[LTP] LTP survey

[Richard Palethorpe]

Hello,

Cyril Hrubis <chrubis@suse.cz> writes:

> Hi!
> I've been asked if we can help to fill in data for the LTP Survey from
> Jonatas send to the mailing list earlier.
>
> So if you haven't done so yet and can spare 10 minutes of your time,
> please go to:
>
> https://docs.google.com/forms/d/e/1FAIpQLSePEESEVOCVtjuAIvm1BkToLJzEcxv_m4HBRiJzuRcQQ4DzHQ/viewform?vc=0&c=0&w=1
>
> And fill it in.
>
> Many thanks on behalf of Jonatas!

I already filled it in once and I don't think my answers will change
much (despite any corrects to the survey).

I will add that I think MI is bullshit, it is not a metric because it
doesn't follow the triangle inequality and it is trying to compress an
N-dimensional problem (where N is very large) into a single scalar value
which is dangerous if people take it seriously. The fact it has
decreased as the LTP's code has improved shows Wittgenstein's ruler at
work. At best it is a heuristic, but it is probably better to use/view
its component metrics directly.

--
Thank you,
Richard.