Welcome to the new version of European Tribune. It's just a new layout, so everything should work as before - please report bugs here.
Display:
Most researchers use a numerical platform but many write lots of unique code to model similar problems by varying methods. I only see the normal process of tool development by research teams with different approaches, which seems to be working pretty well. Having 4 independent estimates of the evolution of global temperature is a pretty good idea.
by Fete des fous on Fri Aug 17th, 2007 at 05:20:44 PM EST
[ Parent ]
Different methods are not what we're talking about here.

I agree that varying approaches leading to different methods and software are of obvious interest, but so are careful scrutiny and tweakings on an existing software and at a much lower cost if the original research source code is available.

The original source code being not available, there is a high cost in doing the above, and so it's not done, all for bad reasons.

by Laurent GUERBY on Fri Aug 17th, 2007 at 05:44:07 PM EST
[ Parent ]
You fail to explain how a very unusual demand for a minor aspect of the research permits you to draw sweeping conclusion about code development and sound science. We shouldn't lose sight of the fact that if not for McIntyre's antics, it is highly unlikely that anyone would have ever wanted to "audit" what is apparently 2 pages worth of arithmetics in Matlab.
by Fete des fous on Sat Aug 18th, 2007 at 01:00:57 AM EST
[ Parent ]

Display:

Top Diaries

Pentecost steam

by DoDo - May 20
11 comments

A Nomad's Life (A Farewell)

by Nomad - May 10
14 comments

Simple Solar Principles

by gmoke - May 17
2 comments

Rail News Blogging #24

by DoDo - May 12
11 comments

Ferguson hates on Keynes

by Migeru - May 6
100 comments

Occasional Series