Responses inline.

Thanks,
-j

On Tue, Oct 9, 2018, at 3:13 AM, Justin Mclean wrote:
> Hi,
> 
> > I've been working on a project for 4-5 years now which I think would make a 
> > good Apache project, at least in terms of it being valuable, high-quality 
> > software. We're using it internally for our production systems at work, but 
> > the code is open (hosted on GitHub). Our process to date has been somewhat 
> > lacking (starting out as a 1-man project, now up to 3-4 regular 
> > contributors). The project is called Indy 
> > (https://github.com/Commonjava/indy).
> 
> The project seems small, but you have managed to grow it and have 
> contributors so congratulations. How large is your user base? Is there 
> potential for users to become committers in the project?

We have a pretty small user base inside Red Hat, mainly concentrated on a 
single deployment. Outside the company, I really can't say, except that we 
don't see a lot of traffic on the GitHub site (PRs are mainly people engaged in 
supporting that internal deployment, some of them hacking from other teams). 
Our internal exposure is growing as Indy learns to do things like cache/track 
content downloaded during a build from arbitrary upstream SSL servers. It looks 
like our install base will be going up, as well as internal users.

But I haven't tried to talk about Indy publicly at all, except for the 
occasional oblique reference to it while discussing other things.

> 
> > With my history working in and promoting the Maven community in the past, 
> > I'm hesitant to say that I can give Indy the exposure necessary to attract 
> > a really thriving, diverse community. This is not a strong area for me 
> > personally, as talking about myself and my work doesn't come naturally. 
> > Also, I've got a lot of existing commitments in life, many of which revolve 
> > around Indy at work, but which don't leave a lot of room for doing extra 
> > promotion work.
> 
> There's more than one way to promote a project, but it can be a lot of 
> hard work and time and effort. Another Apache project I’m involved in, 
> is seeing growth after a year of talking to a lot of people, speaking at 
> conferences, writing articles and q whole lot of other work. Can any of 
> the other contributors help you in promotion?

As it stands now, the other contributors would probably be interested in 
promoting Indy at local JUGs and such (as am I), but will probably have limited 
ability to get on the conference circuit.

> 
> > Does the Incubator have some facility or capability to help project teams 
> > attract a broader community?
> 
> Not explicitly, but often being part of the Apache ecosystem and 
> interacting with other Apache projects can get more people interested in 
> your project and become part of your community.  Are there any other 
> apache projects you see synergy with or that could integrate with your 
> project?

Fundamentally, Indy is intended as a means of organizing the sources of content 
used during builds and build-like activities. We have an ability to do more 
than that, but IMO the fit could become somewhat awkward the farther afield we 
go (from build systems).

Increasingly, as large enterprise integration and CD-oriented projects start 
offering automated builds of things like Maven projects during their runtime 
workflows, it seems like building Indy into the mix would offer a natural way 
to insulate from network failures outside the bounds of the immediate 
environment. We dabbled with a trimmed-back version of Indy specifically 
targeted at embedding, and it wouldn't be hard to open up that line of work 
again if someone had an interest.

I would think that build tools like Maven, Ant, Gradle, PyPI, NPM/Yarn/etc. and 
things like that would benefit from Indy. Obviously, not all of those are 
Apache projects, but I'm sure you see what I mean. We refactored the core of 
Indy to facilitate support for different package types (not just Maven 
repositories) about a year ago, but we haven't done too much with it yet beyond 
NPM proxying / hosting support. I'd say we haven't yet proven the Rule of Three 
for interfaces yet, but we're close.

> 
> Thanks,
> Justin
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscr...@community.apache.org
> For additional commands, e-mail: dev-h...@community.apache.org
> 

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@community.apache.org
For additional commands, e-mail: dev-h...@community.apache.org

Reply via email to