08:00

Another cool feature of Talk just shipped. Check out the link over on the right -- if I'm online, you can chat with me, even if you don't have a Google Talk account.

However, I won't be online for a while since I'm leaving for Hawaii early tomorrow morning :)

Cool bot!

2008-02-13T07:04:00.000-08:00

A reader of this blog, Bruno Nery, sent me a link to a very cool bot he wrote called rcc. It lets you register yourself as an "expert" on various topics and routes questions about topics to the experts on them. It's an excellent example of a simple, useful tool built on top of XMPP -- I'd love to see more things like this done by third parties. Google Talk is a small team and we don't have time to do even a tiny piece of the cool stuff that XMPP enables -- and the developer community at large has many more ideas about cool stuff than we do!

Google Ski Trip

2008-02-03T08:47:00.000-08:00

On the bus ride up, the dangers of many engineers in close confines:

Tour operator/guide: "...and they are planning to install some sort of transportation device, the longest of its kind, between Whistler and Blackcomb peaks. I'm not a skier, so I don't know the name of it, tram? No, maybe it's a French word?"

anonymous voice from the back of the bus: "Trebuchet?"

Pictures!

2007-12-25T15:05:00.003-08:00

Amelia opened bushels of presents last night... here are some pics:

Christmas 2007

It's been a while

2007-12-17T18:40:00.001-08:00

And I still don't have much to say... in the meantime, enjoy this video of Amelia & Lindsey:

A housewarming present to myself

2007-04-10T07:23:00.000-07:00

Lots going on lately... we've moved again, but hopefully for the last time for a while. To celebrate, I commissioned a piece from an artist whose work I've always admired, William Schaff:

It came out better than I could've hoped.

I need to decide where to put it now... I'd love to put it in Amelia's room, but I also want to show it off a bit. It might work in our living room. Hmm... not a bad problem to have.

Cool hack

2007-03-20T09:26:00.001-07:00

If you use orkut & Google Talk & Firefox, check out this hack.

Smurf!

2007-03-14T16:45:00.000-07:00

The guys in the cube next to me just launched something pretty cool:

My gtalk account is moishel -- add me to your roster if you want.

Uh-oh

2007-02-11T20:40:00.000-08:00

I'll be a happy man if I get two shiny new belt buckles this August.

But it's gonna be a hell of a week.

Less programming, more skill?

2007-01-31T22:08:00.000-08:00

I just picked up a copy of Programmers at Work (disclosure: that's a paid Amazon link) and I've been looking forward to reading it. I've only read the Charles Simonyi chapter; I got distracted by other books and other things, but it's sitting there on my nightstand waiting.

So it was pretty cool to see Jeff Atwood quote the book on his blog entry about becoming a better programmer by not programming. I agree with the quote:

Does accumulating experience through the years necessarily make programming easier?

Bill Gates: No. I think after the first three or four years, it's pretty cast in concrete whether you're a good programmer or not. After a few more years, you may know more about managing large projects and personalities, but after three or four years, it's clear what you're going to be. There's no one at Microsoft who was just kind of mediocre for a couple of years, and then just out of the blue started optimizing everything in sight. I can talk to somebody about a program that he's written and know right away whether he's really a good programmer.

And I definitely agree with this from Jeff:

You won't-- you cannot-- become a better programmer through sheer force of programming alone. You can only complement and enhance your existing programming skills by branching out. Learn about your users. Learn about the industry. Learn about your business.

I think he limits it too much, though.

The best thing I ever did for myself was walking away from programming, as best as I could, for 4 years. In 1998, after 3.5 years at Microsoft and 5 years of high-stress programming before that, I was burned out, unsure of what I wanted to do with my life, and fed up with the work I was doing. I cashed out my stock options, put all my belongings in my friend's basement, and set out to hike, bike, and finally get that college degree I'd always wanted.

The day I quit, I started driving to LA. I hiked in Joshua Tree, then went to Canyonlands and hiked for days in the Maze, where I swear I was surrounded by the ghosts of the Anasazi. I flew back to Massachusetts, saw my parents, then took the train to northern Georgia and started hiking home on the Appalachian Trail. I didn't quite make it, but detoured to Washington DC for the 4th of July, hiked some more in Vermont, then went to Olympia for school.

At school, I took classes in forest ecology, evolutionary biology, neuroscience, geography, linguistics, and so on. I learned a bunch, despite the hippies.

Between classes and during breaks, I rode my bike in the mud of Olympia, augmented with road trips to the sandstone of Moab and the dust of Nevada. I met my wife, and during summer vacation, we rode our bikes from northern Montana to Wyoming; we saw bears, and almost got struck by lightning, and saw amazing views and played pool at every dive bar that would let a couple of smelly cyclists in.

Through all of this, I missed programming. I dreamt about it, in good ways.

I didn't miss the long hours or the stress, but I missed being surrounded by lots of really smart people, and sitting down in front of a computer to just crank out code. I did a little bit of contract work, and after a few years I joined a startup with some friends, then ended up back at Microsoft.

Now, 9 years after trying to leave programming behind, I'm working on stuff that gets me as stoked as I was when I was 11 years old and learning Logo, or 13 years old and learning to do psychedelic animation with palette switching, or 16 years old and learning assembly language. There is always something new to learn -- always a new way of looking at a problem you've known about for a while, always a new problem to discover.

But Jeff is saying something a little different than what I'm driving at, I think:

Passion for coding is a wonderful thing. But it's all too easy to mindlessly, reflexively entrench yourself deeper and deeper into a skill that you've already proven yourself more than capable at many times over. To truly become a better programmer, you have to to cultivate passion for everything else that goes on around the programming.

Of course I agree: if you have profound knowledge of the cloud of problems around programming, of the higher level problems you're trying to solve, you'll do a better job programming. This is obvious, and explains why the best programs are always the ones programmers write for themselves, eg. emacs, Google, Visicalc, etc. And it conversely explains why so many programs are so uninspired, too: if you'll never use the feature you're writing, you're not going to imbue it with brilliance, because the brilliance will never occur to you.

And beyond that, if you truly love the entire experience of programming -- not just the act of sitting in front of keyboard, but the planning, and the bug triage, and the promotion of your ideas -- of course you'll do a better job. And yes, this cloud takes some time to master, and for many programmers it's a huge barrier.

But that stuff is necessary but insufficient. If you're gonna implement the killer features you want, you need the chops to bang 'em out. And if those killer features are also gonna be used by a million, or a hundred million, other people, you'd better have the chops to bang out that code rock-freakin'-solid. Those chops don't come in 3 years, as Bill Gates claims, and they don't come from the 97th percentile of programmers, as Jeff alludes:

Good developers are good at programming. Really good at programming. You might even say fanatically good. If they're anything like me, they've spent nearly every waking moment in front of a computer for most of their lives. And naturally, they get better at it over time. Competent software developers have already mastered the skill of programming, which puts them in a very select club. But if you're already in the 97th percentile for programming aptitude, what difference does a few more percentile points really make in the big scheme of things?

The really killer code that's used to build the really killer features come from the programmers with a Z-score of 5 or 6, not 3. Being a natural will get you most of the way there, but unless you're a true prodigy, you need to continually refine your skills over many years. For most people, I think that cultivating expertise and interests outside the scope of programming helps this, as it does with any creative endeavor: as an analogy, if you want to write, read lots and write lots, but also live lots. For me, staying psyched about programming definitely requires lots of extrinsic inspiration; I'm a better programmer when I have time to ride my bike, and run, and play with my daughter, and laugh with my wife.

Being an excellent programmer requires being excellent at the cloud of things directly related to programming -- dealing with people, anticipating problems, having vision about products and features -- of course it does, and of course you should cultivate that. Beyond that, I truly think it requires external stimulation; really, really not programming will let you recharge and be better at programming. I doubt this is limited to programming.

A human being should be able to change a diaper, plan an invasion, butcher a hog, conn a ship, design a building, write a sonnet, balance accounts, build a wall, set a bone, comfort the dying, take orders, give orders, cooperate, act alone, solve equations, analyze a new problem, pitch manure, program a computer, cook a tasty meal, fight efficiently, die gallantly. Specialization is for insects.

-Robert A. Heinlein

Dreaming in Code, take 2

2007-01-29T23:37:00.000-08:00

True story: when I was 18 years old, a series of unhappy accidents left me the only programmer on a massive software project. This project was an upgrade to the cash cow for the company I worked for; if it didn't ship, the company stood to lose lots of money and indeed might've gone out of business. I worked 80-100 hour weeks for months to try to make the thing work right, and every time I fixed a bug it felt like I introduced 2 others.

One night, in the midst of this, I had a vivid nightmare. In this nightmare, I felt sick and feverish; my limbs didn't work correctly and my brain was fuzzy. As I stumbled about, I realized that I was the project I was working on: I wasn't dreaming about working on it, I was dreaming about becoming it. I was a huge collection of code, I was ill and nothing worked right and I had no idea how to fix myself.

When I saw the title of Scott Rosenberg's book, I thought: yes, that's right, that is not nearly so idyllic as it might sound.

And, as it turns out, the subject of the book is not a transcendent glimmering example of software genius. Chandler is, rather, one of those projects that, I suspect, inspires the sort of dreams one might be glad to wake from.

Another reminiscence, if you will indulge me: when I was 22 years old, I got a job at Microsoft. No bullshit or exaggeration: my dream job. I worked on Outlook, with about 20 other developers and maybe half as many program managers. Most of us were recent college grads or at least (in my case) the correct age to have just graduated from college. Our product manager, Brian, had a huge vision for Outlook, and I think we all shared it, and we worked our butts off to make it happen, to the extent that it could. Brian's vision was that Outlook should be an über-PIM, the place to which all your data flow, and the place from which you can see everything. Sound familiar?

Anyone who knows Outlook well must've shared the déjà-vu I felt when I read about Chandler. For instance, its "revolutionary" idea of applying arbitrary views to data was implemented by Outlook in 1994. All the views in Outlook are data-type agnostic; you can apply a calendar view to your files, if you'd like, or a timeline view to your email.

Of course, this sort of abstraction leads to endless feature creep. For instance: did you know that, until Outlook 11, you could browse your files in Outlook, as if it were the Windows shell? It was possible, through some tweaking, to browse the web with Outlook and might still be, for all I know. Did this fit into Outlook's Grand Vision? Absolutely. Did it add unnecessary complexity and make it much, much harder to ship? Hell yes. Chandler is clearly facing similar scoping problems.

Beyond the functional overlap between Chandler and Outlook, there's something else that's interesting. When I first booted Chandler, it took about 45 seconds to start. If you ever used Outlook 97, you may remember a similar experience.

I was on the team responsible for improving performance for Outlook 98. Right after Outlook 97 shipped, my lead printed out every single function call made when Outlook started up. It was a stack of paper about a foot high. He spent about a week going through this printout with a highlighter, looking for "stupid shit", as he called it. Turned out, there was plenty of it. It further turned out that most of this "stupid shit" wasn't a result of any one programmer making a dumb decision; most of it was a result of an architecture and a mindset which tried to prevent developers from shooting themselves in the foot by gratuitously abstracting away "dangerous" things like memory management. Again: needless abstraction will bite you in the ass if you're not careful. My lead and I spent months going through the Outlook code exorcising "stupid shit" -- removing code which hid what it was actually doing, getting the code cleaner and closer to the machine, making everything more explicit, tighter, and less generic. I suspect Chandler is going to need someone to go through its code with a fine-tooth comb in the same way before it's viable.

As a programmer, I find abstraction a beautiful idea. In practice, abstraction is useful only when its benefits are concrete. Here is today's Advice To Young Programmers: Create a class you can instantiate first; then, only if you need to create a similar class should you create an abstract base. Create an application which solves one problem really, really well; then, if you see that the problem you solved is similar to some others, you'll have a nice framework to expand on.

Now I've been tagged

2007-01-28T21:02:00.000-08:00

Justin tagged me, so here goes. I'm supposed to tell you 5 things you don't know about me. I don't know my audience, exactly, so some of you may already know all this stuff:

My mom says that when she decided to name me Moishe, she saw my name floating in the sky with rainbows and stars floating around. She didn't mention unicorns, but I sort of wish she had. My dad has a less drug-induced version of where my name came from, but I believe my mom.
When I was 17, I rode my bicycle from my parents' house in Massachusetts to Los Angeles. When I got home, I was skinny and bearded, and my little sister Mahala (who was 6) had no idea who I was. Apparently I scared the hell out of her by picking her up to hug her.
I met my wife at a poker game at a house in Olympia. The house was named Thelma's, after its previous (?) owner. This house had two incredible features. First, it had a "fire sculpture": the house overlooked Puget Sound and sticking up out of the mud and the water, about 100 feet off shore, were a bunch of pipes about 5 meters tall. Inside the house was a switch. Turning the switch would cause gas to flow through the pipes and an ignitor to fire on their tips. This created a truly awesome effect. Second: in the front yard of this house was a bathtub. Next to the bathtub was a pile of firewood. If you wanted a hot bath, you filled the tub with water (via a pail), piled the wood under the tub and lit it. The effect was like a cartoon-cannibal stew.
With my friends Ansel and Scott, I "stole" (maybe Ansel had permission? It's still unclear) Isaac Brock's shitter Honda motorcycle from a parking lot on Capitol Hill. We stuffed this thing into Scott's VW cargo bus, drove it to Oly, and put it in my garage, where it stayed for at least two years. When I moved out of my duplex, I abandoned the unloved motorcycle in the woods.
During the aforementioned bicycle trip, my traveling companions and I arrived in a town in Texas after a very long day, as the sun was setting. According to our map, this town should've had a grocery store, a campsite and even some people. It had none. So, we started looking around this weird abandoned town for a place to stay, and saw what looked like an abandoned building with an unlocked door. So... we opened the door, saw what looked like the interior of an abandoned bar. Why not sleep in an abandoned bar, right? I mean, if you're a bunch of scrawny east coast liberals and one weird Australian, what harm could possibly befall you in an abandoned bar in an abandoned town in Texas? So we started moving our crap in, and on our second trip outside we were confronted by a GIANT Native American dude, who asked if he could help us. We stammered out what we were doing, and he replied, "wait here". What the hell else were we gonna do, right? So wait we did, and giant Indian dude came back with a littler guy and a mean-looking pit bull and a GIANT CASE OF BEER. And a camp stove, and STEAK. And we fucking raged that abandoned bar, drinking beer and eating steak with the locals. I am no way clever enough to make that shit up, but if you don't believe me I have photographic proof. And, the best part: the dog's name was Asshole. It's all true.

Now, I tag Lindsey, Hosie, Steve, Steve, and Deonn. Special bonus: Dad

Book recommendation

2007-01-18T18:10:00.000-08:00

The Old New Thing: Practical Development Throughout the Evolution of Windows

I've been reading Raymond's blog for about 3 years now, and it is fan-freakin'-tastic. If you're a Windows developer, you should read everything he writes. He gives great, gritty details about the Win32 API: how to use it, how not to use it, why certain things are the way they are, etc.

Also, search for "Raymond Chen" on Google - the ads are cool :)

(the above link is an Amazon Affiliates link; if you click it and buy the book, I'll make a small amount of money)

Book recommendation

2007-01-16T22:55:00.000-08:00

Dreaming in Code: Two Dozen Programmers, Three Years, 4,732 Bugs, and One Quest for Transcendent Software

I just got "Dreaming in Code" by Scott Rosenberg. It's really good so far. I will write more when I'm done reading it...

(the above link is an Amazon link; if you click it, then buy the book, I'll get a small amount of money. I'm going to make it a habit to link to books I like on this blog)

Vegan for a week

2007-01-16T22:17:00.000-08:00

Eric and I have decided it would be fun to be vegan for a week.

This is progressing much as you'd expect: I'm hungry, grumpy and mean. But I ran my best-ever 1.5 mile tonight (though I won't mention the time, as it's rather embarrassing).

I have tentatively committed to this vegan thing until the Big Sur marathon, which is on April 29th, and which I signed up for today. I've got a 17-mile run on Saturday down in Capitol Forest which I may or may not do, depending on if I randomly decide whether Capitol Forest is still deep under a blanket of snow come Friday night. Failing that, I'll go for a 17-mile road run.

Even more pictures

2007-01-07T14:12:00.000-08:00

Grandma Denise sent Amelia a wagon for Hanukolstice. Pictures!

Little Red Wagon

New pictures!

2006-12-13T12:15:00.000-08:00

Late November/Early December

Finally I've uploaded some of the pictures from our trip to Massachusetts & our time so far in Memphis. Check 'em out!

Oof

2006-12-10T22:47:00.000-08:00

Today I was going to run a 1/2 marathon at Squak Mountain in Issaquah. I got up, started getting ready, and just started feeling less and less motivated. I drank my coffee, waffled on whether I should do it, ate a yogurt, waffled some more, re-read the description, continued to waffle, then finally decided I just wasn't in the mood to do an organized run, and decided to do a run I've always wanted to do instead. Said run is at Tiger Mountain, and I've biked it tens or maybe hundreds of times: up Northwest Timber Trail, up the fireroad to Preston Railroad Grade, then up Preston Railroad Grade to another fireroad, then back down. 16+ miles total, wickedly technical, muddy, rocky, rooty , steep and gnarly.

So I did that instead. And I am whooped. Tore up. Wrecked.

But damn I feel good now. And my god it was fun, most of it anyway. As I got off of Preston and back onto the fireroad, a light rain started to fall, and I was feeling strong, and the rain dripping off the dense green trees and the grey gloomy sky and the puddles in the road and the mud splashing off my shoes all felt so right. I was overwhelmed with a feeling of, "this is home". So many rides and runs in this weather, and for all the attendant difficulty of cleaning up afterwards and getting ready beforehand, the damp grey northwest is a beautiful place to be in the wintertime.

Software and craftsmanship

2006-12-05T23:07:00.000-08:00

A few months ago I bought a Laguiole knife for myself. It took about 3 months to arrive, and when I unwrapped it from its packaging it smelled like wood and smoke and oil. It's not perfect -- the engraving on the back of the blade isn't exactly symmetric, and the wood of the handle has a tiny gap against the butt of the blade. Functionally it's like many other knives, and I could use something which cost about 1/10th what this knife did to the same end. But it's handmade and it's beautiful. The blade is forged and the handle carved and the whole thing assembled in the same town, and when you use it you can think about the fact that the blade was made by one guy, and he's probably been doing it his whole life, and he spends enough time on just that blade to sort of get to know it. There is a particular pattern in the handle, made of brass rivets, which is impossible (or at least prohibitively hard) to duplicate by machine, and the guy who carved the handle placed those rivets; again, he had time to think about this particular handle as he was doing it.

I also own a bicycle frame which took about nine months to arrive. One guy designed the geometry and mitred and joined the tubes and sprayed the paint. He had time to think about me while doing this, because he had built another bike for me many years ago, which I rode into the ground and broke, and he saw how the old frame broke and where. When I ride this bike, when I don't completely forget it exists beneath me (because the cliché is true, it really does disappear), I can think about the time and pride that goes into building something like this.

What does this have to do with software?

The software you use every day is in a very real sense hand-crafted. It is simply impossible to mass-produce code because the software is the direct result of the person or people who built it; there is no factory, no assembly line, no repetition of creation. Software is labored over in an analogous way to the brazing on my bicycle or the blade on my knife; the person doing it has done similar things many times but none exactly the same; each problem solved in code is unique, as each piece of steel is unique.

Writing code is and should be an exercise in pride and craftsmanship. It is no exaggeration to say that the best software is written with love. Some of my favorite examples are detailed on Andy Hertzfeld's folklore.org site; for example see the description of the original Mac's sound driver.

As usual, Burrell's new design was very clever. The Macintosh was already continuously fetching data from memory to drive the video display, interleaving memory bandwidth between the display and processor in a similar fashion to the Apple II. But every 44 microseconds, there was a "horizontal blanking interval" where no video data was needed, so Burrell used that time to fetch data for the sound. That gave us a sample rate of 22kHz, which would allow us to do frequencies up to 11kHz, which isn't too bad.

Read that again. What he's saying is that the computer is busy grabbing data from memory to display on the screen most of the time. But if you think about how a CRT is designed, there's an electron gun which scans across the display, then moves down and back to the other side. There's a tiny little gap after every line is drawn when the computer doesn't need to do anything related to drawing stuff on the screen -- so the programmers used that little slice to deal with the sound. Ideas like that come from people who are artists, who obsess over their craft, and who truly love doing so.

Of course not every problem you get to solve as a programmer is so magical; most are mundane. But the mundane problems add up, and if you solve each mundane problem with an eye towards the end, you will transcend the mundane. Incredible software -- my personal list includes but is not limited to the original Mac OS, NeXTstep, Excel, and the new Google Reader -- is the result of thousands of mundane tasks done with pride in addition to the few huge magical leaps and bold strokes of genius.

Going home

2006-12-01T00:51:00.000-08:00

I'm on a plane from Boston to Seattle right now, trying to decompress from what might charitably be called a hectic week.

Lindsey and Amelia and I had a great time visiting my folks: Amelia finally got to meet her great-grandparents Maggie & Jerry, and her great aunt & uncle, and her 2nd cousin and all manner of other relatives. She took her first steps (with just a little help), and got to spend lots of time with her grandparents. Said grandparents of course, as always, amazed me with their generosity and kindness; my mom babysat for Amelia countless times so Lindsey & I could go out to dinner, or go run, or just have some time to ourselves. Plus she cooked for us, gave us gifts, and generally spoiled the hell out of us.

The only downside was that my parents' house is small, and full of people and animals, and consequently Lindsey & I ended up sleeping on the floor in the living room, subject to the seemingly-without-cause mewling of my parents' ancient cat and completely unpredictable behavior of their arguably retarded and perhaps senile English Setter. So I don't think I've got more than 5 hours' of sleep per night since Monday when we arrived on the red-eye.

Of course I was subject to some publicity which I would in retrospect much much much rather have completely avoided. And there was the good hectic-ness of visiting so many people in so little time (and still not getting a chance to spend as much time with anyone as we really wanted to). And I got to experience what it's like to commute two hours each way to work which turns a simple 8-hour work day into a massive 12-hour slog.

The result is that I am thoroughly exhausted and simultaneously wired and of course completely unable to sleep.

Lindsey & Amelia are in Memphis now, visiting Lindsey's family, and I miss them so much already. For the past three days I've only seen Amelia for 20 minutes or so a day and that was killing me; I can't imagine what it's going to be like to not see her for a couple weeks. She will be so different when I see her next, and I'm not sure in what ways. Smarter and cuter, of course, since those traits have been increasing steadily for over a year and I can't imagine they'll stop doing so.

Now for the next two weeks, until I fly to Memphis, I plan to live a more ascetic lifestyle: wake up, run, go to work, go to the gym and lift and steam and soak, go home, sleep, repeat. Please note 'blogging' is not part of that equation, unless it's maybe random notes about my daily runs. I want to be well-rested and happy and stress-free. I'm doing a trail half-marathon in about a week which I am very psyched for and I've got some other long training runs planned.

Good engineering problems to solve at work, a comfortable bed and a good night's sleep await me. I'm out-

Large companies and 'A' talent

2006-11-29T06:40:00.000-08:00

Dare Obasanjo has a post about the culture at big companies vs. small companies, and links to Auren Hoffman's post which claims that it's hard for large companies to attract 'A' talent.

I've worked at a startup, a small company and a couple of large companies.

Each of those companies afforded me fantastic learning opportunities and I can't believe that any was intrinsically better than another type of company. It's not a question of startup vs. big company: it's a question of particular company 'A' vs. particular company 'B'.

I had excellent and in retrospect superhumanly patient mentors at Timeslips, who taught me everything from how to use source control to how to do pointer arithmetic in Pascal and how to debug 8086 assembler code. Not to mention what it means to actually develop software products, instead of just programming.

When I got to Microsoft in 1994 I was like a kid in a candy store. Literally, I guess, if you count endless free Coke as "candy" and 22 years old as a kid.

My lead for the first couple of years was a guy who went on to be Office's director of development. He managed a large-ish team of developers (7 or so?) and knew in intimate detail what each of us was doing, to the degree that he'd answer random questions with a precision that made me think he'd already implemented the feature I was working on and was just letting me do it as an exercise. I was surrounded by other people whose accomplishments and abilities simultaneously inspired me and made me feel like a complete schlub: I worked with one of the guys who wrote Excel for the Mac back in '83, before anyone outside of Apple and a select few outside even knew the Mac existed. I worked with graduates of Harvard and MIT, I worked with people who worked on the original version of Word. And that was just in my hallway: I could wander around campus and bump into the original authors of Windows and those toiling away on Windows 95 or NT4 (the new hotness, which we all developed on, and was at the time super-secret); I could walk over to the original buildings and see Bill's office or occasionally the man himself.

This environment, for someone who hadn't even gone to college, was heaven. Beyond the obvious (getting paid lots of money to code!), the immersion in geek culture was awesome. Suddenly I got to be one of the Cool Kids, more or less (though by saying that I'm gonna get a bunch of comments from my co-workers saying, dude, no you weren't, we just let you hang out out of sympathy).

The downside of the immersion was a monoculture of ideas and software; when you work at a company that makes everything you need, it's easy to lose track of what's going on outside your bubble.

Working at Recipezaar got me out of that bubble. I learned to use Linux and it took me back to the early days of working on the command line in CP/M on my dad's VT180, but it was so fast and so powerful. The immediacy of development in this environment was matched by the immediacy of "shipping" -- the freedom and excitement of writing a feature and shipping it the next day was staggering. Brainstorming ideas in a 3-person team was fun and fast, generally, and being involved in every aspect of running a business from fixing the code at 3am to answering customer support questions to going on business trips and making presentations was the perfect antidote to Microsoft's hyper-specialization of roles. Beyond that, there was the possibility of making lots of money, or at least making enough money to let me live where I wanted and work from home and work on a product I wanted to work on.

Unfortunately I hit the downside of a startup: risk. After 18 months without a salary, no big deals on the horizon, and the ad market depressed, I went back to Microsoft for a steady paycheck. At this point I was a canonical example of someone who wants to work at a big company for the wrong reasons: I craved the steady paycheck and would be a mercenary and write software as a means to an end -- rather than out of love -- if necessary. Probably because of this, I got exposed to many negatives of working at a big, mature company: politics, at which it turns out I am a complete retard; stultifying drudgery and bureaucracy; the need for self-promotion, and etc. Despite this, I ended up having some fun; I eventually ended up on the MSN Desktop Search team where once again I was surrounded by people way smarter than me -- the best environment for learning. Like no other team I'd ever been on, this team emphasized the beauty of the Hack: want to embed search results in Outlook? OK, figure out how to co-opt their HWNDs. Want to make that crusty old Windows Indexer work right? Go to it, 15 years of code be damned, just make it work.

But then I got an offer from Google and after a little bit of waffling (I was having much fun with the hackers) I started there back in January.

And holy shit I hope I can convey to you what sort of geek heaven I'm in now.

Above I talked about NT4 being the "new hotness" back in '94 -- the guys who made it that way sit right next to me. In the same office. And that sort of expertise is everywhere here... it seems like every office is occupied by at least a couple of industry leaders, guys whose names you'd recognize if you're even a casual observer of geek culture.

Google's culture values independence and transparency of communication in ways I didn't think were possible at a large company. We've of course got our 20% time, but beyond that there's a sense that everyone here is competent enough and trustworthy enough to be clued in to many parts of the business -- not just engineering -- which would typically be hidden. That trust nets huge gains in loyalty and excitement.

There aren't many places in the world where you could can come up with the idea for a feature or product, implement it, and launch it to an audience of millions, with the infrastructure to support it. Yes, you can do it at a startup or on your own, but getting eyeballs and servers is non-trivial. For every YouTube there are hundreds of sites nobody's heard of.

Certainly startups have their benefits, but so do some big companies. As cliche as it is, your co-workers are going to make the biggest difference in your satisfaction and learning; find a place with excellent mentors (no matter how much experience you have) and you will do well, whether that place is a start-up or a large established company or something in between.

Apple design process

2006-11-28T05:26:00.000-08:00

Arno has an article about the Shutdown feature and how it was implemented at Apple.

He has some excellent insights about using the physical power button instead of a menu item. Interestingly, I remember one of the initial incredibly cool things about the mac was its software-instantiated shutdown, as contrasted with Windows' "It is now safe to turn off your computer". Which wasn't Windows' fault at all, of course, but an artifact of not having the tight control of the hardware that Apple does.

Let's talk about something else

2006-11-27T19:12:00.000-08:00

I got a wristwatch style GPS a little while ago, on the advice of my friend Gerald, and it is simultaneously one of the geekiest and awesomest things I've ever owned.

Look: MotionBased! I am so slow!!! And I have so much data available to me!

In other news, I've been working out of the Google Boston office while I'm in town visiting my folks. It's fun: I walk about a mile to the commuter rail in Hamilton, take the train in to North Station, then walk across the Longfellow Bridge to Cambridge. This evening when I got back to my parents' ranch house in the suburbs and took off my jacket and put down my briefcase (um, actually my messenger bag) and picked up my little daughter I felt like it was 1950. I may need to buy a suit just to feel like I'm doing it right for the next few days.

Brief followup to yesterday's post

2006-11-25T18:32:00.000-08:00

Whew. I, um, seem to have struck a nerve.

Some things which were implicit in yesterday's post but I should make explicit: I didn't work with any morons; indeed, everyone I worked with was extremely smart. The problems I saw weren't because of any incompetence on the micro scale or even anything attributable to just one person; the slowness and indecision were emergent properties of the system.

Further, as massive as Vista is, it's still just a small piece of Microsoft. One thing which always impressed me about Microsoft was how each team truly did have its own personality and distinct development style. I worked on teams which used Scrum, teams which were small enough to subscribe to no development process at all, and teams which made medium-scale development work very efficiently. I think and hope that my particular nook in Vista was a pathological case which makes for good illustration but whose faults were particularly exaggerated and obvious.

Finally, people have asked what's different at Google. I refer you to my esteemed colleague from the 4th floor for a treatment which is surely better written and more engaging than anything I can say.

The Windows Shutdown crapfest

2006-11-24T04:57:00.000-08:00

I worked at Microsoft for about 7 years total, from 1994 to 1998, and from 2002 to 2006.

The most frustrating year of those seven was the year I spent working on Windows Vista, which was called Longhorn at the time. I spent a full year working on a feature which should've been designed, implemented and tested in a week. To my happy surprise (where "happy" is the freude in schadenfreude), Joel Spolsky wrote an article about my feature.

I would like to try to explain how this happened.

I worked on the "Windows Mobile PC User Experience" team. This team was part of Longhorn from a feature standpoint but was organizationally part of the Tablet PC group. To find a common manager to other people I needed to work with required walking 6 or 7 steps up the org chart from me.

My team's raison d'etre was: improve the experience for users on laptops, notebooks and ultra-mobile PCs. Noble enough. Of course the Windows Shell team, whose code I needed to muck about in to accomplish my tiny piece of this, had a charter of their own which may or may not have intersected ours.

My team had a very talented UI designer and my particular feature had a good, headstrong program manager with strong ideas about user experience. We had a Mac [owned personally by a team member] that we looked to as a paragon of clean UI. Of course the Shell team also had some great UI designers and numerous good, headstrong PMs who valued (I can only assume) simplicity and so on. Perhaps they had a Mac too.

In addition to our excellent UI designer and good headstrong program manager, we had a user-assistance expert, a team of testers, a few layers of management, and me, writing code.

So just on my team, these are the people who came to every single planning meeting about this feature:

1 program manager

1 developer

1 developer lead

2 testers

1 test lead

1 UI designer

1 user experience expert

8 people total

These planning meetings happened every week, for the entire year I worked on Windows.

In addition to the above, we had dependencies on the shell team (the guys who wrote, designed and tested the rest of the Start menu), and on the kernel team (who promised to deliver functionality to make our shutdown UI as clean and simple as we wanted it). The relevant part of the shell team was about the same size as our team, as was the relevant part of kernel team.

So that nets us an estimate -- to pull a number out of the air -- of 24 people involved in this feature. Also each team was separated by 6 layers of management from the leads, so let's add them in too, giving us 24 + (6 * 3) + 1 (the shared manager) 43 total people with a voice in this feature. Twenty-four of them were connected sorta closely to the code, and of those twenty four there were exactly zero with final say in how the feature worked. Somewhere in those other 19 was somebody who did have final say but who that was I have no idea since when I left the team -- after a year -- there was still no decision about exactly how this feature would work.

By the way "feature" is much too strong a word; a better description would be "menu". Really. By the time I left the team the total code that I'd written for this "feature" was a couple hundred lines, tops. [edit: note that that are tons of other more complicated features that support this menu, like the control panel, the additional kernel work, etc., whose code was huge compared to mine. Note also that these features weren't, by a long shot, the only thing all these people were working on]

But here's how the design process worked: approximately every 4 weeks, at our weekly meeting, our PM would say, "the shell team disagrees with how this looks/feels/works" and/or "the kernel team has decided to include/not include some functionality which lets us/prevents us from doing this particular thing". And then in our weekly meeting we'd spent approximately 90 minutes discussing how our feature -- er, menu -- should look based on this "new" information. Then at our next weekly meeting we'd spend another 90 minutes arguing about the design, then at the next weekly meeting we'd do the same, and at the next weekly meeting we'd agree on something... just in time to get some other missing piece of information from the shell or kernel team, and start the whole process again.

I'd also like to sketch out how actual coding works on the Windows team.

In small programming projects, there's a central repository of code. Builds are produced, generally daily, from this central repository. Programmers add their changes to this central repository as they go, so the daily build is a pretty good snapshot of the current state of the product.

In Windows, this model breaks down simply because there are far too many developers to access one central repository. So Windows has a tree of repositories: developers check in to the nodes, and periodically the changes in the nodes are integrated up one level in the hierarchy. At a different periodicity, changes are integrated down the tree from the root to the nodes. In Windows, the node I was working on was 4 levels removed from the root. The periodicity of integration decayed exponentially and unpredictably as you approached the root so it ended up that it took between 1 and 3 months for my code to get to the root node, and some multiple of that for it to reach the other nodes. It should be noted too that the only common ancestor that my team, the shell team, and the kernel team shared was the root.

So in addition to the above problems with decision-making, each team had no idea what the other team was actually doing until it had been done for weeks.

The end result of all this is what finally shipped: the lowest common denominator, the simplest and least controversial option.

I have no idea how much of the rest of Vista ended up like this. I think (indeed hope) my team was a pathological case; unfortunately it's a visible one.

edits: fixed link, removed some strong language, fixed math

Talk & orkut

2006-11-07T15:21:00.000-08:00

Finally I can talk about one of the things I've been working on lately: orkut & Google Talk are integrated now. Orkut shows presence and your orkut friends get automatically added (if you choose) to your Talk roster. Talk's also got some additional orkut-bling, like links to your friends' orkut rosters and scrap pages, and notifications when you get scraps.

Here's my orkut profile.

Donida Farms 'cross report

2006-10-22T21:18:00.000-07:00

Donida Farms 'cross

A couple of weeks ago, Cousin Steve sold me some gorgeous 'cross wheels for way less than they're worth. I got them earlier this week, re-glued 'em (actually paid a shop to do it. If you'd asked me when I was 20 if I'd be paying a shop to glue my tires I don't think I'd've said yes), and put 'em on my bike last night.

Actually, the last phrase of the above sentence is a profound oversimplification.

I also bought a single-chainring conversion kit from CycloCrossWorld, and that showed up on the same day as the wheels.

And when I picked up the wheels from the shop, I also bought a new cassette and chain.

And I had some wider handlebars in my basement, and some pristine celeste bar tape which was just begging to be used.

And I had a shifter-less brake lever to replace the left STI lever.

Oh, and I have a wife who's a bike mechanic and said (really, she said this): "take the baby for a few hours, I'll hook up your bike."

And hook up the bike she did. It is freakin' beautiful now! I think it's probably a good 2 pounds lighter, looks better, shifts better, rides way faster, and so on.

Despite the stigma attached to doing major bike overhauls the night before a race, this one seems to have worked out fine; in fact, I had my best race in a long time today. We got to the course with plenty of time for me to get registered and do a lap to warm up and check out the extremely fun and mostly flat course. There were some nice long gravel-road stretches, an agonizingly hard section along the wet sand of a horse track, lots of bumpy grass, a steep but not-horrendously-long run-up, and a good gravel sprinter's hill. The race -- particularly its flatness -- played to my meager strengths, and I finished 14th, a significant improvement over the 22nd I placed in the prior two series races.

The most stressful part of the race came after I finished. Lindsey had about 20 minutes to get her stuff together, warm up, and get to the starting line for her race, giving me immediate baby duty. I was sweating like crazy, still on that aggressive racing high, trying to change poor Amelia's diaper in the back of the car, which might've been fine but Amelia was totally losing her shit and trying to squirm away, likely because I smelled terrible and was dripping sweat all over her. Once that was taken care of, Amelia was fine and she sat on my shoulders and cheered for her mom (and other racers, and random spectators and birds).

Good times.

St. Helen's

2006-10-20T21:11:00.000-07:00

St. Helen's

I climbed St. Helen's today.

It turned out to be as close to perfect as I'd like.

I'm going back to drinking beer and watching TV now.

A belated weekend-of-cross report

2006-10-05T07:36:00.000-07:00

Saturday I raced Star-Crossed, Sunday I raced in S. Seatac.

Saturday's race was hectic and HUGE. I think there were over 100 people in my class. Unfortunately I started near the back (they arranged us by our numbers, which were handed out in random order) and never moved up that far. I got pulled after I think 4 grueling laps. The worst part was demonstrating my complete lack of remount skills in front of the beer garden, vis: this.

I had a much better race on Sunday. When I pre-rode in the morning, I was full of fear and malaise: the course was very running-intensive and technical with LONG sand pits, some short very steep run-ups, and one very long, loose run-up (though not nearly as bad as Steilacoom). However, once the racing started, I felt good and held on for 22nd (the same position I finished in at Steilacoom), but this time I beat Fabien which felt pretty good.

This weekend there's a race at St. Ed's, about 5 miles up the street from me, which I may or may not do. It might be a good weekend to watch Lindsey race...

'cross fun, continued

2006-09-24T22:35:00.000-07:00

I raced the Steilacoom 'cross race today, the first in the Seattle Cyclocross Series. Once again it was ridiculously fun and once again I flatted when I was almost done. This time I just rode in on the flat, which sucked because I got passed by about 5 or 6 people on the straight, flat road to the finish. This particularly sucked because it was the one part of the course I would ordinarily (that is, with inflated tires) have been OK at.

Thus, onto the parts I was definitely NOT OK at. Firstly, a FUCKING RIDICULOUSLY BRUTAL AND LONG run-up. I believe this is the run-up that Ryan Trebon, of all people, referred to as "retardedly hard." Seriously. Steep, loose dirt for far too long. It's of course packed with spectators. So now form for yourself this mental picture: Last lap, I'm flying towards the right turn into loose dirt to start the run-up, I go to dismount and totally bust ass at the bottom. OK, no problem, get up and start... walking. Yeah, for what seemed like half an hour, in front of all the spectators, covered with dirt. Ugh.

Thankfully, once you reach the top of the run-up, there's more climbing. It actually was fun and I consistently passed people on this part. Then a sketchy-as-hell loose, rocky downhill with one particularly awesome right-hand corner in loose gravel that I hope I made appropriate motorcycle noises through, given the amount of sliding that I was doing.

Also I should mention there were two sets of barriers which proved to be excellent demonstrators of my incompetence. I gotta learn some dismount/remount style, seriously. STANDING at the other side of the barriers trying to get your pedals right is probably not the winning strategy.

All the above said, holy shit that was fun. Looking forward to my cousin Steve's tubies, and I think I'm gonna single-chainring-and-cross-bar my bike in time for the upcoming double weekend. Look out!

New favorite sport

2006-09-05T06:15:00.001-07:00

Why have I never raced 'cross before? Holy shit that's a fun sport.

Lindsey & I raced the Labor Day cross race in SeaTac yesterday; I got up early to go do the Masters C race at 9:30am, then Lindsey & Amelia & I went back so Lindsey could do the women's race at 1:30pm. A full day at the races, in the warm sunshine -- so much fun!

My race went well until the last 1/2 lap (of 5) when I got a front flat that made steering difficult. I could have run in, I guess, but instead I walked back to the parking lot & DNF'd. No biggie; it wasn't like I was gonna place, and the race wasn't part of a series so there were no points. I think (sorta) that I was top twenty-ish out of maybe 40 or so riders. But I had SO MUCH FUN! Holy shit, again, seriously. I had no idea that tearing through singletrack glued to someone's wheel on sketchy little tires down in the drops would be such a blast. I can't believe the next race is 3 weeks away; I want to race again tomorrow!

Lindsey's race went OK: after not wanting to race and whining about how out of shape she is and blah blah blah she went out and got a close 2nd, on her freakin' singlespeed no less! Damn that girl can ride! There was a little post-race drama because they screwed up the results initially but it all got ironed out and everything was fine. Amelia & I had a great time cheering for her (and the rest of the women).

Pictures are here.

Also, some more pics of Amelia.

Happy birthday to Talk

2006-08-25T22:19:00.000-07:00

We got a mention on the Official Google Blog and our logo has changed for a few days.

Singlespeed Devil's Gulch

2006-08-20T08:43:00.000-07:00

I finally got out and did one of the "classic" Northwest rides on my singlespeed. By "classic" in this context I mean: in the mountains (not Capitol Forest), long initial climb, and huge downhill.

Anyway: this was a great ride for the singlespeed. The climb at the start is long -- nearly 12 miles -- but mostly gradual. I ran my 32x18 and, while I had to walk occasionally, I probably rode 95%. There was one section of loose sand leading into a moderately steep pitch which I unequivocally had to walk; the rest of the walking was due more to laziness or lack of fitness. I think if I'd swapped to my 19 I could've ridden the whole thing.

The descent was of course fantastic, though I didn't have quite as much fun as the first time I did this ride on my (now Gerald's) Kikapu. The downhill trail is pretty much perfect: not too steep, but steep enough to get great speed, mostly smooth with some rougher stretches to keep you on your toes, a little bit of semi-scary exposure, some stream crossings to get your feet wet, and lots of beautiful views. It's funny: the climb is unspectacular and the descent is completely spectactular, but: I took some pictures on the uphill and once I started down the last thing on my mind was stopping to get my camera out.

Here are those pictures.

Code Jam

2006-08-15T08:17:00.000-07:00

Have you stumbled across my blog while looking for "Programming Interviews Exposed?"

Here's a better way to get a job at Google: Code Jam.

Yep

2006-07-20T14:31:00.000-07:00

Who can argue with this logic?

New Amelia pictures

2006-07-08T22:41:00.000-07:00

Hosie & Mahala came over and we ate yummy salmon & veggie burgers with Amy (Lindsey's sister) & Diane (Lindsey's mom).

Hosie sculpted Amelia's hair:

smugmug
picasaweb

I haven't decided whether I like smugmug or PicasaWeb better yet... I'm a little bummed that smugmug doesn't have a Universal Binary uploader yet.

LJ goes Jabber

2006-07-07T07:03:00.000-07:00

This blog entry discusses LiveJournal's Jabber (XMPP) service and their plans to federate to other XMPP services. The XMPP world just grew by about 10 million users (modulo the number of LJ users who were already using Google Talk or other XMPP services).

This is so good:

yes, we'll be federating. you'll be able to talk to GMail/Gtalk users, jabber.org users, etc. Anybody else using Jabber will just work, without special business relationships, just like email works.

Hopefully this marks a larger trend.

Picasa Photoweb test

2006-07-06T07:45:00.000-07:00

One of the coolest toys that comes with the MacBook Pro is its built-in video camera and PhotoBooth software. Eric discovered the "effects" buttons while he was here for the 4th. Look:

Pictures

Also note that this is hosted on PicasaWeb (aka lighthouse). I don't think I'll be giving up smugmug anytime soon, but I do like PicasaWeb's presentation and UI.

More interview advice

2006-07-05T19:36:00.000-07:00

Ace the Microsoft interview!

Google hybrid benefit

2006-06-21T06:30:00.001-07:00

Kent asked,

Will Google give you money off a bike if you ride one instead of driving a car?

A Litespeed Blueridge makes a very nice commuter.

Not yet. However, there is a program which gives you money with which to make charitable donations based on the number of days in a quarter you ride or walk to work. We have bike lockers and, yes msfties, nice high-quality towels in our locker rooms. Also Google buys your locally-grown organic fuel while you're at work.

The idea with the hybrid program is, I think, "if you're gonna buy a car anyway, buy a hybrid, and we'll pay you back the difference in price". I think what it mainly does is promote a corporate culture of giving a shit, which is nice. It's cool walking around the parking garage down in Mountain View -- there are more hybrids (and full electrics) than expensive sports cars, despite the high millionaire:employee ratio.

And it's inspired Lindsey & I to go from two cars down to one hybrid, which is a good step.

Recovered

2006-06-08T08:32:00.000-07:00

Yesterday morning I went out & did 3 repeats up the Seminary Dr. side of Holmes Pt Road. It's a delightful little climb up from Lake Washington to the top of Juanita -- less than 400 feet or so, at 7% (this information gleaned from EricGu's awesome Seattle Climbs page, which I think was derived from the late, great Larry Kemp's data.

So, anyway, I feel great. Apparently base miles are good, even if you get them all at once.

In other news: work continues to be very, very fun. I've got my grubby little fingers all over the next release of Talk, which makes me very happy indeed, and the release after that is gonna be incredibly cool. You heard it here first.

And of course Amelia continues to be the cutest baby ever made. She is becoming the squirmiest, too, which is cute but introduces some new complexities. Like: don't leave her on the bed for half a second, even right in the middle, because she'll start squirming towards the edge. She also grabs my glasses which inspires me to maybe buy some big chunky hipster plastic ones which are less likely to put her eye out.

150 miles of "why?"

2006-05-29T16:16:00.000-07:00

Here are some stats:

10 laps

150 miles

17.5 hours on the bike

90 minutes of sleep (where did the rest of that time go? Apparently it takes a long time to prepare food and get out of the tent -- even when you're awake -- when there's rain pounding outside)

1 broken chain

1 bobcat sighting

18 scoops of Perpetuem

2.5 bagels & lots of peanut butter

3 Cokes

1 mocha (the best part of the race, I think)

5 chocolate Gu's

1 blister

1 chapped ass

I had a really weird race. First off, my Chris King hub gave up the ghost on Wednesday night and I couldn't get it fixed in time, so I had to switch categories and race on my geared bike. I love my Kula, but I hadn't been riding it regularly for like 18 months, so that was a little weird. And then on the very first climb (literally right out of the campsite) my chain broke. So I had to run (again) down to the campsite, where I had it fixed by a very friendly and incredibly competent mechanic. And he undercharged me, no less. AND we talked singlespeeds the whole time he was fixing it.

After that, though, I had a low-grade lack of motivation for what seemed like the whole race. The first 4 laps were really hard; I was going slow and steady and felt OK, but not great. Honestly, my 4th lap I was totally dragging ass; Lindsey & Amelia had just left to go back to the hotel, I was at a low point of motivation, the hours to go seemed very long indeed, etc. However: after that lap, I took about a 30 minute break, ate a bagel & peanut butter, did a shot of Gu, and my next two laps were phenomenal.

Unfortunately, at 11:30pm, I was out of battery power (or at least close enough that I didn't want to risk a lap without a recharge). So I decided to take a 2-hour break while my batteries recharged, which might've been a great idea except that I was so jazzed on 2 great laps that I couldn't sleep, so I lay in my tent for 90 minutes, fell asleep for 30, and was awakened by my cruel alarm clock and the pitter-patter of rain on the tent. I set out for what must've been my slowest lap: overdressed, tired enough that I couldn't steer (but not so tired I was hallucinating: I know I didn't push very hard on this race because I never saw faces in the trees or ghosts in the woods), and thinking enviously of Lindsey & Amelia in their warm, dry bed.

So that lap was hard, and I came back and dozed off in my tent long enough that I wouldn't need lights on my bike for the one after that (in retrospect an unnecessary rationalization, but whatever), then had another two OK laps, and that was that. Seeing Lindsey & Amelia at the end of my final lap was great; even though it was a lot of work & complexity for them to come along, it was worth it just for that moment. Also Lindsey made me a very cool little "Go Daddy Go!" sticker for my bike, which was inspiring.

So I did fewer laps than 2 years ago, but the laps were a little harder because of the additional singletrack and wet conditions. I was never mentally into the race, and I don't know how that happened; I hope it doesn't happen again.

That said, I'd like to choose a good long race to get truly stoked for, sometime this fall. The Soul Ride down in AZ sounds like a lot of fun (and cheap!), or maybe Moab... who knows. Right now I'm just looking forward to a summer of good early-morning road rides, long nights at the track, occasional epic solo mountain bike rides all alone in the wilderness, and watching Lindsey race with Amelia hangin' out in the Bjorn. And letting my ass heal.

72 until 24

2006-05-24T07:05:00.000-07:00

On Friday morning, I leave for the Spokane 24-hour race. I think I'm ready, though I'm not sure I'm in as good shape as I was when I did this race 2 years ago. And of course I'm never as ready as I want to be, but nothing to be done about that now. Just rest, get some easy spinning in and get stoked.

Weather forecast says high 50s and a 40% chance of rain. The temperature's perfect but I'm a little nervous about the rain, just because it adds an additional level of complexity to everything: clothes need to be changed more often, supplies need to be put away every lap, glasses get dirty (actually that's the part I hate the most), mechanicals are more likely and brake pads wear out, etc. I'd like to think I get some sort of advantage because I've been training in the rain all spring, but then again everyone else in my category is local (or from AK) so they've been doing the same.

2 years ago I did 12 laps, or 170 miles. I'm hoping to at least match that this year; if I'm not quite as fast, I'm also better accustomed to going without sleep thanks to Amelia.

Lindsey & Amelia decided to stay home while I go race, just because it would be a logistical nightmare for both of us if they came: for Lindsey, 24 hours of standing around in the rain with Amelia doesn't sound like too much fun; for me, having them leave with the car (and who knows what supplies I left in it) at inopportune times could also be a monkey wrench.

Update: just checked the start list, there are EIGHT guys racing Solo Singlespeed! Twice as many as 2 years ago, holy crap. And 34 solos total. So cool!

Whidbey weekend

2006-05-22T08:56:00.000-07:00

Here are some pictures from Lindsey's race weekend on Whidbey, where she whupped her competition and nabbed 1st place and a sweet set of tires. She also netted a noseful of dusty boogers, a big bruise and some blood. Nice one, kid!

Amelia & I had a bunch of fun gettin' rowdy and cheering her on from the sidelines. Soon enough Amelia will be out there too -- I can't wait!

bike trivia

2006-05-13T08:17:00.000-07:00

Lindsey & I went on a ride together last night for the first time in months. We rode St. Ed's (of course) for about 90 minutes, which meant the same trails a bunch of different times, but it was very, very fun. Lindsey is getting SO STRONG... she dropped me on every flattish windy section and on the part from the QFC parking lot to Bastyr she was a good 30 seconds ahead of me, and I was emphatically not just relaxing and taking in the scenery. I wouldn't be at all surprised to see her upgrade to Expert in the next few races.

She's also sportin' brand new tires, she's got a SPV rear shock for the Juli on order, and some other bling. And she's all Velo-Bella'd out with her arm warmers & jersey & shorts & socks -- So Pro!

I've (finally) got my track bike figured out: I'm getting a Flyte frame after all. It's an aluminum main triangle with carbon seatstays & chainstays, and the geometry looks great. I also got a Reynolds Ouzo fork for it... pretty pimp. Now I've just gotta figure out my parts; I've got a sweet Nitto stem & stainless steel keirin bar that I love, and I'll probably use the Campy wheels from my old fixie (though I'd dearly love to get those rebuilt with sew-ups). I'll scour eBay for some sweet cranks and dig up a seat & seatpost from my "morgue" of bike parts (thanks Steve for an awesome word for our parts pile) and get a chain and be good to go, I hope.

Amelia still doesn't have a bike. But I did find a sweet John Hinderliter (the artist whose work is above) drawing of a little kid on a tricycle in shin guards going huge off a quarter-pipe, so that'll be thumbtacked to the wall in her room for a little bit.

I love the suburbs

2006-05-09T19:25:00.000-07:00

I will preface this by saying that I grew up in one of the richest, snobbiest towns in Massachusetts (which is saying a lot): it featured Sunday afternoon polo, of all things, and the queen of England even attended once. So I know from snobby.

Anyway, Lindsey was talking to our two-houses-down neighbor about babies or something and our neighbor piped up with, "So, do you live here or just rent?"

Sweet.

New photos

2006-05-05T19:21:00.000-07:00

Lindsey's out mountain biking and I'm hanging out here with Amelia. It's beautiful and sunny outside so we lay in the sunlight in the living room and I took some pictures.

While I was uploading those, Amelia was lying on the bed without her diaper on. That was a mistake. No pictures of that, thank god.

Anyway, here's my favorite picture:

She looks like Tina Modotti!

One of those days

2006-04-20T22:54:00.000-07:00

This morning I woke up a little late and decided to go for a ride anyway. As soon as I got on my bike, all I could think was, "crap, I wish I could skip work today and just ride for another 4 or 5 hours, then go home and hang out with 'meils and Lindsey for the rest of the day in tired, blissed out endorphin-induced euphoria."

I decided to go to work anyway, after a truly phenomenal ride. The sun was out, the trails were dry, I felt fan-freakin'-tastic. I arrived at work to once again find myself a victim of my own incompetence; I don't understand how I manage to peak my ability to miss stupid bugs right when it matters the most. Happily, I fixed everything quickly and the rest of the day was good and full of theorizing and designing and other engineering goodness. And lunch w/ JR at Cactus which was good despite the circumstances which prompted it.

Trying to decide if I should race on Sunday at S. Seatac. Lindsey's racing singlespeed... V has offered to babysit... there is no reason in hell I shouldn't race. Other than sometimes I don't enjoy being on the edge of puking for two hours. I dunno. It's weird with Amelia: before, if I raced really hard and totally knocked myself on my ass for the rest of the day, nobody cared. Now I have RESPONSIBILITIES and the like, and if Lindsey races too and Amelia gets cranky on Sunday night, we'll both be in a world of hurt. Though I have to say Lindsey's recovery after her race last month was amazingly fast. I swear being a mom gives you super powers.

Bike status

2006-04-14T22:48:00.000-07:00

OK, track season's approaching. I want to get a Flyte track bike, since they sponsor JCV, but haven't heard anything about when they're actually gonna ship one... so I'm hoping to instead get this (with my tax refund):

Nice spec, tubbies, and a good stiff Al frame which I flatter myself by thinking will help me.

I've been all single-speed, all the time, for about 5 weeks now and I'm starting to feel strong, if neither fast nor fit. But I'm getting those weird ancillary muscles back which only singlespeeding seems to really touch, like lower back and obliques and delts. Who needs weightlifting when you're pedalling 15rpm?

Crazy little baby

2006-04-14T22:45:00.000-07:00

Tonight Lindsey went out for a minute to run a quick errand and while she was gone I tried putting Amelia to sleep. I did this completely succesfully and with no drama at all last night.

Tonight, however, Amelia totally lost her shit. Nonstop screaming as soon as I brought her into the bedroom. She knew just what was happening and wanted nothing to do with it. Binky? No help. Food? No help. Swaddling? Absolutely fucking not.

I brought her downstairs and she calmed right down. Lindsey got home, brought her upstairs, and she was asleep in 10 minutes. Mom knows best, apparently.

One other thing

2006-04-11T21:01:00.000-07:00

Somebody asked, in reference to my earlier post about interviewing, if I could recommend any good books on the subject.

I can. Or perhaps you'd enjoy this one.

Seriously, no book specifically about interviewing will do a good job preparing you for a well-conducted technical interview. Reams of programming experience and smarts will help, and sometimes even that's not enough.

That said, you can check Gayle's interviewing page and Steve Yegge's blog, specifically some of his interviewing ponderings.

Missin' my baby

2006-03-21T23:48:00.000-08:00

I'm down here in silicon valley to spend a couple of days at the Googleplex, which will be fun, but I really miss Lindsey & Amelia right now. While I'm looking forward to an uninterrupted night of sleep (Lindsey will kill me for rubbing that in), I'm sad I didn't get to kiss Amelia goodnight and that I won't get to play with her before I go to work tomorrow morning. Now that she has a personality I miss her a lot more. Also, I spent a ton of time with just her this weekend while Lindsey did errands and raced her bicycle, and then today I hardly saw her at all (just long enough for her to spit up all over my shirt).

Alright, off to sleep.

Lindsey races...

2006-03-19T21:25:00.000-08:00

...and wins $40! 1st place women's sport singlespeed at the "Dash for Cash" on Whidbey Island. Pics here.

Not inspired lately

2006-03-18T21:19:00.000-08:00

Just haven't been feeling like writing, don't know why.

Work's good, cruisin' along now. I don't expect to stay in this comfort zone for long, though; there are some pretty cool things coming down the pike that are going to require me learning lots of new stuff. That's good; I have no desire to sit still.

Uncle Jonathan's interviewing at Google in Mountain View on Thursday, and I'm flying down on Tuesday night to do some work. It'll be great to be back in MV and get the Silicon Valley vibe again, and of course it'll be great to see Jonathan.

I've been riding a bit, all mountain biking. I re-discovered St. Ed's/Big Finn -- it's so much fun to just zip up for an hour in the morning. The technical stuff is hard enough that it makes me think but not so hard that it scares the crap out of me. In other biking news: I'm doing the Round & Round 24 Hour race in May. Solo singlespeed (of course!). Lindsey & Amelia will come out and stay at a hotel while I race (perhaps "plod" is a better word) through the night. My only goal is to do one lap more than 2004, so 13 laps. That's 184 miles. Stretch goal is to break 200 miles, which means 15 laps. I think it's do-able -- I slept for a long time last time and I have a little more practice with sleep deprivation now, thanks to 'meils.

Lindsey's racing at Whidbey Island tomorrow -- I can't wait to watch her race! 'Meils and I will be patrolling the course with camera supplied by me and cuteness supplied by 'meils. Good times.

Alright, it's almost 9:30 and I'm exhausted. Good night.

The salt mines

2006-03-06T18:46:00.000-08:00

Unfortunately I missed this trip to the Lego store but I'm sure it won't be the last.

It's been a while

2006-02-26T09:27:00.000-08:00

Sorry for the long hiatus -- my days have consisted lately of: wake up at 7, get to work by 8, work (where "work" means "write code and think of cool code to write", not "slog through politically-mandated busy work") until 7, get home at 8, play with Amelia, read a little bit, fall asleep. I'm not complaining at all. Remind me, next time I'm thinking about switching jobs to one that's more interesting and more fun and pays better, to just go ahead and do it. Why was that decision so hard?

Our big news is that we're moving to Kirkland. Tomorrow, in fact. We're going to rent a house in the East of Market neighborhood which is walking distance to downtown Kirkland and my office. It's going to be so nice not to have to worry about traffic, and not to waste 1 to 2 hours of my day stuck in it. Now I can wake up in the morning, ride my mountain bike up to St. Ed's, play on the trails, ride home, shower, and walk to work. Killer! Then at the end of the day, I don't have to play games with staying late to miss traffic -- I can just walk home when I'm done with work.

Amelia is amazing. She smiles and laughs and coos and grabs my beard. It seems like every day when I come home from work she's doing something new. Watching her learn how to control her arms and hands is incredible -- last night she grabbed her binky out of her mouth by the handle and started waving it around and laughing, like "aha -- look what I finally did!"

And Lindsey is amazing. She's managed to pack up 90% of our house (I'll give myself 10% because I'm feeling generous) while taking care of Amelia and being a great mom. I don't know how she does it.

Hopefully some pictures of our new house and new Amelia pics coming later on today/tonight.

Sweet

2006-02-06T23:27:00.000-08:00

Gmail/Talk

Something funny

2006-02-04T10:10:00.000-08:00

If you have a gmail account, go to the "Spam" folder and check out the RSS links on top.

More Amelia photos

2006-01-29T21:37:00.000-08:00

Pictures.

Also note that you can get an RSS feed if you're into geek stuff like that.

Also

2006-01-28T17:59:00.000-08:00

More Amelia

Trip to Mountain View

2006-01-28T16:58:00.000-08:00

I spent the week before this one down in Mountain View for Noogler training. I'm completely assimilated now. I think I (finally) made the right choice about what job I should be doing.

Training started on Tuesday and I got to Mountain View on Saturday, so I went down to Monterey to hang out with Steve and Stacy. We had dinner in San Jose with Rose (Usa) and Dean and little Talia who is the 2nd cutest baby I've ever seen. On Sunday Steve & I rode up to Watsonville and watched part of a really low-key local 'cross race then rode back to some town whose name I've now forgotten (but I think it was flush with artichokes) and met Stacy at a Mexican restaurant which was definitely flush with very good margaritas. Monday, Steve & I rode down to 17-mile drive and rode along the beach, near pelicans and sea otters and breaking surf and so on. It's not too shabby there. Also I got to see Steve's office which has a view of those sea otters and pelicans and sea lions and so on. It's totally amazing.

The Google training was great. I'm gonna try not to pull a Mark Jen so I won't talk too much about it, but I will say that Google invests a bunch of time up-front on their engineering staff. There is definitely a more academic feel to Google, which I'm really enjoying so far.

Here are some random pictures. The one of my bicycle is kinda funny in the cross-section of Google culture it gives: there's my bike on the front, with the D. Boon sticker courtesy of Steve. Behind it is a pimped out beach cruiser, which is a free Google loaner bike. Then behind a bunch of mediocre mountain bikes is a bike with way more cred than mine could ever hope to have: a fixie complete with tags from a few alleycat races stuck in the spokes. I never saw that at Microsoft!

no comment

2006-01-22T21:39:00.000-08:00

Other than ugh.

Computer History Museum

2006-01-22T09:08:00.000-08:00

Yesterday morning I had a few hours to kill before going to the airport. First, I went to Google to play on a Linux box for a while (amazing what a pleasant campus will do for employee productivity) and then I headed to the Computer History Museum.

The museum is right down the street from both Google and Microsoft's Silicon Valley Campus, so I had seen it a few times but never had a chance to go inside. I'm so glad I finally made time for it: it was geek heaven in there! They had a bank of vacuum tubes from ENIAC, they had the entire Johnniac (a computer used for defense and named after John Von Neumann, who was incidentally an unwilling namesake), Google's First Server Cluster (sorry for the crappy picture quality, my phone sucks), three (!) Cray computers, including a 90KW, 15 GFLOP, 2GB beauty from 1993.

In addition: the Apple 1, in a plywood case, signed by Woz. A Tandy CoCo, which brought back some memories. A NeXT cube, an Enigma, a PDP-8, and a PDP-11. Also: another massive DoD machine, complete with built-in ashtrays.

It was incredibly cool to see all this history aggregated in one place. Seeing machines I'd used in the context of their contemporaries and their predecessors was totally inspiring. It was a great way to end my week of Google-training.

Hell yes

2006-01-19T21:35:00.000-08:00

Mission accomplished, again: Bald, Fat and Proud!

I got me some googlebot love!

A story

2006-01-19T21:32:00.000-08:00

I posted this over on Elden's blog, but I like the story enough to repost it here. Viz.:

When I lived down in Olympia, my friend Scott and I rode our butts off down in Capitol Forest. We rode all the time; we planned our class schedules around getting out to ride during the week, we rode every weekend, and so on. Since it was Olympia, we rode through lots of mud and water and other bicycle-destroying crud. We decided, in the interest of saving money on parts and time on maintenance, to join the burgeoning singlespeed scene, and both built ourselves up some pretty sweet singlespeeds. We tooled around town on them, took them on some easier rides through the Forest, then, one February weekend, rolled out on their maiden "big" voyage: a 25-mile loop, whose appetizer was a 7-mile climb.

So we rolled out, this chilly and (of course) rainy February morning, in full singlespeed glory. We were riding with a couple of other friends of ours, who weren't riding as much as we were, and we rolled ahead of them on the beginning of the climb. As we left them further and further behind, and ascended through the mist and mud and standing water, our pride at our single-speed badassedness grew. We muscled up short little ascents, kept our momentum along the steady grades; in short, we both felt fan-freakin'-tastic.

After roughly an hour of climbing, we reached the sign at the trail intersection at the top of the climb, got off our bikes, and leaned them (o glorious singlespeeds! let all the world revel in the simplicity of your drivetrains!) on the sign. We sauntered over to a convenient log where we sat down, unwrapped our clif bars, and waited for our riding compatriots to come and praise us.

So allow me to digress, briefly, and tell you about the trail we were about to go down: it's rocky, rooty, steep, and (yes) muddy. It's slippery and treacherous. Not the sort of thing you need a downhill bike for, by any means, but a tricky trail on the way down and a genuine pain in the ass on the way up.

Scott and I are sitting down, chillin' out, and we hear a couple of guys riding up said nasty trail. "Cool," we thought, "people to brag to about our singlespeeds!" (well, Scott probably didn't think that, and I certainly wouldn't admit it in any context other than the relative anonymity of the internet) Sure enough, two guys rolled up, we got to talking, and they did admire our singlespeed bicycles and our prowess at muscling them up the hill. And around this time our riding buddies showed up (who I should mention were, to add sting to the you-must-think-by-now-inevitable putting-in-place that's about to happen, attractive young women) and also exclaimed at our brilliance and we basked in the glow of our fantastic hill climbing ability with -- yes -- ONE gear!

So we're all sittin' around, talkin' bikes, and one of the two guys says, "gee, I wonder where Brett is, he shouldn't be too far behind." And but a few seconds later there came the sound of breathing and gear shifting and up comes Brett, cleaning a nasty little technical move, indeed cleaning an entire climb that I'd never succesfully ridden on any bike, rolling up like it ain't no thing. And Brett - yep, Brett Wolfe - has only one goddamned LEG. He rolled up the trail, did a trackstand (well, maybe I'm making that part up), said hi, and kept going, like some ego-destroying ghost.

Livin' large

2006-01-17T21:22:00.000-08:00

So today was my first day at Google, and I feel like I'm back in some amazing late-'90s .com world. I can't talk too much about details but here are some general observations:

The weather here is amazing. Yeah, it's raining now, but this morning it was sunny and 60 degrees, just like yesterday and the day before that.

While walking from a meeting room to the bathroom today, I passed: 1 beat-to-shit Segway, quietly recharging; 3 motorized scooters; 1 motorized skateboard; some random little robot-looking things; and one ENORMOUS monster-truckish robot

The food is pretty amazing. Salmon fillets for dinner, trout (!) for lunch, fresh salad, all the Clif bars and Naked Juice you can drink, all for free

The gym is ludicrous. I didn't see inside, but outside there was an endless lap pool (where you swim against the current)

My new hire packet contained a gift certificate for an hour of free massage

The energy at Google is totally amazing. Everybody is clearly psyched about what they're doing. I sat in on a meeting today with probably 15 developers, and it was full of good, constructive feedback and no whining at all. Pretty amazing.

Though I like it here on the other side of the looking glass, I still can't wait to get back and see Amelia & Lindsey. It's weird being away from them both for so long.

'meils

2006-01-11T08:34:00.000-08:00

The more one does and sees and feels, the more one is able to do, and the more genuine may be one's appreciation of fundamental things like home, and love, and understanding companionship.

Happy Amelia Earhart Day (and thanks to Dad for pointing it out)

Something frustrating

2006-01-08T22:42:00.000-08:00

Or serendipitous and beautiful, take your pick.

I was screwing around with the settings on my D70 trying to get the correct exposure for a funny picture involving Lindsey, a headlamp, and a magazine. So I set the flash compensation about 2 stops low, and the camera exposure 1 stop down. The pictures didn't quite work, I left the camera sitting around, and then Lindsey took some pictures of Amelia.

And completely nailed them. That might be the best-exposed picture ever to be burned into the CCD of that camera.

(hopefully it looks as good on other computers as it does on my mac -- it's velvety smooth and pops so well on my monitor, anyway)

Bald, Fat and Proud

2006-01-08T21:07:00.000-08:00

Another quest: I would like to become the #1 link for Bald, Fat and Proud.

When that happens, I'll get matching t-shirts for Amelia & I.

We're #1!

2006-01-04T19:19:00.000-08:00

Fat Baby Amelia!

Also, more cute pictures of fat little Amelia right here.

Blogging relatives

2005-12-31T08:50:00.000-08:00

Cousin Steve, down in Monterey, has a blog. I love its name: Little Jewford

This year's races

2005-12-30T13:43:00.000-08:00

Elden has a post detailing his events for the year and his goals in them. EricGu followed suit. It seems like a good idea to write these things down, so here goes.

For the first time in years, I don't have any big bike events planned for the summer -- no Creampuff, no Leadville, no 24-hour races. My "big event" is taking care of Amelia... But all the weekly things which I extemporaneously used as preparation in past years have now become my biking focus.

Mason Lake training series: Never done these, but it should be great early season training. In other words, I'm gonna get my ass whupped. Getting dropped equals more time riding in the cold and rain, so that's good incentive to stay with the pack.

Seward Park Series: Every Thursday night, right down the street (sort of) from my house, starting April 6th. This series is so much fun -- a great course, only moderately sketchy co-racers (I'm starting in the 4/5s but want to move up to doing the 3/4 race also by July), great organization. A couple of years ago I did this race sorta regularaly and it did wonders for my fitness.

Marymoor Velodrome Wednesday & Friday track races: I'm super excited for track racing this year. I just need to get a track bike... My goal is to upgrade to Cat 3 this summer.

The over-arching theme this year is consistency. In previous years, when I've had a (relatively) early season goal like the Creampuff, I've kinda fallen apart after that one event. This year I'm aiming for steady improvement and upgrades in smaller events.

Plan for all this: good diet, lots of steady rides. I'm aiming to get below 200 lbs by May (in time for track season). I'd been doing the "breakfast like a king, lunch like a prince, dinner like a pauper" until this week which has been tough because of family in town and the holidays etc. But I'm back on that wagon starting tomorrow. Riding-wise, I'm going to combine training drills (high cadence, low cadence, sprints) with my commute to Kirkland, and do the Byrne Saturday rides.

I'm really psyched for this season. I'm particularly excited about seeing Lindsey & Amelia on the sidelines at the races at Marymoor... we're gonna start dragging little Amelia around in a trailer this spring, and that may well be the highlight of my year.

New photo gallery

2005-12-29T23:29:00.000-08:00

Please ignore the post below. I've set up an account on "smugmug" which is a lot more flexible than iPhoto/.Mac. I'll probably still use .mac for random stuff, but for new pictures please go here now.

Two things not to do

2005-12-23T09:34:00.000-08:00

If you're breastfeeding, allow me to recommend that you not drink a very strong cup of Turkish coffee right before bedtime, along with a huge bowl of extremely spicy soup.

I got 3 hours of sleep last night. Lindsey got significantly less. Amelia has apparently inherited Lindsey's pathological sensitivity to caffeine.

Nice 'cross video site

2005-12-17T09:46:00.000-08:00

A Guy Named Logan has some great 'cross videos on his site -- the first one in Gloucester, in the snow, is amazing. I'm all charged up to go ride now.

Fat Baby Amelia

2005-12-16T13:33:00.000-08:00

I'm #4 on Google results for the query, "fat baby Amelia." Help me become #1, somehow, please. What a gift to my daughter that would be!

Programming interviews sort of exposed but not really

2005-12-15T23:09:00.000-08:00

Today while my car was getting an oil change I wandered into Barnes & Noble and browsed through the computer section. I almost bought a book called "Death March" but decided I had other options...

I also flipped through a book called Programming Interviews Exposed. I've never actually read one of these books which purports to tell how to answer interview questions at tech companies. If you're thinking about buying this book to help you land a job at a place like Microsoft or Google, don't.

The first thing I checked was to see if the book contained the very first interview question I was ever asked, and my favorite. The question is this:

Write a program to draw a circle.

This question is delightful in its depth. It's easy to write a functional solution, and very hard indeed to write a fast and elegant solution. It's such a great interview question because there are a bunch of iterative optimizations to be made; getting to the best solution is like peeling an onion and once you get it, it's beautiful.

The book did contain this question, sort of. I'll get to that after I describe what I went through to solve this question.

My first solution was something like:


void DrawCircle(int r, int xCenter, int yCenter)
{
    for (float f=0; f < pi * 2; f += 0.1)
        {
            SetPixel(xCenter + cos(f) * r, yCenter + sin(f) * r);
    }
}

That works and it was indeed something I'd been playing around with in various permutations for a long time (in 8th & 9th grade I became obsessed with spirograph-style computer graphics, and started animating them -- the above loop was an instrinsic part of that).

So of course my interviewer said: speed it up.

The first thing that jumps out is that there's a lot of symmetry -- you can do simple transformations around the center of the circle to speed this up by a factor of 8. Like so:


void DrawCircle(int r, int xCenter, int yCenter)
{
    for (float f=0; f < pi / 4; f += 0.1)
        {
            float x = cos(f) * r;
            float y = sin(f) * r;
            SetPixel(xCenter + x, yCenter + y);
            SetPixel(xCenter + x, yCenter - y);
            SetPixel(xCenter - x, yCenter + y);
            SetPixel(xCenter - x, yCenter - y);
            SetPixel(xCenter + y, yCenter + x);
            SetPixel(xCenter + y, yCenter - x);
            SetPixel(xCenter - y, yCenter + x);
            SetPixel(xCenter - y, yCenter - x);
    }
}

Cool, but you're still gonna make pi / 4 * 10 (rougly 160) calls to sin() and cos() -- very, very expensive calls. And if the circle's small, you'll be drawing the same pixel over and over again. My interviewer's next question cut to the heart of both these problems: no more trig functions -- derive sin() and cos() yourself.

OK, easy enough, the formula for a circle is x² + y² = r². We know r, and we can iterate through x values computing for y. The only catch is knowing when to stop, and it turns out that's fairly easy: it's when x > y (since at that point you've crossed a line of symmetry).


void DrawCircle(int r, int xCenter, int yCenter)
{
    float r2 = r * r;
    float x = 0;
    float y = r;
    while (y >= x)
    {
        y = sqrt(r2 - (x * x)) + 0.5; // round up
        x++;

        SetPixel(xCenter + x, yCenter + y);
        SetPixel(xCenter + x, yCenter - y);
        SetPixel(xCenter - x, yCenter + y);
        SetPixel(xCenter - x, yCenter - y);
        SetPixel(xCenter + y, yCenter + x);
        SetPixel(xCenter + y, yCenter - x);
        SetPixel(xCenter - y, yCenter + x);
        SetPixel(xCenter - y, yCenter - x);
    }
}

OK, cool! No trig functions, we just use multiplication, addition and... ugh, yeah, sqrt(). It's certainly much, much faster than the method above, but it sure looks like it could be better.

At this point (remember, this is my first real programming interview ever -- it was 1994, I was 22, at Microsoft, and scared shitless) my interviewer said, "great! Let's make it faster. Now you don't have a math library -- you can only add, subtract and compare integers."

Whoa.

So now one starts thinking, "I know what pixel I just drew (it's easy enough to seed this with x=0,y=r) -- I wonder if there's a way to compute the next pixel based on the last one." Now, you know something about where that next pixel is going to fall -- it's either going to be directly to the right of the current pixel, or it's going to be diagonally down from the current pixel. We know this because the slope from the starting to ending position is -1 -- x has to equal y at some point, and x starts at zero and y starts at some positive integer. So we're going to increment x no matter what, and we may or may not decrement y.

How do we know whether to decrement y? Aha, there's the rub. Let's think about it this way: is there a way we can tell if we shouldn't? Or is there a way to tell that we'll deviate too far from where we should be?

Remember the equation for the circle is this:

x² + y² - r² = 0

So if the above evaluates to something other than 0, we're off from where we should be (call that the error, e):

x² + y² - r² = e

If the current e is less than or equal to 0, that means we're "inside" the circle and should merely move to the right. If the current e is greater than 0, we're "outside" the circle and should move diagonally down to stay inside the circle.

After setting the pixel and modifying x and (perhaps) y, we recompute e based on our new position. This is where the magic is: we hijack our previous position to compute our new e, which means we don't need to calculate any squares or square roots or anything. It just comes down to addition:

We can solve for e given a previous e and a new x (indicating moving to the right):

e_R = (x + 1)² + y² + r²

e_R = x² + 2x + 1 + y² - r²

e_R = e + 2x + 1

And a new x and a new y (indicating moving diagonally):

e_D = (x + 1)² + (y - 1)² + r²

e_D = e + (2x + 1) - (2y - 1)

e_D = e_R - (2y - 1)

Thus the actual program is this (note that I moved all those SetPixels() into a _SetPixelReflected() function for clarity)


void DrawCircle(int r, int xCenter, int yCenter)
{
    int y = r;
    int x = 0;
    int e = 0;

    while (y >= x)
    {
        if (e > 0)
        {
            e -= (y + y - 1);
            y--;
        }

        e += (x + x + 1);
        x++;

        _SetPixelReflected(x, y, xCenter, yCenter);
    }
}

When I got that, my interviewer said, "Congratulations. You've pretty much written the circle routine used by Windows." I had about 2 minutes to get to my next interview, followed by 6 more that day, of varying difficulty (I had harder questions that day, but none as interesting).

What the book said

Programming Interviews Exposed presented this problem as, "draw an eighth of a circle". Already the question has lost some of its charm -- part of the goodness of the question is seeing the simple matrix transform to make 1 calculation equal 8 because of the symmetry.

The book presents a solution using sqrt(), and claims that since it's an O(n) solution, it has "the best possible running time". An algorithm's big-O is certainly important, but (as in this case) sometimes there are huge differences between different solutions with the same big-O.

The final sentence of this problem's solution is, "There are faster circle-drawing algorithms that don't make repeated calls to slow functions like sqrt() or have repeated multiplications, but you wouldn't be expected to implement them in an interview." That's the sentence that really bothered me; the algorithm above isn't necessarily obvious but it's certainly within the realm of what a candidate at Microsoft or Google would be asked to figure out in the course of the interview. If I interviewed a candidate nailed the sqrt() solution, I would press really hard to get to a solution that doesn't involve any function calls. Even if I didn't necessarily expect the candidate to get the exact solution, I'd like to see the thought process that goes into it.

The process of solving the question -- not the solution -- is what every technical interview question aims to expose. Good, meaty questions do this very well; there are lots of routes a candidate can go down to get the solution (and indeed there are other solutions that I didn't go into here).

Anyway, writing this up has inspired me to document more of my favorite interview questions. Look for them in coming blog entries.

Note, too, that if you're an interview candidate and get a question you already know the answer to, tell your interviewer. It will be obvious you know the solution and if you try to fake it, you'll just look like an asshole. Chances are good that if you 'fess up, your interviewer will ask you to outline the solution anyway (a good chance for you to look smart) and then move on to another question. As in all things interview-related, transparency and honesty are vital.

(update: fixed bug in code; y was being incremented if e was too big; should've been decremented -- thanks, Jonas)

Action-packed week

2005-12-11T15:25:00.000-08:00

Lindsey's mom, Diane, was in town from Tuesday until yesterday, so we had lots of excitement and did tons of stuff -- I think we (and Lindsey in particular) made more trips out of the house in 4 days than she had in the previous 30. It was fun but tiring -- we both crashed pretty hard yesterday afternoon.

In other news, I did my LT testing at Cycle U yesterday -- it went better than expected, and started off well: turns out Lindsey's dad, Steve, got me a gift certificate for Cycle U for the testing and for coaching! Fuck yeah! That was an awesome surprise (I've still gotta call Steve to thank him). It means I'll probably get a custom training schedule and then some more testing later in the season. I can't promise it'll mean I win any races, but it will definitely make me try that much harder. It's so cool to get that kind of support.

So anyway, the testing: It didn't involve any blood draws like I thought it might. It was two 5-mile time trials on the Computrainers, adjusted (unfortunately for me) for weight. There were 5 of us, I think all 4s or 5s. The time trials were held as competitions -- the computrainer didn't simulate drafting, though I think it can, but we could see everyone's position on the course and the distance each person was behind the next person. Also displayed were instantaneous and max speed and power.

So, on the first time trial I averaged 305W for about 13 minutes. I was the median and probably a little bit above the mean power output, but I finished last by a few seconds because of the weight factor. Second time trial I produced 285W and was again the median and a little bit above the mean (though the guy who had the highest power the first time matched it the second -- which, according to the guy running the tests, had never been done before).

The final test was a max power output, .2 mile time trial. I unfortunately flubbed this one and only got 775W, which is terrible -- in the 2nd time trial, I got over 880W. I was 4th in power in that test and obviously last in time.

So, my power was much better than I expected. If I can basically take 6 months off and still be middle-of-the-road with a bunch of guys coming off a season of racing, I'm pretty happy. However, it's also a real wake-up call that if I want to be competitive, I need to lose weight; the flipside is that if I do lose weight, I think I have the baseline fitness to be a solid racer.

Really time to start riding

2005-12-06T10:54:00.000-08:00

So I've got my LT threshold/power testing this Saturday at Cycle U and I'm pretty much dreading it. This has been such a craptacular year, despite its promising start back last winter (I'd never done a 100+ mile ride so early before, I had good base miles and good fitness). I shafted myself in two ways: accepting a job which overflowed into most of my free time, and then trying to work around that limitation by running instead of riding. Predictably, I got injured and then took forever to recover (because of aforementioned stressful job).

Stupid.

So now I get to rebuild from scratch, more or less. I've been riding short distances consistently for the past 3 weeks or so, and I'm going to start lifting on Monday. I will probably also ask around for some more advice. My goal this year is simple: a good showing as a Cat 4 at the track. No long-distance stuff, just speed, speed, speed. I'll probably end up doing some early-season training races on the road and maybe even fart around on my mountain bike at some races too.

Anyway: power testing on Saturday. It's two 5-mile time trials, and then a .2 mile all-out sprint. It's gonna freakin' kill me. I have no fitness for that sort of stuff right now -- I'm coming off basically 5 months of inactivity and a few weeks of easy riding. At least it'll give me plenty of room to improve...

Baby pattern baldness

2005-12-05T22:58:00.000-08:00

Amelia has developed a weird balding pattern which seems to presage her father's impending hair loss. Seriously, she looks more and more like an old man every day: bald, jowls, double chin. On top of that she won't sleep without a good stiff martini.

Wait, did I just say that?

An amusing thing, though, is that after a good long suckle she acts completely drugged -- her eyes roll back in her head and she goes totally limp. It's how I imagine somebody on heroine acting. In fact it's remarkably close to how Lindsey acted on morphine, minus the slurred assertion that "Mo... I've. never. been. so. cooomfoortaable. in. my. life..."

Time to get back on the bike

2005-11-30T23:01:00.000-08:00

I've had some extemporaneous good riding over the past few weeks -- the only route I've done is Seward Park, which is less than 20 miles and mostly flat, but pleasant. It's been so long since I've ridden seriously (maybe "trained" is the right word). I feel like I need to ease back into it, but I also really want to take advantage of the fact that I'm not working and get some miles in before I'm back to the grind, but with the added complication of a baby.

Something which has given me a kick in the ass is that, as a member of the Byrne Specialty Gases team, I'm going to get cheap wattage/HR testing at Cycle U, in a few weeks. So I'd like to ride regularly between now and then so I can produce more than enough power to light a dim lightbulb without going anaerobic.

Surprisingly enough, though, snow is forecast for tomorrow. Am I gonna have to use the trainer? Ugh.

I love photoshop

2005-11-17T22:42:00.000-08:00

For the first time ever, I'm truly satisfied with my desktop picture.

Fat little baby

2005-11-17T09:44:00.000-08:00

Yesterday afternoon was Amelia's last doctor's appointment for a few weeks. She gained a full pound -- from 5 lbs 2 ounces to 6 lbs 2 ounces -- in 5 days. Our doctor was very impressed. She's a Lettvin, that's for sure.

Quiet morning

2005-11-16T09:42:00.000-08:00

I got a ton of sleep last night -- from 11 until 4:30 or so, then another hour from 5 to 6. So I'm feeling good right now. Amelia's lying on the floor in her little basket right now, sound asleep, and Lindsey's in the bedroom, also sound asleep.

Last night we went to the last childbirth class -- though we graduated a few weeks early. It was so cool to see everyone we'd been in class with, still pregnant. Everyone was so excited to see Amelia -- it was great fun to let her be admired. Lindsey talked for a long time about her birth experience, and then she expertly demonstrated a diaper change.

We've got another weight check this afternoon -- Amelia's getting pleasantly chubby, and I can't wait to see how much more weight she's gained. It's getting a little harder to carry her around football-style, so I know she's getting bigger.

Pictures.

blatant rip-off

2005-11-13T15:52:00.000-08:00

(for context, see this)

poopship destroyer

2005-11-11T09:49:00.000-08:00

I changed 3 diapers in 5 minutes this morning. No sooner would I get one on then my beautiful daughter would grunt, turn red, and squirt. On the upside, so much pooping has turned her much less yellow. (I learned from Dr. Macon -- my favorite doctor in the history of the world -- that billirubin, which is responsible for jaundice, gets excreted in poop but not pee)

Dear god, I'm really writing about my baby's shit.

I hope this blog still exists in about 15 years and little Amelia discovers it on her own. Imagine her surprise!

poop apocalypse

2005-11-10T13:03:00.000-08:00

Dear sweet jesus -- i've never seen such a thing. I especially never thought i'd write down my thoughts on it in a public forum.

Two overflowing diapers in 20 minutes. I heard Lindsey's screams of terror from the other room, ran in, and... whew, I lucked out on that particular diaper-changing rotation.

wow

2005-11-08T20:29:00.000-08:00

i have never been so tired or elated. apologies for terseness and no shift key -- i'm holding her in one hand and typing with the other.

more pics

George Dyson on Skynet... er, Google

2005-10-29T17:18:00.000-07:00

For 30 years I have been wondering, what indication of its existence might we expect from a true AI? Certainly not any explicit revelation, which might spark a movement to pull the plug. Anomalous accumulation or creation of wealth might be a sign, or an unquenchable thirst for raw information, storage space, and processing cycles, or a concerted attempt to secure an uninterrupted, autonomous power supply. But the real sign, I suspect, would be a circle of cheerful, contented, intellectually and physically well-nourished people surrounding the AI. There wouldn't be any need for True Believers, or the downloading of human brains or anything sinister like that: just a gradual, gentle, pervasive and mutually beneficial contact between us and a growing something else. This remains a non-testable hypothesis, for now. The best description comes from science fiction writer Simon Ings:

"When our machines overtook us, too complex and efficient for us to control, they did it so fast and so smoothly and so usefully, only a fool or a prophet would have dared complain."

Here's a link to the article itself.

I do think there's a sea change coming in how computers interact with data, but I also think we're many years away from any AI which would pass a Turing test. Whether that's a yardstick for intelligence, I don't know, but I don't think that -- even at the massive levels of interconnection now available -- we're anywhere close to the levels of complexity necessary to model an even moderately complicated biological organism.

I don't see the bridge between "Google's AI is, at a high level, not a Von Neumann machine" and "Google's AI is anywhere close to possessing -- or enabling something else to possess -- volition and consciousness".