Wednesday, November 29, 2006

How not to blog

I just stumbled over a few micro-blog entries, like 'what is time'. Cool title, completely disappointing content. The guy is just asking his commenteers exactly this questions.

Let me put it this way:

If you have something to say: Blog. (And 'that URL is interesting' is not interesting by itself, thanks.)

If you have somthing to ask: Google! (Instead of additionally providing misleading google fodder.)

Saturday, October 28, 2006


I wrote the beginning of the previous entry twice. The old mozilla managed to crash just while editing that post (and googling for the URLs). Ok, if you can't even remember what you wrote then it's not worth it anyway. But being a programmer I do not like repeating myself, or repeating the refinements of text; I even notice that it bugs me when I actually write code fragments on paper and can't just insert the argument I forget and have to rewrite the whole line.

So forgo the blogger edit window and use the friendly local emacs, where your idle process types ^X^S anyway after a few seconds' timeout.

Teardrops in the programmer's eye

Tears are a sign of pain. Or of being deeply touched, the soul's indication of a profound truth (and pain of internal readjustment). I had three pages recently doing that to me. The first was a the end of LFM and LFSP, about people stopping coding because they were forced to use langages inadequate for their mind. I feel the beginning of that myself.

Second: Transformation by Steve Yegge. The guy can write, and does so a lot, but this piece has an absolutely beautiful final point. You may not agree with his conclusions but he is interesting reading. And read the book first.

Third: This piece. My detector slightly beeped all the way through--there must be something to it.

Wednesday, October 18, 2006

Rigging your Poll, #2781

A majority of germans would accept more video surveillance in railway stations (german) to fight terrorism. Says the headline. Now, the actual poll asked whether said surveillance would increase security. Note the difference?

I am actually quite convinced that monitoring every fart of the population would will provide less opportunity to do stupid things. But this is a far cry from whether I would accept cuts to my personal freedom, which I don't. Just look at the numbers. Getting innocently killed in a car accident is still must more probably than getting killed in a terror strike, even with 9/11. But for some reason nobody wants to cut driver's freedom to make traffic safer. Strange, isn't it?

Also, I fear to some extent that massive surveillance will not actually prevent terrorist acts. Even the israeli don't succeed in that. What will be easier is to identify the terrorists after the fact which is pretty pointless with the current crop of suicide terrorists.

At least I'm not the first one to notice, so I don't need create an account with them to be able to file a comment.

Friday, October 13, 2006

You shouldn't be reading this

Well, this blog is just of the kind of dual-use that Steve Yegge had in mind: Making up one's mind and training writing. So as much as I would like readership I can hardly expect them. I like reading Steve's blog; usually they are long essay-style texts carefully crafted, and also happening to be in my line of thought yet inspiring. And he gets to points from seemingly unrelated starting points which I think is much better than just stating the point. At least when it's written in an entertaining way, which he usually does. No idea how much work he puts into that stuff.

Most other blogs feature rather shorter texts that just pound out a point or two (or even just an annoyance without any actual positive ending); I seem not to like those as much. Good thing about them: technical solutions presented there are preserved for googling, which sometimes made me want to write up stuff that I found out myself and isn't quite in any howto, for example.

For me, blogging work seems to happen primarily at late weekend evenings (when the next day need not start early). Don't be fooled by the posting times, by the way; I'm probably not sitting in the same time zone as you are.


I just had to sign a few of these. You know, the kind saying 'John Doe has successfully slept through "Incompatibility obfuscation complication library"'.

I could hardly keep a straight face through this. For, first thing, I just basically gave a talk along some slides, and, unaware of the certificates beforehand, I did not do any tests. (Wouldn't have if I knew, either.) Only indication of achieved understanding were the questions I got underway, which for some people were interesting, for other largely nonexistent, which again has many possible reasons.

And what does 'successful' mean? Has been there and awake the whole time? May be able to clone the demo examples? Has grokked the purpose and intent of that library? They first two are somewhat easy to verify; the third would take me many hours. But only that kind of understanding is useful, only it does not come immediately when presented with the whole set of concepts; it usually takes time to 'get' it.

And foremost, what are these certs good for? Since they are generally only vaguely correlated with true understanding, they are only useful for people who have no idea on the matter themselves (think recruiters-using-grep); in an job interview you could more easily find out by asking a few questions if the candidate knows that library or would easily learn it.

But in this case the course topic was a proprietary library that no one outside the company uses, so the certs are as shiny as they are exactly useless.

Also you wouldn't really want to work for someone who asks for the cert instead of asking you whether and how much you know the topic. They show that they don't trust you, and instead they risk getting someone who learned just for the cert. If you ask for crooks you get them.

The only advantage gained by looking for certs is the same as by buying software: you have someone else to blame if it doesn't turn out so good. John Doe should have known the iocl; Microsoft should have delivered a working Word, it's not my fault. Unfortunately finding the guilty doesn't help you at all (unless you are SCO). It covers your seat, but even your seat would have been financed from the project that just went down the drain.

Sunday, October 01, 2006

...forced to find out

I just stumbled over why there is a word verification (aka captcha) which keeps robots from posting comments to these blogs even though only registered users can post at all: There were enough of those that post automated 'nice article, here's my blog'. State of the humanity: Disgusting?

And yesterday we were shopping for shoes. We came over a nice pair that was unfortunately only available in 36 (yeah, we have strange sizes over here) and not in her size. We then decided on another pair, and I wonder if we later will find out that we don't actually like them that much and instead have only transferred our liking of the first pair to everything in the shop. Well, at least they costed less than the lunch beforehand. (We also have strange habits regarding relative opulence of lunch and dinner.) State of the humanity: Easily fooled?

And would you believe that people willingly buy stuff that is more expensive and less functional just because it is a bit better looking? Ok, it happens with shoes all the time, but I'm talking of glasses now. Last round I selected an open frame, which means no glass lenses because of danger of damaging. So, plastic was required, and the clerk quickly talked me into aspherical lenses which he said were thinner, due to being aspherical and also due to being available with a higher refractive index. A lot of time went by until they arrived (which is another story and not related to any complex production process). And then they turned out to be simply shitty. The regular glasses had a field where everything goes blurry, and the sun glasses had a bit of that nearly everywhere except when looking exactly straight through. Now, I intend to do viewing around by moving my eyes, not my head.

As it turns out, 'aspherical' is not the same as in photographical lenses. There the surfaces are not exactly spherical because their properties can be improved by slight, but properly computed deviations. In glasses, however, they simple start to flatten out the lenses from the (basically) correct spherical form so that they don't end up being so thick at the rim. This means simply that the lenses do not work properly in some areas! And you get sold that as an advantage: The lenses are indeed slightly slimmer, but more expensive and less functional. I wonder how anyone would want to live with that who has has proper lenses before. (Someone who gets her first glasses may however think that the imperfect correction is unavoidable, never having seen it better.)

Only good thing is that these days the shops offer full refund if there is anything you don't like about the glasses. I had other glasses made for the regular frame (at no extra cost), returned the other, ordered sunglasses in another shop (also spherical), neglected to specify exact kind of glass and thus spent an extra minor fortune on those, got two glasses that work everywhere I can look through them, and learned a lot more about spherical, cylindrical and prismatical refraction, eye distance yaddayadda that I wanted to know. But you need to know to get glasses you want to wear.

Funnily I hear a lot of people complain about colored borders (aka chromatic aberration) but I hardly ever notice those, and I don't mind them.