Lars Wirzenius: Rant, 2004
- September 15: Too little text message spam in Finland?!
- August 16: The price of the C standard, part 2
- August 14: The price of the C standard
- July 26: Referer spam
- June 07: Nokia didn't really spam me, their employee did
- June 05: Nokia spams
- April 26: "Attachment: No Virus found" considered harmful
- March 21: Spammers make valid HTML, why can't you?
- March 01: Finnish "Computer security guide" hacked
- February 17: I hate computer hardware
- February 12: Stupid people shouldn't write programs
- February 01: Things used to be better when I was young
- January 8: Add a LED to your device
- January 6: Firstname.lastname@domain
Wednesday, September 15, 2004
The Finnish newspaper Helsingin sanomat published a news story about a study that concluded that advertising in mobile media in Finland is much less common than in the rest of Europe. Mobile advertising seems to mean, mostly, text messages with ads. Helsingin sanomat titles the news as "Study: Finland is a mobile advertising developing country" ("Tutkimus: Suomi on mobiilimarkkinoinnin kehitysmaa"). That's like saying that Finland is really bad at corruption and murder: we have too little of them, and need quickly to get some more!
There is too much advertising already, in my life, and I don't even watch television or listen to radio. Streets and the city scenery has plenty. Despite my best efforts to reduce it, I even get some via snail mail. The last thing I want is more advertisments reaching me, especially in my phone. Text message spam is wrong in so many ways, it is incomprehensible to me that anyone would even think of doing it.
Monday, August 16, 2004
In my previous entry, I wrote that in the ANSI on-line store the price of the C standard is $270. It still is, if you search with the code 9899:1999. However, if you search with the code 9899-1999 (dash, not colon) the price is $18. I think they're confused. I did not try to buy a copy, as I have one already (courtesy of a previous employer), and was looking for the price for other reasons.
I'm not going to report this to ANSI for at least a week, in case they decide to fix it by removing the $18 price. If you want a copy, then, get it now...
Saturday, August 14, 2004
I wanted to have a look at the C99 version of the C standard. I had the impression that it would cost about $18 at the ANSI on-line standards store. This is what I found:
$270 is way too much for a popular programming language standard. Blah.
The Finnish standards organization only sells a paper copy for about a hundred euros and that is something that has been published in 1993, so it's not even current anymore. Blah.
Monday, July 26, 2004
My Referer log is full of spam links. I assume this is because some sites automatically publish Referers, so spammers use them to create more links to themselves, thus boosting their status in web search results. Can't blame them for being lazy and uninnovative, but they're still a bloody nuisance.
(I know "referrer" is the right spelling, but if HTTP is going to spell it wrong, so should I, when I refer to to their thing.)
Monday, June 07, 2004
It seems that the spammish mails from Nokia were of the type where an enthusiastic person goes ahead and advertises without thinking things through and commits spam almost accidentally. Thus I can't hold the incident against Nokia the corporation. The difference is small, admittedly.
Saturday, June 05, 2004
It seems that Nokia has a workshop about using Python on their phones. For some reason, they have decided that it is a good idea to send mail about this to random people, such as myself. The mail is badly formatted, and markets Nokia's developer forum and includes a 70 kilobyte Word file as an attachment. It is unsolicited, commercial bulk e-mail; also known as spam. For shame, Nokia should know better. Of course, it might be that Tapio T. from Nokia, who seems to have sent the message, just made a mistake, but really, they should know better.
Given the massive proportion of the spam problem in the past year, anyone wanting to send marketing e-mails should think really carefully about it. Not doing so is incompetent.
Since this mail was sent within Finland, Finnish laws about marketing apply, and those laws say that the sender needs to inform the recipient where they have collected the address from. Nokia did not do it.
Given Nokia's lobbying for software patents, and this spamming incident, and the fact that my almost-new Nokia phone has keyboard trouble, I'm finding it difficult to find reasons to not dislike Nokia.
Monday, April 26, 2004
Due to various reasons, I am not currently running a filter on my e-mail. This gives me the pleasure to see the kinds of things spammers and virus writers do these days. This morning, I have seen a virus message with the the following signature (it was probably not the first such virus message):
+++ Attachment: No Virus found
+++ Panda AntiVirus - www.pandasoftware.com
It is good to scan outgoing e-mail for viruses, of course. If your scanner finds a virus, it should stop the mail from going out. If it does not find a virus, however, it should not add a note of this to the mail. Such notes have no useful purpose for the sender or recipient of the message, and only function as advertising for the virus checker. If the recipient is naive or ignorant and trusts the note, they will eventually be harmed by a virus.
The notes make the sender of the mail look stupid. Anyone with a clue understands that viruses will add such notes to the mail they send. Since the recipient can't trust the notes at all, anyone whose filter adds such notes is made look naive at best and incompetent at worst. Is that what you want to tell the world?
I don't want to have this kind of advertising in my mailbox. It wastes my bandwidth, my disk space, and my time. It differs from ordinary spam only in that it is sometimes attached to real mail.
Sunday, March 21, 2004
For various bad reasons, for the past week or so I've been
running without a spam filter. As a result, I've seen quite
a lot of spam. What amazes me is that a number of spam mails
have been valid HTML (usually 4.01 Transitional). I even
checked them with
rxp -Vs and
If spammers can make valid HTML, why is it so hard for the majority actual valid web sites to do that? Especially commercial ones?
Monday, March 01, 2004
It seems that the Finnish government and Microsoft, and possibly some other companies, have put out a computer security guide for normal homes. The online version is at tietoturvaopas.fi. It was broken into and vandalized today. At the moment, the web site merely says "No web site is configured at this address.".
I have not read the original guide myself. It was supposedly distributed to homes for free, but I didn't get a copy. According to Kai Puolamäki's commentary, the guide lacks information about things like backups.
If tietoturvaopas.fi had had proper backups, it should not have taken them more than an hour to set up a new web server (possibly redirecting DNS to another IP) and restore the contents to that server. The admins for the site were not prepared for this, which indicates negligence or incompetence.
The real bad guys here are, however, the crackers. Even if a site is silly and has bad security, there is no justification to break into it. Even so, I can't deny that this one was funny in a perverted way.
Tuesday, February 17, 2004
I really, really hate computer hardware. This weekend I bought and installed a new IDE hard disk. IDE disks are an old, well-established and well-understood technology, I thought, and installing a new hard disk should be pretty much plug and play. In reality, I knew better, but I let my hopes guide me.
Every time I buy a new IDE hard disk I run into some kind of limit that makes the new disk not work with an old PC. In this case, the 120 gigabyte disk was too big for one of my servers at home. The BIOS would not boot with the disk installed and no amount of fiddling with BIOS settings helped. (I'm sure I'm just stupid.)
Installing the disk on my desktop computer, which is of a
newer vintage though still a few years old, worked the first
try. Except for some read errors while doing
-c -c. But I seem to have gotten those sorted out
as well: they're rare and hopefully only indicate a problem
with the kernel device driver.
I'm pretty sure, however, that hard disk installations shouldn't be this difficult. Perhaps things would be easier if the hard disks were all connected over USB2 or Firewire. I assume this because they are connected simply by connecting a cable to a socket. When I install my headphones, I plug in a cable in a socket, and then they work. Surely that means external hard disks would work equally well?
Thursday, February 12, 2004
Stupid people shouldn't be allowed to write programs. I offer myself as an example: I spent today an entire work day trying to figure out why Bluetooth communication wouldn't work in one of the programs at work. I finally figured out that the program didn't open the serial port to the Bluetooth device. I added one line to open it and after that everything worked perfectly.
If I weren't so damn stupid, I would have made it an logged error to write a serial port that isn't open.
Sunday, February 01, 2004
Mike Johnston's column Sunday Morning Photographer for today talks about old technology being better than new technology. This touched a bitter chord inside me. I've adapted the mail I sent to him to this rant, if only because it's been too long since I ranted properly. Gotta give them rant muscles exercise.
I first used a computer, and started learning programming, at the age of fourteen, twenty years ago. Back then, computers were pitifully slow, with puny memories, but people did wonderful, amazing, mindblowing things with them. Partly these things were thrilling because they were done at all, of course. No-one had ever made a computer do this or that thing before and seeing it done at all was the amazing part, it didn't have to be all that good.
On the other hand, the pressure applied on the programmers by the limitations of the hardware sometimes brought out the very best in them, and the results were truly beautiful. Beautiful to another programmer, that is, users mostly saw weird command line interfaces and, just possibly, a few blobs of color.
Today, the programming profession is mostly not limited by the hardware anymore, or not to any significant degree. Computers are so fast, memories so large, bands so wide, that it doesn't really matter what you do, the program is going to be more than fast anyway. Also, a prevalent attitude is that it doesn't pay to use programmers' work time to make programs faster, since computers are cheaper to buy. Programs now bloat like never before. It's still possible to make beautiful programs (make programs that are also art, if only to another programmer), but fewer people bother, they're in it just for the money.
I see a trend in computer books as well. Twenty years ago I practically lived in the public library. I must have read through every computer book in the various libraries in the city of Kotka, Finland, where I grew up. That's not quite as impressive as it sounds, there weren't all that many of them. The books were not all new, some of them were from the early 1970s. Many of them were pretty good and had been written by experts in the respective field. Even a newbie could see a lot of effort and love had gone into them.
Typical computer books of today, however, are quite different. Many of them are written by people who are good at writing books, and who write books on any topic that is expected to sell. They're often competent enough, but the books are lacking that something that makes a book written by a true expert worth ever so much more than one that is merely competent.
They are also thick. There seems to be a marketing assumption that books will sell better the more pages they have. It might even be correct. Thus, books are written with lots of fluff in them, and typeset with every trick to increase the page count. Result: I haven't seen more than a handful of new computer books that were beautiful in recent years.
There is something in my subconscious that associates superb finishing with extreme quality. I cannot remember a single computer book that was beautiful that wasn't also very well written. When I need a new book on a topic (not necessarily about computers) I don't already know well, I may need to choose one of several alternatives. Usually I choose the book that is most beautiful: the one with most class, the most understated and elegant typography, the best paper, the highest quality binding, the nicest smell. (I smell books, don't you?)
I wonder why O'Reilly won't make beautiful books.
Thursday, January 8, 2004
People who design electronic devices should always include a LED to indicate whether the device gets power and is functional. A LED costs almost nothing, but it can be immensely useful when debugging. For example, today we were trying to figure out why a Bluetooth device wouldn't talk to us over a serial port. Knowing that it at least got electricity would have been helpful and helped eliminate one common source of problems.
Not including a LED is, in fact, a sign of stupidity or malice on the part of the designer.
Tuesday, January 6, 2004
Quite many people think that e-mail addresses should be
of the form
firstname.lastname@domain. I think
that is a bad idea. In any big organization there will
be clashes: people with the same name. More importantly,
from my point of view, is that it makes the address next to
impossible to communicate over a phone, or for most people
to remember correctly.
My last name, Wirzenius, is not the most complicated one in the world and yet it is quite common for people to misspell it. My father and grandfather have collected a list of such misspellings, and it has over a hundred variants. Among my favorites are "van Breda" and "Lessick". I didn't get my matriculation photographs from the photographer until they searched through all of them and matched them with my face. When making the appointment, I had spelled my name, letter by letter, over the phone and they had written down "Wicksten".
The times when I have to tell my e-mail address over the phone are usually related to high-stress business situations: a customer screams at me because they think they have found a bug. It does not make things better if they can't get it right the first time.