Salix downloads, forum members, users

[mimosa]

I just saw on Sourceforge that 13.37 has been downloaded nearly 100,000 times - over half in Russia.

Is there a rule of thumb to indicate how many users that translates to?

Even forum members aren't necessarily users any more. But maybe most users never have any problems with Salix so don't feel the need to ask for advice.

I don't know if this is typical but I don't often download other distros these days - very occasionally out of curiosity. But in my distro-hopping days I looked at loads, many of which I didn't get very far with at all.

A wild guess - several thousand active users?

[tsuren]

Much less - I imagine. But honestly, I did not know that Salix was so popular in Russia...

[mimosa]

I think top downloads for 13.1 were in Argentina - thirty percent.

[gapan]

That number is a lie. Sourceforge wasn't registering downloads for a period of several months. I'm not really sure they fixed that bug in their metering service yet. So the actual number is significantly more than that.

[thenktor]

And guessing the active users by the download count IMHO is not possible at all.

[gapan]

And guessing the active users by the download count IMHO is not possible at all.

Quite right.

[Shador]

And guessing the active users by the download count IMHO is not possible at all.

Quite right.

Not to sure about that one. We've been told that asking people about their own behaviour (e.g. addiction, excessive use) can be unprecise, as their own one is usually relatively unreflected. So you just ask them about the behaviour of their friends which is much more reflected. At that point we were asking ourselves how you can jump to conclusions if you only know that at least one of the friends shows that behaviour. You've got no idea how much exactly. But apparently this is possible as it is supposedly in common use for studies.
You probably can't get a relatively exact value as you don't know the "success factor" (i.e. how many users that download actually keep it and possibly for how long). But by sufficiently averaging the "success factor" from other similar data, you should be able to get a very rough idea.
I also think if some group of people is downloading Salix more often, you can conclude that more of those persons use Salix than from the other groups.

[SlackHack]

Is there a rule of thumb to indicate how many users that translates to?

Doesn't Ubuntu put out figures based on how many unique IP's access updates?

[Shador]

Possibly. But that's a) not necessarily completely reliable (my IP changes e.g. once per day and some ISPs/companies/institutions/... use proxies) b) we don't collect such traffic statistics or at least don't use them in this way (there are legal barriers regarding collection/usage/publication of traffic data here in Germany, applies for other countries (in the eu) probably too and finally c) there's not one repository but also many mirrors. And users can setup new mirrors too as they wish. No way to track those and possibly multiple users using them.

I don't think we get much information from the sourceforge download statistics. Even if a and c remain. Not to forget there are also torrents

Anyway even without unique ip information and even more with that, it might offer a rough direction.