The other side of the coin is that I am very dilatory about installing any kind of update to anything because a) experience shows that the chance of an update breaking something in a serious way is something like 10-20%, b) the problem may not be obvious in the first five minutes or the first week of operation.

My wife's PC has now been rendered unbootable TWICE by Microsoft pushing through bad updates. I personally will not install a Mac OS update until I've taken the time to do a local backup to a hard drive, a remote backup to a cloud backup service, and waited two weeks to see if Apple retracts and re-releases the update, and read Macintouch for user reports to see what kinds of problems people are having.

I've already bought a new router once because the manufacturer's firmware update broke it and it was easier to buy another than to troubleshoot it.

I haven't got the time or energy to do that on a dozen household appliances.

The software industry has got to figure out a way to make sure that updates are one or two orders of magnitude safer and more reliable to install than they are today.

The old saw was that when corporation builds fancy new corporate headquarters, it is an indication of an "edifice complex" and a red flag.

The combination of a fancy new building _and_ a self-aggrandizing book seems dangerous to me.

An Wang's "Lessons" was published in 1986, about five years before Wang collapsed...

"The Ultimate Entrepreneur: The Story of Ken Olsen and Digital Equipment Corporation" came out in 1988, the year DEC merged with Compaq... it's a little hard to date the "collapse of Digital."

My wife, who is sort of the idea non-techie user--follow directions, does virus scans, etc. is almost ready to abandon her Windows PC and see how well she can get by with an iPad. She is just totally ticked off at Microsoft. She bought a Windows PC with Windows 8 preinstalled, to avoid any possible upgrade hassles.

She found Windows 8 disturbingly close unusable, but gritted her teeth and started to learn it. Windows 8.1 managed to change enough things to be disorienting, without actually be an improvement. Then her PC was twice rendered unbootable by routine updates--in one case it seemed to be a case of dueling updates between Microsoft and HP, another time it was a faulty update that autoinstalled. (In both cases the "solution" was to boot in safe mode and roll back to the previous checkpoint).

Then came the forced Windows 10 upgrade, which again managed to change enough things to make the system harder for her to use without really improving anything.

Somewhere along the way the bloatware program she used to manage her photo library, which had come preinstalled and automatically associated to jpg files, so she was seduced into using it, stopped being compatible with Windows.

I think 10 to 10.1 has been painless, though.

The whole user experience of moving from Windows 7 to 8 to 8.1 to 10 has been so badly mismanaged that it is easy to see why anyone who isn't forced to use Windows might abandon it for a tablet.

...to do a dozen things I don't understand why it needs to do. It's great that I can at least see what it's doing, but if I only downloaded apps whose permissions I really agreed to, I wouldn't download any.

The majority of them want my location, which I consider to be very sensitive information, for no obvious reason. Now as a matter of fact I have "location services" turned off--and quite a lot of them will lock up or crash if location services are turned off. So I end up deleting them.

The general quality of Android applications is just too low. I want an application to do X, I see ten competing applications to do X, I can't tell which is best--apart from astroturfing, user star ratings reward feature bloat rather than usability... and it's just too hard to download five of them and do personal SQA on them.

I've been wondering about this. We just got back from a nine-week camping road trip in which we visited fourteen states, and so far my record has been 100%: I NEVER was able to use my chipped card in a chip reader. Not once.

Let me be punctilious: at hotels and restaurants I couldn't always see what they did with the card, so I don't know for sure THEY weren't using a chip reader.

A very conspicuous absence was pay-at-the-pump gas stations, and that's a pity because that's said to be a common place to find skimmers. I did run into a pump--major brand at a service plaza on an Interstate--that declined my card when I swiped it. I went into the office, they had a chip reader on the POS terminal but they told me it wasn't working, and swiping didn't work their, either. I called the credit card company, who said there was no problem with my card... they had no record of the purchase and decline... and when I asked about security they said "Oh, you don't have to worry about that because your card has a chip in it."

Given that there was supposed to be a hard deadline of October 2015, yes, "disaster" sounds accurate.

The only sense I can make of it is that the banks don't actually care at all whether the system is implemented, they just want to cost-shift the costs of fraud to the merchants.

My primary card doctor is reasonably young and when I started seeing her, she keyed in notes about treatment plans and such right into the office computer. So I know she's comfortable with computers and that's she's a fast typist.

About two years ago, when she came into the exam room, she was followed by a young person with a laptop whom she introduced as "my scribe!" Her scribe was constantly tapping away at the laptop, taking notes and entering orders and so forth.

I don't honestly know whether this is good, bad, or indifferent, but it certainly is evidence that the burden of data has become so overwhelming that doctors need assistants specifically to help with that.

She works for a gigantic megapractice that is proud of being a Patient Centered Medical Home and an Accountable Care Organization and all that good stuff, so I think they are following current "best practices."

Geezer reminiscence on. When I was a kid, the doctor's office had a big lab, where they had microscopes and hemocytometers and did their own lab work, and a small business office. Now the labs are gone--they send all the lab work out. The business offices occupy a third of the floor space, because they need room for people waiting all day long on hold to talk to insurance companies. And they have to hire scribes to help the doctor with the data entry. Maybe it's progress.

...when they saw it. Under the name "RRS Sir David Attenborough" it will drop out of the public eye, do some good work, and be forgotten in thirty years. Under the name "RRS Boaty McBoatface" it would have been the subject of children's books, stuffed toys, animated cartoons, been remembered for a century, and inspired a generation of kids to become polar researchers.

While not intentionally funny, the HMS Beagle and the DSV Alvin don't have the most dignified names in the world, and the scientific work they did is none the worse for it.

...the update could have detected that it was running on a misconfigured motherboard. It could have issued a warning, containing directions on how to make the appropriate settings... with a PRINT option. It could have refused to install on a system that Microsoft knew it the update would damage.

Microsoft chose to do none of these things. Microsoft chose to hurt people who had paid money for their operating system.

Because it is beneficial to almost everyone in the industry to believe that "everybody" uses only the newest gear, there is a systematic distortion of the facts of what might be called "product demography." I've seen this everywhere I've worked, including several years at a (long-gone) Fortune 500 computer company.

It seems that almost everyone relies on 15 and 20-year old equipment. Everyone scratches their head in amazement at what's in the back of the server room and the unbelievable story of why it is still in service--but it is there.

I've had several conversations with people at the computer company that went about like this.
"We don't need to support that model, it's too old, nobody is using it."
"I think a lot of people are still using it."
"Why do you think that?"
"For one reason, because we still use it ourselves."
"WHAAAAT?"
"Sure. Check with Lewis on the 4th floor of building III. They have three of them."
"What on earth for?"
"Because of [reasons X, Y, and Z]. And they can't get rid of them because the new models [have problems Q, R, and S].
"Oh, well, that's a completely unique situation. Nobody else in the world is using them."
"Trust me, if we're using them our customers are using them. Unless you believe that everyone else in the world is better managed and more up-to-date than we are."

I continue to be amazed at how high the "bandwidth" of a traditional, printed, paper highway map--such as those still provided by AAA, and frequently by the states themselves--compared to anything you can get electronically. Scrolling a six-inch screen is no substitute for a square meter of paper surface printed in high resolution... and with judicious human preselection of points of interest.

For your typical 150-miles-to-a-specific-destination trips I continue to try to make do by printing out relevant Google maps, a small-scale one for the major highway routes to get there and a big one of the neighborhood. It never really works. The GPS and our car's NAV system will get you from point A to point B and show you in very good detail the local roads immediately surrounding your present position, but don't work very well for planning.

Nor are electronic maps very good for sketching, highlighting, or carrying with you. And paper maps don't need to be recharged.

The movie suggested that the "Titanic" wasn't strong enough to support half its weight, levered and elevated, unsupported in the air. In the movie, the ship snapped in two (without the pieces fully separating).

This would have happened somewhere around 2:40 in the simulation video.

I guess this is just another illustration that Titanic buffs disagree with each other.

This professor's simulation indicates it did break--but not the way Cameron's movie showed!

All kinds of AI have teething pains, during which the problems are obvious and comical (the Apple Newton's handwriting recognition being a case in point). At the same time, the achievements of modern AI are amazing--but also troubling.

When I compare AI as envisioned in the 1950s--Isaac Asimov's Multivac, or his robots, perhaps--the assumption was that AI would be closely similar to human intelligence. For example, it was implicit that robots would answer questions by actually understanding them. What we are seeing today evokes an analogy with technologies like the sewing machine. Early efforts attempted to sew the same way humans did, and failed. Singer's brilliant idea was a method of using thread to fasten two pieces of cloth that did not resemble human sewing or even use the same stitch.

A Google search is within shooting distance of Multivac. You type in a question and you get a useful answer. The interesting thing is that most modern AI is shoddy. It goes halfway. It gives you something that's inaccurate, yet useful. But the key thing is that you are expected to use your human intelligence to get the rest of the way and correct mistakes. In the case of Google, you do this by looking at a ten or a hundred search results, for example--and reformulating the question if you don't get the right answer.

Perhaps one of the things that early AI pioneers missed is that modern AI relies more on having huge databases of information than would have even been imaginable in the 1950s and 1960s, and less on AI actually mimicking human intelligence.

This is not a problem when it is all open, the AI is offering you something to look at and not making decisions for you, and it is all in the nature of help or suggestions rather than direct action.

It becomes far more serious when it is happening behind the scenes--when AI is deciding whether you get a loan, or pass an essay test on an exam, or get onto a terrorist watchlist.

"Wouldn't it be great if you could read a novel in an hour or two?"

You used to be able to. Reader's Digest used to publish Reader's Digest Condensed Books. In effect, they did the speed-reading for you. I have to say that they did a very skillful job of the editing, too. Very impressive. But not really that enjoyable to read.

They don't seem to be around any more.

I've seriously wondered if the gradual adoption of more and more standardized icons and emoji is slowly creating an ideographic, common world written language.

I know that a few "fast forwards" and smileys here and there is a long way from verb tenses, and I don't think I've seen people use a string of several symbols to create a meaning that's different from the sum of its parts. But we're only a couple of decades into the process.

"The good news is that you have 30 days to downgrade to the previous version of the OS."

Imagine the average retail user, not expecting an upgrade, not prepared for an upgrade, so perhaps no recent backups and (if it's a home user) perhaps no backups at all. No IT department scan to check application compatibility and peripheral compatibility.

(And does this unsolicited upgrade check to make sure the computer meets Windows 10 system requirements?)

An installation on top of an existing installation, jumping two versions in between (8 and 8.1). But at least going in the direction Microsoft wants, and therefore probably SQAed as well as Microsoft knows how to SQA.

Now suppose the user attempts a downgrade to Windows 7. Another system installation on top of an existing installation, and again jumping over two versions, but this time going in the direction Microsoft thinks is unwanted and unimportant and probably has not tested quite as thoroughly.

What do you think are the chances it works _well_ afterwards?