Tom Pittman's WebLog

Earlier this year

2023 November 22 -- Peter Principle on Steroids

It seems the "Peter Principle" is far more prevalent than Laurence Peter imagined. Not only do people get promoted until they arrive at a job for which they are incompetent in government and large corporations, this also happens among the self-employed, and if nothing else, the job itself becomes more complicated until everybody doing it -- or even attempting to do it -- is inadequate to the task before them.

I first noticed this in my own work some 50+ years ago. The microprocessor revolution had just started and I had enough computational competence (and no secure employment) to jump in with both feet. As an early (and the most famous) developer of TinyBasic, I discovered I did not at all understand the theory behind Dennis Allison's implementation notes originally published in Bob Albrecht's People's Computer Club tabloid, but I was a pretty good programmer. I wasn't first (I didn't have a computer big enough to run it, so I had to borrow time on somebody else's) some guy with an 8008 computer beat me, but my solution was arguably the most elegant: I wrote an interpreter to execute Allison's code essentially unchanged (except he had a couple bugs I fixed). I thought a Fortran compiler would be a fun next project, but I had Clue Deficit Disorder (CDD, aka Peter Principle) and ended up with a mass of spagheti code I could not debug. I went to grad school to cure the problem.

A few years later I tried to build a DragonDrop programming environment as a replacement for HyperCard, which Apple was trying to kill. They succeeded, but I did not: it was too big. Apple also experienced this problem: they built an awesome operating system with a dedicated following, but it was losing market share because Apple did not understand what sells computers, so they made it even more complicated by changing the platform and interpreting the old code. I understood interpreters and compilers, so I thought I could write a recompiler to make the code run at machine speed, but discovered (rather late in the game) what I called "the RISC Penalty": it worked, but ran slower than the interpreted code (because "object-oriented" C++ code was much less memory efficient than old straight code, so it didn't fit in the processor cache, which had to keep reloading). There were at least three of us (including one at Apple) independently trying to do this, and we all ran into that same problem, which was eventually solved by the hardware guys making a bigger cache. It was less of a problem for non-RISC computers like Intel because their code was smaller. Besides, they had no previous version to compare to.

Fast-forward some more years, I decided that I could use my PhD material to create something like a  compiler to translate the Bible. CDD again. I mean, I got BibleTrans (sort of) working, but it was not a one-man job, I really needed a dozen people working a year or more to encode the whole New Testament, then three or four times that to do the Old Testament, and God did not give me the management skills to assemble and motivate a team to do that. I mean, I tried but failed. Peter Principle. Read about it on the website I created:

I've been trying to read the Bible in Greek and Hebrew. I've always been terrible at memorizing things like vocabulary, so I figured I could use some of the data I collected for BibleTrans to build an electronic version of the Bible text with pop-up hints about word meanings and grammar. I just needed a decent dictionary. There are public-domain dictionaries online for Greek, but the only Hebrew one I could find was scanned pages from a book printed in 1848. It's a fine dictioinary, I have the printed book I use a lot, but it's 900 pages. has the scanned pages, but they tried to convert it to a text file using OCR (Optical Character Recognition) software that doesn't know Hebrew. It's unreadable. That was nine years ago, I'm still working on it.

Then an old friend called out of the blue, and sort of invited me to work with him mentoring (his word) high school students in a summer computer workshop. Peter Principle from the get-go. I'd never worked with high-school kids. The first couple years went OK -- after they decided that this old geezer actually knew the technology. But those kids graduated, and the new crop rebelled. Besides, the other guy was trying to get funding, and he was also running into the Peter Principle. Well, sez I, everybody talks about teaching disadvantaged kids STEM (Science, Technologh, Engineering and Math), let's teach programming aimed at those kids. Remember BibleTrans? Same problem. I cannot persuade an Eskimo to buy a fur coat if they don't want to. Besides, the other guy was still "Petered out" getting funding.

So I'm back to building an OCR engine for that Hebrew dictionary.

Oh wait, there's one more, mostly the reason I decided to write this up. After her own children were grown and off doing their own thing, my sister adopted a special needs child. A few times she'd call me "I'm too old for this!" Peter Principle. She grew into it, and when the kid reached maturity, she became his Guardian. It had its own PP problems, perhaps contributing to her early demise. She designated me as successor Guardian. I had done that for  our mother, but she was on Medicare and had savings, so the job was little more than paying her bills and visiting her in the nursing home. The kid was on disability and Medicaid, and I had CDD about how the means-testing worked. All my life, I got paid for doing computer work, and I have a PhD in it, and if I'm good at anything at all, it's computer work (including teaching, which had its own PP, see STEM above). Everybody is born with the same number of brain cells, and most of us focus on what we are paid to do (or else on something entertaining); guardianship is neither of these for me.

There are people who are paid administer Medicaid, and others are paid to suck Federal (Medicaid) dollars out of the system in the guise of helping disadvantaged persons like my ward. Both kinds of people are nominally there to help people -- I would think that includes making sure whatever guardian comes with the package knows enough to keep the Federal dollars flowing. I thought wrong, that kind of "doing your job" falls on the far side of the Peter Principle. The government person is at least aware of the problem, and she probably has the authority to correct it. At least that's her job. The other person, maybe she will recover, maybe not, but as I come up to speed on this, I will be inclined to award her PP (in-)competence with the ultimate capitalistic consequence, which is taking my ward (and his humongous Federal cash cow) elsewhere. Maybe I won't succeed, or maybe it's too PP hard for something I'm not paid to do.

The whole thing leaves a bitter taste in my mouth. I do not believe there is such a thing as "systemic evil" (link TBD), but if I did, Medicaid would be an example of it.

2023 November 2 -- Not Great Literature

Even after careful scrutiny in the library, I still wind up with half the novels I take home are not worth reading. One of those I did go back for sequels took its title from the ironic line, "Plans go wrong, Bad Things Happen, People die." I just finished the third and last of the Harry Dolan books on the library shelf. It was more like a Russian novel (a huge number of characters to keep track of) than its better prequels.

At the end of the book, just before the author bio, was a two-page "Discussion Questions," like they imagined this to be a piece of "great literature." First and foremost, it's fiction, contrary to fact. The Real World doesn't happen that way. Dolan even admits that the Rome, NY of the story was significantly different from the Rome he grew up in. The clincher, Question #10:

... Who do you relate to most in this story?
To which my unequivocal answer is "Nobody." Dolan kept switching out of the first-person thoughts of the not-quite-hero to give us the inner turmoil thoughts of the other characters, or maybe just their PoV and plans. And I kept thinking in each case, "a Christian would not be doing this." I would not be doing this.

The great literature of the past wrote about a world that no longer exists, a world in which people acted as if they believed that Justice is a moral absolute, and the Judge of the Universe would some day judge them personally. People still believe in moral absolutes -- I have yet to see or hear of anybody who does not (see my essay "Moral Absolutes") -- if only to express outrage when some other person violates them. But all the people in this book live like my one-time colleague the atheist supposed of himself, who said "[ethics] is doing what feels good" (which is no ethics at all). Of course nobody can live that way, which is why he is a former colleague. He broke it off, not I.

Bottom line, I feel no disappointment that there are only three Dolan books on the library shelf.

2023 November 1 -- Unsafe At Any Speed

It doesn't take a PhD in rocket sci-- I mean computer science (same diff) which I have -- to know that what passes for "artificial intelligence" these days is in trouble. One of the features in the current IEEESpectrum "What Self-Driving Cars Tell Us About AI Risks" is a case in point. Author Mary Cummings, a university roboticist and sometime policy wonk, offers five insights on "AI failure modes."
1. Human errors in operation get replaced by human errors in coding.

2. AI failure modes are hard to predict.

3. Probabilistic estimates do not approximate judgment under uncertainty

4. Maintaining AI is just as important as creating AI.

5. AI has system-level implications that can't be ignored.

Cummings tells us (I didn't previously know, but it fits with their narrow focus) that self-driving cars use the same predictive technology  used in "Large Language Model" chat programs which have no understanding of what they are saying, they just put words together in probabilistic patterns that match their training data. Which makes them about as intelligent as an earthworm or the card sorter invented by Herman Hollerith for the 1890 Census. Yes, 130 years ago. Bigger and faster, but not smarter.

The result is that self-driving cars exhibit what Cummings called "phantom braking" = stopping for no apparent reason, because they found themselves in a traffic situation not included in their training, which I guess is better than the Tesla that drove full-speed into the side of a truck it had no training to "see." Anyway, because most (human) drivers follow too close for safety, driverless cars have double the rear-end crash rate as human-driven cars.

It seems like it will be a long time before autonomous vehicles achieve a safety rate matching human drivers, let alone being significantly safer. Like Melanie Mitchell, whose book I reviewed a couple years ago, Mary Cummings is a professional in the AI industry and a True Believer. The people making these toys endangering your life and mine have no such qualms. They don't even know how stupid their AI is.

When my income was based on creating embedded systems for computer-controlled products, and before I gave up sarcasm, I often said, "C is a wonderful programming language, and I hope all my competitors make full use of it." Maybe these driverless car developers will ride in their own creations, and thus experience extinction from the Darwinian "survival of the fittest" principle. Or maybe Darwin's hypothesis makes no more sense in the marketplace than it did in the history of the biosphere.

"25 confirmed deaths and hundreds of injuries and instances of property damage" is just the tip of the iceberg. It only takes one or two high-powered lawyers going after the deep pockets of the perpetrators (or another Ralph Nader, likewise), or a couple programmers sent to jail for manslaughter, to put a lid on the whole industry.

2023 October 25 -- Notable Three in ComputingEdge

IEEE ComputingEdge is not the most profound or well-thought-out of all the periodicals that have passed before my eyes, this month being hardly the exception, but for one message that should be shouted from the tech housetops. I have been saying things like this for several decades, but nobody pays much attention to me; the author of this piece quotes others as far back as 1986 with comparable effect. Some samples:
...many companies have embraced "corporate social responsibility," having recognized that supporting employees, customers, and the broader world can positively impactboth reputation and profit.
In other words, keeping the Golden Rule (GR) is good business.
... the C language lacks type and memory safety, "... having learned our lesson from 45 years of use, surely we do not still use C in new projects and in building new brand systems, do we? As it turns out, the evidence suggests we do."

(As a side note, it is insufficient to leverage type-safe languages if the runtimes for those languages are also written in C/C++, as the runtimes for Java and Ruby are, for example.)

For over 20 years, since I started teaching (and then using) Java, I have said that "Java fixed some, but not all, of the problems in C." Unix (including Linux and the unix wannabe = Windoze) is written in C, so the unix security model is and remains broken. All they can do is issue a continuous stream of patches and upgrades -- which may be a short-term financial benefit for the perpetrators, but it violates the GR. The simple solution to this industry dilemma is to create a new language+compiler that looks and feels like C/C++ but refuses to compile code not known to be safe. That is, it should (using existing flow analysis technology) create safe compiled code whenever it can be inferred, and error off when it cannot. I did that for my "T2" version of Java (which is still unsafe in some ways).

The last article in the rag is also by a single author, who imagines himself to be a historian, and perhaps so. Unlike computer programmers, historians are not compelled to produce anything that is necessarily True (in conformance to the Real World), only that it is entertaining. So he can with impunity deprecate "the history of sweatshop labor even in the earliest accounts of our machine's mass-manufacture," and only later (on the next page) admit that what is manufacturable determines what can be sold and used. When you price the labor out of the marketplace, then there is no market for that labor. The dual consequences of that economic truth is that (apart market economics, that is, everywhere Marxist economics is attempted) innovation and the general increase of wealth for everybody is stifled, and also the only employment available for the masses is boring make-work.

When so-called "sweatshop labor" is available, it is because people are willing to work for those wages so that they can have a better life than would be possible for them without that job. Large monopolistic corporations can abuse that delicate balance, but the beginnings of the computer revolution did not happen in large monopolistic corporations, but in small startups where even the owners worked for less than minimum wage. These would not have been possible in a Marxist economy, and in all of history they were not possible and did not happen. The USA is the richest country in the whole world and in all time because market economics works well, but especially where people are taught and believe in the GR. The Golden Rule was promoted only by Christians who were first encouraged to read their Bibles and do what it taught. That's going away in the modern Evangelical churches, which makes Marxist theology credible (but still not workable). Welcome to the third-world USA of the future.

Finally, there is a piece which purports to be about Virtual Reality (VR) but is really a feminazi whine about "diversity." VR is a technology whose time has not yet (and may never) come. It is expensive and clunky and not particularly useful for the important things in life, like making and distributing food and clothing, or communicating and entertaining. I mean, it can be used for those things, but it's not nearly as simple as ordinary video. Programming VR is way harder than ordinary computer programming, so for such a limited market nobody wants to do it. That's kind of like the way it is for computer programming, except in spades. Women especially have numerous career options much more to their liking, so of course they mostly choose those kinds of things over computer programming of any kind. (More details in my blog posting "Feminazi Disinformation" four months ago, and another on a woman's perspective three years before that).

The few women who do get into computer programming and VR, they look around and see so few women -- which is much more important to women than to the guys who like this kind of work -- so, as with most people who find themselves in a position where they are unwilling to invest the effort nececessary to perform as well as their peers (other demographics), they blame somebody else.

This whine would not be worth my time to mention it, except that they do raise a valid criticism about the ergonomics of the hardware. If VR is the hammer and the whole world is a nail, then they certainly should test their hardware and software on the people they want to attract, and not just on the guys who are writing the software. That's true of all kinds of programming and non-programming situations. Like the idiots who put big numerals on the telephone buttons to make them "age friendly" but then made the zeros on the display look like eights unless you get really close. Or the shampoo bottles that fit in your hand only one way, but can only pour out of the side a right-handed person would pour it.

Several decades ago some idiot at (I think it was) Chrysler extended the car fenders forward and back so that the lights could not be seen from the side. I saw that and knew it was an accident looking for a place to happen. Somebody in the government came to the same conclusion, and a couple years later every car sold in the USA had to have marker lights on the sides. Just a little common sense -- the kind proposed in this article -- could have prevented all that cost (and probably some lost lives). It's the Golden Rule, which makes life better for everybody.

2023 October 5 -- 7000 Who Have Not Bowed the Knee

I have family from out of town visiting this week, and while I do not have much in common with them other than the family connection and faith in Jesus Christ (so I don't have much to say over the meal table), just sitting there listening has been enlightening.

As my regular readers know, my understanding of the Christian faith leans heavily on what Jesus and Paul (and Moses, whom they both quoted) said about the importance of what is commonly known as the "Golden Rule" in being acceptable for living in God's Eternity, and how the Relationshipism (follow the links for my definition of the term and why it's not Biblical) taught in pretty much every Evangelical church in America is so utterly opposite to it. They have not used "the R-word" in my hearing, but my family collected here in Ore-gone these two weeks are clearly Relationshipists. What is less obvious, except that it was quite visible last night over the dinner table, is that these people still meet Paul's minimal formula for salvation (confess Jesus as LORD and believe in their hearts that God raised him from the dead).

Long long ago, the prophet Elijah demonstrated God's power over the pagan idolatry of the Israelite religious system, and was told by the reigning queen that his life was toast. He panicked and left town in a hurry. God met him in the desert and asked "What are you doing here?"

"They've torn down your altars and killed all your prophets, and I alone am left, and they're trying to kill me too," Elijah replied. That wasn't exactly true, in an earlier chapter Elijah was seen hiding and feeding 100 prophets in two caves. God met Elijah's depression with work to do and countered his claim of solitude with

"Yet I reserve seven thousand in Israel -- all whose knees have not bowed down to Baal and all whose mouths have not kissed him."  -- 1Kings 19:18 (oNiv)
Paul quoted this verse and extended it to his own time, so I think it reasonable to see it as a general principle: We may not see evidence of faith in God, but it's there (perhaps only in a tiny minority), and the people on the road are still learning the rest of what they need to know -- and do -- to be good citizens of Heaven, and God will bring it about in His own good time.

OK, I can live with that. Got it.

2023 September 26 -- CT Twofer, Not Three

The cover story in the current issue of ChristianityToday is a pretty good discussion of the moral (theological and social) implications of declaring and/or using preferred pronouns. It was better than the average female-written magazine article, but seemed to conclude a middle-of-the-road approach not far from where I found myself before seeing it. I had a work colleague for some months last year who stormed off the job saying things you'd expect to hear from a woman, but whose work skills were more masculine. I had to work with this person, so I consciously chose to use their preferred pronouns -- when needed, which was almost never.

It was after that that I began seeing pronoun selection being offered in sign-up forms, and I decided I did not need to play that game (I leave it blank, same as when they ask for race). I do not always have the choice in racist forms, so my policy the last decade or so is "Put me down as Native American: I was born here," which in today's culture should be as valid as "self-identifying" a gender other than what matches your DNA. Reading this article prompted me to get more specific: "I refuse to say, but if you get it wrong, I will be insulted." At my age it's unlikely to be an issue, and I'm happy to leave it that way. The article offered an escape I had not thought of: Don't use pronouns, just repeat their name. It's a wonderful idea.

I spent most of the last seven years working with a self-proclaimed atheist, and when that fell apart (as both it necessarily must, and I anticipated) I had plenty of data to think hard about what it means to be an atheist. I'm still working on that, but you can imagine my interest when I saw the CT article titled "New Wave Atheism." It was more words than content, so I flipped back to the front to see who the author was, and sure enough the name was female. Basically a zero. sigh

CT has a number of regular columns at the front besides the usual editorial(s), a couple pages of letters and a half-dozen or more of news relevant to the readers of a Christian magazine, some one-pagers by regular columnists, plus an invited (different each month) guest opinion and another invited meditation ("Close Reading") on some Scripture portion in their "area of expertise," both several pages, before they get to the feature articles. Like pretty much every Christian "in ministry" -- that's code for getting paid for it -- Darrell Bock is a Relationshipist (follow the links for my definition of the term and why it's not Biblical) but his meditation on Luke's John the Baptist account is rather remarkable.

Unlike most Relationshipist preachers, Bock looks at Luke's unique narration of John's ministry and sees not relationship (although he insists it is there) but action, doing stuff for other people. He does not say so, but if you analyze these activities, every one of them is a particular case of the Golden Rule, what Jesus called the Second Great Commandment (2C). Both Jesus and (Bock mentions only) John teach the importance of doing 2C as appropriate "fruit of repentance" (John's term). Bock adds, "Surprisingly, God is not directly mentioned in any of John's answers [to the people who came to be baptized and asked what to do next]." Why is that surprising? I think it's because the Relationshipism taught in almost every Evangelical church in America -- the pastors of which are CT's primary target -- and in most of the CT articles themselves, that Relationshipism is (ahem) relationshipal, that is, all about affirming and being affirmed, it doesn't matter what you do so long as you emphasize the affirmation. It's a concept totally foreign to the Bible, but not even Bock can bring himself to say so.

2023 September 25 -- Inner Turmoil

Last week's trip to the library stocked me up with five flicks (the limit) and three novels to read when the flicks ran out. Actually two new ones, because I was already (just started) reading one from the previous week. It was full of inner turmoil, so I dumped it and started the next. Same problem. The third was the sequel of one of the better authors from the previous week (one chapter full of people talking about having sex, but otherwise good clean action). The sequel has a few potty-language "F-bombs" but mostly the same good writing I enjoyed in his previous novel.

The stack of videos were similarly disappointing. The first, I was duly warned that it might be inappropriate by the explanation of the title, but I brought it home anyway, just in case. The theme turned out to be as I suspected, guys making jokes about fantasizing other women's anatomy (not their wife), which could only get worse as the story progressed. Then a historical war movie, obviously intended to demonize our VietNam experience by the liberal use of potty language. I usually enjoy historical flicks, just not when their agenda gets out of hand. Then a couple thrillers, tolerable but not great.

The last flick was Mel Gibson playing Hamlet. It was all about inner turmoil, but I guess if you must do inner turmoil, Shakespeare does it best. One of the "Making of" videos commented on throwing out half of Shakespeare's original dialog (there were a couple places where it was obviously missing some back-story) and still winding up with a 2-hour movie. They had a dialog coach to help (American) Gibson speak the King's English -- as if anybody can do more than guess at what it sounded like 400 years ago. All in all, I found the hour and a half of "Making of" more entertaining than the story itself. Perhaps the inner turmoil had something to do with that, hard thinking is what I do all day long and I rather expect my Sunday to be a day of rest, not more of the same. Oh well, it was still the best of the five.

2023 September 9 -- Today's Spiritual "Journal"

The length of our days is seventy ["threescore and ten" in KJV] years -- or eighty, if we have the strength; yet their span is but trouble and sorrow, for they quickly pass, and we fly away...

Teach us to number our days aright, that we may gain a heart of wisdom. -- Ps.90:10,12 [oNIV]

Three months ago I commented on the "Spiritual Discipline" Of The Month at the church where I park my fanny on Sunday mornings, and I said this weblog is it. Today, anyway.

I was born in 1943. If you do the math, then you know this year I hit the Big Eight-Oh. My father, on his 80th birthday announced, "Last year I didn't feel old. Now I feel old." I thought that happened to me two years ago, but I didn't have a clue. So this year I am thinking a lot about what Moses said in Psalm 90. Moses went on to do his greatest work in the next 40 years, but nobody lives to 120 any more.

I went to seminary right after college, partly so I could learn what it is I believe -- and I did! But that's another story, for a different time -- and partly to learn Greek and Hebrew because I am a distrustful kind of guy and I wanted to know what it really says (the translations are actually very good, but there are some insights in the original, see my "Mistranslated Words" essay). Not long after that I started carrying the Greek New Testament to church to read along. The pastor there noted the distinctive brick-red cover and said "That red book in the hand of a parishoner strikes fear in the heart of every pastor." The pastor here also saw it, and probably thought the same thing, but said something rather less. These two pastors only. I suspect that if you (as a pastor) think that, you probably don't need to worry, certainly not these two pastors.

My Hebrew is much less competent than my Greek, and it was many years later I found a single-volume Interlinear Hebrew Bible I now carry to church every week. Not long after that -- I guess it was maybe 15 years ago -- I decided to do my morning Bible reading in the Original, starting in Genesis. When they teach Hebrew in seminary, they always start out in Genesis because historical text is much easier to read. I quickly noticed that I was unnecessarily looking at the English glosses under each word, so I started covering them up, then peeking only when I needed to. I have gotten better at it, so now I need to peek at maybe one or two words each verse in some parts of the Bible, mostly the historical books (except Chronicles, perhaps the court scribes were showing off their education). David (perhaps with the help of those same court scribes) had a much bigger vocabulary than the post-Exile Psalmists ("sons of Korah") who normally spoke Aramaic in the home, so Hebrew was for them a foreign language.

Today's Psalm (39) had so many words I don't know, I just gave up and read the English glosses. Many of David's Psalms (including 39) are pure Feeler inner turmoil, which mostly is not where I'm at. The Feelers who populate every church in America (see comments and links here in my home page) would feel very much at home in this Psalm, but it's much too negative to ever get preached in church. I read these, but it's not usually for me, where I'm at. And that's OK.

Then I got almost to the end, and the English glosses read "for a-guest I-am with-thee." I wondered what Hebrew word means "guest" and it turned out to be a word I know, 'GeR' = "stranger, sojourner" from the verb 'GWR' ("I. to sojourn, to dwell for a time, as a stranger..." in the Davidson lexicon I am currently working on an OCR to digitize). I think the original NIV (which I have the text of) got it much better:

39:12 ...For I dwell with you as an alien, a stranger, as all my fathers were.
I got to thinking about my late sister, who greatly feared losing her mind (both our parents were pretty far gone the last year of their lives) so she might "stop loving Jesus." God answered her prayer and took her quickly. I can only hope for as much for myself, and my little sister agreed (for herself) when I last spoke with her.

But am I a "stranger" with God? Perhaps by God's standards, but... Hmm, maybe I shouldn't be comparing my self to the people I see. Anyway, now that I'm into this Psalm as maybe relevant after all, I see earlier

Show me, O Lord, my life's end and the number of my days; let me know how fleeting is my life. You have made my days a mere handbreadth; the span of my years is as nothing before you. Each man's life is but a breath. -- Ps.39:4,5
Which looks looks a lot like what Moses said. I find myself singing the "I'll fly away" song, which I never heard in church (except maybe after I moved to Ore-gone 7 years ago). Today for the first time, I see the same line in the Moses Psalm.

I don't often remember my dreams more than a minute or two after I awake, but I remember thinking about the Grim Reaper pounding on my door -- the door of this house; most of my dreams have me in a house I never saw in real life -- and then wondering if "instead I only heard my marbles rolling off the porch." It's happening.

Over the years I learned five languages other than English (and failed to learn three others that I tried), but I never was very good at it because learning a language is about memorizing hundreds and thousands of new words (the average person has a working vocabulary of about 10,000 words in their own language), and I never was any good at memorizing. More and more, my Hebrew readings -- including today's Psalm -- have words I used to know, but forgot. Even worse, 2 Corinthians, which I'm also reading the last couple weeks, the text has a lot more big Greek words I don't know than most of Paul's other epistles. OK, I didn't forget those words, I never knew them. The point is, I often find myself looking the paragraph up in English to see what Paul said. I did that today. The Greek is not as hard as James or Hebrews, but clearly a different amanuensis than Paul's other epistles, including the first epistle to Corinth, which I have no trouble with.

Maybe I'm losing it around the edges. Yesterday I got a lot of help from God with my programming. Actually, I now know that all new code, every invention of any kind, necessarily must come from God (the entropy principle, see "Why Is This Industry So Devoid" last month and "Politics, Not Technology" the previous month), but I didn't think that way so much when I was younger. Instead I think of Psalm 144, which starts off praising God "who trains my hands for war, my fingers for battle." [oNIV]. The Hebrew word translated (accurately, in context) as "war" is from the verb 'QRB' which means "to get close." Hand-to-hand combat requires getting close, but with guns and rockets we don't normally think that way, so the translators did the right thing. Other places (with the preposition meaning "in") 'QRB' means "inside of." The same verb as a hiphil (causative) passive participle ("corban") is translated in the New Testament as a gift, because it means "something that somebody caused to draw near." So I like to read Ps.144:1 more literally, and praise God for teaching my fingers to "draw near" to the keyboard (to write and debug programs) because that is the closest I ever come to anything like "battle." This year and three years ago (and several times before that) some real person chose to be my enemy, but I'm basically non-combatant. I walk away. Ore-gone is the only place I was ever physically assaulted as an adult. That was six years ago and the guy tried to provoke me into casting the first punch, but instead I drove away (that's when he hit me: the top was down, and by the time he connected I was moving as fast as his fist, so I barely felt it, but he did intend harm). He was not trying to be an enemy, he just didn't like the way I was driving in compliance with the law, and he had an anger management problem. A lot of people have that problem with me. God protects me. I have no enemies I know of.

Bottom line, maybe I'm losing my language skills, maybe today was just a hard day for the texts I was trying to read. I still am pretty good (with a lot of God's help) at programming. So I guess I should stop the logorrhea and get back to what God made me good at.

2023 September 2 -- The New Doomsday Machine

Five and six decades ago, everybody thought the world would end in nuclear holocaust. Then Ronald Reagan became President and invited "Mr. Gorbachev, tear down this wall!" With a little help from Reagan's (planned, never implemented "Star Wars") Strategic Defense System, the wall came down and the Cold War was over, no nuclear war needed.

So the "Chicken Little" fear mongers needed another Doomsday Machine to scare the "poor, uneducated, easily led" (their words) people into submission. The Dems hated George Bush for his religion, but they couldn't say so because "Separation of church and state" was their own mantra. They couldn't complain about Education or AIDS because Bush was left of the Dems on those topics. They couldn't complain about the Middle East war because the whole country was behind him on that. So they hunted around for some obscure science where the results were not yet all in, and picked whatever side he wasn't on. So "Climate Change" -- it started out "Global Warming" but it wasn't getting as warm as they predicted -- became our new Doomsday Machine.

Well, the results are coming in, mostly on the side they didn't pick (although they won't say so in public, you have to look around for yourself), so they need something else. Apparently "AI" is taking over as the latest Doomsday Machine. As noted two weeks ago, the American military (the favorite whipping boy of the left-wing bigots) is looking at what passes for "artificial intelligence" and thinking about how to use it in weaponized conflict without killing so many of our own soldiers. As you recall, it was the military that (so they claimed) would be responsible for the nuclear holocaust that didn't happen, so here is another political jab at the right-wing militarists. The editorial (women) at WIRED wrote two "Stop the AI" screeds in the same issue with the military piece.

I'm here to tell you, it's no more of a problem than climate change or nuclear holocaust.

First and most important, I read the last chapter of the Book, and things don't turn out that way. The God Who made the entire universe and everything in it is not so impotent as to let silly humans mess up His plans. Maybe you don't believe that, so let's look at the purely secular reasons why not to worry. Because God is God, and He really does know what He is doing, the secular reasons match the theological 

In the short term, what passes for "artificial intelligence" today is not very intelligent. It's basically about as smart as an earthworm or the card sorter that Herman Hollerith invented for the 1890 -- yes, more than a century ago -- Census; "AI" is very much bigger and faster, but not smarter. Intelligent machines are possible, and they were on the right track to make them some forty or fifty years ago -- that technology is now derided as "GOFAI" = Good Old-Fashioned AI -- but it came down to the hard work of smart people programming them, so they looked around for something they could do with less effort, which is neural nets (NNs), a (very inefficient) computer intensive way to do what we used to call "linear regression" (or something like that). Computers are cheaper than programmers, and the electricity to run them is even cheaper (but it puts carbon into the air, which is another reason to find a different Doomsday Machine). People are beginning to see the cracks in the edges of NN-based AI, and maybe they will get around to making intelligent machines again. Melanie Mitchell, a researcher who is rather more honest than the average NN programmer (see my review of her book "A Guide for Thinking Humans") thinks maybe around 2040 they will start to get it right.

A lot can happen in 20 years, should we be worried? I don't think so. First, there's physics: the Second Law of Thermodynamics says you cannot get more energy out of a system than you put in. Claude Shannon, back before there were computers to distract him from clear thinking, showed that information obeys the same laws as energy. That's not a popular view because it makes Darwinian evolution physically impossible (which actually has been confirmed, see my "Biological Evolution: Did It Happen?" but nobody wants to admit it), but it basically means that nobody can make a computer or robot smarter than the programmers doing it.

Should we be worried? Stupid people using stupid computers can do stupid things. Nevermind what they tell you about ChatGPT, it is mathematically impossible for a computer to program a computer the way humans do, at least not using NNs, because a NN (or any combination of NNs) is a "Finite State Machine" (FSM, the simplest of stupid computers), and modern computer programming languages are all the next level up ("Context-Free") which a FSM cannot generate, and the semantics of all non-trivial computer programs are another level up ("Context Sensitive") from that, and many (perhaps most) useful computer programs are actually "Turing Machines," the top level of complexity. I have a PhD in this stuff, I know these things. The bottom line is, only people can write new computer programs that actually do something, and those people need to think very hard to succeed at it, so nothing really BAD can be programmed into a computer without a lot of people thinking about it. Near as I can tell, all that ChatGPT can do is reproduce some variant of the programs it has already seen (or they wouldn't compile), which is about the same as looking on GitHub (an online repository for open-source programs), which is probably exactly what they trained ChatGPT on. See also my "Now You See It, Now You Don't" blog post a month ago for a more intuitive explanation that comes to the same result.

Yes, stupid people kill people -- and sometimes also smart people trying to stop the stupid people need to do some killing -- and you can buy a drone today that can be programmed to fly someplace and drop a bomb there, but we can figure out who did it and go stop (or punish) them, so it still boils down to stupid people, not smart machines. Worry about the people, the stupidest of them are way smarter than any robots are likely to become in your lifetime or mine.

2023 September 1 -- Why Capitalism Is Good For You

Here we are coming up on Labor Day weekend (in the USA, everybody else in the world celebrates May 1 as Labor Day), which WIRED magazine in their leftist political reason for existence (after abandonning technology as their aim) celebrates with two feature articles (one an interview, the other reviewing the guy's books) profiling two different persons condemning capitalism. Both the authors (guys who should know better and ask the hard questions, but they did not) and their subjects are ignorant of the facts, perhaps because their focus is too narrow.

It's a bit like watching a hit&run fatality and blaming the tires on the car. Yes, if there were no tires on the car, it could not have run over the victim, nor could the perp have left the scene of the crime. But without tires on cars, the victim could not have been there at all, nor the observer, because every one of us rides in cars with tires to get from where we are born to where we are today, and we eat food brought to us (or we go there) in cars and trucks with tires. More good than harm happens by vehicles with tires. And that is why we still have vehicles with tires driving up and down the streets and highways of the country.

So also is capitalism. Let's be perfectly clear: capitalism is not about rich people oppressing the poor, that happens even more in socialist and monarchist countries. Oppression happens when people with power refuse to live the Golden Rule. You don't need to be rich to have power, gangs of otherwise unemployed inner-city youths also oppress other people. Bigger school children (we call them bullies) oppress the weaker kids. Socialist governments oppress their subjects. It's human nature.

The people complaining seem to be really complaining about the disparity between the rich and the poor, and they (unfairly) pick on the capitalists because in a capitalist economy, they are the obviously rich people, who also happen to be making the whole country wealthy. Make no mistake: the poorest people in the USA are richer than most of the people in the rest of the world. We do not have people picking over garbage dumps for something to eat, but other countries do. The complainers actually have more wealth than the poorest in our own country, who cannot spend the travel money and take time off from trying to make ends meet to go protest in another state. And why are they rich enough to do that? Because the capitalist economy makes everybody richer. They are protesting the very mechanism that makes their own lives better than in third-world countries with no "rich capitalists" driving the economy.

Capitalism, like the tires on cars and trucks, does far more good than harm. Make no mistake: the poorest people in the USA are richer than most of the people in the rest of the world because the capitalists got even richer making life better for the rest of us. Other countries have rich people too, but they didn't get that way by making other people richer. That's why the USA is richer overall than every other country in the world.

So maybe you didn't think it through yet, and you are not convinced that capitalism is good for you. I explained it in my new essay "Das Kapital (Updated)". If after you understand my explanation and you still disagree, or if I explained it pooly and you don't understand, tell me. Maybe I'll have answers for you, or maybe you will convince me. But don't waste my time and yours if this is Religion for you, that is, you are unwilling to consider contrary evidence, I usually take that to mean you are ignoring some or all of the evidence and therefore have nothing to offer.

For the record: I am not a "rich capitalist," I got out of the stock market after I realized it's a zero-sum game (contrary to the Golden Rule). I'm a technologist, and I made some money from my technology. I'm still living on it. My technology made life better for some people, and they were willing to pay for it. That's how it works. You gotta do things that people are willing to pay for. If you don't want to do that, you won't get rich, but don't complain about the folks who do. Their wealth "trickles down" and makes everybody's life better, including those who refuse to work for it.

2023 August 29 -- Movies, Good and Not-So-Good

From my remarks here, you probably gathered that I watch old movies when I have nothing better to do. Maybe in another post I'll talk about why there is nothing better (besides going to church on Sundays and programming the other six days). WIRED is usually worthless, but once in a while they have a gem. It is what it is. Today it's movies.

Up until Covid shut everything down, I went to the library each week and stacked up as many movies as they would let me take out, and then a pile of novels to read when the flicks ran out. Many years (more than a decade) ago I read in WIRED -- once in a while a gem -- about the availability of free movie downloads on Good novels at the library are hard to find: Like churches everywhere (see "Why Is This Industry So Devoid" last week), the libraries are run by, and for the exclusive benefit of, women. Women buy books, so if the authors want to sell books, they must write for women. Chick-Lit (like chick flix) is very different in focus from the stuff aimed at guys. Guys are interested in moral absolutes -- Justice, Truth, and Duty -- while women crave affirmation ("love" or "Relationship") and thrive on inner turmoil. I have enough inner turmoil of my own, I don't need to read about other people, (fake) fiction, no less.

The movie people are not stupid. Women don't go to the theaters alone, they need to drag a guy along. The guys don't want weepy women's fare, so the movie industry does epic pieces like "The Great Wall" and "Bridge On the River Kwai" and "Black Widow" (all in last week's viewing, three different sources). But I'm getting ahead of myself.

An ancient historian once described Athens where "All the Athenians and the foreigners who lived there spent their time doing nothing but talking about and listening to the latest ideas." The techies at (like everywhere else) are all from the mystical Athens. Unlike the movie industry, they all write programs to please themselves -- to exercise the "latest and greatest" technology -- rather than to entice the customers to spend money by giving them what they are willing to pay for. Especially on the internet, because there's no money to be made from "free" downloads. The inevitable result is that the website keeps breaking, most recently sometime in July. If you don't have the budget to buy a new computer every year or two (why else would you be downloading old stuff from you are out of luck. After a while it occurred to me that they might have an "API" -- that's geek-speak for hooks that outside programmers can use to access the data on a website, so to mash-up their own presentation. A few years ago APIs were the latest Athenian rage. It turned out I was right. It's kind of a hodge-podge and poorly documented, but it's neither this year's "latest" nor much fun to maintain, so it still works on older computer systems. I'm back to downloading movies.

People upload flicks to for their reasons, not ours, but sometimes there's an overlap. For a while somebody was uploading whole carloads of Iranian movies. Nothing in the descriptors said these were all in Farsi with no English subtitles, but eventually I started to recognize Iranian character names, so to skip over the downloads. Most of the flicks on have expired copyrights -- or were so bad, nobody bothered, like a bunch of Chinese flicks from when China did not honor the international copyright law, basically all martial arts flicks, the same story but different characters -- or else uploaded with the permission of the copyright holder. I think that (sort of) happened with some unexpired American flicks, somebody added Spanish or Portugese subtitles, so they held the copyright on the subtitles and they uploaded the whole (English) movie with the subtitles. It ain't exactly legal (unless they have the copyright owner's permission) but I'm not saying anything. I found these subtitles -- especially the Spanish, because I know Spanish, but Portugese is similar enough to be helpful -- especially useful when the dialog was inaudible or muddled, I could back-translate the subtitles to figure out what I missed. Then came a whole bunch of Japanese flicks with English subtitles, originally celebrating some famous Japanese moviemaker, then quite a few where I didn't see his name. I even learned some Japanese: "Hi" or sometimes "Hah" means "yes" or "OK" or "Hello" or "Come in," the same spoken word in all those places (different subtitles). The Japanese word order is different from English: I could recognize their pronunciation of names (often dropping a vowel shown in English), but they were in different parts of the spoken sentnce than in the subtitles.

Anyway, last week's fare from included several of these Portugese subtitled flicks, often some classics you would not see otherwise, like "Kwai" from the 1950s (unexpired copyright). I wondered about the tune the soldiers whistled, but GKA (Google Knows All) it's the "Colonel Bogey March" and you can even hear (or download) it. Along with "Odessa" and "Zorba the Greek" and "Passage to India" these were all given the same Portugese title "Assista Clicando Aqui" ("Help Clicking Here") so I didn't know what I was getting until I played it. Maybe that was to hide the upload from the copyright vigilantes.

My nephew has been subscribing to NetFlix DVDs, which this month is the end of that -- the profit margins are much greater for streaming than for snail mail -- and his final take is "Black Widow". As I pointed out earlier (see "Feminazi Contempt" two years ago) male authors often genuflect at the Feminazi altar by giving their guys female names and bodies, but they are "Orios" (black outside, white inside) or otherwise denigrated. "Black Widow" was that kind of action, but there was more: this was a chick flick with all the inner turmoil that you never see in a true guy movie or book. When it was over, I had to hunt around some (you never saw so many producers) but sure enough, the lead writer was "Jac Schaeffer" (no 'k' had to be female, and GKA agreed, with a face picture) followed by a guy's name to do all the guys with female names and busts.

Next I watched "The Great Wall" from the library, and I had this Feminazi feeling all over again: the leading Chinese warrior had a female body (and I presume name, but I wouldn't know) over a mostly guy, a little bit of feeling, like what a guy might write for an actress to say, and sure enough all the writers were male. But the lead character was clearly the American, who was all guy. Except for the director, all the important production people were American guys. Like all Chinese flicks, this was not intended to be a money-maker -- I  guess the American production people wanted to make a profit, but -- the sets were huge and splendiferous, the kinds of stuff a self-aggrandizing government like China would (and could) do to make up for the fact that they are not cultural leaders like the USA.

Most impressive (to me) in the library batch was Charlie Chaplin's "The Great Dictator." It was his first talkie, but not at the same caliber as his silent greats, the flicks that made him the highest paid actor of his time, and loved all over the world. The studios wouldn't touch a political flick like that, so he paid for it out of his own pocket. I guess he could afford it. The documentaries on the second DVD said it was his highest grossing movie, but apparently it also contributed to his being invited to leave the USA. Nobody wants to see a political message, not even me. He tried to use humor to ridicule Hitler, but later admitted that he wouldn't have done it if he'd known about the death camps. God uses humor against the unbelievers, and it comes off quite well. But political is too intense.

Besides that, Charlie Chaplin's final impassioned -- one of the documentaries said it was six minutes, but it seemed longer to me -- utopian speech cannot work in the real world. We are all selfish, without even thinking about it we make demands on other people that they don't want to do. Take Charlie Chaplin himself, he paid for making this flick out of his own pocket ($2 million is a lot of cash) and his brother's color movie of the making-of showed him pushing the actors and production people harder than they wanted to be pushed. Actors and production people need to eat and pay their bills, so they put up with the hassle. Steve Jobs did that to the people working for him. Some of them couldn't stand it and left. I think what Winston Churchill said of democracy applies also to our capitalistic economy: "It's the worst of all possible, except for everything else that's been tried." The socialism that Chaplin promoted in his speech doesn't work, and it didn't work everywhere it's been tried. It doesn't even work in China. Let's save that for another blog post, my long computer run just finished, I can get back to work.

2023 August 26 -- I Still Hate OSX

Some 12-15 years ago my Bible translation project ran out of steam -- basically it's like STEM education and climate change: everybody thinks it's a good thing to do, but nobody is willing to actually do anything about it themselves with their own money -- so I started looking around for something I could do to "make the world a better place" and I decided I might be tempted to buy a smart phone if it meant I didn't need to carry a ten-pound book bag (the Hebrew Bible is heavy) to church every week. I got my app mostly working, but I needed dictionaries for Greek and Hebrew, and I needed something to parse out the Hebrew word forms (I already had that for the Greek from the Bible translation project). I figured I could adapt a public-domain Greek dictionary from one of the CDs available at the time. There wasn't so much available for Hebrew, but I had a printed copy of Davidson's Analytical Lexicon I got at seminary (I later found out I have two copies) and has both definitions and parse information, and it was published sometime in the mid-1800s (no copyright date in my copy that I could find) so I could do that. It would take a while to scan 900 pages... Oh wait, has the full text, apparently OCR'ed from their scanned pages, also online. The text was a horrible mess, whoever did it just ran it through an English-language OCR, and the Hebrew part was illegible. But I'm a programmer, I can do this. I always wanted to write an OCR engine, now I had a good reason to.

After a couple or three years working on it I was finding that the resolution I thought it was scanned in was too course to resolve some of the smaller letters. Silly me! I thought the resolution OSX offered for capturing the image was the image native. Silly me, OSX is Unix, and unix makes people stupid. If there are two ways to write a program, and one of them is useful and/or easy to use, the unix programmer will always do it the other way. If there's only one way to do it and it's easy to use, the unixie will still do it the other way. Well, I took some time off to work on another project I thought would help a lot more people than a couple hundred Biblical Hebrew students, but I was wrong again: everybody thinks STEM education is a good thing to do, but nobody is willing to actually do anything about it themselves. Including the guy I thought I was in partnership with.

So here I am working on OCR'ing the Davidson Lexicon again. In the intervening years got a newer, higher resolution scan from the Library of Congress edition (fewer handwritten notes in the margins ;-) so I threw away my previous work and started pulling off 600dpi images from the new file. It's taking a lot longer, because the image files are larger. OSX's Preview program spends a lot of time spinning its beachball.

Today I found out why: the program was making unauthorized changes (corrupting) the pristine downloaded PDF file. It put up a notice "Saving changes" or something like that. I didn't make any changes!! I hit ForceQuit, which is not as fast as it is in Windoze or the old command-line unix, but at least it stopped whatever it was doing and didn't destroy the file. I had to hunt around for it, but I found out how to lock the file, then re-opened it. As soon as I tried to scroll back to where I was, it offered me the option to make copy for my changes or unlock or cancel. NO, I do not want to make any changes, and especially I don't want YOU making any changes, but there was no such option. I expected "Cancel" to kill the program or at least close the file, but it was really the "NO, I do not want to make any changes" option I needed. After a while I figured out what it was doing. There are a bunch of blank pages at the front, before and after the title page and contents, then the preface is numbered sequentially, but that's only the first 10%. The main dictionary is numbered from "I" in Roman numerals. Acrobat just shows the absolute page numbers, but apparently Preview is wasting my time OCR'ing the page numbers in the corner of the preface pages, then completely blanking out the page numbers in their thumbnail index of the main lexicon, making it excruciatingly slow to count off the pages I want for each image file. After the first time I killed it, it opened up with actual page numbers in the thumbnail index, then eventually caught up with the renumbering and blanked them all out. I guess they were saving the OCR'ed numbers in the file someplace, but they didn't change the file date, so I have no clue what other damage they did. I found an option to turn that off, and it seems to run faster, now that I'm basically finished. Whatever.

I Still Hate OSX, sometimes it's even worse than command-line unix (if that were possible). The (original) Mac was scriptable -- it was awesome! But they killed that with System/7 -- and the original unix was (probably still is) scriptable, but OSX is not. They have something that pretends to do scripting, but none of their own software works with it. So I had to do all this work -- waiting for Preview to spin its beachball -- manually. I Still Hate OSX.

"Unix" is a four-letter word, suitable for expressing extreme dissatisfaction, where other people normally talk about their sex life (real or imagined) or what they do on the toilet, or utter a prayer to some deity they hope and presume is not listening, four-letter words all of them. Whatever.

Some glad morning when this life is o'er, I'll fly away.

2023 August 24 -- Why Is This Industry So Devoid of the Christian Faith?

It's an interesting question, with several -- I was going to say "good" but to a question like that there can only be Bad -- answers.

First and most important, it's like asking why there are so few Christians in China or Saudi Arabia or North Korea. There are more Christians than you can see, but the environment is so toxic that most Christians don't have the cojones to become visible. More than any other art form, computer technology is where we most truly make the promise of the Serpent to Eve in the Garden come true, we become gods of our own little universe, the universe(s) that each one of us created out of nothing and gave life to inside the computer. Rather than give glory to our creator God who made this all possible, we prefer to think of our puny creations as representative of the whole universe, and we get to control it absolutely. But ultimately physics is not violated, entropy is not reversed, it's all a gift from the God of the universe we all live in. He made us, and not we ourselves. And as He told Elijah trembling in the cave on Mount Horeb (Sinai?) and Paul generalized to all nations and all places, God still has His "7000 men of Israel who have not bowed the knee to Ba'al, and whose lips have not kissed him." There are Christians in the tech industry who know where it all came from. It's not even as dangerous to say so as it is on the other side of the world, in places where they kill you for saying so. Here they only ridicule you and take your job away, which feels worse in many ways than a quick trip to Heaven. OK, maybe that's two reasons, Why there are so few, and why so many of us don't openly say so. Whatever.

Hmm. When I started to write this, I thought I had three reasons, but then the first one split into two and the other two disappeared. No, I'm just getting old and forgetful...

There is an important reason why tech people are not drawn to the Christian message: The churches in America -- now exported all over the world -- chase us away. More than any other skill set that you can choose as a life career, technology only works in a context of moral absolutes, specifically Truth. Either you make your computer conform to the laws of physics, or it doesn't run at all. Either you make your computer program conform to the instruction set of the computer you are writing it for, or it doesn't run at all. Truth is conformance to reality, and the God of the Universe (so we are told) "cannot lie," because Reality is what He made. That message is all through the Bible, but the people who bear his name in American churches preach "a different gospel," a message of "God's unconditional love" (affirmation) which is nowhere to be found in the Bible God gave us, and is anathema to the tech people whose value system is more like the God of the Bible than what the preachers preach.

And when I try to say that out loud, the pastors forget all about their "unconditional love" (which they cannot live) and throw me out of their church. Yes, it really happened. Twice. These are pastors who say they preach the Bible -- and mostly they do, usually more and better than most of their peers, and they apologize when the text is about Truth and not "relationship" -- but where the rubber hits the road, they cannot live their own message. They know it, and often say so. And the techies, who know you can live a Truth message (because we do, 40 or 50 or 80 hours every week) are unwilling to put up with the hypocrisy. They stay home. Or if they are married to a Relationshipist wife because that's what most women are, they come and sit in back and leave quickly. I know both kinds.

So I don't say this very loud in church. I try to stay under their radar, which is easy because "I'm a Zero" and nobody reads my blog. Except a few people who really want to, they find me on Google.

2023 August 22 -- Internet Ethics

It never ceases to amaze me, what gets called "ethics" in a world so bereft of ethics. So here I am looking at this article in ComputingEdge (originally published a year ago in some other IEEE rag) "A Digital Communication Twin for Addressing Misinformation." First some definitions. A "Digital Twin" is a newfangled name for 60+year old technology that used to be called just plain "simulation," that is, programming the computer to model out the behavior of some physical or social process. I did that kind of stuff as a fresh-out just short of sixty years ago. What the people are doing today is applying a new label to old technology so they can continue to get Federal grant money, and (as the Dean told me when I was trying to play it) "Grantsmanship" is the name of the game. "Misinformation" is, as I hinted at last year in my blog post "Information Literacy vs Misinformation," seems to be code for "ideas I happen to disagree with." At least that appears to be the working definition in this article. Most of the people trying to combat "misinformation" have politics in mind, perhaps this author does too, but he demurely claims it's about anti-vaxxers. Being one myself (see "I Didn't Kill a Baby Today" two years ago) helps me to understand his agenda as fighting ideas he happens to disagree with, rather than as a purely virtuous goal.

Then there is "ethics" (the superscription on this article). There is such a thing as ethics, and it is necessarily based on moral absolutes, in particular the notion that what you don't want people doing to you, you should not do it to them, and that principle applies to all people everywhere and in all time without exception. And pretty much everybody has a very good notion of what that is, and why they wish other people would do it, and also why they don't want to do it themselves (except maybe when people are looking).

Assuming, for the sake of the argument, that his sole agenda is opposition to vaccination, let's look at that issue on its merits. Why is he so opposed to people refusing vaccination? Maybe they have moral grounds, like me. Maybe it's religious, like the Jehovah Witnesses. Maybe it's political, people don't like being told what to do. I don't think he ever bothered to try and discover whether this is a slam-dunk black-and-white issue or perhaps a he-said, she-said difference of opinion. Most states (including where I live) recognize that doing things that doctors consider healthy is a personal choice, and should no more be coerced by law than doing things preachers consider virtuous, and indeed his whole computational solution is based on the premise that people can and do choose otherwise than he (the author and his eleven colleagues) thinks prudent. He never gives the slightest hint of recognizing that maybe their reasons are valid. The law here -- and probably also in North Carolina, where his team are on faculty at three branches of the state university system -- gives the person being vaccinated the benefit of any doubt. His failure to give them the same courtesy makes his battle against "misinformation" somewhat disingenuous. Let's call a spade a spade here: how is the product of his "Digital Communication Twin" (DCT) different from the misinformation it allegedly combats? He doesn't say.

Yes, this is an ethical question, but methinks Professor Arnav Jhala (and his selected colleagues) has missed the core ethical issue here: Does he want some other person or group of persons labelling his DCT as "misinformation" because they don't like his moral choices? Then don't do it. That, Ladies and Gentlemen, is the essence of ethics, and this guy has missed the mark.

2023 August 21 -- Military (A)Intelligence

When the military tells you they are using AI, you should worry. YOU should worry. Me, I'm too old to be harmed by anything they start doing today. Besides, I read the last chapter of the Book, the USA is not there. Maybe their AI will fail, maybe it will be something else, but I doubt I'll live long enough to see it happen.

Anyway, so there's this obscure article in the current WIRED about the military using off-the-shelf AI, and as you know, I think the letters "AI" stands for "Artificial Stupidity." The article was written by a guy (where'd he come from? WIRED articles are mostly written by women) so he asked some of the hard questions that female writers never do, but he knows no more about what's wrong with what gets called "AI" than the military.

So here's why it won't work. You won't see this anywhere else. First of all, let me make it clear, there are smart machines out there, nowhere near as smart as the humans who programmed them, but way smarter than the neural net (NN) stuff that drives
ChatGPT and the image recognition stuff that capture all the headlines. Anyway, IBM's "Deep Blue" did beat Kasparov (apparently by cheating -- see my "End of Code" post seven years ago -- but even getting close is a commendable feat). The point is, computers do what they are programmed to do, the NN computers no less than the computers that were programmed by humans before NNs were invented. And when the computer encounters something it wasn't programmed to do, then it does something bizarre -- or nothing at all -- but never intelligent. Like autonomous cars running into obstacles they were not programmed to recognize, which author Will Knight did mention (like I said, male writers do ask the hard questions). The difference between the Kasparov chess game and war is that there are more smart (human) robbers out there trying to make war on peaceful people than there are smart cops (programmers) reprogramming the drones to deal with the surprises that Kasparov threw at the computer.

Case in point. Knight visited one of the manufacturers of the drone technology the military is buying. The drone took off, then "hovered about a foot from [Knight's] face" before flying off to do what it was supposed to be doing. "It's checking you out," the engineer said. Maybe it was, but if so, it's because it was programmed to do that. Note, the engineer was not surprised like Knight was, he knew what the program would do, and why. War isn't like that. As soon as the Bad Guys figure out how the drones will respond to what they throw at them, they will keep trying different things until they find something they can attack. Even that one-second hover could be fatal if the Bad Guys know about it and can shoot it down while it is paused like that. The programmers are not out there in the field reprogramming the drones on the fly like they were in the Kasparov chess match, these are catalog items sold to the military for use thousands of miles from SanDiego where the programmers are.

There's another reason why AI is a loser in war: How did that joke go, "Why did Israel win the Six-Day War?" It helps if the other side are Arabs. Overtly or covertly, the other side in the next war will be Chinese. Maybe they are not smart enough to invent the AI tools our military is using, but they certainly are smart enough to steal it. And they have enough factories to replicate it way faster than we can. Even without AI technology, they (or whoever it is trying to replicate 9/11) have enough people to overwhelm whatever drones we throw at them.

When the Final Battle comes around, and we are not there, but let's suppose whoever is there attacking Israel is using autonomous weapons invented for the USA military but subsequently sold to the highest bidder, and remember, God is fighting for Israel. I think one novelist imagined that God could start a hailstorm with basketball-sized hailstones. Hailstones that big can take out conventional fighters, let alone smaller drones. Those hailstones could also cripple the autonomous sea vessels (God has perfect aim!) and with nobody on board to repair the damage, the autonomous weapons are quickly out of the action. Of course the USA is not thinking that far -- hey most of them don't even believe it will happen -- but the Chinese can and will beat the drones. Iran or Pakistan might try their own drones -- and get trampled by the American and/or Israeli drones -- and then switch to humans. They have ten times (together, 30 times) the population of Israel. Israel beat Egypt and dinky Syria in the Six-Day War, the more bellicose countries are farther away. When China goes after Taiwan... Did you ever play the board game "Risk"? The biggest army wins. Drones may win the first couple battles, then they are out of the game and "boots on the ground" (or real pilots in the sky) will win. Real people can be reprogrammed in a few minutes or hours; drones take a lot longer.

China against Taiwan may happen in my lifetime, but China has too much investment in the USA for them to consider that a smart move. They are trying really hard to get stupider, maybe they will succeed. North Korea is already there (stupid), but they don't have the population nor the wealth to pull off anything more than a gnat's sting before disappearing in one giant mushroom cloud.

Basically, it's not my problem. Besides, I live in a fly-over state. We may die of starvation and/or government stupidity, but not from direct foreign attack.

2023 August 12 -- Consumer Reports Innumerate

When I was a kid, my parents read TIME magazine and Consumer Reports (CR), so like a good apple-falls-near-the-tree kid, I subscribed too when I moved out of the house. Pretty much all mainstream media these days (my whole lifetime) are edited by left-wing bigots -- they probably prefer to call themselves "liberals" but "liberal" means open and accepting, and these people are knee-jerk heterophobic anti-Christian left-wing bigots (basically their own terms, with the polarities reversed) -- so when after several decades reading their tripe, they neglected to send me subscription renewals, I didn't feel bad.

Pretty much everything published, the people doing it get paid in proportion to the number of readers, either directly from subscriptions, or else (usually) by advertizers who pay for the number of eyeballs seeing their ads, CR and TIME being one each. TIME, being much bigger, didn't feel the pinch from my departure, but CR did and came asking "We want you back." Maybe their left-wing politics (which they deny) and the irrelevance of their reviews has moderated in the ensuing years, so I opted for their "risk-free" trial issue. Same old same-old in the issue that came this week. Only more so.

Only one of the two political parties who trade off running this country every four or eight years has as policy that the people running the government are considered smarter than the people who voted for them, and that might be true of politics in general, but it certainly doesn't run to what kinds of things people should be allowed to buy. So when CR engages in political action to create what is called "restraint of trade" (and is illegal when companies do it other than as forced by the government), they are basically politicking for the one party that likes to do that kind of thing. Me, I have a higher opinion of both my own intelligence and that of the people, so I try not to vote for that party. It helps that I also don't like their policy of denying civil rights to the people least able to speak for themselves (because they are too young), but the last four or five elections, the other party didn't offer much better, so I just voted "None of the above."

Anyway, the latest CR issue has several items urging the readers to lobby the CR agenda for restraint of trade. I don't need to be paying for that brand of politics.

CR has always spent most of their pages reviewing high-ticket items I (almost) never buy. Duh. Only the people with enough disposable income to buy those things can afford to pay for the CR subscription. I spend my money in other ways, including not at all, which enables me to be more choosy about when I work and what I do as work. I think I bought maybe four refrigerators in my life, the first I cannot remember anything about why that one. The other three, I got what was cheap, but I had some choice on what was available on the second-hand market for the last one, so I looked up the choices in CR. My previous refrigerator had a single temperature control, so in summer my veggies froze, and in winter my ice cream melted. I wanted the next one to have a separate control for the freezer, but none of the CR reviews said anything about it. I bought cheap and it turned out to have separate controls, no help from CR. My first two cars I got second-hand from somebody I knew, but the current car I bought because it was said to be designed in the USA to resemble my first (and favorite, except that British cars spend more time in the shop than on the road). Only later did I see that CR rated its frequency of repair better than usual, which is nice to know but too late to be helpful. Most everything else expensive enough to get rated, I bought on the cheap, ratings irrelevant.

Frozen pizza is more in line with low-income budget, except that they rated the name-brand products -- I looked at them in the grocery a few years back, after I moved away from a state that has really cheap (Aldi frozen pizza was $2, cheaper and better-tasting than I can make it myself) and they were all more than I wanted to spend on a home-cook meal. So now I make my own, CR-rated products again priced out of reach. Nothing in Ore-gone is as good as RoundTable (in Calif), but Domino on coupon is cheap enough and better than my own -- my sister used to work at a pizza place, and she told me it's the hot oven, hotter than you can get in a home kitchen -- and Domino has that advantage over frozen and is almost as cheap.

The feature article in this freebie issue talks about the rising cost of prescription drugs, and how to get the insurance companies to pay for them, and how to pressure the government to force them to. I think the editor(s) of CR -- mostly all women except for cars -- are innumerate (that's the numerical version of illiterate, that is, they can't -- or won't -- do math), which is a common failing in women. The most important thing to remember about money is: There Ain't No Such Thing As A Free Lunch (TANSTAAFL). Insurance is the problem causing rising costs in health care, not the solution. Every dollar in health care that the insurance company pays out, somebody (mostly the insured, that would be you) must pay in $1.20, the extra two dimes go for operating costs and profits. The only advantage of insurance is if you are too stupid to put the money away that you would otherwise spend on insurance, or if you are too stupid to live healthy and keep your medical bills down, then everybody else pays for your reckless disregard of the public good. But health insurance is worse than that, it gives everybody the impression that it's "free" health care, so the stupidity is encouraged. The food book that came as part of this CR come-on has only one message repeated over and over, "Eat fruit and veggies and whole grains, and stay away from red meat, white grains, and processed foods." Do you think anybody is going to worry about that? After ObamaCare made insurance mandatory, everybody -- apparently including the editors at CR -- seems to think the health care is (or should be) free, so eating healthy can be compensated by more (free) pills and hospital visits. They somehow detach in their minds that the cost of those pills and doctor visits is higher on the consumer than when they paid out of pocket. Still is, I can tell you, but the out-of-pocket payers pay more also. That's why I was so annoyed when ObamaCare passed. Fortunately, I got my bionic eye before ObamaCare happened. It is what it is, but I certainly don't want my money going to support that kind of idiocy at CR. I took the "cancel" option. Maybe if I do that often enough, they'll stop sending the offers.

Postscript, They sent the next issue before they got my cancel. Good thing I didn't pay for it. Now I remember, a couple years ago I did pay for it, and most of the issues were a complete zero, so I did not renew. Maybe if I accept their freebie and cancel often enough, they'll stop sending the offers. They obviously don't pay attention to anything I say. But then nobody does, not even the people who say they care.

Earlier this year / Later this year
Complete Blog Index
Itty Bitty Computers home page