Saturday, July 28, 2012

On the negative side of duties

Yesterday Lennart Poettering presented a talk with the title "The UI and the OS". The main idea of the talk was simplification of the architecture of the whole OS. On one of the slides, there was approximately this seemingly-obvious text: "The distributions' job is to distribute your software".

This phrase (while completely appropriate for the slide in question) has reminded me one of the first lessons on philosophy that I received from G. V. Boldygin in the Urals State University while studying there. On that lesson, we discussed the role of science (including theoretical science) in the society. The obvious idea immediately proposed: theoretical science leads to technological progress by telling what to do that nobody else did before. The surprising fact given to us as students was that there were significant practical technological advances (achieved by trial-and-error) even before theorecical science was formed. The conclusion was that it is almost always incorrect to refer to only one side of something's role of duties (i.e. to something that is done, as opposed to something that is prevented, or vice versa). Indeed, we were told that science not only provides technological progress directly, but also saves us from wasting money and time on certain "scientifically impossible" projects such as trying to build a perpetual motion machine.

So, let's apply this lesson to refute Lennart's phrase: distributions exist not only to distribute your software. Indeed, as illustrated by the Windows world, one doesn't need any help for this. The other side of their duties is that they reject bad (in their opinion) software by not distributing it, thus creating an anchor of trust for software they do provide to users.

No comments: