suggestion to increase community involvement

[Denis Akhiyarov]

Sometimes I see questions and issues posted on github, SO, and groups, which do not receive enough attention due to commitments of developers and users.

Mostly people add +1, but some of the issues may even drive away the early adopters to alternative technologies.

Is there any way to get Numfocus to support most outstanding items to resolve these questions or issues for the benefit of Numfocus projects?

Please let me know if you would like me to send few examples?

Thanks,

Denis - NumFOCUS supporter

[Arnaldo Cristián Antuña Cepellotti]

Mr. Akhiyarov:

I don't know if this is exactly what you are adressing, but I used to answer questions about Cython on SO, as a way to try to help using the technoology. My experience is that, in projects that are 0.x, sometimes the reason for not answering a question is not uninterest, but rather that the question refers to some feature still undeveloped in the technology and the best you could provide is a workaround that might or not be acceptable to other people.

As an example, at some point in Cython you could not declare a pointer to a method as a type (something like, say, ( int ) (MyClass::* method) ( int ) in C++) so my workaround was wrapping it inside a class in C++ that exposes that same pointer - but then every client must have a method that makes it compatible with receiving the pointer that way :/.
Probably not the best, surely not elegant, but it served its purpose.

There were a lot of critics to answers like this, because it involved a lot of C++ writing (which was trying to be avoided in Cython), it was rather hyperbolic, and so on... And in fact, those were valid critics, but, referring to an still undeveloped feature, it is natural some controversy as what is the best way to solve that issue - but that is a discussion I would expect between Cython's developers, not users.
Eventually, as every answer led to pointless discussions I just stopped answering.

I don't really have a suggestion to improve these kinds of situations, but I wanted to point out that, even if interested in answering the questions, sometimes adressing young projects is difficult. I'd be really glad to hear from people who dealt with this issue in a more effective way than I did.

Regards,

Cristián

[Denis Akhiyarov]

BTW, looks like Cython is dropped from NumFOCUS projects.

[Andy Ray Terrel]

Its not exactly dropped but we deleted the projects from that webpage that were only supported through our general efforts and not FSAs. I'll throw the other projects up now until we get that portion of the webpage fixed completely (instead of only half way as is now.)

--
You received this message because you are subscribed to the Google Groups "NumFOCUS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to numfocus+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Denis Akhiyarov]

What I'm asking for is not individual people to answer these difficult questions by all means, but rather have concentrated and funded efforts to resolve such problems by NumFOCUS. 

Something more powerful than bountysource.

On Thursday, August 13, 2015 at 8:24:09 AM UTC-5, Arnaldo Cristián Antuña Cepellotti wrote:

[Anthony Scopatz]

Hi Denis,

As a volunteer organization, I think that if there is a direction that one would like to see numfocus go in, the most effective mechanism for such people to volunteer :). Positions for the board are now being solicited.

However, there are a lot of roles that folks can play without becoming a board member.  I forget their exact title, but their are "vice president of X" sort of positions that you can also inquire about.  Generally speaking, if x doesn'ts exist you can create it and submit it to the board for approval.

There are many many things to do, and a lot of will to do them.  The bottleneck is human resources. I encourage everyone to help out in the way that they can.

Be Well

Anthony