Ecoterrorism

This is an odd story: some luxury homes were burned down and it’s been suggested that ELF was the culprit.

Okay, so that’s among the least-scary, most goofy acronyms I can imagine. But it stands for Earth Liberation Front, and it’s apparently a decentralized group of environmental arsoni… err, activists.

So I sought to find out a little more information about them. This may be their site. I say “may” because, although it’s full of news about the elves and that ELF is a decentralized group with no formal leadership, the title of the page is “Viagra Sample Packs” and the banner ad is for a swingers convention. It also reads like a page that’s lost its stylesheet? So overall I’m just totally confused.

Of course, this will get them the PR they want, but I’m left wondering if it really does any good. I now view them like PETA and Greenpeace, in that they have a good cause, but are so horrifically insane in carrying it out that no one takes them seriously. But if people already think you’re a bunch of loonies, having a webpage partially about your environmental conquests and partially about Viagra, with an ad for swingers clubs, really doesn’t help.

Confirmation Bias

Have you ever noticed that someone mentions something, and all of a sudden you notice that thing multiple times? One professor mentioned that very phenomenon one week. Early the next week, I was touring Waltham’s 911 dispatch center, and noted the software they used for  (Computer Aided Dispatch) CAD, intending to look more into it. The next day, I had that professor’s class again, and, for reasons I still don’t understand, somehow slipped in a mention of knowing someone who worked at that company. So I mentioned this to him after class, and also pointed out the incredible irony of the fact that the week before he’d mentioned the “someone mentions something, and all of a sudden you keep seeing” it phenomonen. {Meta: this phenomonen may, itself, be related to a confirmation bias, but I digress.}

So Kyle mentioned confirmation bias in one of his recent posts. The next day someone on Ask MetaFilter asked why black people were afraid of his dog. Most answers bordered around it being a confirmation bias. (Although it got interesting when several people reported having noticed the same thing, suggesting that maybe, just maybe there is some sort of racial difference in how dogs are viewed.)

The other day there was a lunar eclipse. Around the same time, we had several short problems with Internet access, where traffic wasn’t leaving Bentley. (Incidentally, I had the same experience I’ve had with most of my tickets–I only open then when I run into an issue I can’t solve myself and am very confident that it’s a problem, so I usually include enough information that the front-level have no idea what’s going on an forward it to an expert… So I ended up speaking with our “Network Engineer,” a position I didn’t even know we had, and we had a good talk about exactly whta the problem was… It appears to lie with an upstream company performing some maintenance.)

So today my professor–who teaches nothing related to electronics nor astronomy–randomly interrupted her lecture saying, “Thursday was a lunar eclipse. Did anyone else experience technical problems because of the lunar eclipse?” People had several problems that day–several brought up the Internet disruption. Someone had their computer crash. She explained that the lunar eclipse tends to cause those things.

If ever there were evidence of a confirmation bias, that’s it. I don’t profess to be a total expert, but I’m pretty certain that the moon being obscured by our shadow isn’t why our Internet link went down for a couple minutes. And yet she’s found plenty of evidence to support what’s probably an urban legend.

This Is My Hobby

I want to start a “meta ISP.”

When you sign up with your ISP, you’re paying for transit. They carry your data from one network to the other.

But now let’s say that I’m a mediocre residential ISP. I buy connectivity from a couple different upstream providers, and use BGP to make sure your data takes the fastest route. This is what most people do. It works.

Let’s further say that you run an extremely popular site, maybe one of the top 100 sites out there. You have a mediocre IT team. You have enormous bandwidth, coming in from three different carriers. You, too, use BGP to make sure that your outgoing traffic takes the quickest route.

So everything works. Traffic flows between the two networks. What’s the problem?

Well, it turns out that you, Mr. Big Site, have some of your core routers in a major data center out this way. And I, Mr. Big ISP, also have a few core routers in that building. This is really pretty common–there’s a (very aptly-named) network effect with transit. When several big guys move into a building, all of a sudden, more people want to be there too. So you get sites like One Wilshire, a thirty-story building in LA full of networking equipment. They’re very confidential about their tenants, but “word on the street” is that every network you’ve heard of, and plenty you haven’t, is in there. (When viewing that picture, by the way, it’s worth noting that these wires don’t go to some secretary’s PC. Each is probably carrying between 100 Mbps and 10 Gbps of traffic between various ISPs and major networks… Also an interesting note to the photo, they supposedly keep an elaborate database and label each wire, so that this huge rat’s nest is actually quite organized.)

Since we’re both huge companies, we’re each paying six figures a month on Internet. But when one of my customers views your site, they go through a few different ISPs, and across multiple states, before it arrives on your network. It’s asinine, but that’s how the networks work.

So we wise up to this. I call you up, and we run a Gigabit Ethernet line between our racks. And all of a sudden, life is peachy. Data travelling over that line–my customers viewing your site–is free. My bandwidth bills drop, and speeds improve, too. This is the world of peering. And, strangely, the mutually-beneficial practice is rarely done.

I think there’s a market for a big middleman here. The last mile (that would be a good book title, if a telecom magnate wanted to write his memoirs) is difficult–running lines to consumers’ homes. Similarly, it’s hardly trivial to become a Tier 1 ISP, a sort of ‘core backbone’ of the Internet. But an intermediary broker? Easy enough to do.

So you’d get space in the major exchanges, and peer with popular sites. Google, Yahoo, MSN, Youtube, Facebook, eBay, Myspace, Amazon, Akamai, etc.

Understand?

I had to read this headline about eight times before I understood it: Boyfriend on roof punches weaving driver.

And even after the eighth time, my mental image still didn’t match what I found when I read the story.

Actually, even having read it, I’m not sure I understand it. I mean, I understand it, but I don’t understand it. Why was he on the roof? Why was he punching her? Why didn’t she stop? Why was the car’s air bag inflated? (And a second “Why didn’t she stop?” is in order here.) And tell me “She eventually stopped the car and hit him with it, police said” isn’t unclear. It’s yet another, “I understand… the words” case. We can infer that she stopped the car, and then started it again to hit him. But it seems like poor reporting all around to rely on the reader to make these assumptions.

Lottery

I lost the lottery last night. Actually, I never even entered. The drawing was for $270 million (30-year annuity), or $164.3 million cash. The stakes were high, but I still didn’t dare risk life and limb to go buy lottery tickets. Further, I’d have to clear my car off and I was feeling pretty lazy. I hoped it would roll over. Sadly, this was not the case.

But some of us were talking about what we’d do if we won. Not the “I’d buy an awesome house” or lease a 10GigE line. But financial planning stuff. First, we started the debate over whether you take the cash or the annuity. $270 million over 30 years is $9 million a year. Not to knock $9 million, but if I won $270 million, $9 million a year would seem pretty pathetic.  So I pulled out my trusty old financial calculator.

Let’s call the $164.3 million a nice even $150 million. You take $14.3 million off the top to get your indulgences out of the way. At 4% interest (realistic enough for something like a treasury note) over 30 years, you’d make $500,000 interest a month. (Assuming that you took the $500,000 out each month to spend.) That’s $6 million a year, or two-thirds of what you’d be getting if you took the annuity. Except you’ve already paid cash for a house overlooking Hollywood and shared your riches with your friends and family bought a couple cars too. Conversely, if you didn’t take your $500,000 a month out, you’d have just shy of $500 million in the bank at the end of 30 years.

So I’d definitely have taken out the cash.

The other thing I overlooked in the past is the concept of using annuities. I might like to host a scholarship, for example. I’d want $50,000 each year to fund it. Assuming the same 4% interest, you’d sock aside ($50,000 / 0.04), or $1.25 million. And then each year take out the $50,000 accumulated interest.

But, alas, I didn’t win. So no 10GigE to my home (probably not feasible anyway–transit costs are much less than I expected, but the cost of a fiber circuit to your home isn’t accounted for, and I’m not sure you can just call up Verizon and ask for them to light up a 10GigE link to LAIIX for you…)

Criticizing Web Apps

As long as I’ve posted a lengthy diatribe about how awful the library room-booking web interface is, there are two more that drive me nuts.

We have a way of putting in work orders for maintenance. Last semester I tried to open one of our windows and it just fell out. This semester, we had three different light fixtures burn out in 2 days time. So you go online and put in a work order. This is a great thing to have web-based. Except they picked this insane system that opens multiple browser windows, resizes your browser (?!), uses copious JavaScript requiring you to double-click on links… And it only works in IE. Oh, and there are irritating things that could be fixed with one line of code… You log in with your student ID, which is eight-digits that inexplicably have an @ sign in front of them. So they have this big note on how you cannot use the at sign, you must only use your eight-digit number. One line of code could just strip it out if it was included.

Much like booking library rooms, submitting help tickets is a Programming 101 exercise. In fact, it’s easier than the library interface, because you don’t have to do time calculations. You have an employees table, a clients table, and a work table. Tasks get entered into work by the client, and the staff assigns an employee to it. And when it’s done, you set work.status to “Complete,” a simple ENUM field. This is like 45 minutes of coding, although I’d probably spend more time prettying up the interface.

Then there’s the computer help desk, another web app. For one thing, all the links to it point to an http:// URL. But if you actually use them, it barfs up an error that you must view it over a secure channel. Being a web dork, I just take “s” onto the end of “http” and life is golden. To someone who’s not so good with computers, and who’s already at wits’ end with their computer, they’re probably going to break down and cry, because even the help desk webpage doesn’t work for them.

This, too, only works in IE. In this case, they didn’t have copious bizarre crap (like requiring double-clicking on links), so I set Firefox to pretend it was IE. The page loads okay, but looks terrible, with nothing lining up right. IE and, well, the rest of the world, have differing views on how lots of things are done, but requiring IE really isn’t the best solution. Oh, and as an added bonus, they control your mouse cursor, preventing it from indicating links in any manner. This means that someone took time to write code that does nothing but decrease usability.

But worst of all, even if you use IE like they demand, if you actually try to click on any tickets to view them, you get taken to a random system with a long canonical hostname, which just throws you “HTTP 400 – Bad Request.”

So last night, I submitted a help desk ticket indicating that the help desk is broken. Because, frankly, it doesn’t work. All of its internal links take you to the wrong server (or, seemingly, the right server but with the wrong hostname), and that’s assuming you’re smart enough to get in, by understanding the error indicating that you need to use HTTPS, not HTTP.

Most of these things are sold as turnkey devices, it seems. Maybe I should start a company making them. Apparently, no technical expertise is required to do so.

Arresting Firefighters

This is wild. The fire truck pulls up to a car crash on the highway, and parks the truck to “shield” the emergency workers as they extract a victim from the car. The cop yells at the firefighter driving the fire truck to move so as to not obstruct traffic. The captain, who was actively working on the patient, yells for the fire truck to stay put, pointing out that they very deliberately parked that way for the safety of anyone involved. So the cop pulls the guy away from the patient and arrests him.

Of course, not all firefighters arrested are innocent.

Cold War

Anyone who’s learned about the Cold War will be familiar with the chilling fact (no pun intended…) that we came very close to a nuclear war.

But after reading things like this article, mixed with other anecdotes, I’m left wondering how on Earth we didn’t go to war… Accidentally. Both the U.S. and the Soviets, on multiple occasions, “detected” launches of nuclear weapons by the other, and came within seconds of retaliation before someone noticed something out of the ordinary.

Fortunately, the U.S. was very thorough the first time around, and quickly proved that the first “attack” they witnessed was caused by some guy inserting the wrong tape… In the case of the Soviets, the only reason they didn’t launch a counter-attack after their own false alert, it seems, was because the guy who was supposed to press the button disobeyed orders and went with his gut. (And boy are we glad!)

And there’s a further set of coincidences, really. After a flood of nonsensical data, officials discovered some problems. Apparently, one detection system was alternating between reporting some 2,000 incoming missiles and 0 incoming missiles. Because of the conflicting data, they turned to alternate systems, which also reported 0 incoming missiles, and it was traced to a hardware malfunction, with the 2,000 number just happening to match, by sheer luck (or lack thereof), internal checksums.

So they wrote some code to compare results from multiple systems. And not more than a few months later, the problem with the training tape occurred, when one of the systems began reporting more believable numbers of incoming missiles. (Apparently, a steadily increasing number.) The data “made sense,” but, because of the newly-implemented code to compare with other systems, they realized that it was just one system, and quickly isolated it to a case of someone sending “training data” as if it were live data. It’s almost a case of two wrongs making a right–had the first error not occurred, the safeguards wouldn’t have been implemented to catch the second error.

Oh, and there exists a slightly-creepy website dedicated to the Russian who decided to trust his gut over the myriad indications that we were attacking

Radio

I’m a long-term radio geek, and I’ve realized that the technology interests me more than actually using it. Having worked with lots and lots of radios (I realized that I have three sitting on my desk, all of which I have used in the past 30 minutes), I’ve concluded that I’d like to start a radio company. Our motto would be, “Our radios don’t suck.”

One of my radios is a ham radio, which is front-panel programmable (FPP), meaning that you can punch in frequencies on the keypad. This is pretty common with ham radios. By contrast, land-mobile radios (things that, say, a police officer would carry) very rarely have FPP capability; in fact, the FCC frowns on certifying radios with that capability, except for certain federal agencies that need to be able to reprogram their radios in the field. However, it’s often offered as a software add-on. But even using the ham radio, it’s really hard to use. Part of the problem is that the radio’s probably a decade old, and the print on the keypad has worn off. So I’m guessing at what buttons do.

There are very few radios with a graphic LCD. Dot-matrix LCDs almost seem cutting-edge in the radio world. By contrast, try to find a cell phone that doesn’t have a big color LCD on it. I have an old Garmin GPS III, and still admire that screen. I think it’s four shades of gray, and fairly high resolution. It’s a nice graphic LCD. It’s so much easier to use, and introduces stuff like the ability to “arrow” around a screen, as opposed to trying to use obscure key combinations. I’d actually love to see something like a 2″ by 2″ e-ink display (which, in addition to looking amazing, would reduce power usage), but it’d be a pain since it’s slow to redraw.

Motorola’s MDC1200 technology is practically ubiquitious in the public safety industry, transmitting a 1200 bps data burst containing a four-digit identifier. This could be so easily improved. Put a little $20 GPS chip in it, and have it transmit GPS coordinates on each transmission. (You could also include stuff like battery level, if on a portable, and information on received signal strength. The latter would be useful to run in the background and plot a map of the radio system’s reach.)

Programming is always a pain. Some of Motorola’s radios are programmed in ways that are so obscure that they border on comical. (I think the goal there is security.) I want to write an XML file for my radio. Put a USB port on the side of the radio. Let me hook it up to a computer, or just plug a thumb drive in and reprogram from that. But consider bigger problems, though. Boston PD switched to an “improved” channel lineup last year. Apparently they worked for weeks to pull radios in at the end of a shift, load up the new set of data, but leave the radios set to old configuration, until all the radios had the new programming in them. And then, at a quiet time one day, they broadcast a message telling officers how to switch to the new configuration. Over-the-air programming is possible, but it’s generally used in some specific situations. (OTACS, Motorola’s Over The Air Channel Steering, to direct a radio to switch to a particular channel, and OTAR, Over the Air Rekeying, to send new encryption keys to the radios.) Why not let the system send out bursts of programming data when the radio system is idle, loading up new programming data in the background, until they’re ready? Obviously, all of these programming things need some security constraints, but that’s trivial to implement.

I’m pretty confident that software-defined radio is going to become ubiquitous in the next decade, but no one’s really making use of it yet, except for uber-geeks in labs. APCO’s Project 25 digital voice (IMBE) has emerged as a standard in digital voice, but it’s meant to be made obsolete in the future by a “Phase II” implementation. Various other technologies have come and gone, such as Motorola’s VSELP. And there exist myriad trunking protocols for larger networks. I want to embrace SDR and use it in everything, “future-proofing” radios. (Of course companies have an incentive to not future-proof their hardware, forcing people to upgrade… But you can still make your money on selling software upgrades!)

Oh, and put an SD slot on the darn thing. Record the audio it receives, letting people play back transmissions they miss. Or host applications. (Or, permit programming!)