From: Eric Niebler (eric_at_[hidden])
Date: 2006-06-12 10:45:05
Arkadiy Vertleyb wrote:
> Eric Niebler wrote:
>> I suggest we stop tracking the download counts
>> and remove the "top downloads" page until
>> the numbers can be made reliable. As they are,
>> the download counters are meaningless and misleading.
> I don't know if this kind of popularity contest
> is a good idea in the first place. First, it is not
> even accurate: it highly depends on how often
> the new version of file is uploaded. Second, all
> it may do is to help popular libraries to become
> even more popular.
> However, as I said before, I think the download counter
> is a very useful tool. It helps the library author to
> estimate the amount of interest to the library, that
> sometimes cannot be determined by the response on the
> mailing list. And this usefulness is not affected by
> occasional drop of the counter.
> So, I would suggest removing the "most popular" page,
> but leaving the counter -- it's very useful.
Then the counter should be changed from "total downloads" to "downloads
per day" or some such. Currently, it gives a false impression. Right
now, the download counter for xpressive stands at a whopping 3. What
potential user in their right mind would trust code that has apparently
been downloaded only 3 times. It's off-putting.
And yes, the "top downloads" page must go.
-- Eric Niebler Boost Consulting www.boost-consulting.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk