No Surprise

Published at 15:45 on 6 May 2019

In the least surprising news development since the Sun rose at the forecast time this morning, it turns out that Alexa and Siri are, in fact, home eavesdropping devices.

George Orwell was an optimist. In Nineteen Eighty-Four, everyone had a telescreen in their home because the government forced them to. In today’s USA, people agree to it because advertisers have convinced them it’s personally convenient.

New Software Won’t Fix the 737 Max

Published at 07:18 on 4 April 2019

Disclaimer: I am not an aircraft engineer. But I am a software engineer, one who looks at my own field with a critical enough eye to see how software is often used inappropriately, and I see the signs of the latter all over the place in this latest story.

The original software didn’t fix its fundamental unairworthiness, so why should new software be able to? The problem with the 737 Max isn’t that it has buggy software, it’s that it should never have been built in the first place. Its safety should come from its airframe being compatible with its engines. It can’t come from a software-and-sensor kludge that tries to compensate for an unsafe physical design.

In an article in today’s Washington Post:

Boeing said it would take about an hour for technicians to load a software update for the planes. The company’s software fixes will change the way the MCAS receives information, requiring feeds from both outside “angle of attack” sensors, rather than one, before it is triggered.

The system will also have more limits on how often it will engage, and Boeing will make changes that prevent the anti-stall feature from angling the plane’s nose too far downward in its attempts to correct for a possible stall.

Let’s take the fix of requiring both sensors to concur. We know the angle of attack sensors are unreliable, because they sometimes falsely indicate an excessive angle of attack. Being unreliable, it seems reasonable to presume that they also sometimes fail to indicate an excessive angle of attack. So this “fix” will actually fix nothing. It will merely trade one form of unsafe behavior for another.

The second fix is in fundamentally the same category as the first: like the former, it makes the system more conservative in deciding when to engage. That system was put there for a reason: the attempt to compensate for an unairworthy plane, whose airframe mismatches its engine size and placement. The physical plane will remain as unairworthy as before, only with less software compensation for it. Again, one problem is merely being traded for another.

Instead of tragedies caused by planes falling out of the sky because MCAS engaged in error, we will have tragedies caused by planes falling out of the sky because MCAS didn’t engage and they stalled.

I strongly suspect the only fix for these planes will be to scrap them and sell their bodies to recyclers, who will turn them into new metal stock from which fundamentally safe planes can be built. Those “fundamentally safe planes” will mostly be Airbus A320neo’s. Boeing’s attempt to get out of the corner they found themselves in the cheap and devious way is going to end up costing that company a lot.

Android Smart Phone First Impressions

Published at 18:17 on 30 November 2018

Yes, I’ve bought one of the things, at long last. Not because I really want one to be part of my life, but because I’m thinking of being a technology assistant (focusing on senior citizens) as one of my self-employment ideas. Whether I like it or not, smartphones are one of the most common and most complex things people interact with on a daily basis, so it behooves me to familiarize myself with them before I start claiming I can help others with them.

Furthermore, the majority of web browsing is now done on smart phones, and it’s very difficult to to design a good mobile-friendly web page unless you have a mobile device to test it on.

I decided to purchase an Android, because those cost approximately 2.5 times less than iPhones, yet have about five times the market share. I expect I will acquire an iPhone as well, because I want to be familiar with both models.

Much of it is as expected. I don’t like touch screens and probably never will. The overall impression is of a device that wants to hijack my entire life and take it over. For example, it begs for email addresses then automatically configures its mail client, with notifications enabled, for any accounts you entered. So every fucking email you receive will make your phone emit a noise and disturb you. In fact, the default is for notifications to be on for every application on the phone. That’s right, the standard configuration is for the device to be as intrusive as possible. This can be defeated, of course, but it’s quite telling.

The Data Saver mode could be better. This tells applications to put a lid on the data usage unless you’re connected to a WiFi network. It’s not as good as booting them off the Internet entirely, but it’s a start. Ideally, one should be able to set apps to be in one of three mobile data modes: unrestricted, restricted, and disallowed. The latter mode should be enforced rigidly at the system level; networking calls would basically fail if it is set. Why should an app be allowed to hijack my phone service and force me to pay against my will to feed it data? Because it’s a device built on the premise of hijacking my entire life, that’s why. In a way, I’m actually surprised there’s any sort of Data Saver mode at all. An imperfect limit on network usage is still better than no limit at all.

It’s an unlocked phone. Enabling it was a simple matter of transplanting the SIM card from my existing Nokia 3310 3G to the new phone. Disabling it will be a simple matter of of reversing the SIM card transplant. I plan on doing that after carrying the new phone with me for a few weeks to better learn it.

Then the phone gets powered down and wrapped in foil or put in a metal box. Why? Because (cue scratched gramophone record), it’s a device built on the premise of hijacking people’s lives. You can’t fully turn it off with a physical on/off switch. You can’t remove the battery, either (at least not with special tools). This is standard for smartphones. As such, there’s no way of being certain it’s not being used as an eavesdropping device, unless you put it in a Faraday cage.

New Internet Service

Published at 23:21 on 6 November 2018

I’ve gotten a few mailings from the phone company advertising DSL internet service. The prices seemed very attractive (about one third the existing cost I’m paying the cable company). So I called to investigate, and found out that there was basically no bait-and-switch; it would be that much cheaper. It’s not as fast as the cable company’s service, but my calculations indicated it would be fast enough.

The vastly cheaper rates aren’t even the best part. That’s being able to finally sever my relationship with Comcast, which has a very well-deserved reputation for being the most hated company in the USA. Mind you, I’m replacing them with the phone company, which is it’s typical bureaucratic and inefficient self. I’ve had to call and go through phone trees for things that there should be self-serve options online for.

But, it’s still vastly better than dealing with Comcast. At least Century Link’s service options are simple, up front, and understandable. Want phone service? Order phone service. Want internet? Order internet. Want TV? Order that. Order more than one service and get a volume discount. No special “packages” that make it cheaper to order TV service you don’t need for 12 months (then you have to negotiate another deal, and it’s always confusing, and you can never just pay for what you want and leave it at that). No commitment in advance to subscribe for a minimum period of time.

Tomorrow I call Comcast and tell them goodbye. Can’t happen soon enough.

More on That in This Post

Published at 19:28 on 1 October 2018

A couple posts ago, I wrote:

Part if it is that I may be moving further west and simply not visiting this particular cranberry-harvesting spot in future years (more on that in another post).

Today, it became rather more likely that I will be doing just that. At my most recent job, I interviewed for a software developer position. I was informed that there would be some on-call duty to support mission-critical software in those cases where front-line people can’t resolve the problem, but not much.

While I positively loathe on-call duty, I’ve managed to shirk it in the past by taking pains to release only well-tested software, and engineering in reliability to the code I write (e.g. designing things so that if components fail, the consequences of the failure tend to be less severe and self-healing). My code would sometimes fail over a weekend (nobody writes perfect code), but never badly enough that I’d get called to put out an emergency fix.

Such shirking works if you’re a developer (and employers love it; it means you’ve written reliable software). But for a sysadmin, it’s basically impossible: emergency response is a core part of the job. It’s one of the reasons I got out of systems administration and became a developer. One of them: I also simply find the creative aspect of designing and writing code to be intrinsically fun in a way that messing with system and network configuration parameters never can be.

Anyhow, it turns out that the position which had been advertised as a developer job (and which I had been hired for) had morphed into a systems and network administration one in the months between when I interviewed and when I was hired. Or so my boss said this afternoon, and I have no real reason to doubt him; he comes across as basically an honest guy.

I just wish he hadn’t assumed I’d be OK with that just because I mentioned having been a systems administrator in the past. I never mentioned the part about getting burnt out doing it, because I didn’t want to appear negative.

I’m resigning the position. There’s really no alternative. When I burned out on systems administration in 2002, I was so thoroughly burned out that I adopted what I call Rule No. 1: no more systems administration, no matter what. It’s a good rule, and a necessary one: I’ve come to despise systems administration so much that any stint of it I do, I’m fated to be resentful and do a terrible job. I’ll just end up getting canned for poor performance within a year, anyhow. Then I’ll have to recover from that. Better off to nip the problem in the bud and get out now. I call it Rule No. 1 for a reason.

I guess the moral of the story is that there is simply no good way to mention past experience in systems administration in an interview. Either you signal a whiny, negative attitude (if you mention being burned out on it), or you signal a willingness to do systems administration. Damned if you do, damned if you don’t.

At this stage, it’s becoming increasingly clear that it was a mistake for me to get a computer science degree so many years back. I’ve almost never had good high-tech jobs, and the few good ones haven’t lasted. As the old saw goes: the definition of insanity is doing the same thing over and over and expecting different results.

So it’s time to get truly busy with the difficult process of moving on from high-tech work. What that will be exactly, I don’t know yet, though I do have more ideas than I’ve had in the past.

What I will say is that it probably doesn’t make much sense to continue living in the Seattle area:

  1. The availability of tech work is why I decided to move back to this area, and that point has now been mooted.
  2. The Seattle area has become increasingly expensive; given how my new income is going to be significantly reduced, I’m better off living someplace more affordabole.
  3. The above is particularly the case given how I don’t think Seattle is really that great a city; it suffers too much from too many decades of poor planning and lack of vision. There’s not enough large parks near the urban core, and Seattle’s mass transit is decades behind most other West Coast cities.

All in all, I’d love to live in a place like Portland, if my allergies weren’t so bad there, that is. Portland has Forest Park, and great mass transit. I can’t have both the city and nature like that in Seattle; I must choose one or the other. If compelled to choose, I will choose nature every time. Conveniently, that’s also the option that involves a lower cost of living.

So it’s likely I’ll be moving further away from the big city, probably to the Olympic Peninsula, though it’s still very early in the visioning and decision process and that could easily all change.

If So, I Move to Linux on a Commodity Machine

Published at 08:06 on 6 August 2018

If Apple is really going to dump real keyboards, I will stop using their products.

Per the article, yes, Apple’s current keyboards have very little travel. As such, they have very poor tactile feedback and I find them unpleasant to use. I haven’t bought an Apple keyboard for a desktop machine in years, and if I’m using my laptop on a desk, I will plug it into a real (classic IBM Model M) keyboard using an adaptor.

If I’m traveling, I put up with the suckiness, because basically all laptop keyboards suck. The laptop form factor dictates the small travel that makes them suck.

But if Apple makes its laptops emulate the awfulness of a stupidphone, game over. The lack of real keys is one of the big reasons why I refuse to get a stupidphone.

And yes, “stupidphone” is a much more accurate term for the things, considering:

  • Non-existent tactile feedback, as already mentioned,
  • Very limited battery lifetime,
  • Bulky, awkward size,
  • Poorly-coded, software-based user interface that makes use as a phone more awkward than a traditional cell phone.

And I’m Back

Published at 18:00 on 30 July 2018

Since the main focus of the trip was for volunteer work, which is actually work (just because work is unpaid does not mean it is not work), there is paperwork to fill out.

Encouraged by others, I decided to give the electronic versions of the paperwork a try. They’re all Microsoft Word documents, of course. Despite Word being a mediocre typesetting program, and being properiety, and being expensive, it is what virtually everyone uses. And may I add much to my annoyance due to the previously mentioned factors.

“Just use Open Office” they said when I mentioned not having (nor wanting to squander money on) Microsoft Word. I should have realized what the outcome of that exercise would be, but me being generous to a fault, I gave it a try.

Naturally it was an exercise in revealing just how low most people’s standards are when it comes to document layout. Neither Word, nor Preview, nor Open Office rendered the forms properly. All made messes of the layout which rendered any attempts at filling them out ambiguous. That may not matter to most, but it does matter to me: I don’t want my submitted forms to simply result in back-and-forths with multiple clarifying questions (or, worse yet, incorrect data being entered).

Three strikes and electronic document submission is out. I ended up printing the form out and filling it in by hand, as usual.

Silicon Snake Oil

Published at 21:00 on 18 June 2018

This morning, I ran across two stories that perfectly illustrate the concept of silicon snake oil, the all-too-often-believed line that technology is going to change everything (for the better, of course) and you really have no choice but to be an enthusiastic adaptor of all of it.

  1. Duck Unchained, an article in Dissent magazine about a French newspaper that continues to be very successful despite being extremely judicious about the technology it adopts. It has no online edition; its web presence is limited to a small site that allows customers to purchase a (print) subscription online.
  2. The Tesla Model 3 cost $28,000 to build, German engineers say—and it still may not be profitable, an article in Quartz magazine about how Tesla’s enthusiastic and insufficiently-questioning embrace of technology it hurting its bottom line.

I’ve linked this article and made the point before, but in many ways the Amish are one of the most technologically sophisticated groups in the world today, because they neither unquestioningly reject new technology (as many think) nor unquestioningly embrace it. They evaluate it, then decide if it is a net benefit or a net loss to their overall society.

It’s why I’ve personally decided I want nothing to do with a “smart” phone, a “smart” home or “smart” appliances: any benefits I’d get would be extremely likely to be overshadowed by the harm caused by increased complexity and decreased reliability.

Well, That Sure Was No Surprise

Published at 14:52 on 1 May 2018

The culprit for the failing aftermarket back-up camera in my truck was the solderless quick-connectors I used (per the recommendation of the camera manufacturer) to wire the camera to power.

This was so much not a surprise that I didn’t even bother to do any troubleshooting to pinpoint the culprit. I simply removed the solderless connectors, cut and stripped the wires in question, twisted them together, soldered them, and taped them.

Then I started my truck and put it into reverse. Bam! Fixed.

This has been consistent with the (crap) performance of virtually every solderless quick-connector I have tried in the past forty years, which is why I felt so confident attempting this repair without further troubleshooting. For some reason, I was willing to give them a try again when I installed that camera. Lesson learned: never again.

A soldered connection firmly and securely bonds two connectors together at the atomic level. Assuming copper wires, the molten solder actually partially dissolves the surface of the copper conductors before it solidifies, resulting in one seamless conductor (transitioning from copper to solder back to copper) that is virtually immune from oxidation or vibration induced failure. Nothing else even remotely comes close to this reliability, except a seamless connector with no splices whatsoever.

I am convinced that the only reason quick-connectors exist is: a) manufacturers who want to cut corners on their assembly lines in order to pad their profits, or b) people, usually do-it-yourselfers, who don’t know how do solder and who are unwilling to learn how.

The Key to Printing 19th Century Modern Serif Fonts

Published at 21:10 on 26 March 2018

Use the highest resolution you can. Do not trust the defaults at a print shop to be reasonable. Do not trust the defaults for your software’s PDF generator to be reasonable.

Both defaults might well be reasonable for most of the fonts popular with contemporary tastes in typography, but the fonts popular in the 19th century were crafted in part to show off how the ink and paper technology of the day had progressed to the point where the fine details they employ were possible.

I found that when using Monotype Modern, the thin parts of the strokes showed up so poorly with 10 point body text at the default printer resolution, that the readability of the resulting text was seriously compromised. This might be part of the reason why such fonts have a bad reputation for readability: modern print technology can fail them.

Do everything at the highest resolution possible. An output of 1200 DPI is the bare minimum, with 2400 DPI being better (letterpress printing with hot type had an effective resolution of around 2000 DPI). By “1200 DPI” I mean 1200 DPI in both axes, on a black and white printer. (Color printing uses clusters of 4 dots, and printer makers use weasel wording to flatter their products, so a “2400 DPI” color printer has only the resolution of  600 DPI black-and-white one.)

Using the highest resolution the printer can print should not typically cost more; most shops charge the same per-page fee whether you tell their laser printer to print at a degraded resolution or its best resolution. If you can’t even get 1200 DPI, take your business elsewhere; the shop you are using has substandard technology.