Justin Dolske’s blog

I currently use IRSSI, a traditional text-based client for IRC. I’ve used various GUI clients in the past, but found them all to have bugs and quirks that drove me crazy. IRSSI’s been very good to me, but it’s lacking one major feature that every GUI client has — notifications. And so unless I actually *look* at my IRC terminal, I can’t tell if someone’s trying to get my attention. To make things even worse, I run IRSSI under screen on a Mozilla server (so it’s always online even if my laptop isn’t). That makes it tricky to get a notification on my local system when the trigger is on a remote system.

Back in 2010 Justin Dow blogged about one way to get this working. It’s a little complicated… You first configure IRSSI to log notifications to a text file, the run one normal interactive SSH connection to screen+IRSSI, and another SSH connection to pipe the IRSSI log back to a local script that feeds it into Growl. I didn’t want to do that.

I tried using iTerm 2 for a while. Among its many bells and whistles is the ability to do actions (like notifications) upon a regex match. Neat little feature, and was easy to get working for someone saying “dolske” in whatever channel I had active. But background mentions don’t send any obvious string, so that was a huge limitation. (I tried regexing for the ANSI control codes that hilight an IRSSI window number, but that didn’t work reliably.)

I recently read that as of OS X 10.7, the default Terminal.app includes a simple feature — whenever it receives a bell character, it will bounce the Dock icon and annotate the icon with a number. Perfect! I got this all working, so here’s what you can do:

1. Configure Terminal.app. Actually, you don’t need to. This works automatically. But Preferences –> Settings –> Advanced –> Audible Bell may be something you want to disable if you dislike terminals making noise. Or enable it for debugging this.

2. Configure IRSSI to beep on private messages and hilights: /set beep_msg_level MSGS HILIGHT

3. Configure Screen to allow passing through beeps. You can put “vbell off” in ~/.screenrc and/or toggle it interactively via Control-A, Control-G. An easy test for this is to enable Audible Bell (see step 1) and use IRSSI’s /beep command. If you hear a sound, you’re good to go.

Note that IRSSI also has a “/set bell_beeps on|off” toggle. It’s not needed for the above; it seems to be for filtering out bell characters from channel messages, but the docs are somewhat vague (and a quick skim of the source wasn’t any more enlightening).

There’s a nice blog post up today talking about the typefaces used in the current presidential campaign. (Along with a link to an interesting study about their impact on perception.)

I couldn’t resist making the following:

(References: [1], [2], [3], [4] ;-)

Enhance 224 to 176. Enhance, stop. Move in, stop. Pull out, track right, stop. Center in, pull back. Stop. Track 45 right. Stop. Center and stop. Enhance 34 to 36. Pan right and pull back. Stop. Enhance 34 to 46. Pull back. Wait a minute, go right, stop. Enhance 57 to 19. Track 45 left. Stop. Enhance 15 to 23. Give me a hard copy right there.

It seems demos of the new WebRTC getUserMedia() are all the rage these days. The bug’s bitten me too, so I took Tim’s green-screen demo and hacked it up to my own ends…

A common technique in photography — especially astrophotography — is “image stacking.” The stuff you DON’T want in your image is transient and random noise, whereas the scene you DO want in your image can be reliably and repeatedly captured. So, the basic idea is to take a bunch of photos, changing as little as possible, and then use image processing to combine/average (“stack”) them together. Once you start thinking about capturing photons this way, it’s possible to capture images that far exceed what one would normally expect.

I’ve implemented a simple-and-dumb version of image stacking using HTML5′s getUserMedia() and canvas. Let me illustrate with some pictures.

First I pointed my super-cheap USB webcam at a thing on my desk — which was dimly lit and quite stationary. Here’s a typical frame of captured video:

Doesn’t look very good; it’s a typical poor-quality webcam image. Next, using my little hack, I then captured 50 frames at 640×480 and averaged them together:

Yum. Much better. The image is overall much cleaner; the random-color “static” is suppressed in favor of flat colors and smooth gradients. (A little too much so, making it look cartoonish. I’m not sure if this is because of my cheap camera, a dumb algorithm, or something else.). But this image isn’t just smoother — it’s also sharper. Lines and edges are now crisp instead of blurry and mottled. Text labels that were barely readable before are now easily readable. This is particularly evident in tiny “◃SCALE▹” and “◃POSITION▹” labels just above them.

The differences are even more obvious if you boost up the brightness of the above images in Photoshop. The already-bright areas are now washed out, but detail in the darker areas is easier to see. In particular, the area from the column of dark buttons to the lower-left corner of the pic is much more detailed than in the single-frame capture:

Now, just point your space telescope at a dark patch of the sky, use similar techniques to stack up 23-days worth of exposure, and you get this. Neat.

If you’d like to play around with this demo yourself, try it out in your browser here. I’m curious if people can find improvements to the averaging (in either speed or quality).

(or how I learned to stop worrying and love the Sun)

This is going to be a longish post, the ravings of an obsessive-compulsive mind. So here’s the tl;dr… I took a picture using a telescope (indirectly). It’s Venus transiting across the Sun. It took some frantic last-minute work, but it was worth it because it won’t happen again for 105 years. Here is my picture:

The black dot is Venus. The whitish circle is the Sun. Behold, the glory of orbital mechanics!

OK, most of you can safely skip the rest of this post. Geek stuff follows…

The story begins 16 days ago. The annular eclipse of May 20th 2012. These are relatively infrequent (every 2 or 3 years), but what was special about this event was that the optimal centerline-path of the eclipse was only about a 3-hour drive away. So I packed up my gear and hit the road. Yes, I also took a picture then:

I learned a few things from this event. The first was that — damn — these things sneak up on you fast. I’m a science/astronomy geek, and knew this thing was going to happen, but it wasn’t until a few days prior to the event that some random news article reminded me that is was going to happen omgnow. So I had to scramble. The second lesson was that it really helps to do a few dry-runs for your observation. I was prepared with a pinhole projector and pinhead mirror for observing the eclipse, but I built (!) both of them the day of the eclipse and was in a rush. It turned out ok, but I glitched the math on my pinhead mirror focal length and so the image was more blurry than I had been hoping for. No big deal for the eclipse, though, since most of the experience was more atmospheric (pardon the pun)… The relaxing drive, the slowly darkening skies, and the unexpected sights along the way. The saying that “the journey is the reward” was spot-on.

Taking these lessons to heart, I decided to be more prepared for the then-upcoming Transit of Venus. After much armchair research, I figured that the absolute best viewing experience would be with a specialty telescope fitted with an H-alpha filter. The views are astounding, as is the price — it jumps into the $thousands before you can even blink.

So after careful deliberation (especially given the once-in-a-lifetime opportunity), I did what had to be done. Yep. I got a $10 #14 welder’s filter and $5 pair of eclipse glasses. (Looking at others’ expensive pictures? Priceless.) I do have a decent conventional telescope, but it’s a pain to lug around and I was thinking I wouldn’t really need it.

But I found that the eclipse glasses seemed useless for a transit — the naked-eye diameter of the Sun is surprisingly small, and the tiny dot that is Venus would be challenging to see. I also did a test shot of the Sun with my camera thought the #14 filter, and scaled a shot of the 2004 Venus Transit to that size to create a mockup of what I might be able to capture on Transit Day:

Unimpressive. Venus would be about 4 pixels — at best. I’d get better results with a longer zoom lens on my camera (Lumix GF3, 20mm); but I tend to prefer wide-angle shots, thus spending $hundreds for a lens I might not use very often was unappealing. And so I basically left it at that, figuring whatever I might manage to record would be just fine as a token momento. After all, NASA and professional astronomers would capture the event in stunning detail, so why bother trying, right?

Timeline: Tuesday June 5th. 10:30am.

I started off my morning with a weekly meeting with my boss, Johnath. I don’t recall exactly how we got to the topic of astrophotography and Venus, but we did. And something (read: Johnath’s nefarious plan) sparked my brain into thinking that surely it couldn’t be that hard to something with my telescope for the transit. And from that point on I knew my day would have one, intensive, singular focus.

The ideal way to view the Sun through a telescope is with a special H-alpha filter (expensive, as noted) or much cheaper general solar filter. Both basically block 99-point-lots-of-9s of the light entering the telescope, making it safe to look through directly with your eye. Alas, I had neither of these filters. (See above, re: prepare and practice.) Instead I fell back to an idea I had heard about, which involves physically blocking most of the telescope’s aperture except for a small opening. This would reduce, in my case, the light-gathering power of an 8″-diameter Schmidt–Cassegrain ‘scope — which is normally great for viewing dim galaxies and faint nebulae — to a much smaller opening. Say, a half-inch to 2 inches.

First step — measure twice, cut once. I was very precisely cutting a form from foam board (say that four times fast), because I wanted to ensure that it fit snugly and didn’t leak light. A full-aperture telescope pointed directly at the Sun has an overwhelming tendency to do things like cause blindness, burn things, and damage delicate optics. That would be bad. On top of the first foamboard layer would be a sheet of foil (because foamboard isn’t very opaque to light), and then a second layer of foamboard to help protect the foil and sit flush with the outer lip of the telescope body. A little bit of gray fleece on the back and flat-black paint helped cut down on reflections. Ta-da, done.

I also built a holder for the #14 welding glass I have, but didn’t end up using it. It’s too dark to use for projection, and viewing through an eyepiece just didn’t seem to work as well as projecting. I also ended up entirely skipping the creation of some kludgey adapter to attach my camera to the telescope (I have a camera fitting, but it’s for the ancient Pentax-K mount).

Finally, after all that, I set up everything to see if it worked. I got the sun projecting onto a small piece of foamboard, and my first thought was to wonder what I has screwed up to get such a big black dot in the image. Oh, wait. Right. That’s Venus. I wasn’t watching the time, and the transit was already underway. Hooray, it works!

After this I couldn’t directly see the sun from my patio any more, so I headed off to a park near work with an open view of the sky. Matt and Jared helped out with moving equipment and holding the projection surface (again, formboard) while I fiddled with the telescope. This is how we were set up for capturing the photo at the top of this post:

Afterwards, just for fun I took an image from NASA’s SDO satellite and scaled it down to compare with my own contrast-stretched shot. I will be the first to admit their $850 million toy captures a far superior image, but I’m still pleased with my own result. (To be clear, mine is on the left. :)

And that’s how I spent most of Tuesday.