XO-1 touchpad once more [Devel Digest, Vol 64, Issue 30]

[Yioryos Asprobounitis]

> Date: Tue, 14 Jun 2011 09:05:46 -0400
> From: Paul Fox <pgf at laptop.org>
> Subject: Re: XO-1 touchpad once more [Devel Digest, Vol 64,
> Issue 28]
>  > One is running os860 while the other different OLPC
> 11.2.0
>  > development builds.  Both are also running Puppy
> linux from an
>  > SDcard.  The data record re-calibration events
> and in most cases
>  > CPU load and memory use, in 5 or 7 minutes intervals.
> 
>  > 
>  > When running puppylinux there are also some logs
> where the touchpad
>  > is powered cycled in 5 or 7 minutes intervals during
> a sudo
>  > anti-RSI step that appears to improve touchpad
> behavior [1].
>  > 
>  > Unfortunately the data are not processed in any
> meaningful way but
>  > the frequency and the extend of the events is
> evident.  Maybe it
> 
> i guess i don't see it.  just skimming your logs, i'm
> not having
> trends jump out at me.  

The sense of "improvement" is still mostly subjective from my experience as a user. 
Actually interpreting the data is a bit tricky because there is no good way to detect user activity. 
CPU load is a good indicator of *total* inactivity but a lot of times the XO may be downloading a new build or olpc-update with no user input or just used for reading with minimal touchpad input (at least in my case).
In other instances when there is intense touchpad use, it may go nuts in under a minute of use.
 
If there was a way to monitor overall touchpad use, the results would be much more relevant. 
In the current form they are only good for extensive time chunks that should even out use variations, and in a comparative rather than absolute manner.
So there are not really to quantify touchpad misbehavior is absolute terms but rather to illustrate the problem.