We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.
Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]
|
I just did a bit of copying and pasting Of this:
Born in New York City in 1945, Tesler eventually studied computer science at Stanford University before working in the school’s artificial intelligence research lab in the late 1960s. He moved to Xerox in 1973, where he devised the time-saving concepts to cut, copy and paste in computer systems.
“Your workday is easier thanks to his revolutionary ideas,” Xerox tweeted Thursday to honor Tesler.
Oh, it goes far beyond that, for me and surely for many others. My whole life was made possible by this sort of stuff.
With thanks to Instapundit. It’s little postings like this that keep me going back there. If all there was there was politics, I’d keep going back, but surely not so often.
Imagine having to copy out just the two links above, letter by letter, number by number. Absurd.
A BIT LATER: I put together all of the above for my personal blog, but then I thought: this should go to Samizdata. So I copied and pasted it across. Took me about two minutes.
|
Who Are We? The Samizdata people are a bunch of sinister and heavily armed globalist illuminati who seek to infect the entire world with the values of personal liberty and several property. Amongst our many crimes is a sense of humour and the intermittent use of British spelling.
We are also a varied group made up of social individualists, classical liberals, whigs, libertarians, extropians, futurists, ‘Porcupines’, Karl Popper fetishists, recovering neo-conservatives, crazed Ayn Rand worshipers, over-caffeinated Virginia Postrel devotees, witty Frédéric Bastiat wannabes, cypherpunks, minarchists, kritarchists and wild-eyed anarcho-capitalists from Britain, North America, Australia and Europe.
|
The concept of cut-and-paste was cut and pasted from the process of physically cutting and pasting of typewritten words on paper. But taking advantage of the properties of bits transformed the concept. Bits can be copied as well as cut, and pasting them is not completely analogous to pasting snips of pigmented paper – in many ways, it’s the “good parts” version.
May this good man Rest In Peace.
This post needs to be saved in perpetuity.
Wheres my hammer and chisel?
“Imagine having to copy out just the two links above, letter by letter, number by number. Absurd.”
That part wasn’t the innovative element. It was mostly about the merits of verb-object word order versus object-verb – a more subtle and far cleverer point.
The earlier way to do it was based on the English language approach of verb-object. To move text, you said ‘move’ (where the computer now has to give some sort of visual indication it is in ‘move mode’), then specified the source location, then the destination location, then told it to do it. It was confusing because it led to an explosion of ‘modes’, in each of which the input was interpreted differently, which was hard to learn, it made it more complicated to fix mistakes (if you realise after that you didn’t want ‘move’, you wanted ‘copy’, you had to go back to the start and select your text again, it meant queueing up a chain of operations to be carried out at the end instead of each operation being followed by immediate feedback, and it required a different tool to be developed (with a different mode) for each job, rather than using combinations of the same small set of tools for doing many jobs. Say you want to copy a piece of text into ten different places. With a copy command, you have to reselect the source and destination text after every copy: copy-select-select, copy-select-select, copy-select-select,… or do a special copy-n command and then select all n destinations (with frustration if you make a mistake) before executing. With select-copy-paste, you can more easily turn that into select-copy-paste-paste-paste-…
It’s a subtle bit of psychology. It was all about usability, and task complexity, and how we think about tools. Break down complex tasks into a sequence of steps, and build a small set of simple tools for doing each step, which the user can combine, instead of having the programmer have to predict and build a very large number of tools for doing each of the large number of complex tasks a user might want as a monolithic unsplittable block. It was a significant philosophical shift.
It’s the small things we take for granted and don’t notice that often make the biggest difference.
Both approaches are English-language oriented. The question is whether the text is an object or a subject: do you move (or copy or whatever) the text or do you command the text to move itself (or copy itself or whatever). Selection-first makes the text the subject of whatever verb-command you then issue, whereas the other method makes you the subject (who had better know what you intend to do before you start selecting the text that will be the object of your already-issued verb).
Nullius in Verba
I have been using HP calculators forever.
Does Reverse Polish Notation fit into this somewhere?
i.e. enter two numbers first then choose what to do to them.
Seems similar to me.
Reverse Polish Notation works by a similar principle, although it is designed more for the convenience of language designers and computers than it is for programmers or users. It’s a stack-based language, which gives a particularly simple way to implement tree-based algebraic expressions without the need to do complex look-ahead and operator-priority parsing that human-like languages use.
When it comes to language, humans have this super-sophisticated parser already built in, so it’s actually easier for humans to comprehend programs and algebraic expressions written like a language than it is if the operations are listed sequentially in order of execution. But when it comes to planning how to use tools to carry out a complex task, we don’t have quite the same built-in machinery. So the simpler sequential atomic-operation approach works better, for the same reasons stack-based languages are easier to code. It takes less working memory.
As it happens, the cut-and-paste operation was based partly on a version of a prior program called TVEDIT, ported to the PDP-10 by Pentti Kanerva, that had a clever error-correction method, where ‘deleting’ text moved it to the top of a stack, and then a second operation (called ‘oops’) moved whatever was on top of the stack to a selected location. Although intended to allow unintended deletions to be corrected, breaking it down into two steps with any sequence of other operations allowed in between them also allowed moves to be done more easily.
Programming languages are often designed around the limitations of the hardware. Different limitations emphasise different language features. One of the most important is working memory, the number of bits of information you can have instantly accessible at once, rather than being kept in higher-capacity but must slower to access main memory. Computers have only a small number of registers – early computers might have had only 7. And the human brain has been said to be able to remember only five or six things at once, before new items push old ones out.
But humans have a separate system for parsing language, with a higher capacity. Even so, there are limits. Consider the problem of centre-embedding. “A man that a woman that a child that a bird that I heard saw knows loves.” To parse the sentence, you have to store up multiple incomplete parts until their completion comes along. You can do one or two without effort, but four is noticably difficult. “I heard a bird, that was seen by a child, who is known by a woman, who is loved by a man.” That puts a lot less strain on the memory. You have to structure it the right way for your hardware.
The example at the end of Nullius in Verba (February 21, 2020 at 11:51 pm) above is unusual in English but would be less so in classical German. Bismark sometimes wrote classical German sentences that filled an entire page and had 15 verbs all at the end. Cummings notes Bismark’s considerable political skills, but a girl I knew at Oxford doing a German degree had a hard time warming to him – those sentences made demanding translation exercises.
And since I respect Cummings, whom I mention in my post above, it follows that I the Brexiteer that the statesman that the girl that I knew at Oxford disliked notes respect. 🙂
Have we uncovered the cause of Germany’s wretched political culture? 🙂
On the other hand, if Bismark could do this sort of thing in a sentence with 15 verbs, he was clearly no fool – though maybe a bit of an elitist.
I may have mentioned before the Berlin conference history tour in which the guide said she was relieved to guide an English party because when guiding Germans, there were people you could simply praise and others (one Adolf, for example) you could simply blame but when you reached the years of Bismark you had to look very carefully at the average age and style of the German party before deciding whether to present Bismark negatively with a few good points or positively with a few bad points.