On Sun, Sep 29, 2019 at 8:34 PM Eric Bresie <ebre...@gmail.com> wrote:

> Sounds interesting,
>
> However there may be some privacy concerns which but as long as it's
> optional (and has to "opt-in") then that may not be a big issue.
>

Glad to hear that, thanks.
Sure, everything would be optional, configurable, anonymous with publicly
available data.
Rough guess is that if we could get 30% of users out there, we could have
pretty good idea whats going on


> I am wondering if some of these sorts of metrics might be easier to gather
> from the "download sources" of features, plugin's etc.
>
I thought I remember seeing someone (maybe Laszlo?) posted on the mailing
> list showing some metrics but I think that may have been some linux snap
> type metrics.  Given there may be chance that some of the downloads may be
> from maven central and other repositories, not sure if that is fully
> possible per say but may be interesting.
>

These could also be included, and track by different channels. The idea
would be to put all the metrics about different things in one place to get
insights about tool, users, community and development.And then keep adding
some advanced analytics and AI over time.
First step would be to recreate existing usage tracking infrastructure
under Apache. Does anybody knows whats needed to make that happen, and
would that be possible?

Zoran


> Eric Bresie
> ebre...@gmail.com
>
> On Fri, Sep 27, 2019 at 4:18 AM Zoran Sevarac <seva...@gmail.com> wrote:
>
> > Hi,
> >
> > One of the things we talked about during our discussions at Code One is
> to
> > set up some
> > anonymous, open data, usage analytics for Net Beans . We could use and
> > extend existing usage information gathering and that could be optional.
> It
> > would help us to better understand how users are actually using NetBeans
> at
> > the moment, which features, technologies etc. and maybe make better
> > decisions about its development.
> > For start it could be basic statistics but we could include more stuff,
> > figure out some useful patterns.
> > In a long term it could also include info about development, issues, bug
> > prediction etc.
> > Maybe that would also attract data science/ai/ml community and we could
> > possibly expand in that domain.
> >
> > Any interest to make that happen?
> >
> > Best
> > --
> > Zoran Sevarac, PhD, Associate Professor
> > Department of Software Engineering
> > University of Belgrade, Faculty of Organisational Sciences
> > <http://www.fon.rs>
> > Deep Netts <http://www.deepnetts.com>  Co-founder & CEO  | Oracle
> > Groundbreaker Ambassador | Java Champion
> > <https://community.oracle.com/community/java/java-champions>
> > Open source: Neuroph <http://neuroph.sourceforge.net/>founder, Apache
> Net
> > Beans <https://netbeans.apache.org/> contributor
> > Homepage: http://www.zoransevarac.com
> >
>


-- 
Zoran Sevarac, PhD, Associate Professor
Department of Software Engineering
University of Belgrade, Faculty of Organisational Sciences
<http://www.fon.rs>
Deep Netts <http://www.deepnetts.com>  Co-founder & CEO  | Oracle
Groundbreaker Ambassador | Java Champion
<https://community.oracle.com/community/java/java-champions>
Open source: Neuroph <http://neuroph.sourceforge.net/>founder, Apache Net
Beans <https://netbeans.apache.org/> contributor
Homepage: http://www.zoransevarac.com

Reply via email to