I caught an interesting observation on O’Reilly’s Radar. In my opinion there are two problems that web users should pay attention to. One, as noted on Radar, is privacy and the other is reliability of your service provider.

There are limits to what we can do about either of them as a company. We built Marela on a fault-tolerant system using solid hardware and think user’s data is reasonably safe from data loss and intrusions. Like others we can’t rule it out completely, but I think it was worth it to build a safer system than our competition even if users often don’t care (until it’s too late).

We also don’t collect any personal data we don’t actually need and don’t store what we do collect a minute longer than absolutely necessary. However, there’s little we can do about our legal obligations. Even though legislation in Slovenia is not quite the same as the one cited in the article, it’s still true that it’s easier for law-enforcement institutions to legally access your data (unknown to you) if it’s not residing on your personal computer. We are eating our own dog food, so this doesn’t make us happy either.

I still think it’s dumb to have all your data stored only on your personal computer, especially since vast majority of computer users seldom if ever make backups. Encryption can go a long way in protecting data from prying eyes, but what I wish most is simply for people to be more aware.

Without awareness there’s little hope for improving the legislation and practices of our industry.

The pixel debate is back . Internet explorer out of the box doesn’t permit resizing text measured in pixels, which at the same time is the only reliable way of setting dimensions that produces same result everywhere.

The crux of the matter is simply who should have the final word in how the page is displayed, designers or users?

Personally I think it should be users. After all, it’s about them and if there’s a way for them to make their experience more pleasant or useful without degrading it for others, why shouldn’t they be allowed to do so? Good designs serve, not command.

However, I find it more interesting how many people have no idea what px unit actually means . Pixels, as defined by W3C, are not the same as pixels used to define graphic resolution of a computer, even if they usually seem to be.

For example, if we ever happen to get high-density monitors with say 300 dpi pixel density instead of these days common 96 dpi, then a 15″ display would have a graphic resolution of 3600×2700 dots, but it would still have only 1152×864 pixels in CSS model. Therefore one web pixel would be represented with around 9 graphic pixels.

Well, that’s the theory. In practice there’s a lot of cheating. Monitors have various sizes which are only an approximation of the stated one (15″, 17″ etc.). Operating systems try to pay attention to pixel density, but often don’t or can’t. And so on.

The result of this is that size of displayed text and images is device dependent even when in theory it shouldn’t be. So the only way is also good only up to a point.

Does anyone find margin collapsing useful and if yes, why?

So far they’ve only caused me grief, but there must be something I’m missing.