Package download statistics

[Chris Turner]

On 06/02/13 07:32, John Marino wrote:
> The bigger question, of course, is *why* pkgsrc users are likely to build from source?  Could it be that they had lots of issues with that in the past?  Who wouldn't want something fast and quick if it were trustworthy?  The "indication" alone is telling.

My point was, more generally, that one shouldn't read too much into
the numbers.

To play the contrarian to your point, solely for the purpose of
further showing why it is silly to read too much into the numbers,
it could just as easily be speculated that:

a) people are initially testing dports, and it's faster to do so
    with binaries

    For 'ports like' platforms, I often 'bootstrap' a platform with
    binaries and then rebuild from source 'on the flipside' later on
    once I get a build environment setup, and this is just 1 use case.

b) only 'neophyte' users use binary packages, and since dports
    has been put forth as quasi-offical-in-beta in various
    announcments, any such users are much more likely to use dports

and other various spurious and presupposed self-gratifying
foregone conclusions.

I prefer to build from source, no matter what, period, and, while
not necessarily the most common type of user, am not alone in
this method.

Again - my point is not that package system A or B is better, but
that the ratio could be explained by several factors, and so reading
into the numbers is not well advised, in any pro/con/neutral
direction, and so any speculation given the data as presented
is just that.

Cheers,

- Chris