[e2e] was double blind, now reproduceable results

Jon Crowcroft Jon.Crowcroft at cl.cam.ac.uk
Wed May 19 22:44:37 PDT 2004


I am not sure I want people to give away everythign to everyone - however, there is a strong
precendent for allowing a program committee or esitor to see evidence behind a paper even when
it isnt published - for example stuff that is under patent application

look, if the medical world, with all the finanical and ethical pressures on them
can manage with this problemm (esp. viz genomics, protemics, patent fights, patient privacy, and
pharma indutry), then we are pretty sad if we can't figure it out

as i said, my goal is not to force people who want to publish to give away all the state secrets

it is to engender a culture of openness as far as possible, which encourages people to think in
terms of offering results and methods in a way that others can verify so far as possible - a
side effect is to put a barrier in place to publuishing work that couldnt in principle meet such
a requirement (people could put their secret data and methd in escrow somewhere so that it could
be re-run by others who are prepared say to sign appropriate NDA - such has been done with new
drug trial results where some patent application is underway)

hey if the internet is critical infrastructure, why is it any different from our health?


In missive <E1BQbE9-00047z-00 at smeg.dsg.stanford.edu>, Jonathan Stone typed:

 >>
 >>In message <6.0.1.1.2.20040519194147.05b61bc8 at 127.0.0.1>"David P. Reed" writes
 >>>At 04:11 PM 5/19/2004, Jon Crowcroft wrote:
 >>>>well this doesnt argue that we should publish part of the experiment,
 >>>>but even more strongly then that a unique observation should be
 >>>>captured more completely.
 >>
 >>>I don't disagree with making all the data available, and that is almost 
 >>>always possible, even where reproducibility is hard. 
 >>
 >>With the notable exception of whole-payload packet traces.  If you can
 >>get permission to collect them at all, its well-nigh impossible to
 >>redistribute them. For obvious and excellent reasons.

 cheers

   jon



More information about the end2end-interest mailing list