|
2009.04.03 in code
The FOSSCamp at UDS : Hardy Heron was, more or less, my reintroduction to the Linux world... I'd been solely using Mac OS X up until that weekend. Robb dragged Gino and I (and Sarah) down to Boston to hang out with everyone there, and see what was going on. Since, I've become somewhat more involved in said world, mostly through contributions to Seed...
Fast forward to this year: Mom and Dad are helping (for Christmas) to send me to UDS : Karmic, so I'm going to Barcelona!!
I've been looking around at hotels and flights and stuff; most places in the area only offer doubles, so if anyone's going sans sponsorship and is looking for a roommate (and doesn't mind Hotel Rey Juan Carlos' 200€-a-night prices), I promise not to be too bad (really, I just might talk about JavaScript a lot... that's all!)! Email me if you're interested (or, even better, if you know of somewhere in the area that's cheaper!)...
2009.03.30 in code
Chromium is surprisingly usable on OS X at the moment. There's some quirky bits in rendering (which are funny because of the whole it's-really-WebKit thing), and you can't reorder tabs at the moment (they drag out and don't reattach to the window). I haven't crashed it yet...
In any case... I'm rather impressed with how far it's come so far, what with nearly nobody ever actually talking about it. Nate is convinced it'll be a worthy competitor to Safari, but I don't understand what a browser could offer that Safari doesn't already :-)
Oh, also... Google's SVN server is a million times faster than Apple's. I checked out all of Chromium (including a whole checkout of WebKit) within an hour. Which is significantly faster than the half-a-day it takes to get from Apple's SVN or Git. And build was really fast, too. But I've never built WebKit+Safari, so I have nothing to compare with (still, I'd say the overhead of Chromium+V8 over what WebKit normally takes to build was minimal...).
Try. It. In a few months...
P.S. Wrote this from Chromium!
2009.03.29 in school
I got all signed up for classes on Thursday. Besides the below, I'm also signed up for Robotics I, but I'm probably going to drop it because it's at 8:30AM three days a week (and is by no means required), and I have plenty of other class. I'm also TAing Inventor's Studio next semester, so that'll take up a good chunk of Wednesday mornings.
2009.03.26 in school
Here's a quick Keynote to set down "what we need to learn" for some people who want to write some iPhone apps. I'm TAing "Inventors Studio" for a few weeks (and next semester), just in order to be a resource who happens to know Objective-C and Cocoa and stuff...
PDF
Keynote
I've staved off our storage problems for a few more months...
2009.03.21 in code
I spent a bit of time trying to get the Intervalometer working, after Dad and I carved up the case while I was home. The board is really a mess; unsoldering the posts I put on from before turned into a real mess, so that's not going too well. We'll see — in any case, I put that off for a day when I'm not soldering at 1AM.
Instead, I got angry at the amount of crap required to get a standalone ATmega working (and programmable) on a breadboard. Even though it's only a handful of components (3) and wires (a bunch), the most annoying part is getting the ICSP cable attached. Right now, I've been just poking wires into the cable, which is annoying because there's no way to unplug it without looking up what goes where, again and again... very annoying.
So I sat down and started putting together a little gadget, based on one of the tiny little bits of protoboard I got from SparkFun. It's got a ICSP header, 16 MHz resonator, reset switch, and power LED, as well as the reset pullup resistor. There are 9 pins on the bottom (one more than optimal, really), in two rows (3 and 6) spaced at normal IC-spacing (0.2"?), so it drops right into a breadboard. Now, it's just 9 (optimally eight) wires to the ATmega, and you've got everything working great (except analog input, but that's just one or two more wires), and can easily remove the programmer without issue, and everything remaining is contained on the breadboard.
It's not as elegant as the little board that slips over the ATmega that I saw on that Arduino blog a few months ago, because the protoboard I had wasn't wide enough to span all of the necessary pins. Perhaps revision 2!
I'm not posting a schematic because it's obvious (and because it's 3AM), but if I revise it to be more interesting, I might...
Lots and lots of bits and pieces:
Just before I headed home for spring break, Mom and Dad decided that Dad and I should go to Florida: STS-119 was supposed to launch on Wednesday night, and there's only a year's worth of shuttle launches left before we never get a chance to see one. We decided to fly down Wednesday morning, and back up Thursday afternoon, and stay overnight at my grandfather's house, in Fellsmere.
Our flight out was at 6 AM, so we had to get up at 4 and head on over to BTV; the security people were irritating as usual, and decided to pat me down in order to find my entirely non-metallic wallet (or is it? duct tape? I don't know... Dad got through with his without any trouble). The flight was mostly uneventful; my ears aren't horribly happy with me, with the pressure changes while I'm coming off a little cold... We had a layover in JFK for an hour, and then the remainder of the flight. Again, uneventful.
We got to Pepa's after a long drive down the highway, just in time to discover that the launch had gotten scrubbed. DARN! Apparently it's a good thing they scrubbed it, though, apparently there was so much hydrogen leaked around the engines that it would have caused an enormous explosion...
After we all took a nap, we heard that it was delayed more than a day (initially they said they'd try again the next day), so we couldn't delay our flight and hang out in Florida until it went. NOOOO! So, unfortunately, that part of our trip was lost. Still, we decided to go to Kennedy Space Center the next day, take the tours, look around, see what we could see...
That's Dad in front of the next-generation Orion capsule (part of the Constellation program that's replacing the Shuttles). There are lots of other cool pictures of stuff over on Flickr, including a picture of Discovery's fuel tank peeking over the scaffolding out on the pad, which was, unfortunately, all we could see.
We flew back Thursday afternoon, after a many-hour delay due to the plane we were supposed to leave on having hit a duck on its way into Orlando... they found us another plane, and we left at about 10:30, though we were supposed to leave at 7... got back at some time after 1 AM...
All in all it was a fun trip — an adventure off in Florida with Dad, and I got to visit Pepa's new house in Florida, and hang out with him and Bernice for the evening.
Battlestar Galactica: everyone at school has been raving about it and watching it, Reddit is obsessed, the Internet is up in arms, and I hadn't watched any of it up until recently. During the past week, I've run through the whole first season, and most of the second, and I have to say: if you haven't been watching it, you must. It's not a TV show, or a movie, or a book. It's a story, and it's amazing.
Jayne mk. II is up and running. I decided not to change the name because coming up with a new name is really hard, and because it shares so many parts with the original Jayne that it really doesn't deserve a new name. All is going well, and I've not yet had any more temperature/power/whatever problems. Awesome!
Anyway, it's back to Troy for me, tomorrow...
It was becoming very clear around the last time I posted about Jayne's troubles that I needed to do something about the seemingly ailing server that hosts this site, as well as Carol's, Matt's, and Robb's, as well as tons and tons of media and backups for the five of us... it's really unacceptable to have your backup machine going down every few weeks (or quicker) with disk errors!
Last week, I ordered the first shipment of parts for Jayne's replacement. I, sadly, had to give up on building my "ideal" machine, a tiny little thing, realizing that such computers are for thermal engineers and Johnny Ive to design, not a mere mortal like me! So I've ended up with an absolutely gigantic case. It's actually a silver version of one of Mike's cases, though I didn't know that until it got here on Friday.
The motherboard, power supply, and video card (oooold NVidia 7000 series, for cheap, since I really only need console graphics but the motherboard doesn't have anything onboard) are coming on Monday (though I don't expect the mailroom to be open then, so Tuesday!!).
My main conundrum right now is what I should do with the old case/motherboard. It's clear to me that the motherboard would be perfectly fine without any storage in the machine, as a net-booted compile/render server, and I could get a new CPU for the new (unnamed as of yet) computer... a Core 2 Quad, or something... but that seems rather counterproductive to the whole having-money-for-potential-late-Spring-early-Summer-plans-thing (UDS?!). That would leave Jayne as a compute node without any storage, which would be cool, but expensive. And considering the fact that Jayne's load average is currently 0.01, it doesn't make much sense (though I'm careful, because of the potential heat problem).
It's unfortunate that it's either that, or leave the board/case completely unused. Oh, well...
Anyway, I'm pretty sure I'm going to move Jayne's CPU to the new machine, for now. Maybe after the summer I'll work something else out.
Well that was a disconnected, early-morning rant. I'm off to sleep now!!
I got a "PACKAGE RECEIPT NOTIFICATION" email today. Not being sure what it was, I figured I'd pick it up on Tuesday, when the package center reopened. However, walking to lunch today, I noticed they kept it open specially, for Valentines Day (or Singles Awareness Day, or whatever, depending if you're Nate, Carol, or the rest of us)... I stopped by, and now I'm the proud owner of two copies of the second revision of the Intervalometer board!
I got it all soldered up today, and it's working wonderfully so far! I made the jump to 3.3V (with one of Sparkfun's new 3.3V LCDs); I haven't gotten output hooked up yet, and won't, today, since I have to finish some stuff with the secondary microcontroller that's in charge of the trigger and output and stuff... including actually acquiring the exact chip I'm going to use there.
Here's a shot of it, all clean-like, with the LCD attached, and power from a 3.3V regulator:
The code/schematic/etc. is all in bzr, the pictures are all on Flickr!
I'll keep writing, especially once I make a box (which I can do now!! finally!)
A few years back, I got the idea that I should archive my email to disk every few years; mostly because Apple Mail used to be sluggish with lots of mail. I suppose this might have been a result of Trillian not having much memory, or Mail just being crappy at the time (noo!), but whatever it was, it wasn't Gmail's fault!
A few days ago, Robb was moving his mail from RPI's webmail to his Gmail, happily reorganizing, reading old emails, and reminiscing, and I remembered the archives sitting on Jayne (and on CD, somewhere...). I copied them down, and poked around for a way to try to move them back into my live Gmail account.
It turns out Mail can't import its own files... you need to convert them first. Anyway, after a long conversion session, I spent the better part of the last few days slowly copying tens of thousands of emails back up to Gmail.
Now, I have a mess... a single folder, with more than 52,000 emails in it; everything from 2004 on to now. What's really amazing is instantaneous search both in Mail and in the Gmail web interface... I'm impressed that everything is holding together!
Anyway — if you wanted to know how much mail Mail can handle, it's well over 52,000, at least on a 2007 MBP...
|
|