For anybody who follows LJ more closely than Facebook or Twitter: I finally finished my PhD! (Okay, technically it was a couple weeks ago and I'm being late posting it here, but what's a week or two over the course of 6 years?)

Just had my first two golf outings this weekend... 18 holes at Fountain Springs and 9 holes on the par 3 at Pipestem State Park. I shot a charitable 120 at Fountain Springs -- there was nobody behind us so my dad and brother gave me several mulligans, and ultimately on hole 17 I dropped on the fairway after hitting four balls into the weeds, never to be seen again. (I think I ended up only taking 6 on the hole, because it was a par 3 and "you can't record more than double par," but the actual count would've been more like 12 -- like I said, a charitable 120.) And I shot a fair, if not particularly respectable, 46 on the par 3. I think 4 of the holes I had to record a 6, 2 of which were pity for going over and 2 were legitimately 6s.

Also, somehow I managed to lose my pitching wedge... had it and the putter near the green, must've left it on the side when I went to putt out, and then somebody picked it up behind us.

Lessons learned:

- If I could consistently hit the ball anything like straight, instead of 30 degrees off to one side, I could do okay.
- I can't putt for anything. On the par 3 course, it was usually 2 to the green and then 3 putts.
- On the par 3 I had terrible drives, except for the one time I decided to hit too big a club, connected hard and straight, and completely overshot the green. But I managed to make up for it by hitting pretty good 8 and 9 irons to chip onto the green most of the time.
- It's just about impossible to lose a pink golf ball.

Random nerdy post of the day: Today I wanted to have my code dump its maps at the end of execution, but I was being too lazy to actually open /proc/self/maps and read it myself to write out. So instead, I forked cat /proc/self/maps. Clearly, this was not correct...

It just occurred to me that I didn't make any followup post about my interviewing/job hunt. I accepted a position at Microsoft working on Failover Clustering in Windows Server, so I should be moving out toward Redmond and starting there some time in the fall.

Yesterday's interview was somewhat shorter. One of the interviews' technical question was more abstract about what you have to do if you take a page fault on an instruction fetch. The first programming question was how you determine the next node in an in-order binary tree traversal, given only a pointer to the current node and a node with left/right/parent pointers, and then do an analysis of the runtime. The second was basically about how to delete a node from a binary search tree such that you get back a valid BST. I totally spaced on how to write the code for that one, which was disappointing, although I think from the discussion of the issues I convinced them that I knew how to reason my way through it and that I could probably write it out with no problem if I could just get my head on straight.

I had a phone interview with Corensic today. Just one technical question, on which I did well enough that they decided to fly me out to Seattle at the end of the month for a full on-site interview. The question was given two sets implemented as sorted integer arrays, find the intersection and discuss the run time. Then how would you change it if you know set 1 is always going to be much smaller than set 2. Then how would you change it if set 1 and set 2 are of similar sizes, but they are not sorted. It took me a minute (and a little prodding) to come up with using a hash table, after which point my interviewer mentioned "For future reference, probably any programming interview question that it seems like you should be able to do faster, the answer is 'Use a hash table.'"

Random thought: I'm a little surprised that the RIAA doesn't issue takedown notices for the copies of the takedown notices on ChillingEffects. After all, unless the host takes down all the links instead of just the mentioned ones, the site list on the takedown notice is a great shortcut to finding MP3 blogs.

For anybody who didn't know (see on Facebook/Twitter), I interviewed at Microsoft today.

For anybody who's interested, these are the whiteboard programming questions I was asked:

Implement binary search on an array (the "warm up, stretch your legs" problem)

Implement strtok (which has an annoyingly weird behavior, it turns out)

Implement an iterative postorder traversal of a binary tree, where the node includes a parent pointer. How would you synchronize this? How would you synchronize it if you knew that concurrent reads happen all the time, and a concurrent write happens maybe once a week?

(there was some weird question here about lock-free concurrency that I just totally couldn't follow)

Implement a sparse bit array in a memory efficient manner (this was pretty fun, although I had a "typo" on the board, and the guy also pointed out where I could've saved some memory and removed a conditional block)

"There's a mathematical theorem that if you take any two digit number, reverse the digits, and add them, and then keep doing this with the sums, eventually you'll get to a palindrome. What is the maximum number of iterations it would take to do this, and what numbers needs that many iterations?" (On this one, after going down the wrong path for a minute, I started going the right way, and the interviewer noted that we were running short on time so he said "I can see from where you're going with this you're on the right track now and you'd be able to implement it with enough time.")

Got my first hate mail as a music critic today. Oddly enough, it was for a review from five years ago, and not even (close to) the worst review I've written. Some woman tracked me down on Facebook and asked if I'd heard of the band Control_Z, and when I said yes, asked if I was the one who's slammed them in a review. When I said, "Quite possibly, what's your point?" she replied "That you're an ASSHOLE...that's all!" She then proceeded to complain that I'd written a "novel" about a band that I knew nothing about.

For reference, here's the review in question: Control_Z - System Failure @ Rocknworld.com

Random thoughts about the "success" of the CD I released back in May...

( behind a cut so as not to fill your friend page )

( Wherein I rant about getting screwed by the bank )

Just went to my first Polish class. The professor seems entertaining... very hyperactive. The class will probably teach me something if we stick with it. We're just sitting in right now and have been given the ultimatum to come or not come but to pick one instead of dropping in on random classes.

Back from the honeymoon... back to the grind.

Honeymoon trip planned! Although Kar wasn't paying thinking when she put in my name when buying the plane tickets, so the name on the tickets says Dan while my driver's license/passport say Daniel. Guess we'll have to call up the airline and fix that.

Now to decide what kind of camera gear I need to buy before going 8-)

Got a spot today on Day In Rock's Singled Out:

Singled Out: Synaptic Disturbance's Gallows

Should've asked somebody to come over and water our plants while we were in LA. Either it didn't rain as much as we were expecting, or it was just hot enough to dry it up. Our hanging tomato looked miserable, completely shriveled when we got home, but it seems to have mostly recovered after we dumped water on it. Some of our berries, which are prone to drowning if you over-water them, also looked pretty miserable.

On the other hand, one of our tomato plants finally has fruit on it!

All of our plants are starting to get yellow leaves near the bottom. Root rot, lack of nutrients, something else? Certainly not due to underwatering them...

Since I've got it out for review and just got a physical copy in the mail today, I guess it's time to start announcing it other places than just Twitter. I "officially" released my new CD this week:

Synaptic Disturbance - They Are Killers @ Bandcamp

( pics behind cut )

Explain to me why we have to sign copyright over to IEEE (or ACM, as appropriate) when publishing a paper?

( rant/context )

I'm sure there's some good reason, though... anybody know what point(s) I'm missing that make it totally reasonable?

I've never seen so many people opposed to the idea of killing a groundhog.