www.intermod.org

SETI@home and distributed computing

When I was sophomore in high school I got my first computer, a Commodore 64.   It had a laughable amount of processing power by today's standards, but as my first computer that wasn't really an issue.  Here I had a machine that would do what I told it, constantly, 24-7.  After learning basic programming, I figured out an algorithm to brute force prime number calculations and set it to work.  I'd watch it race through the early numbers 1 2 3 5 7 11 13 17 23 29 31 37 ... 41 ......43 ......... 47 ........... and then slowly be brought to its knees.

About 5 years later I bought my second computer, a very nice 286 (12 MHz with "zero wait state").  Probably within the first day I had it crunching prime numbers.  There's something satisfying about telling a computer to go do something and then watching it work furiously on that task.  I remember being in college and instead of shutting it down at the end of the night , I would run the prime numbers program (say, starting at 1,000,000) and see what it had found in the morning.

Another 5-6 years and another computer ... but by then it was 1993 and the internet was starting to crash through the doors.  But in due time the internet and brute force mathematics would combine in a new way.

In 1997 the distributed.net organization (just a bunch of computer nerds) started to think seriously about distributed computing.  Their idea was to take a very very large task, split it up into digestable pieces, assign those pieces to many computers (using the internet), and then watch the results come in.  The very very large task they were tackling was codebreaking.

Important and sensitive data, such as in finance transactions or military communications, is encrypted using encryption algorithms.  The distributed.net guys were trying to make the point that computers had gotten so fast that the old encryption algorithms were quickly becoming worthless.  What good is a password if your computer can figure it out in 30 seconds?

Looking at d.net's own history page I think I got involved in summer 1997.  At that point they had started a serious attack on the RC5 algorithm, again just to show how quickly it could be broken.  A few thousand people, myself included, ran software on their computers that basically crunched numbers while the computers were idle.  Like a screensaver, it just kicked in when you weren't using the computer.  I definitely remember being involved in this by the summer of 1997, because I remember showing it to an engineer in Saudi Arabia, and I was there in June 1997.

As the months wore on, the project got more popular, got more participants, got more press, and eventually they partnered with the EFF and cracked a 3DES message in 22 hours, putting the nail in the coffin of the U.S. government's venerable Data Encryption Standard and paving the way for wider acceptance of the Advanced Encryption Standard.  D.net moved on to the search for Golomb rulers, Mersenne primes,and other mathematical endeavors, but I lost interest.

By that time the nerd news vine was chattering about the upcoming SETI@home effort.  Berkeley scientists were going to take radio noise collected from space and see if anything interesting could be heard in that noise.  They had received seed funding from The Planetary Society and Paramount Pictures, who were about to release the movie treatment of Carl Sagan's Contact (the one with Jodie Foster).  In which, by the way, prime numbers feature prominently!

And so, much like with my more recent Tesla obsession, in early 1999 I began regularly checking for word about the software release.  Every day I'd check the SETI@home website to see if they'd released their software to the public yet.  And so it was one day, as I was having lunch in my office, that I idly went to the SETI@home website to check on it, and they had publicly launched their software that morning!

Within minutes I had it running and was starting to benchmark the speed of my computer.  It was sobering to see what a massive task each SETI@home work unit entailed.  I watched the percent-complete number tick up from 0.000%, 0.001%, 0.002% ... Gulp.  This was going to take a looooong time.

And so the search was on for other computers to run it on.  I ran a small tech demo room that had a couple terribly under-utilized Sun workstations, and within days (perhaps hours) I had them running the software too.

That was May 17th 1999, ten years ago today.

The guys at SETI@home (and they remain a very small team) are this week celebrating their decade of work.  Despite their towering profile in the nerd community, they have minimal funding and it has been a constant struggle for them to keep up with the demand, especially as computers have gotten faster and faster and have placed more demands on the infrastructure they maintain to keep handing out chunks of radio noise.  Recently they released software (called CUDA) that takes advantage of the graphics processor in your computer.  Pretty much every computer today ships with a extremely fast graphics engine, designed for the purpose of rendering fancy 3D graphics on your screen, but which only a small percentage of people actually use  The SETI@home guys are figuring out how to tap into that wasted computing power.

So this Thursday I will be watching as the SETI@home guys celebrate their 10 years of work, and look forward to their reports and presentations on their progress:

• SETI@home website
• anniversary background info
• May 21st 2009 program agenda
• May 21st 2009 program streamed online

And, of course, I wait for the day that they find little green men!

120 seconds that could kill the space program

On Monday, space shuttle Atlantis is scheduled to launch STS-125, a mission to the Hubble space telescope.  This is the fifth and final mission to Hubble, and is a mission that almost didn't happen.  After the Columbia accident in 2003, NASA's top management decided it was too risky (no ISS lifeboat capability) and canceled it.  That produced an uproar in the scientific community, and after a couple more years of study and contingency planning, the mission was resurrected, and here we are just hours from liftoff.

This mission is incredibly important, but just as much for the entire space program as for Hubble.  This mission has been driving the space shuttle program's planning for 5 years now.  They worked to reduce the foam shedding off the tank (which is what doomed Columbia).  They created new inspection techniques to be carried out in orbit, using the robot arm, so that they'd know if they'd damaged the protective skin of the shuttle during ascent.  They added sensors to the wings so they could "feel" the impact of any debris during ascent.  They upgraded all of their ground cameras for launch views.  They developed and tested repair techniques, to be executed by spacewalking astronauts.  And they figured out a way to rescue this Atlantis crew from orbit if they deemed that it was damaged beyond repair.

Which is why space shuttle Endeavor is now sitting on the other launchpad, ready to go within a few days.  It's going to sit there while Atlantis is up, and once they deem Atlantis in good shape and able to survive the fiery re-entry from orbit, Endeavor's emergency mission (officially called STS-400) will stand down and they will resume preps for what Endeavor really has next on its plate -- a mission to ISS in June.

The first 120 seconds of launch are going to be critical.  During those first two minutes, any debris (such as foam coming off the tank) could hit the shuttle and punch a hole in the skin.  After the first two minutes, any debris shedding doesn't matter because they are up out of the atmosphere and there's literally no air rushing by to slam the debris into the shuttle.

Now, the thing is, we won't know right away.  Atlantis will probably get through those 120 seconds just fine and continue all the way to main engine cut off (MECO) 8.5 minutes after liftoff.  But that's when the imagery analysis begins.  Here's what to look for in the news:

1.  Immediately after the shuttle reaches orbit, amateurs on the net will start looking at the footage and pointing out possible debris events.  In the first two flights after Columbia (in 2005 and 2006), there was quite a bit of debris so there was a lot to talk about.  But in the past 2-3 years it's been very good.  We'll have this amateur data within 30 minutes after launch.  Update 2:45pm: launch looked good!  No debris events noted in camera replays.  The first 10 seconds were exciting through when a transducer (pressure gauge) failed, creating some spurious alarms.

2.  About an hour after launch there will be the post-launch news briefing, and the reporters will ask the managers what they saw.  I gaurantee that the managers will say make an empty statement like "we're going to go take a look at all the data and see what we see", but you will get a sense from them whether anything jumped out at them as a problem.  Update 3:21pm:  managers in press briefing also saw no debris events of note!

3.  The day after launch, aka Flight Day 2, the crew on orbit will do the inspections with the robot arm and will be able to see if there's any obvious damage.  We'll see that inspection video coming down live so we'll see anything when they do.   Update 12-May:  looking good!  There are some nicks in the tile but they don't think it's serious enough to trigger any actions.  However, an important part of the flight control electronics failed during the launch, and while they can continue the mission for now, if they take another failure while in orbit the flight rules dictate that they must come down immediately -- like within hours.  So everyone has their fingers crossed that nothing further will fail.

4.  On Flight Day 3 or 4 we'll start to hear officially about any serious problems.  And that is when they make the decision whether to proceed normally with the mission (capture Hubble and get to work) or whether they literally shut down the lights and start saving power so that they can survive long enough for Endeavor to get up there.  Update 17-May:  the "Thermal Protection System" has been officially declared fit for entry; they still have to do late inspection at the end of the mission but it's looking very good.  Meanwhile there's been day after day of nail-biter spacewalks and I'm exhausted just from watching :)

I will update this post with the status of each of the 4 stages above as they are cleared.

If they end up having to rescue Atlantis, that means we will be down to just two shuttles, and NASA has said that they can't run the program with just two.  They certainly wouldn't build another one, as these orbiters are 30+ year old designs and about to be retired anyway.  So triggering the rescue would be the implied end of the shuttle program, forever.

Regarding the actual mission, I wrote back in September about the mission and pointed to Dr. Heidi Hammel's speech in particular.  Of course, this was 7 months ago, just before the anomaly on Hubble that forced them to scrub the Oct 10th launch.  The equipment to repair that latest failure is now on the shuttle and they're ready to go up and fix that too.

As I said back in September, I would encourage anyone with even a fleeting interest in science or astronomy to take the time to download and watch the entire briefing, as it is truly fantastic stuff they're talking about, and these guys do a great job of explaining it to regular people.  If you don't have time to watch the entire 90-minute briefing, at least watch her 10 minutes.  Links are in the post I made back in September.

Update 13-May-2009: Sadly, I see that the links to the Hammel speech in the old post no longer work. I have not been able to find the briefing on Youtube, and the repeat briefings from a couple weeks ago did not include Dr. Hammel. FORTUNATELY, I did find most of Dr. Hammel's speech incorporated into a nice 5 minute video right here. Please check it out!

Update 16-Jun-2009:  And with the "Houston, wheels stop" call from the commander on May 24th, Atlantic completed her mission and everyone involved breathed a huge sigh of relief.  And later that week, a Soyuz launched three men to ISS, finally bringing the space station crew up to 6 people for the first time ever.  We've been anticipating the Hubble mission for 5 years, but the 6-man crew on ISS has been anticipated for nearly two decades! It didn't make headlines, because it all went so smoothly, but this is the heyday of space flight, RIGHT NOW. That last week in May, with the completion of the Hubble servicing mission and the beginning of serious science operations on the space station, represents perhaps the biggest milestone in the US manned space program since the first launch of the space shuttle in 1981. Congratulations NASA!

Cheap High-Def TV

or

Chris's Guide To Getting Free Digital TV

or

How To Flip Off The Cable Company

This guide is intended primarily for people I know who are having to tighten their belts, and are looking at that big expensive cable bill every month and wondering ... do we really need this?

It really comes down to this:  do you really care about all the basic cable stations (like Bravo, USA, ESPN, TNT) and their crappy programming to pay all that money?  Do you really need access to 50 different mind-numbing reality shows or 30 channels of procedural drama repeats?  Because if you think you'll be happy with about 15 stations (the 4 major networks plus your local independents), I can show you how to get them all in high definition for free.  HIGH DEFINITION FOR FREE.

If this sounds intriguing to you, then by all means read the rest of this guide, but please do the first step right away.  Do not delay one day -- do it right now.

Here's how to cut the cable, save a ton of money, and get gorgeous digital television.

Step 1:  Get your $40 DTV coupon.  By act of Congress 3 years ago, every household in the country is entitled to two $40 coupons to pay for a little converter box.  Don't worry about what the box does (I'll explain later).  Don't wonder if you really need it (you might, might not).  Don't worry about getting on some mailing list (you won't).  DO GET IT RIGHT NOW, because the deadline is approaching and there's going to be a mad crush for them again.  You may have heard about a June 12th deadline; well, if you wait until then, you're too late.  Go to www.dtv2009.gov right now and take 60 seconds to fill out the form.

Update June 13th: Well, the analog shutdown happened yesterday and now you're really going to be in a crush of procrastinators who are trying to get their coupon and converter box.  The funding for this program isn't going to last long, so you still need to get your request in ASAP.  Next week might be too late!

Step 2:  Evaluate your existing TV.  If you have a newer "digital" TV, the kind that's big and flat and probably expensive, then you do not actually need the converter box, and you could skip the next step but you shouldn't.   If you have an older "analog" TV, the kind with a big glass curved front surface, then it will work fine but you will need a converter box, per the next step.

Step 3:  Get a converter box.  These are officially called "coupon eligible converter boxes", and what they do is convert the new digital signals that are now on the air into the old analog signal that your old TV understands.  Even if you have a digital TV and shouldn't need a converter box, you should go ahead and get at least one coupon and buy a converter box, because you might find out later that you need one.  And you won't be able to buy them later -- the $40 coupon program is only available this year.  So, apply for the coupon, receive it in the mail in a week or two, and go to any electronics store and buy a converter box.  It's OK to buy this online, but not the antenna (more on this below).  Which one?  Eh, it doesn't matter much.  Look for a feature called "analog bypass", but I think they all pretty much have that now.  If you really want to get the best one, ask me and I'll explain about analog bypass, AFD, smart antenna control and EPG passthrough (or plow through the data yourself).

Step 4:  Get an antenna.  This is going to be the trickiest part, but hang in there.  Most likely you will be able to get by with just a simple indoor antenna near your TV, like the old rabbit ears.  But it's not quite rabbit ears you need -- those work best in VHF, and you need an antenna that works in UHF.  The big metal loop part of most indoor antennas is what's actually important for digital TV.  So go to any electronics store and look for a small TV antenna.  Don't buy a big one yet -- you probably won't need it!  A couple examples are the RCA ANT111 and the Zenith Silver Sensor.  DO NOT buy online, because you might need to return it if it doesn't work for you, and doing that by mail sucks.  Just go to a real brick-and-mortar store and buy whatever small TV antenna they have.

Step 5:  Set up the antenna and connect it to the converter box per the instructions that came with the box.  Connect it to your TV.  Follow the instructions to do a "scan" of the airwaves.  This means the box finds the stations that are on the air.  In the new digital TV world, it's not as simple as just going to Channel 2 and watching -- the first time you set this up, you have to let it scan and find what channels are out there.

Step 6:  Enjoy free digital television!  You'll quickly discover that there's a lot more on the air than just the big networks that you expect.  Most stations have their primary channel and then also have subchannels.  So for example, WXIA/11 has their NBC programming on 11.1, but a weather forecast channel on 11.2 and a sports channel on 11.3!  WGTV/8, one of the two PBS stations in Atlanta, has three simultaneous programs running, including GPB Knowledge on 8.3 which runs documentary-style material 24-7.  Nerd heaven!

And that's it!  You're now set up with a free digital TV signal.  If you're not able to get all of the local stations, DON'T GIVE UP.  You just need to be patient and fiddle with the antenna a bit.

But you've got to do the first step right away.  Scroll up to Step 1 and do it, even if you're not sure.  Time's running out.

Here are a few more things you can do:

Optional Step 7:  Get a better antenna.  The new digital picture should be absolutely perfect, even when it's raining or windy.  If the picture is breaking up at all, then you need to get a better antenna than the cheap little guy that I had you buy in Step 4.  So return that one to the store, and step up to an attic-mounted panel antenna like the Antennas Direct DB2 or the ChannelMaster 4220; two places to find these in Atlanta are Fry's Electronics and Dow Electronics, both in Duluth.  If you don't have an attic to put these in, or don't want to run the cable, try getting a "smart antenna", like the Apex SM550 or the RCA ANT2000; however for these you will need a TV or converter box that has smart antenna capability.  Ask me about that if you get this far.

Optional Step 8:  Get your VCR working again.  Got a VCR for taping shows off the air?  It won't work after June 12th.  If you want to keep using it, you'll need a converter box for it.  Now you see why I said get the box even if you didn't need it?  You might need TWO!  But the truth is that setting this up to work with your VCR is going to be a monster pain in the ass.  You really should just upgrade to a DVR ...

Optional Step 9:  Get a Digital Video Recorder (DVR).  And this is where you reach Free TV Nirvana.  If you've got some money to spend, please do spend it on a Tivo HD ($250) and the service ($12/month); you won't regret it and in the end it's still far cheaper than cable.  The only other device on the market worth considering is Echostar's DTVpal DVR; it has no monthly cost but is much more expensive, and further is actually buggy and compares horribly with a real Tivo.  And a real Tivo will quickly learn what kinds of things you like, and start recording shows that you'll probably enjoy -- and suddenly it's like you've got an On Demand system, with a box full of shows at your fingertips, ready to play at any time.  Oh, and Tivo's do Netflix and Youtube too.  All in HD.

Optional Step 10:  Change your internet service to DSL.  If you're doing all this to be able cancel your cable bill, but you've also been getting your internet service via cable, then you're going to have to change that too before you can completely cut the cable service.  DSL via your phone line works great, but there will definitely be a hassle in store for you as you migrate from one ISP to the other (e.g. new email address, if been using a comcast.net address or similar).  Just something to brace for if you're getting your internet via cable.