From: Doug Sutherland (wearable_at_earthlink.net)
Date: 2002-02-05 16:57:18
John,
> i put the idea forward as an alternative to paste as someone said it was
messy.
Don't take it the wrong way, all ideas are useful, and I have
shared plenty of my own. The point was that the electrode
is so critical, because we are measuring miniscule amounts,
that the trade-off in going away from traditional paste/prep
will probably be noise. I tried some flextrodes which are
pasteless, but I got better results with paste. It seems that
the active electrodes, with amplification at the source, may
be a way to avoid the goo, but I think we should get the
basics working first.
Aside from software engineering, I have done a fair
amount of systems analysis and design and project mgmt.
Some would say that programming is the easy part. It is a
challenge to get really good specs hammered out. I once
designed a system to automate work for a dept that had
never used computers, it was challenging! Trying to get
even objectives nailed down was hard. Trying to get the
detail on how they actually do their work was like pulling
teeth. That is why I say that functional specs are a good
way to think about this EEG software problem. Best case
scenario would be if we could pick the brain of an NF
practitioner. Next best scenario would be to look at some
existing software. The FFT + filtering is one thing, but
getting the thresholds/configuration right is another.
Namaste,
Doug
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:38 BST