(Search & Sitemap)
> Writing >
Journals >
Musings
Previous
1
2
3
4
5
6
7
8
Next
Printer-Friendly Version
Last night, a friend and I spent a long time trying to use the GIMP (GNU Image Manipulation Program) to perform a simple task (swapping colors in a two-color submit button). I came away from the frustrating experience with a new appreciation for what Unix's arcane interface is like to a newcomer.
Technical support people (and sometimes other hackers) have an acronym PEBKAC, short for Problem Exists Between Keyboard and Chair. From the jargon file:
PEBKAC /peb'kak/
[Abbrev., "Problem Exists Between Keyboard And Chair"] Used by support people, particularly at call centers and help desks. Not used with the public. Denotes pilot error as the cause of the crash, especially stupid errors that even a luser could figure out. Very derogatory. Usage: "Did you ever figure out why that guy couldn't print?" "Yeah, he kept cancelling the operation before it could finish. PEBKAC."
With a great many apparently technical problems, the problem exists between the keyboard and chair. The question I was thinking about is, which keyboard and chair?
On Mac and Windows computers, and to some extent on the web, an alert box will pop up with some snippet of text and a button. The messages that pop up are often not very important -- something like the warning labels attached to many products -- and, as such, alert boxes carry a nonverbal message of "I am interrupting your work because I have something to tell you, probably not very important, and you can't use your computer until you click my button. Once you have clicked on this button, you can go about your business." When there are a great many alert boxes like this, it is not a stupid thing at all to habitually click the button when the alert box appears... except that, on a small minority of such boxes, the habitual response cancels your print job. The problem with this system exists between keyboard and chair, but not the user's keyboard. The problem with this system exists between the designer's keyboard and chair. A great deal of stupid user errors are not stupid user errors at all, but the results of bad interface design by software developers who did not design with human-computer interaction factors in mind.
That's the thing about people who think they hate computers. What they really hate is lousy programmers.
Larry Niven and Jerry Pournelle in Oath of Fealty
PEBKAC.
I've thought of a hacker's game. Here's the core idea: Alice writes a program. Bob modifies the program in such a way that he can tell the difference between them. Alice wins if she can discern which is which. Bob wins if Alice can't tell, but he can tell. The round is a draw if they both can't tell.
This basic idea is in need of refinement and rules for both parties, roughly speaking in order that there are no obvious and cheap ways for either side to win. (Neither of them should be able to perform direct tests on the compiled programs, for instance, and things like "Click on the upper lefthand pixel of the applet window, and then hold shift and click on the bottom righthand pixel, and a smilie face will appear" aren't the kind of cleverness that is desired.") If such rules are formed, it will take a community's work over time. But I think this could be a good programmer's game.
I was saddened to learn of the demise of Canada's Rhinoceros Party, a satirical political party with platforms like "Coast from coast to coast!" (after your car has been raised to the top of a giant, Canada-wide ramp), "My platform is the one I'm standing on," and "Legalize pot. And pans. And spatulas. And other kitchen utensils." It's defunct as of the last two elections, and learning of its demise (when doing a web search, because I wanted to show Rhinoceros Party information to some of my coworkers) was saddening, like a child's finding that all the fairies were dead -- a learning that a shining part of the world has gone out.
The U.S. still has Dave Barry and his year 2000 presidential campaign (I'm taking an educated guess, as I'm waiting to hear the results of the Florida recount in the U.S. 2000 Presidential Elections, that the final difference between Bush and Gore will be less than the number of votes Dave Barry received), but that was saddening news.
There is an image I've had (partly from my own experience, partly from other sources) of someone very bright who is off in his own little world, and when he talks with other people, he tries to answer as faithfully to his own world as he can, and people just don't get it. What I realized in my Gospel reading a few days ago is that this happened with Jesus. He spoke from his world, and people tried to interpret his words as what they would have meant from their world, and there was a glaring absence of connection. Examples of this are threaded throughout John's mystical gospel account in particular; one conspicuous example is where Jesus is on trial before Pilate and they are talking about whether Jesus is a king. Jesus is trying to bring Pilate up to his plane, and Pilate is equally trying to understand Jesus's words without leaving his own plane, and there is conflict.
Seeing this in the Gospel accounts, and having things click, gave me a feeling of being in good company.
Make-believe is a kind of illusion that implicitly depends on being recognized as illusion. I was thinking about this basic phenomenon in some matters related to my Halloween costume this year. My fun was spoiled when I realized that at least one of the children had literally believed I was Blajeny, that the illusion of my costume had not been recognized as illusion.
A while ago, I was having a conversation with Robin (techie) and another friend (Bob, non-techie). We were talking about making custom modifications to software, and I mentioned that a few decades ago, it was common to have computers with their own instruction sets. Robin immediately saw the point I was trying to make; to translate for Bob, I said that for each computer to have its own instruction set would be like each book having its own alphabet.
In places where we've gotten used to standards, breathing them is second nature. There are rare exceptions where it is desirable to break good standards -- off the top of my head, I can think of the beautiful Elvish script in J.R.R. Tolkein's Lord of the Rings trilogy -- but, in certain areas, standards can be quite helpful. The computer industry is moving towards increasing standardization at higher levels of abstraction -- and this is a good thing. Dealing with a locally suboptimal standard solution twenty times involves, among other things, significantly less cognitive strain than dealing with twenty locally optimal nonstandard solutions.
Where I see this argument as applying (technical areas, and human cognitive strain), this is not a death penalty on nonstandard approaches -- many of the best technical ideas have been highly nonstandard approaches. I do believe, however, that things should be done in a standard manner unless there is good reason to do otherwise. For something meant for humans, doing something nonstandard means potential confusion and a probable learning curve.
Web pages that are not designed with a first-time visitor in mind are a prime example of material that breaks this principle.
I have been occupied recently, and have several ideas jotted down, but not taken the time to write them down. I wrote a letter and received an invitation to join a very high-intelligence mailing list; I spent a good deal of the past week worrying about whether I've bitten off more than I can chew. Today, I felt that still, small voice saying, "Get back to writing your musings."
The musing I've been carrying around for a while, has to do with the monoglot and the polyglot (the person who speaks only one language versus the one who speaks several). I debated whether it was worth writing down, and decided for a while that it wasn't worth writing until I came across something in George Steiner's Errata. He mentioned the distinction, talked about his own polyglot background, and then poetically and emphatically argued that polyglot is the condition to be in -- at one point, he said that the monoglot does not know even his own language.
That bothered me; to explain what bothered me, I would like to bring two brief images to mind: an American who is devoted to his country and holds the kind of patriotism Lewis extolls towards the beginning of The Four Loves, and the American who is devoted to his country and considers the natives of other countries to be unfortunate second-class world citizens. The first is laudable (and compatible with respect for the patriotism of other nations); the second is not. It is the second condition which parallels Stiener's exaltation of the polyglot condition and (unnecessary) denigration of the monoglot condition.
To explain where I stand on this question, I would like to begin with a lunchtime conversation with my best friend (Robin), and an old friend of his (Morris). I surprised Robin by saying that I preferred to read texts in English translation when I had the option -- preferably a free translation -- rather than reading them in a non-English original.
It's not that I'm afraid of learning another language. There have been times when I found thinking in French to be easier than thinking in English, and there has been a span of several years where my French sounded closer in relation to a typical native French speaker than my English sounded in relation to a typical native English speaker. If one counts mathematical and computer languages, I've worked with more languages than the number of years I've been alive, and this will probably remain true for the rest of my life. I've had the experience of not recognizing which language a text was written in, but still being able to read it. I've lost count of how many languages I've dreamed in, and I occasionally have dreams where my mind makes up a new language on the fly.
Why, then, would I prefer to read texts in English? In a single word, comprehension. I came to realize at one point that my knowledge of French at its best has been a rough equivalent to a native proficiency, but that I will never speak another language as well as I speak English, not if I am immersed in it for the rest of my life. The proficiency I have in English is something beyond what is normally meant by 'native'. There is an additional cognitive strain -- so I am spending energy trying to interpret the text (in the direct and mundane sense) rather than on interacting with its meaning (in a deeper sense). I'll understand a good free translation a lot better.
More broadly, proficiency in multiple languages takes mental energy that could be used to other purposes. There are people that can afford that expenditure of mental energy, and there are definite benefits to knowing two or more languages -- the ability to compare ("The better you know another world, the better you know your own." -- George Macdonald, Lilith), the ability to communicate with more people, the improved ability to pick up other languages. For all that, there is a consolidated energy that comes of having spent your efforts on learning one language and learning it well -- and there are a great many people in the world who do not have the excess mental energy to have spare room to learn extra languages.
Bloom, in the introduction to his translation of Plato's Republic, argued for making strictly literal translations. The essential argument is that the translator, however great a scholar he may be, must have the humility to realize that the student who reads his text may be a greater mind, capable of deeper understanding. As such, the translator should provide the student with what the words say, rather than confining the student to his interpretation. He proceeded to give several quotations from free translations of the Republic which, in trying to make the text accessible to a contemporary reader, succeeded in producing something accessible, albeit inappropriate as renderings of the text. I forget exactly what they were, but they would be comparable to portraying Martin Luther's crisis of faith as a postmodern midlife identity crisis.
I do not believe that choosing between literal and free translations is a choice between a flawed and a near perfect rendering model; a student who wants to really understand a text (which is written in a language he cannot read well) should probably peruse several translations, varying in how literally/freely they render the text. And, if I want to know a short text or excerpt well, my rendering of choice will be a heavily footnoted literal translation.
For large-scale reading -- for the kind of reading comprehension that can be sustained for numerous pages -- there is a different phenomenon. The danger in free translation is that it can confine the reader to the translator's interpretation. The danger in literal translation is that it can confine the reader to not understanding the text at all. A woodenly literal text, one that's read for dozens or hundreds of pages, brings a cognitive strain and consumes energy that could be used in thinking about the text. And, for that reason, if I can only choose a single translation, I'll take my chances with a free translation.
I read a book recently called Please Understand Me. It was a valuable resource to read, but it's something I'd prefer to give to others with a complimentary grain of salt.
It's about different personality and temperament types, and one of the central theses is that people have fundamentally different natures, but engage on a Pygmalion project to reshape others into copies of themselves. It is written in such a way that a reader who is persuaded of the legitimate point (that temperaments are not right or wrong, just different, and it is inappropriate to try to change a person to a temperament that he's not) will (in a similar fashion to the Green Book in C.S. Lewis's The Abolition of Man) come to the unjustified and illegitimate conclusion that there is not right or wrong in much of anything, and that it is wrong to try to change a person on any score. It never draws this conclusion in so many words, but a transposition of key takes what Lewis said about the Green Book and fits it (quite well) to Please Understand Me.
The book is worth reading, if you can resist the conclusion that the flow of the text pulls you towards.
(Search & Sitemap)
> Writing >
Journals >
Musings
Previous
1
2
3
4
5
6
7
8
Next
Printer-Friendly Version