From: Dave Fisher (dfisher_at_pophost.com)
Date: 2002-07-09 03:27:13
On Sun, 7 Jul 2002 00:50:36 -0400 (EDT), Michal Wallace wrote:
>>dave seems to be working on something:
>>http://groups.yahoo.com/group/buildcheapeeg/message/2768?expand=1
LOL! Yeah, I'm getting that 'something' done, just not as quickly as I would
like. :) I will be checking the work that I have done into the repository this
week, or early next week at the latest. My web site is a bit out-of-date
concerning the most current status of the LinuxEEG library, but a basic
overview can be found here:
http://www.psychosensory.com/biofeedback-library.html. A *fairly* recent UML
class diagram can be found here:
http://www.psychosensory.com/linuxeeg/linuxeeg.gif. The one thing that I have
not adhered to is the internationalization (L10N) aspect of the design, but
that can still be retroactively accomplished. (Oh, and I won't be calling this
LinuxEEG anymore, either. :)
My focus is going through a bit of an evolution, also. I spent a week with Dr.
Jeffery Thompson late last year, and am going to be doing work with
psychoacoustics and heart rate. I will also still be working with brainwave
patterns, but for the moment, that is secondary. What is important is that my
most immediate direction is clarifying. (Finally! :) In October I will be
going back to school full-time for my licensing in massage therapy, which I
will complete in April, 2003. Before then I would like to finish (or at least
have functionally done) a working version of the biofeedback library as well as
the sound component (used for feedback).
>* software
>dave points out the cross platform commonc++ stuff:
>http://groups.yahoo.com/group/buildcheapeeg/message/2244?expand=1
>http://sourceforge.net/projects/cplusplus
Actually, Andreas convinced me off-list that the Common C++ may not be the best
choice of support libraries. He tried to compile it under MS Windows, and the
last couple of versions generated several errors. Combine that with me losing
two days of work as I tracked down a bug that took me deep into the Common C++
internals of their thread implementation, and I was only too glad to agree to
leave this library behind. I have been removing any dependencies on this
library from the code the past couple of weeks. I have now written my own
Serial class (which gives us more control over that process, anyway), replaced
the threading and logging functions with counterparts from wxWindows, and am
now focusing on replacing the last couple of sections with non-Common C++
alternatives (configuration file parsing and command line processing).
Right now the library is fully functional in terms of interacting with the
biodevice hardware (Thought Technologies ProComp+ in my situation). But other
devices can be added. Tomorrow I will post a log of the output demostrating
the library's interaction with the device. This will help illustrate the way
in which objects are used to manage the data flows. What I do not have is a
functional GUI interface. I have one started using wxWindows, but it will take
work to flesh out.
When I check this into CVS, I'll put the biofeedback SDK into one branch of the
repository (which can be called OpenEEGlib, thanks Andreas :), and the
test/monitoring program into another (which can be called something like
BioTest or BioMonitor). In this way we can work with a tangible implementation
of an SDK, continue development work on that, and eventually have several
"spin-off" applications. Those applications can be part of OpenEEG or not, as
the most important part is the foundational OpenEEG SDK. This is an important
distinction (SDK vs. application) since the exact nature of the application for
the OpenEEG group is still open to definition.
Also, I want to thank both Jim Peters and Andreas for their input and
suggestions as I fleshed out design and implementation issues, both on and off
list.
Looking forward to good things to come!
Dave.
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:44 BST
