[Neuroimaging] When to use neurostars, and when not to

[Matthew Brett]

Hi,

On Tue, Aug 25, 2015 at 6:40 PM, Michael Hanke <michael.hanke at gmail.com> wrote:
> Hey,
>
> On Tue, Aug 25, 2015 at 7:15 PM, Matthew Brett <matthew.brett at gmail.com>
> wrote:
>>
>> Hi,
>>
>> On Tue, Aug 25, 2015 at 6:10 PM, Michael Hanke <michael.hanke at gmail.com>
>> wrote:
>> > Hi,
>> >
>> > please allow me to hijack some of this thread and ask for feedback on
>> > how to
>> > guide users to find help. For the upcoming relaunch of neuro.debian.net
>> > we
>> > have drafted this (example for nibabel):
>> >
>> > http://neurodebian.ovgu.de/pkgs/python-nibabel.html
>> >
>> > If you click the "help" button you can see how we envision this
>> > guidance.
>> >
>> > Here is a more featureful example:
>> >
>> > http://neurodebian.ovgu.de/pkgs/fsl-core.html
>> >
>> > You can see that neurostars will be recommended for most cases. I'd be
>> > glad
>> > to get your feedback and any suggestions for improvements.
>>
>> Why yes - I do have a suggestion - I recommend you refer people to the
>> relevant mailing lists!
>
>
> I agree. The problem is that it is not always clear what mailing list is the
> right one. Currently,
> we use this formulation:
>
>
> | If you have evidence that your issue not specific to NeuroDebian, your
> report is best directed to the developers of the respective software.
> | Information on how to contact the respective developers is available at
> this location , or on the project homepage.
>
> where 'this location' and 'project homepage' are links to the project sites
> that we have on record
> in the package meta data.
>
> Needs improvement?
>
>
>>
>> But more seriously - we have a serious issue here which is about the
>> balance of convenience and accessibility against the quality of the
>> community that results.  We don't have enough data to know how this
>> will pan out yet, but my strong instinct is that the risks of moving
>> user discussion and questions to neurostar-type forums are rather
>> high, and not worth the gain.
>
>
> I'd be interested in the outcome of this discussion and will adjust our
> acting accordingly.
>
> We (NeuroDebian) have to balance the goals of not "stealing community", and
> not "flooding upstream with
> Debian-specific  bugs".

The question for me is not about stealing community - but about what
kind of community results from the different interfaces.
Unfortunately we (tech) people don't often consider "soft" questions
like this - or reject them from consideration as being in poor taste.

See you,

Matthew