[Networkit] Taking my leave

Raphael C drraph at gmail.com
Fri Apr 22 13:56:12 CEST 2016


Dear Chris,

As a normal (npn-dev) user I would first like to say thank you for all
your work. I would also like to say how valuable and rare high
performance graph tools designed to work with massive datasets on
single (although multi-core) computers are.  In python, as you will
know, there is igraph and networkx but the former is slightly moribund
and the latter is not designed to scale.  Other than that one is
forced to use spark but this can he hugely inefficient when graphs
with billions of nodes can be processed on a single multi-core
computer.

In short, I think there is a strong need for a tool such as networkit
and I am very grateful it exists.

Best wishes,
Raphael

On 22 April 2016 at 09:04, Christian Staudt <christian.staudt at kit.edu> wrote:
> Hi all,
>
> I finished my thesis this week, which seems like a good moment to officially take my leave as maintainer of the project. First of all, thanks to all contributors, who have invested countless hours in sum.
>
> For NetworKit we have followed a model that is not (yet) the norm for academic software - we have invested time into usability, documentation, packaging, distribution and support. As a result, NetworKit has something like a small user base, judging from the couple of researchers from all over the world sending us the occasional feedback.
>
> No doubts that we've created a useful resource to build research projects on. I wish I had something like NetworKit when I started my PhD research - such a code base could have saved me months.  On the other hand it has been very educational for everyone involved to build this from the ground up.
>
> I also think that such a code base is in fact, in the long run, a depletable resource that requires regular maintenance to stay useful.
>
> At this point in the development, there is still a backlog of various ideas for improvement. That includes yet another algorithm that could be added, as well as overall improvements that make NetworKit a better data analysis tool. However, personally I'm not convinced that there is enough demand out there that justifies working hard to take the tool suite to the next level now, since we are operating in a relatively narrow academic niche. That's why I'm going to turn to other projects.
>
> Thanks again to all contributors for making this work,
> Chris
>
>
> _______________________________________________
> NetworKit mailing list
> NetworKit at ira.uni-karlsruhe.de
> https://lists.ira.uni-karlsruhe.de/mailman/listinfo/networkit
>



More information about the NetworKit mailing list