Sunday, September 30, 2012

Sunday: blah?

My sushi exploded. :(
Remember last week when I made fun of WF for displaying the dried mushrooms priced by the pound?  They fixed it:
Which makes me think that maybe WF reads my blog.  Stop stalking me, WF!
Who keeps bread in a pot?  That's just stupid.  And if you do that, why would you need two?

Thursday, September 27, 2012

Thursday: Memory or processor

Today I found a bug that I can either fix by calculating a bunch of numbers twice (which is expensive), or by saving those numbers in a big NxM array so I can (effectively) do sums along both dimensions.  I'm fairly sure that the memory way is the "best" way to do things, but it's going to be a bit clunky.

Interesting trick I came up with today:  I have really crappy earbud headphones at work, and I travel with a better pair.  The biggest problem I've had is that they're identically shaped, so it's hard to figure out which is the left and which is the right.  This is doubly hard on planes, since they tend to be dark.  The solution: take the left earbud, and tie it into a simple knot around the cord to the right.  This makes the left cord shorter than the right, so by pulling to the end, the left earbud hits your hand first, so you know which one is which.  Problem solved, and it no longer requires looking at all.

Eh, close enough.

Wednesday, September 26, 2012

Wednesday: Decade old data

Technically, it was a decade ago that I received the data.  It was taken a decade before that.  Yay, science? In any case, other than a larger outlier distribution than I'd really like (but don't want to have to look into too closely because there are like 5000 fits to check), the core Gaussian distribution says my code is accurate to about 2%.  Cool.

It turns out, yes, the data wanted some science in its face.
The crazy thing is, this analysis took less than two minutes to run, compared with the original analysis, which took a summer.

Yay, computers?
I did not use the squirrel tech support system.
Nor did I use bird/bread tweet/keyboard.

And, just because:
"Bark bark bark!"  "Dude, ok, I'm going to be the better guy here, and just go, because I'd really feel bad shattering your body with this paw.  No, paw, no.  Don't wind yourself up.  We're just going to let this dog think that he's the greatest thing ever.  You're a jerk, dog.  I don't know how you sleep at night.  Me?  I'm going to sleep wonderfully tonight, knowing that I'm not you."

Tuesday, September 25, 2012

Tuesday: Totally still Tuesday


You're a jerk, water balloon guy.  Little squirrel doesn't need that.

Wait, no.  This isn't right.  Those aren't proper squirrels.  All wrong.

I've met boston terriers.  This doesn't seem like the best idea. Or the most efficient use of the space budget.

Monday: I'm going to math at you

Copying comments from a source file is totally the same as writing a blog post, right? Has anyone really considered the full-width N-max of a gaussian before, and then extended that parameter to non gaussian data?  Here's how they would start, if they did:

/* Discussion on FWHM for Gaussian lines:
   Given a gaussian distribution defined by PDF = 1 / sqrt(2 * pi) * exp(-0.5 * u^2),
   then you can define a fractional max location by:
   PDF / PDF_MAX = exp(-0.5 * u^2) / exp(-0.5 * 0^2);
   PDF / PDF_MAX = exp(-0.5 * u^2)
   sqrt(-2.0 * log(R)) = u
   So, for a given ratio of the maximum R, the half-width R-max is located at u.
   This leads to the full-width R-max location being at
   FWRM = 2.0 * sqrt(-2.0 * log(R))

   Since we've used u, we can add the sigma dependence back in by noting that u = x / s:
   FWRM = 2.0 * sqrt(-2.0 * log(R)) * sigma
   The standard value is the FWHM, or R = 0.5:
   FWHM = 2.0 * sqrt(-2.0 * log(0.5)) = 2.0 * sqrt(2.0 * log(2.0))

   Do this for quarter max, R = 0.25:
   FWQM = 2.0 * sqrt(-2.0 * log(0.25)) = 2.0 * sqrt(2.0 * log(4))
        = 2.0 * sqrt(2.0 * 2.0 * log(2.0)) = sqrt(2) * FWHM
   And for 3/4 max, R = 0.75:
   FW3QM = 2.0 * sqrt(-2.0 * log(0.75)) = 2.0 * sqrt(2.0 * log(4/3))
         = 2.0 * sqrt(2.0 * (log(4) - log(3)))
         = 2.0 * sqrt(2.0 * (2.0 * log(2.0) - log(3)))
         =~ 0.64423 * FWHM

   If a distribution is "wingier" than a Gaussian, then FWQM will be inflated, such that
   FWQM_wingy > FWQM_Gaussian


Here's a sleepy bear.
And here's Neil Armstrong making pizzas before flying to the moon and becoming super cool.
  • OCD dog.
  • Snorlax is best.
  • Comics are run by morons now, so it's good that someone I've liked is getting out of the thrashing.
  • Also on comics: really? Facepalm. Followed by washing to get the dirt off.
  • Politics (also fluid dynamics): Really?  Facepalm.
  • How long will it be before people in power realize that dragging people into the street to charge them with minor traffic violations is not the way to crush dissension.  Forcing them to come together and present a coherent set of goals with a means to obtain them is far more likely to crush the rebellious feeling.
Mind you, don't take that as a disparagement of the occupy movement, which I fully support.  I'm merely commenting that if I were to want to put such a thing down, I'd do it not by police action to arrest people, but rather by forcing them to deal with impossible situations and come up with solutions that are likely to not fully solve the problem, thereby marginalizing the dissidents as ineffectual.  That's the true way to defeat dissent.

Sunday, September 23, 2012

Sunday: Do I have a telecon tomorrow?

I think no.  I think I had one last week.  Let's say no.

I saw this guy yesterday.  I'm bad at surreptitiously taking pictures of guys who have clearly devoted their lives to dressing up like Elvis.

At this point, I think you can just round it up to $300/pound.
 Pictures I didn't take:
Clearly some people are complete idiots.

"Mark my words, human. Once my horse gets me over there, I will devour you."

Friday, September 21, 2012

Friday: Fuck yeah, cider!

Fuck yeah!
 Yes, I did buy this half gallon of cider this evening after work, less than three hours ago.  Yes, I paid like $6 for it.  Yes, it was fucking delicious.
So it's a blend, and not a single apple special.  It's fucking cider and I'm drinking it, and it's delicious.
 I hope that that "As Fresh As An Apple all year round" thing means that this is going to be available here all year round.  I guess if it's still around in December, it's probably a permanent addition.  Until then, I'd best buy as much as possible and drink it all, just to be safe.

"Huh.  This is awkward.  I forgot that I invited you."
Also, Fuck you, Mitt Romney.  You paid 14% of your income in federal taxes.  Remember when I did my taxes this year and ended up paying 14.6%?  We can also factor in the fact that adding payroll taxes to your bill doesn't change things much, but it does change my effective rate quite a bit.  This is why you're out of touch.

And how the hell is the median worker paying so much less in taxes than me?  Is that the child and mortgage credits kicking in to help them?

Today was going to be "Draw a Kitty" day, but I don't feel like it right now.  Let's move that to tomorrow.

Thursday, September 20, 2012

The first thing I thought

Thursday: Headache

Surprise sandwich!

Surprise bear!

Wednesday, September 19, 2012

Wednesday: Not so good leftovers

So that chorizo taco stuff I made yesterday didn't reheat very well.  Crap.  Now I'm trying to figure out if I just want to trash it and make something else for tomorrow dinner.
Making supremes of lime is really easy to do.
I finally got around to setting up my mail correctly this evening, as well.  Now I get a copy automatically forwarded to my computer at home, and my phone is able to sync against the work server copy via IMAP.  This probably means I'll have to manually mark things read on both sides, but it gives me an easier way to read mail.  Mostly this is a problem with mutt and my phone fighting over IMAP locks.


Tuesday, September 18, 2012

Tuesday: Problem solving!

I felt better today, so I made it in to work for the afternoon.  I decided to switch to problem B, which is similar to problem A, but different.  It turns out that this was the correct thing to do, as I was able to solve problem B, and use that solution to discover the solution to problem A.  That puts me far ahead of my expected weekly progress, even including missing yesterday.  Woo.

I've been thinking about making tacos for dinner this week, but couldn't remember what spices I'm out of.  I decided to skip this problem by making chorizo tacos.  A google search led me to the realization that this is apparently common with potatoes.  While doing so, the first image on this page came up during my search.  Who the fuck models t-shirts without pants on?

The tacos came out ok, but I'm going to need to think up ways to perk them up a bit.  I simply fried an onion, three small potatoes, and two things of chorizo (which turns out to be 20oz, not 1lb as I thought).  This works ok, but is a bit too monotone.  I bought an avocado, so I'll add slices of that, and I also picked up some limes, so maybe some lime juice, or thinly sliced lime supremes?  It needs something.

This sounds like an analogy for something.

The cutest omurice ever.

"Jerry, just stop. They're never going to come down far enough for you to get them."

Monday, September 17, 2012

Monday: Teh Dethh

I woke up this morning feeling kind of crappy, made it in for the morning telecon, and then went back home because the pounding headache and scratchy throat made me think that work wasn't going to happen.  I stopped off to get the required sick day supplies (soup, juice, something sweet to snack on when I felt better), and then crashed on my couch.  I woke up for lunch, and then again for dinner.

I still have the scratchy throat, but when I checked, I don't have a fever.  I think I'll have to see how tomorrow works.

I did get to see the news to see that this guy is never going to be president. Because he's apparently a vindictive douchebag.

"Oh man, I love cello music!"

"Just pass him! Just pass! Hey! Get out of the way! I'm in a hurry!"

  • Never going to be president.
  • Two different ovens?  Maybe one is a microwave?  Is that a radio built into the wall that the coffee pot is plugged into?  Plus, that fridge eats up what could be a decent sized island.  With doors on both sides, it has to be horribly inefficient.  Plus, what's to keep you from pushing a ham in the front and shooting a cake out the back (because it's too full, not because it's some magic meat<->sweet converter).  Finally, those kids look like mom is forcing them to attend omelette school.
  • tawpie. Totally a real word.
  • I kind of want to make burgers after reading this article.  I want to make one with a crispy crust.

Sunday, September 16, 2012

Sunday: I did absolutely nothing today

Successfully, though, so at least the day had that going for it.

What are you doing, squirrel?

Cheer up, bear. I'm sure next week will have better internets.

Saturday, September 15, 2012

Saturday: I still need a menu plan

I thought I was doing reasonably well coming up with the plan for today: go to Wendy's for lunch, and then to Target.  On the other side of the island.  Sometimes it's just nice to drive.
Yes, the frosty is there just for dipping the fries.

One link?

Friday, September 14, 2012

Friday: Seriously slow week

I spent part of the evening attempting to find a post that I thought would be searchable under "microwave," but it turns out blogger search sucks, and I only used the word "microwaves."  WTF, blogger?

Anyway, I discovered today that if I press the "Add 30 seconds" button on my microwave, it immediately starts cooking, allowing me to add subsequent 30 second amounts until I reach the desired time limit.  Therefore, my microwave has essentially become a single button device to me.  I do not require my foods to be cooked with greater than 30 second precision.

Yes, that was the most exciting thing to happen today.

I apparently only keep a one day surplus supply of bear photos.  Hopefully the internet will be fun tomorrow.

Thursday, September 13, 2012

Thursday: slow week, I guess

Although yesterday was the Day of Second Impact.
Unless you believe the cover-up story, then it's today.

Wednesday, September 12, 2012

Wednesday: I have nothing except pictures

Did I post this one already? Walrus dog. 
"I'm on vacation. Everyone knows bears can swim."
"Can't fly, though."

Tuesday, September 11, 2012

Tuesday: I drew this picture yesterday

It's supposed to be the projection of two images onto a distortion corrected frame, with two super pixels that have failed the interpolation and so have bad values.

Also: it looks like a robot with a silly hat.

  • Squirrel!
  • Seriously? It's just mustard. I don't need to join a fancy club for fucking mustard.  I have like eight varieties in my fridge right now, Grey Poupon. You're not even the best.
  • Reading the comments on this story is painful.  The main arguments seem to be "teachers don't work over the summer, so fuck them," "teachers have a decent retirement plan, so fuck them," and "fatcat teachers don't need to worry about being fired on a whim, so fuck them."  Whatever happened to solidarity?  When did it become a bad idea to make sure that the people who make sure children don't grow up stupid are taken care of?

Monday, September 10, 2012

Monday: How can a ten minute nap totally reverse my sleep?

I spent most of the day about to fall asleep.  Then, upon getting home, I did fall asleep. For like ten minutes.  Now I feel totally awake and anticipate problems getting to actual bed.  Great.

Also: how's this for a problem/algorithm: Given a data set that is contaminated in one direction by bad values, determine the best mean level if the contamination were not there.  First choice is probably going to be a median, but if the contamination is sufficiently prevalent, you're likely to end up biased in the direction of the contamination.  Alternatively, you can take the assumption that the noise in the data set is normally distributed, and therefore is symmetric.  This leads to the suggestion that any asymmetry in the observed distribution function of the data is due to the contamination.

Ok, so how to symmeterize the data (yeah, that's probably not a word)?  Construct the median subtracted CDF, and then construct the a CDF (reflected around x = 0 and y = 0.5).  CDF1 tells you what the data says the distribution is, and CDF2 tells you what the opposite side of the median looks like, telling you how asymmetric the observed distribution is.  So you can just reweight the values on the contaminated side by the ratio of the CDF values at that point CDF1 / CDF2 (properly accounting for CDF2 == 0.0).

However, this will basically make things left of the median look like the right of the median, resulting in a median that's the median again.  So, let's switch the median to a weighted mean.  This should give a result closer to the unbiased distribution peak.

One concern with this is that you can't iterate it.  Iterating will just increase the fraction of low weight values, eventually pushing the "mean" to the maximum value.  This is a bit of a concern.  Maybe I should run further with the assumption of normality and come up with something that's erf based?

Hey look! Bears!

That's totally a super villain lair.