On Thu, Mar 15, 2012 at 9:18 PM, Patrick O'Neill
<synapse.and.syntax@gmail.com> wrote:
> The topic of scientific robots is a popular one on this list, but I'm
> more interested in the issue of workflow planning: can you take a
> natural language command like "amplify insulin" and translate that
> automatically into a detailed protocol?
you might be interested in reading some stuff from the archives:
http://groups.google.com/group/diybio/browse_thread/thread/19df187ef1014735
http://news.ycombinator.com/item?id=2191105
http://88proof.com/synthetic_biology/blog/archives/290
http://groups.google.com/group/openmanufacturing/browse_frm/thread/a8d8ee245aaae97d#
http://groups.google.com/group/openmanufacturing/browse_frm/thread/b2e21ccc953d6328#
The language technology is easy. What I would like is to then
automatically generate a microfluidic lab on a chip that implements
the protocol, or generate beautiful documentation for the workflow
that I could give to a team of people (or, er, just myself). But most
people just use gradstudents, yes...
> It sounds like an impossible problem, but there appears to have been
doesn't sound impossible
> What ever happened to this stuff? The first generation of programs
> was hampered by the problem of large knowledge-bases, which restricted
> them to trivial experiments. But dealing with large knowledge-bases
> is a problem we've gotten a little better at in the last thirty years.
problems with knowledge-based stuff:
1) figuring out a good format/represetation
2) collecting it without killing ourselves in the process
3) finding something useful to do with it (sometimes this can pay for
the other two steps)
> Is anyone familiar with more recent efforts at this problem? Does
> anyone use anything that sounds vaguely like what I described?
biocoder ain't it
- Bryan
http://heybryan.org/
1 512 203 0507
irc.freenode.net ##hplusroadmap
--
You received this message because you are subscribed to the Google Groups "DIYbio" group.
To post to this group, send email to diybio@googlegroups.com.
To unsubscribe from this group, send email to diybio+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/diybio?hl=en.
0 comments:
Post a Comment