Tried to reproduce a recent ASPLOS result. Got 100x better performance than their baseline/before setup with something quite naive.
-
-
Comparing to that history psychology, we're engaging in "2006-era thinking", at best, and we might still be in the 80s/90s.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Gelman has a specific take on CS at the end of this post: http://andrewgelman.com/2016/10/13/30222/ …
-
The comments on that post are sad (and sadly, also completely expected)
@nothingspic.twitter.com/TNqeYvOVCq
End of conversation
New conversation -
-
-
Artifact Evaluations - http://artifact-eval.org/ , http://ctuning.org/ae/ - seem encouraging in this context, http://janvitek.org/pubs/cacm15.pdf
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
have you seen ioannidis' recent take?https://twitter.com/hirojin/status/813778497621688321 …
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
you're doing pretty well if the code is even available to reproduce results.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
we are even worse because it's possible to make the research easily reproducible, but we choose to not do it.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
There is good research into CS reproducibility! http://www.craigstuntz.com/posts/2016-06-17-Andreas-Stefik-on-PL-Human-Factors.html … (supporting your POV)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
maybe we need a “standard computer” like the standard KG and you can reserve time on it to run benchmarks
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.