Dave Andersen's Blog

I was going to title this post "hacking teeth", but that has a particularly horrible image associated with it. So, to the point: This is what happens when your Ph.D.-holding geek friend gets worried about his oral health and catches a cold that gives him a few days to do research on the always-reliable Internet. If you want to skip the blather, the bottom line is: About five months ago, I started using a (brushing, rinsing, etc.) regime that its inventor calls, generically, clean white teeth. Before I jumped into it, I spent an embarrassing amount of time reading some of the background research behind it, and came to the conclusion that, in theory, it should work. My dentist visit today was zippy fast, things looked great, and I'm now convinced this regime works. If you want the rationale and links to the underlying research, read on.( Full DetailsCollapse )

A few days ago, I was looking through one of my server's logs while trying to debug an unrelated problem, and noticed the best source address yet for a brute-force password guessing attack against my machine:

Nov 21 11:35:20 moo sshd[55732]: error: PAM: authentication error 
for illegal user j from 213.250.36.254

Who, might you ask, is 213.250.36.254?

inetnum:      213.250.36.248 - 213.250.36.255
netname:      URJV-SI
descr:        Uprava RS za jedrsko varnost
descr:        Slovenian Nuclear Safety Administration scope
country:      SI

Doesn't it make you feel comforted to know that there's a nicely compromised computer within the Slovenian Nuclear Safety Administration that's trying to spread its malware around the Internet? (cough).

The last few years have seen a resurgence of DIY and hardware hacking, fueled by the Internet, remarked upon by slashdot and boingboing among many others, and propelled (and capitalized upon) by magazines like Make.

A key question raised by this resurgence is: Do you have the capability, the access, and the right to hack the hardware devices you've purchased? Apple and the EFF are fighting this question out over iPhone jailbreaking. In the past, Apple sent DMCA notices to people trying to write software to load music onto their ipods. At the root of this debate, of course, is what rights an owner has to modify the hardware they've purchased, and whether the DMCA's "anti-circumvention" provision covers such use.

In a strange turn, the same battle is emerging on a new front: calculators. If this gives you a vision of a teenage boy with a soldering iron prying open an old touchpad calculator, reconsider: today's calculators contain full-fledged microprocessors (TI's latest is believed to use a 150Mhz ARM CPU, placing it somewhere between an iPhone and a toaster oven) and large graphical displays. They're affordable, light, and a ripe target for experimentation.

What's the issue? One might think that a calculator company would be glad to see people using and finding new uses for its products. But when an inspired programmer announced that he'd cracked the key needed to install your own operating system on a TI calculator, TI responded by sending a DMCA 'takedown' notice to anyone relaying or even linking to the details. TI's position very likely comes from their calculator's position as "blessed" for use on the SAT and related tests, and they likely fear that the ability to write a custom OS might permit cheating on these tests. Obviously, a better solution would have been to grant the ability to install a custom OS freely, and to signal this to the test-givers---in that event, circumventing the protection measure would only promote undetected cheating. But that's not the case here: the mechanism prevented any third-party software modification of the hardware. Do we really not truly own our hardware?

[Warning: Geek content ahead.]

Phew! Back from summer vacation, managed to contribute a bit to our HotNets and Infocom submissions, and now I'm very psyched that the final camera-ready version of the FAWN SOSP paper [pdf] is online.

The path to this one was lengthy: We submitted a version of it to OSDI last year, and got back surprisingly supportive reviews that said, in a nutshell, "this will be great - now go finish it!" The reviewers were right. We let the paper bake for a year, even skipping the NSDI deadline, and sent a paper I'm really happy with to SOSP. The students on the paper (Amar, Jason, Lawrence, and Vijay) put in a boatload of work, even rewriting the network communication parts of the system to use protocol buffers to eke out some extra performance and slash out a few thousand lines of code from the system.

Now we're just waiting for some systems using Intel's next-generation Atom to come out; the plan (at least, as much as one can plan when it's not released yet! :) is to use these as the basis for the next generation FAWN node. It should be zippy fast, and we're expecting to more than double the energy efficiency that we get out of our (now over four-year-old) current nodes. The extra CPU capability should let us push FAWN into more difficult analytical tasks that chew on huge streams of data vs. the small-object seek-bound workloads we've targeted thus far.

Exciting times! At HotOS, Margo Seltzer dropped the phrase "One Cluster Per Child" -- and it might actually be possible to do so with a FAWN cluster. Now we just have to worry what they'd do with it...

Just returned from a few days in Switzerland for HotOS XII (the workshop on Hot Topics in Operating Systems). Good workshop, good hiking, horrible airplane experience - really, who could ask for more? :)

link to all photos if you just read blogs for the pictures.

The workshop: Matt Welsh covered a lot of the technical meat in his blog already, so I won't repeat it here. Had a great continuation of the yummy-extravagant-dinners-at-conferences theme. Very enjoyable technical content and discussions. Keeping in mind that this list reflects my technical interests more than anything else (you'll note a strong multicore flavor in here), I'd particularly recommend taking a gander at:

• Hierarchical File Systems Are Dead
• Your computer is already a distributed system. Why isn't your OS?
• Hera-JVM: Abstracting Processor Heterogeneity Behind a Virtual Machine
• Reinventing Scheduling for Multicore Systems

I may be a bit biased on the last one, because I chatted about the broader work at length with Silas while hiking after the workshop. Still, it was nice work. And that brings us to...

The hiking: On Wednesday after the workshop, Silas and I wandered west from Monte Verita to a little mountain called Corona Dei Pinci and then to Alpe di Naccio. Great little afternoon hike.

On the following day, I decided to stubbornly forget that any form of transit other than foot existed, so I wandered to the nearby city of Locarno, and from there, went up an impressively steep staircase to a small hamlet called Monte Bre. It was staircases all the way up - about 900 meters of elevation gain.

From Mte. Bre, I found the wrong staircase that ended in prickly bushes and an intimidating electric fence that wrapped entirely around an impressive looking estate---which I decided to not photograph on the grounds that the occupants seemed to be zealously protective of their privacy already. And most Swiss houses are well-armed. But after that, the mountains opened up, I even found some snow, and got to the top of a rather nice mountain, Cima della Trosa (1900m) . Which, like every easily reachable peak in the area, seems to be decorated with an impressive crucifix.

I then wandered down the mountain and found that other people had figured out a better way down. Drawing inspiration from them, and already having experienced the joy of walking up the huge staircase, I hopped on the tram from 1300m to 400m to the upper reaches of Locarno.

Of course, there were still plenty of stairways , all of which were full of lizards.

Very cool trip overall. And, fodder for another post some time, I even made it back to the states a mere 5-ish hours late, despite a canceled flight from Zurich.

I figured I'd follow Michael Mitzenmacher's sigcomm PC post-analysis with a few ruminations of my own. Michael noted that the quality of submissions seemed down, and I'm one of the other PC members who agreed with this. I'm also on the SOSP 2009 program committee, and the quality gap is surprising, even compared to previous years. Comparing the papers I reviewed in sigcomm 2008 and 2009, where the scale is roughly logarithmic and "1" is bottom 50% and 5 is "accept or i will scream":

2008:
1:  1  2  3  4  5  6  7                (28%)
2:  1  2  3  4  5  6  7  8  9  10 11   (44%)
3:  1  2  3  4                         (16%)
4:  1  2  3                            (12%)

2009:
1:  1  2  3  4  5                      (31%)
2:  1  2  3  4  5  6  7                (44%)
3:  1  2  3                            (18%)
4:  1                                  (6%)

The total number of submissions was down slightly (288 in 2008, ~275 in 2009). There's a small bias because the number of later-round papers I reviewed was lower than in the previous year (reviews in round 2 omitted some of the lowest-ranked papers from round 1, which were "quick rejected"). In comparison, in my first round of SOSP reviews, I've already assigned one 5-equivalent.

Every PC is an experiment in how to run a PC, and this year's SIGCOMM was no exception. For background on this ongoing dialogue, it's worth peeking at some of the papers in last year's Workshop on Organizing Workshops, Conferences, and Symposia for Computer Systems -- we continue the debate about how to organize reviewing, double vs. single (or non)-blind review, accept papers, etc. Like last year, the PC was split into "light" and "heavy", where only "heavy" attended the PC meeting. The 2009 PC was larger (25 heavy, 35 light) than the 2008 PC (22 heavy, 24 light), and the per-reviewer workload was lighter (16 vs 25 papers). This is a useful experiment - one common complaint about serving on PCs is the time commitment. Unfortunately, I don't think this one worked out: my impression was that the room was a bit too full and individual PC members hadn't read enough of the papers to get as good a sense for the overall ordering of papers. Unfortunately, it seems there's some good that comes from a painful reviewing load in terms of having a good feel for the papers. I definitely like the way this and last year's PC structured things with heavy and light---those with more workload read a greater proportion of good papers, because the bottom 25% were rejected after the first round. This is a great way to preserve the sanity of your program committee.

There was (apparently; I didn't hear too much of it) some grumbling last year about the number of PC papers accepted. As a result, out-of-band papers were handled using a different website, along with some number of randomly selected additional papers. While this is a pretty sensible way to try things, it turned out to be really awkward - I couldn't login to the OOB website for weeks, and my guess is that the "randomly selected" papers took a hit by being reviewed separately from the main papers. It's clear that we need ACM (or some third party) to run a hosted instance of the HotCRP conference management software where the sysadmin and conference chairs are separate, so that the reviews are integrated but the chairs can't see who's reviewing their own papers, perhaps with some small tweaks to HotCRP to allow reviews to be assigned by an "OOB Chair". I note that USENIX has already started heading down this path.

Finally, I strongly encourage the next local host of the Sigcomm PC meeting to follow Brad Karp's lead and have an expert barista with a good espresso machine parked oustide of the meeting room. Simply awesome!

A while ago, I blogged about my Seattle espresso exploration, and lamented that I couldn't find such beasties in Pittsburgh. Charlie mentioned 21st Street Coffee and Tea.

Yum!

They've got the best espresso I've had in Pittsburgh, and it was the equal of the best espressos I had in Seattle. Rich and creamy, with a thick crema and a tingly acidity, and not a hint of bitterness. They use intelligentsia's Black Cat espresso blend, and it comes out very tasty. For the brewed coffee fanatics out there, they've also got a Clover machine. I highly recommend this place. Now we just need to convince them to open a copy of themselves near or at Carnegie Mellon.

I've extended the grow-lights to try to keep some plants going over the winter a bit better. Last year, my indoor plants stayed alive, but were never really happy. Theory: Not enough light, and they didn't like my 45 degree living room. No guarantees this will work. I'm using both the LEDs and the previous year's CFL, and have wrapped the ensemble in plastic. (Odd as it sounds, a combination of plastic wrap, of which I have oodles from a costco trip three and a half years ago, and some non-sticky window sealer plastic I haven't put up inside yet.) The bottom is wrapped in a mylar space blanket to reflect some of the light, and there's a mirror and some aluminum on top for the same reason.

---

You're in Calgary at the (quite fun) HotNets 2008 workshop. Your flight leaves at 10:45pm, promising to be a brain-draining three-flight red-eye back home. You want some exercise before the flight, but you've checked out of your hotel already. Combine with a bit of a habit of enjoying going places under your own power, and the solution seems obvious:

Walk to the airport.

It turns out that, in Calgary, it's possible. You might even take this route. The first 6 miles are great. Through Nose Hill park on trails through knee-high grass, with a great view of the city and lots of friendly people walking happy puppies. After that, though...

Advice 1: Don't try to cut through the under-development extension to the airport blvd. Climbing through a (stationary) train is a bit nervous. Getting stopped by a river afterwords just sucks, because you know that you then have to climb through the train again. And before you ask, the river was too wide to jump, and it smelled like cow poop. I met the cows a bit later when I went around the area to the north. They looked very surprised to see a human walking past on the freeway, as about fifteen pairs of cow eyes swiveled to intently track my progress.

Advice 2: Just ... stop at the 6 mile spot and call a taxi. The rest of the route gets very freeway-like. The overpass over highway 2 is particularly noteworthy, with its knee-high guardrail and rushing traffic inspiring mild vertigo even in a climber.

Advice 3: Ignore the above. You'll be chuckling about the time you walked to the airport in Canada for years to come...

A few other photos.

(Many thanks to Carey Williamson for suggesting a much better route than the one I'd initially picked!)

The trailing end of Hurricane Ike knocked out power for 85,000 people. As the old saying goes - no power, no crepes. (September 15th)