It's probably (most likely) not the best place to post this so please
excuse my annoyance.
Some time ago I made one of the projects I've been working on
available to everyone through Github.com and PyPi. There is seems to
be some interest in it -- according to PyPi download statistics last
version of the package was downloaded few hundred times. I was
expecting some feedback like bug report or feature request. I was
surprised that there was none beside one nice guy posting a patch to
fix a bug in it.
I'm not complaining, I'm far from it. I just want to understand what
is it making people to download every new version of your package and
not giving any feedback. Honestly, even "you suck" would be great at
this moment :)
Frustrated contributor
P.S. PyPi does not privide such details but I'm starting to suspect
all those downloads are made by some kind of Web crawlers that scan
PyPi database every day in search for new /updated packages. That
would explain everything I was wondering about.
I feel your pain, but you are doing amazingly well to get your reported ~1% contribution rate. My advice is that you make sure your projects are worthwhile or useful enough to you, personally, to justify the effort even if you do not have any other users. You will feel better and it will keep you from adding features just because an imaginary "someone" might need them.
> It's probably (most likely) not the best place to post this so please
> excuse my annoyance.
> Some time ago I made one of the projects I've been working on
> available to everyone through Github.com and PyPi. There is seems to
> be some interest in it -- according to PyPi download statistics last
> version of the package was downloaded few hundred times. I was
> expecting some feedback like bug report or feature request. I was
> surprised that there was none beside one nice guy posting a patch to
> fix a bug in it.
> I'm not complaining, I'm far from it. I just want to understand what
> is it making people to download every new version of your package and
> not giving any feedback. Honestly, even "you suck" would be great at
> this moment :)
I'd say you're doing well to get a patch with only a few hundred downloads. I've found the contributor-to-user ratio to be about 1/500 or so.
there are a large number of automatic downloads from web crawlers that will index or re-host the packages.
some of them are legit, others are for advertising farms ( they do a lot of SEO optimization against your project , to overtake the real origin in search rankings - then advertise against the traffic ) or marketing scams ( every time i upload something, i get a new "Your package was included in this prestigious directory!" form letter )
otherwise, I've had the same experiences as you.
i've discovered that if you want feedback/etc - the best option is to make a link in the PyPi record and the homepage to a GitHub page ( or make the homepage on github ). aside from tapping into their community / tools , it just makes it super easy for someone to comment "this is great" or ticket some bugs/feature requests .. vs emailing you or emailing a public list. i've also had people randomly fork, rewrite and push code back to me - without hearing from them. and i've done the same too -- finding it faster to just fork , fix and pull-request than to wait on a debate.
Thanks for your reply and the advice. That project is part of my paid
job so it's definitely worthwhile to me. I am a lucky one that's being
paid for what he loves to do.
- Alex K
On Aug 16, 1:26 pm, Daniel Holth <dho...@gmail.com> wrote:
> I feel your pain, but you are doing amazingly well to get your reported ~1%
> contribution rate. My advice is that you make sure your projects are
> worthwhile or useful enough to you, personally, to justify the effort even
> if you do not have any other users. You will feel better and it will keep
> you from adding features just because an imaginary "someone" might need
> them.
I understand that it was too optimistic to expect people making any
feedback until project reaches certain level of usability and
popularity. It's great to have someone who went that same way and
shares valuable information.
Thank you
- Alex K
On Aug 16, 1:43 pm, Chris McDonough <chr...@plope.com> wrote:
> > It's probably (most likely) not the best place to post this so please
> > excuse my annoyance.
> > Some time ago I made one of the projects I've been working on
> > available to everyone through Github.com and PyPi. There is seems to
> > be some interest in it -- according to PyPi download statistics last
> > version of the package was downloaded few hundred times. I was
> > expecting some feedback like bug report or feature request. I was
> > surprised that there was none beside one nice guy posting a patch to
> > fix a bug in it.
> > I'm not complaining, I'm far from it. I just want to understand what
> > is it making people to download every new version of your package and
> > not giving any feedback. Honestly, even "you suck" would be great at
> > this moment :)
> I'd say you're doing well to get a patch with only a few hundred
> downloads. I've found the contributor-to-user ratio to be about 1/500
> or so.
> there are a large number of automatic downloads from web crawlers that will
> index or re-host the packages.
> some of them are legit, others are for advertising farms ( they do a lot of
> SEO optimization against your project , to overtake the real origin in
> search rankings - then advertise against the traffic ) or marketing scams (
> every time i upload something, i get a new "Your package was included in
> this prestigious directory!" form letter )
> otherwise, I've had the same experiences as you.
> i've discovered that if you want feedback/etc - the best option is to make
> a link in the PyPi record and the homepage to a GitHub page ( or make the
> homepage on github ). aside from tapping into their community / tools , it
> just makes it super easy for someone to comment "this is great" or ticket
> some bugs/feature requests .. vs emailing you or emailing a public list.
> i've also had people randomly fork, rewrite and push code back to me -
> without hearing from them. and i've done the same too -- finding it faster
> to just fork , fix and pull-request than to wait on a debate.
