Happy New Year and various news

[Grigori Fursin]

Dear colleagues,

cTuning foundation and dividiti wishes you
a very happy and successful New Year!

We would like to share some news about upcoming events:

=== ADAPT workshop final program ===

http://adapt-workshop.org

18 January 2016, Prague, Czech Republic, 10:00-18:00

* KEYNOTE "Benchmarks vs The Zombie Apocalypse: A comparison"
by Ed Plowman (Director of Performance Analysis Strategy at ARM)

* Final Program: http://adapt-workshop.org/program2016.html

* Note, that we have successfully finished our open publication
model experiment where authors submit article via ArXiv and
have a possibility to share related artifacts at the same time
with their submission. We then had a fully open pre-reviewing
of a paper and validation of experimental results by the community
via Reddit (before ADAPT PC prepares the final program):
* https://www.reddit.com/r/adaptworkshop

We now think that such approach can help involve an interdisciplinary
community to collaboratively solve reproducibility issues
in computer systems' research. It can also change current
reviewing process into a positive cooperation
between authors, reviewers and readers!

=== Artifact Evaluation for PPoPP/CGO'16 ===
We have successfully finished 2nd artifact evaluation
for CGO and PPoPP'16.

All accepted artifacts are listed here:
* http://ctuning.org/ae/artifacts.html

We will have a public CGO-PPoPP Artifact Evaluation discussion
on the 14th of March at 6pm - everyone is welcome.

=== Collective Knowledge Tutorial at CGO'16 ===

We will also have a tutorial on Collective Knowledge Framework at CGO'16:
* http://cknowledge.org
* https://github.com/ctuning/ck/wiki/Tutorial_CGO2016

We now actively use this framework with our colleagues to:

* share any artifacts as reusable and customizable components
with a unified JSON API and distributed UID via GitHub
(so that others can easily build upon them) - it allows us to use
agile development methodology based on open source practices;
* use them to assemble and share customizable experimental workflows;
* replay or crowdsource experiments (such as multi-objective
autotuning) on the latest software/hardware
(rather than using quickly outdated Docker/VM images);
* apply statistical analysis to analyze variation
of empirical results (similar to physics);
* enable interactive articles.

We now have a growing community of users sharing benchmarks,
data sets, experimental workflows and predictive models in this format:
* http://github.com/ctuning/ck-clsmith
* https://github.com/ctuning/ck/wiki/Shared_repos
* https://github.com/ctuning/ck/wiki/Shared_modules

=== DATE 2016 ===

We will present Collective Knowledge framework at DATE'16:
* http://bit.ly/ck-date16
* http://www.date-conference.com

=== New year resolution ;) ===

Overall, we hope to make a considerably progress this year towards
truly collaborative and reproducible computer systems' R&D
- join our community effort!

Take care,
Grigori