nodalpoint.org - Book Reviews - Comments

For non-fans of soccer...

Neil — Mon, 12 Jun 2006 21:08:10 -0400

I enjoyed the famous curves. If soccer* is really not your thing, I recommend this YouTube video (via BoingBoing).

* since moving to Australia, I've slipped into the habit of calling football "soccer", as "football" here means "any game where you kick a ball unless it's spherical".

Life in science

Neil — Wed, 08 Mar 2006 03:39:25 -0500

I've always thought that the lives of working scientists could make great material for a novel or film. Not in the sci-fi tradition, but with the focus on the people - their lives, feelings, problems. Yes, scientists have personal lives too! This is often forgotten, I think.

Anyone recommend a novel or movie that they feel comes close to an accurate portrayal of the "tortured complex scientist"? Curiously, one of my favourites is a cult B-movie titled "Darkman", about a guy who invents artificial skin.

The slow pace of our own work

PedroBeltrao — Mon, 06 Mar 2006 05:39:44 -0500

Probably most of us have talked about this with friends and colleagues. One of the most frustrating things about working in science is how slow our own work moves. Specially when compared to the pace of science as seen in the journals every week. This is for me one of the big challenges, to keep interested with the work that I am doing even if moves along so slowly. It is so much easier to just start something new after getting all excited with something we just saw somewhere recently. This is specially true for me when trying to finish a project, to continue working after the initial effort to establish if the idea has potential or not.

XML Databases

dopey — Tue, 17 Jun 2003 11:24:17 -0400

> .. there are also XML databases
One that I've heard of, which is supposedly quite good (however, it was one of the authors talking about it, so YYMV) is SODA , from UNSW XML Database Group

Visualizing Networks

rainer — Sat, 14 Jun 2003 05:38:23 -0400

I also like Linked. The (not uninteresting) danger with such books is that inevitably the author's personality and political views and also the field's political "surrounding" somehow creep in. Scientific authors often (unconsciously?) introduce philosophical and political aspects without reflecting on them appropriately (which they would with scientific aspects). Doesn't the "the rich get richer" algorithm imply that social order arises from a free market?

Pathway DBs: I didn't use them computationally yet. But I agree with Neil in his 3 points of criticisms. I personally like EcoCyc and aMAZE best, but only judging from brief looks and the paper of PD Karp (EcoCyc) and a talk at the ICSB 2001 from Shoshana Wodak (aMAZE). After this talk the discussion was mainly about including stoichiometry, reaction rates etc. into the pathway DBs, so they would allow to retrieve mathematical models directly.
Not only storing interactions but also linking to different kinds of data (also sequence and structure data) will make such DBs a powerful tool for biological knowledge management. They can represent our current understanding of biological systems and thus should be able integrate all data that led to and underlie this knowledge: visual displays of interaction networks (and causal relations) as graphs (or even animations of simulation results?), were nodes and edges are hyperlinked to data (and editing) interfaces on multiple hierarchy levels; and of course computational interfaces to retrieve large datasets of a specific context.
Pathway DBs could naturally reproduce what has been worked on a lot in different fields: knowledge maps etc.
The Pubmed Browser is an intersting example were graphs can be used.

Combining this with Blog-like knowledge exchange and discussion (and publishing?) of pathways and models could be an interesting future direction to go for projects like nodalpoint.

No, but I'm interested in your experiences...

Greg — Sat, 14 Jun 2003 00:37:00 -0400

In a similar vein to Barabasi's Linked is Mark Buchanan's Ubiquity which I would also recommend. This was my first introduction to the concepts of scale free networks.

WRT relational databases vs. object oriented databases, I'm familiar with the usual RDBMS suspects e.g. MySQL. However I have only ever used Zope DB object store. And I'm not sure if this is the same as an OODBMS. I'm none the less interested in hearing about your future experiments with OODBMS systems :)

Interesting

Neil — Wed, 11 Jun 2003 23:37:44 -0400

Can I be of no use whatsoever and say no, I know nothing of OODBMS either. I'm a big fan of MySQL, but if I had the time I'd like to look at postgresql and I believe (Greg?) there are also XML databases.
The world definitely needs a much better pathway database - I hope it's yours. In my opinion KEGG is about the pick of the bunch at the moment, but a lot of its emphasis is on graphical views. I find current pathway databases have 3 major problems (1) they are inadequately cross-referenced to other major databases, (2) a lack of methods for analysis or backend files that you can download and play with, (3) the frontends are frequently horrible to look at and it's not clear what you are actually able to do at the site. Those are things I'd like to see improved.