indeed! Now having those latter tabulations on hackage-server  (or a hint for new haskellers about where to look) would be dandy

i'm happilly over the "who does/likes what" hump myself, but it is valuable breadcrumbs for folks getting started. That said, asking via cafe  / reddit / irc is also valuable because you can get peoples *opinions* about when two libraries are better in what use case. Not all problems can have a "canonically best tool" (as much as we'd like to strive for such tools)


On Mon, Nov 4, 2013 at 5:04 PM, Dan Burton <danburton.email@gmail.com> wrote:
And these steps are done!

* Download count is already there on Hackage, though it's relatively new so it may take some more time for these numbers to have real weight.
* Revdeps are calculated and provided here: http://packdeps.haskellers.com/reverse

-- Dan Burton


On Mon, Nov 4, 2013 at 1:39 PM, Carter Schonwald <carter.schonwald@gmail.com> wrote:
Honestly the first step is making it easier (and efficient) to look at reverse dependency information plus download count.  Those numbers are both manipulable,  but can still yield some informative basic info. 


On Monday, November 4, 2013, Ben Gamari wrote:
Charlie Paul <charlieap@gmail.com> writes:

> This has been proposed many times before, and each time everyone agrees
> that something like this is a good idea. However, like many good proposals,
> no one has put up code.
>
I also think that many Hackage improvements like this one were blocked on
Hackage 2. Now since this is finally in the wild it should be a bit
easier for people to pick up this sort of project.

> Also in this particular case, the devil is in the details. How do ratings
> transfer between versions? How do you account for the effects of bitrot?
>
Certainly there are tricky details to work out but I think a lot of the
work will be simply getting to the point where we can collect ratings
and stuff them into a database. After this there would need to be some
experimentation to work out the finer points you mention.

In my mind a rating would consist of some numeric rating (1-5, for instance,
perhaps along multiple dimensions, e.g.: quality of documentation, type-safety
of interface, performance) for a particular package. The user, date, and
current version number should also be recorded.

A zeroth-order approach for accounting for bit-rot might be to use a
simple temporally-weighted average. This would be simple to implement
and might even produce marginally useful results. Even if not, it's a
place to start.

Cheers,

- Ben

_______________________________________________
Haskell-Cafe mailing list
Haskell-Cafe@haskell.org
http://www.haskell.org/mailman/listinfo/haskell-cafe