phik

For years I have put off reading The Rise and Fall of the Third Reich, afraid that it would be a dry tome, failing to deliver literary justice to the most horrific and astonishing years of last century. I was also mildly intimidated by its 1200-page bulk — too big to finish before I left town again, and too big to take with me. Last month I decided that enough truly was enough, and I can now say without hesitation that it is an uttery fascinating, phenominally well-written, grippy tale.

I’m about 80% finished, just past the point where the Nazi army storms across the Eastern front into Russia, thoroughly dominates the brave but woefully underprepared troops, and manages to get within 15 miles of Moscow when winter saves the capital’s (and perhaps the world’s) bacon. Despite all kinds of warnings — which, knowing Hitler as they did, should have tipped them off — it is incredible how surprised the Russians were by the attack:

Almost equally weird to [the German chief of staff of the Fourth Army] was that the Russian artillery did not respond even when the assault began. “The Russians,” he subsequently wrote, “were taken entirely by surprise on our front.” As dawn broke German signal stations picked up the Red Army radio networks. “We are being fired on. What shall we do?” he quotes one Russian message as saying. Back came the answer from headquarters: “You must be insane. And why is your signal not in code?”

A few hundred pages before was the story of Western reaction to the world’s first experience of blitzkrieg tank warfare. Britain and France had done precious little aside from declaring war on paper, in response to the utter extermination — tanks against horses! — of their supposed ally, Poland. This led the German man in the street to coin a new phrase for the purpose of mocking the Western effort — a sitzkrieg.

If you want my advice, you will begin a 1200-page sitzkrieg of your own. I cannot recommend it highly enough to those interested in modern history; and why isn’t everyone?

On and off for the last few weeks I’ve been reading from A Review of Criticality Accidents, a document that contains short reports about every known worldwide nuclear criticality accident prior to the year 2000. Some of these occurrences, and the situations surrounding them, are astonishing!

I learned that in the early days of working with fissile material, people could dismiss unusual happenings as unimportant — surprising to me, even considering their lack of criticality training (from process accident #1):

Immediately after disconnecting the hose, the operator noticed foaming and violent gas release from the vessel. … Neither of the two operators had any training in criticality safety and neither recognized that a criticality accident had occurred. They did not expect that there would be any health effects and they elected not to report their observations.

As with every single accident, the investigation showed that multiple procedure violations were the cause; the operator lost both his legs due to acute radiation poisoning.

I learned that the margins between critical and subcritical can be razor thin, and that the most seemingly innocent actions can make the difference (from process accident #3):

Then three of the experimenters manually lifted the vessel and began to move it (in order to directly pour the contents into containers) when the excursion occurred.

They immediately noticed a flash (due to Cherenkov radiation), and simultaneously, fissile solution was violently ejected, reaching the ceiling about 5 m above. … The combination of additional reflection from the three experimenters and the change in the geometry of the solution volume was sufficient to cause the system to exceed prompt critical.

The three experimenters died within a week.

I learned that “unfavorable geometry containers” were contributing factors in very many accidents. I eventually learned that favorable geometry describes a container that prevents criticality by simply not allowing concentration of too much solution in too small a space. A spherical container, for example, is the worst. Depending on the circumstances, a thin cylinder or a shallow flat container could be considered favorable — but the depth margins can be razor thin (from fissile solution systems #4):

The reactivity of such shallow, large diameter assemblies is very sensitive to the solution depth but quite insensitive to changes in the diameter. For this system [0.76 m in diameter, filled to a depth of 130 mm], the estimated difference between delayed criticality and prompt criticality is only 1 mm of depth. … It is thought that the falling scram, a cadmium sheet slightly deformed at the bottom, set up a wave system that must have converged at least once and created a superprompt critical geometry.

I learned that, once again, the cover-up was worse than the crime (from process accident #17, emphasis mine):

The shift supervisor insisted that the radiation control supervisor permit him to enter the work area where the accident had occurred. … In spite of the prohibition, the shift supervisor deceived the radiation control supervisor into leaving the area and entered the room where the accident had occurred.

The shift supervisor’s actions were not observed by anyone. However, there was evidence that he attempted to either remove the [60 litre] vessel from the platform, or to pour its contents down the stairs and into a floor drain. Whatever his actions, they caused a third excursion, larger than the first two, activating the alarm system in both buildings.

The shift supervisor, covered in Pu organic solution, immediately exited the room and returned to the underground tunnel.

For his efforts to conceal the multiple procedure violations, including improvised setups and unfavorable geometry containers, he received a dose of about 2,450 rem and died a month later.

It is a macabre document to be sure, and I’m more than a little surprised that it was even produced, let alone released publicly. But it’s largely understandable to the layperson, and an interesting insight into accidents that occurred in some of the most secretive places in the world.

I believe that tonight I witnessed a defining moment in American cinema. A film so powerful, it will unite a generation. I have never before been part of an audience that was so thoroughly and lustily indoctrinated before anyone had seen the film, and it was, to my great surprise, not without merit.

Samuel L. Jackson was absolutely right: Snakes on a Plane. You read the title and you either want to see it, or you don’t. Nobody in America is going to leave the cinema disappointed because they didn’t get what they were expecting.

Apart from Samuel’s zingers — which I looked forward to with great zeal — my expectations were pretty low, but I’m pleased to report that it is a genuinely entertaining film. It is what it is. We are not talking about Tuesdays with Morrie here, we are discussing a ridiculous summer thriller that excels by not overreaching.

Go see it before the opening-weekend enthusiasm wears off.

At first glance, Tracks looked like quite the nifty tool for managing tasks a-la GTD. Unfortunately, thighs slick with AJAX juices though it may be, I found the Tracks workflow to be very much at odds with my own.

I’m writing in the hope that others will be spared the trouble, or perhaps that the Tracks developers welcome the feedback. I should confess up front that I am not completely certain about these points; it is possible that I gave up too soon, or that these are largely known issues.

It gets off to a good start by directing you to the Tracks wiki. If you want links to other people’s blogs, where you need to digest the comments to learn what was insufficiently clear in the main text, you’re in luck. If you want a single straightforward “How to Install / Basic Configuration” page, I cannot point you to one. I know that this is not commercial software; I’m just saying, I would want to make my software as easy to install as possible.

More substantively, many of my Projects contain a large number of items which inevitably have internal dependencies, or are simply sequential in nature. When I view the dashboard or a given Context, I almost never want to see all of my Project tasks, I only want to see the ones that are actually pending right now. I’ll be the first to admit that this can be tricky to design properly, but the current view is cluttered with information I don’t want to see. If nothing else, give me a +/- button that will expand/collapse the project tasks into a single line item reminder.

Next: Is there a simple way to reorder Actions? Whether it’s for mental clarity or sometimes as a reflection of priority, order matters to me. A lot. I should be able to drag and drop Actions to reorder, or move them between Projects or Contexts.

Small nit: when you’re entering new Actions, I really like that you can just keep typing titles and pressing Enter and it does the right thing. But I wish the Context or Project selected in the drop-downs were sticky, instead of reverting to the top of the list each time.

Between the fixed ordering and the clutter that all of my Project tasks were generating, I had to give up before I could learn much more. Very promising! But not ready for my workflow yet.

(Also I upgraded wordpress, so your rss readers will claim that every entry is unread. You’re welcome.)

I made some pickled beets this afternoon, and while they are undeniably delicious, they lack the crunch, sourness, and zip of the Lebanese pickled beets of which I’m so very fond. Crunch I know how to handle. But the zip?

For once, Google has not bailed me out. Dear readers, do either of you have recipes for Lebanese-style pickled beets?

Update: In a stroke of genius, Rob realized that I was leading everyone astray; my objets d’affection are in fact Lebanese pickled turnips, flavoured with beet juice. Twenty days is a long time to wait before I know if this recipe is any good, but these are the sacrifices we make in a time of perpetual war against an ill-defined enemy with unspecified success conditions.

While I was in Toronto last week, I was afforded the opportunity to have dinner with the always spectacular Graydon Hoare, during which I discovered that contrary to my self-centered beliefs, most of my faraway friends are out of touch with my major life events. It should be explained for the historical record that I have some noteworthy experience at being a terrible long-distance friend, which will no doubt be chronicled extensively in the official biography.

Having just assumed that everyone who was interested already knew, it seems odd to write about this so far removed from the actual event.

Effective April 1 (almost five months ago!) I resigned from my positions as a director and Chief Executive of Cluster File Systems — which is to say that although I remain an owner, I’ve given up all operational involvement. After more than four years, I had taken CFS about as far as I could under the circumstances, and I felt more than ready to start thinking about new challenges and a change of pace. It was an incredibly (but not surprisingly) difficult decision to leave a company that I practically co-founded, but the more time passes, the more I think that it was the right choice. My own burnout was imminent, and the company was clearly stable and mature enough to handle the transition. If not then, when?

My intention was to spend the summer and perhaps autumn unemployed, as a way of making up for my failure to take any kind of serious holiday over the last five years, and to think about what I want to do next. But almost immediately — since it actually took well into June before everything was completely wound down — these plans were waylaid by someone I thought was my friend. I was tricked into a whirlwind summer contract to improve the management of something I knew nothing about (I like to think that I have not completely failed at this task), but I am pleased to report that September 1st will mark the commencement of the Autumn of Phik that we all so richly await.

So there you have it! The first question everyone asks is what I’m going to do after my time off, and I honestly don’t know — that’s at least half the purpose of the holiday. I will be watching with interest to see where Lustre and CFS go from here, but my labours will almost certainly be totally unrelated. I’m exploring an interesting part-time opportunity for the winter months, which I’ll write about if it becomes something real, but I’m hoping by the spring to be refreshed and ready to start full time on something new.

Two weeks ago, I completed the first phase of what would be accurately described as a life-long dream: I became a private pilot.

To attain a new certificate or rating, one must pass an examination (inf: checkride) to the satisfaction of an FAA inspector or designated examiner, according to the requirements outlined in the FAA Practical Test Standards (PTS). The exam includes an oral portion followed by a flight test (this order is now mandated, to avoid the concern that some examiners, after witnessing an excellent flight test, might be inclined to let slide a sub-par oral exam). Going into the test I found only a couple thorough descriptions of the practical exam experience, and having found them helpful and reassuring, will now contribute my own experience to the canon. If the details of flight training do not interest you, you may wish to return on another day. But for the non-pilots who want more information, I’ve linked in a few places to some background information for concepts or terminology that wouldn’t have been obvious to me before I started flying.

The weather was lousy that morning, and my examiner called while I was driving to the airport to suggest that we cancel. He was already at the airport, though, and I was halfway, so we elected to go ahead and complete the oral exam, so we could get straight to flying next time.

My examiner, a Mr. Ray Collins, is a highly seasoned aviator, presently serving as a 737 Captain for Delta out of Boston. He was professional, calm, and reassuring, explaining that he would brief me about what was expected, that he wouldn’t try to trick me, and that by law I was the pilot in command when we fly. I could ask for his help, such as being a second pair of eyes looking for traffic, or even asking him to fly in a true emergency, but I must remain in control and make the decisions from the moment we reach the tarmac until we’re safely parked on the ramp. He warned me that there are some things that simply must not happen; for example, our airport sits underneath Boston’s busy airspace, which you need a clearance to enter and begins (there) at 4,000 feet. If it comes to that, he has to sit there quietly watching the disaster looming, and if I reach 4,000 feet 6 inches before I’m clear of the airspace, he’s required to fail me. So don’t let it happen.

After we signed into the FAA’s relatively-new online application system, he reviewed my logbook for the required training and instructor endorsements — the least favourite part of a DE’s job — and we verified that everything was in the computer properly. The FAA is as strict about forms as any US Government agency, so the online system is a big improvement, eliminating a lot of the common mistakes in form-filling that apparently spelled the end of many checkrides before they even started.

We spent between 60 and 90 minutes on the oral exam, and it seemed to be much more about breadth than depth. He asked a question or two about each topic and moved on; presumably if my answers had not been satisfactory, he would have probed until he could determine how fundamental was my misunderstanding. For example, I misremembered the emergency radio frequency as 123.5, but mentioned in my answer that I wasn’t sure because I knew it was on my kneeboard. He replied that he is a strong advocate of having this kind of rarely-used, important information on the kneeboard, reminded me that it’s 121.5, and we moved on.

If memory serves, we began with weather (TAFs, METARs, and area forecasts), some basics of the airplane we would be flying (how much and what type of fuel it holds; some V speeds; the meaning of the coloured markings on the airspeed indicator; the pitot-static system), airworthiness directives, time between inspections, Emergency Locator Transmitter maintenance, night flying (position lights, airport beacons, runway lights), recovery from a spin, and oxygen requirements (day vs night, crew vs passengers, and critical altitudes). We reviewed my assigned flight plan from Bedford to Glens Falls, NY, and weight and balance information, then spent a long time on the sectional chart: airport markings (”Tell me everything you can about this airport, just from looking at the chart”), restricted and military operation areas, temporary flight restrictions, wildlife preserves, VOR and RCO information, controlled and uncontrolled fields, minimum altitude markings, obstructions, and airspace. That led us to cover the basic VFR requirements for flight in controlled and uncontrolled airspace, the requirements for operating in each class of controlled airspace, and four or five subtly different situational questions about whether or not a given conversation with air traffic control (ATC) provided sufficient clearance.

For example: I want to enter class C airspace, so I call ATC, and he responds with “Tomahawk 9182A, standby”; may I enter the class C? Yes, because he addressed me by my callsign, thus establishing the required two-way communication, and did not specifically instruct me to remain clear. Conversely, if he had responded “aircraft calling 10 miles to the east, standby,” then I must remain clear. The rules are subtle and differ between the classes of airspace, but they are sensible and uncomplicated; my instructor had covered this before I embarked on my solo cross-country flights, so they were no problem.

Finally, we covered pretty thoroughly two items that the FAA has been aggressively targeting, and are included in the PTS’s Special Emphasis Areas: runway markings / incursion avoidance, and wake turbulence.

Finished with the oral exam, we checked the weather again, and lo and behold the ceilings had climbed to about 2500 feet around Bedford — marginal for most real flights, but enough for a checkride. With the cold front passing through, it was windy: 17 knots gusting 25, and I was thankful that I’d been able to spend Saturday practicing maneuvers and crosswind takeoffs and landings in a stiff breeze.

He sent me out to preflight while he completed some paperwork, and once we were settled in the cockpit explained the beginning of the exam: we’d start with a short-field take off, then a soft-field touch-and-go to a soft-field takeoff, and he’d brief me on the next steps once we were back in the pattern.

Because of the weather the pattern was unusually quiet, so there was no hurry as I taxied to the very end of the massive 7,000-foot runway to maximize my available distance on this “short field”. One notch of flaps down, brakes on, full throttle, good static RPMs, release brakes, keep it on the center line, rotate at 55 knots, climb at Vx, and clean the airplane up after we’ve cleared the usual simulated 50-foot obstacle. No traffic makes for a quick pattern, and I’m soon easing it gently down for a soft-field landing, then getting quickly back off the runway and into ground effect for the soft-field takeoff.

On the downwind leg he tells me that we’ll do a no-flaps touch-and-go to a normal takeoff, and then begin the navigation part of the test. The touch-and-go was uneventful, and I requested a northwest departure from the tower. The flight that he asked me to plan to Glens Falls, NY is, by chance, very close to the route that I’ve flown several times to Keene, NH — so the visual checkpoints on the ground are very familiar to me. We crossed I-495 right on schedule, then the large (but abandoned) two-runway airport, and he decided that he’d seen enough — this is typical. If you get further away from the airport than two or three checkpoints on your private pilot checkride, something is not right. He reached over and pulled the throttle down to idle. The engine failed right over an abandoned airport; permit me to feign surprise at this fortunate circumstance!

He tested my emergency procedures in a way that I didn’t expect, but completely understand: he was more interested in my organization, multi-tasking, and decision making than in whether I could actually put the plane down on target. After I trimmed for best glide speed (intended to maximize the distance we can glide, now that the engine is out), went through the emergency checklist, and simulated trying to restart the engine, radioing for help, and using an emergency transponder code, he’d seen enough. Before I could even get set up for a simulated landing, he restarted the engine and told me to dial in the Lawrence VOR.

Once established on the VOR, he took the controls and told me to put on the hood, which prevents you from seeing outside the plane, so you can fly with reference only to the instruments (this page has more information about the basic instrument training we receive as VFR pilots). Under the hood I spent a few minutes maintaining straight and level flight, left and right turns to a specified heading, climbing and descending to specified altitudes, and one recovery from an unusual attitude. In the latter, he takes the controls and has you put your head down while he puts the plane at a pitch and bank that would lead to disaster if it were not corrected, then has you take the controls and recover with reference only to the instruments. My instructor prepared me well, by putting the plane in extreme situations that required dramatic and immediate correction (this is unsurprisingly one of the more troublesome aspects of flight training for those with weaker stomachs). By comparison, Ray’s unusual attitude was surprisingly gentle.

After that, the hood came off, and we did some maneuvers: a turn around a point (not as easy as it sounds, if you have a stiff breeze), steep turns (one normally turns using small amounts of bank, and the airplane naturally wants to return to level flight; steep turns use enough bank that the airplane wants to keep rolling if you don’t prevent it), a power-on stall in a clean configuration, a power-off stall with half flaps, and a right turn at minimum controllable airspeed.

At that point, all I had to do was get us home safely. I picked up the Bedford ATIS, contacted the tower, set up for a three mile right base entry, and landed with a small crosswind. An uneventful taxi to the ramp, and he went in to complete the paperwork while I put the plane away.

I use this until the FAA gets around to sending me a real card

Thinking back, it went with barely a hitch; my instructor had prepared me extremely well. There were one or two instances where I neared or slightly exceeded my altitude restriction (usually plus or minus 100 feet of the assigned altitude), but I was quick to correct it, and that’s most important. The turbulence from the cold front was always present, but was less troublesome than I expected. It was probably the best I’ve ever flown, in conditions which would normally have caused me to scrub the flight as a student.

It is often remarked that this is a license to learn, and the process of getting this certificate was (at least for this author) nothing if not humbling. I have already made my share of mistakes in the cockpit, but each was a learning experience, and I’m pleased to say that I haven’t made the same one twice (and ahem, for the benefit of those readers that will be among my future passengers or sharing airspace with me, I should note that these were not critical imminent danger-class mistakes).

Hanscom Air Force Base is a great place to learn; it’s a moderately busy airport, mostly general aviation and corporate jets, with a few daily commercial flights thrown in for good measure, and the occasional Air Force flight (although vastly fewer than I expected for an active base). It has a control tower, so you get used to interacting with ATC from the beginning. It sits underneath Boston’s class B airspace, and in a generally very dense area of the country, so you get used to paying close attention to airspace borders. It has enough traffic to keep things interesting and varied, but rarely so much that you waste a lot of time (and money). If it gets too busy — and when it does, fair or not, jets and turboprops tend to get priority over your 112 horsepower trainer — you can always fly to Nashua or Lawrence and practice there. And it’s a 30 minute drive from Cambridge, door-to-door.

If you were interested enough to sit through all that, then you’re either already a pilot, or you should be. If you’re in the Boston area, then I cannot recommend instructor John Nutt at East Coast Aero Club highly enough (although I do so at the risk of having you unwashed masses make it even more difficult for me to get on his very busy schedule in the future). Not everyone wants the same teaching style, but John is patient, friendly, and phenomenally experienced — he’s been doing it for decades, and with 11,000+ flight hours, it’s clear that he teaches because he loves it, not as a stepping stone to an airline career. To get his time during the busiest months, you need to plan your flights a couple weeks in advance, but it’s a small price to pay.

For a few weeks I’ve been using Akismet to tackle spam on the site. I’ve tried other plugins with various degrees of success, but this has been almost perfect. I think one false negative and zero false positives in those weeks.

Yesterday evening, the Akismet developers made an upgrade to their server software to start collecting better data about the service, and in the process confused our plugins by accidentally changing the protocol. As a result, all spam was getting through.

I have a hard time being mad at Akismet — software is hard, and it’s not like I pay for the service — and I’m sure they learned some valuable test cases for their next upgrade. But it’s a measure of the company’s dedication that I can get a support request turned around, as a non-paying customer, in less than three hours, on a Sunday afternoon.

It’s how you turn casual users into people who suddenly evangelize for you on the web. (ahem)

This weekend I was planning to visit a friend in Portland, Maine, and it turned out to be such a beautiful Saturday that I could hardly resist. There was just one problem with the plan: the President was visiting his parents in Kennebunkport, which means all but shutting down almost 3,000 square miles of airspace in the form of a temporary flight restriction (TFR in the lingo).

The white part is the circle-of-having-F15s-scrambled-at-you

Permit me to digress in a way that, for most of you, should be entirely predictable: I have a real problem with this TFR. Quite apart from the fact that it inconveniences me as a pilot, I have issue with the idea that our leaders are being elevated so far above the people that they ostensibly serve, whether it’s closing down busy airspace, being exempt from parking tickets, or collecting large federal pensions and benefits. There’s a permanent one-mile no-fly-zone over the Bush compound that will probably remain in place until the end of time, and I argue that nobody is that important.

But until you elect me and I can do away with them (phik in 2016), the TFR was in place, so it was a good opportunity to learn how to operate within it. The TFR prohibits a very large number of useful and productive activities: all flight training, IFR practice, aerobatics, gliders, parachuting, ultralights, balloons, crop dusting and other farming, and banner towing are all prohibited anywhere within the 30-nm radius. Furthermore, there’s a 10-nm inner core no-fly-zone, with exceptions only for flights associated with the Secret Service, other law enforcement, or emergency medical flights — that 10-nm ring includes Biddefort airport, effectively closing it for three days. And anyone in the busy Portland area could forget about their flight lessons during this beautiful weekend. You’re welcome!

Actually operating within the TFR is not too difficult: you need to file a special flight plan, squawk a discreet ATC-assigned transponder code, remain in radio contact with ATC at all times, and go directly to your destination. Frankly, I would have done all of these things anyways for reasons of safety and efficiency, but I did have to alter my route to avoid the no-fly-zone and keep from finishing a very bad day in prison.

Once in Portland we flew to Laconia (KLCI), which was absolutely stunning. It’s such a beautiful airport to visit, because it’s surrounded by lakes everywhere that it’s not surrounded by hills, and it makes you think seriously about parking the plane and having a picnic, or renting a boat. Maybe next time.

And so it was that I carried my first passenger, who seems to be no worse for the wear. Since I started flying the Warrior this week, he was also treated to some exceptional landings, if I may be permitted to sound my own bugle. By training on the Tomahawk, which has no shock absorbers on its main landing gear, I had to work overtime to land without compromising my spine; every tiny bump was transmitted directly to one’s brainstem. The Warrior, however, has oleo struts on all three wheels, so these Tomahawk-trained landings are greasers every time. It doesn’t hurt that the wings are also huge by comparison, helping you settle oh so gently down, like a feather landing on a kitten, asleep on a pure down pillow.