[E-voting] End-to-end validation (side-issue)

Dr J Pelan J.Pelan at gatsby.ucl.ac.uk
Wed Mar 10 13:28:00 GMT 2004


Does anyone feel that the lack of an *end-to-end* validation test is worth
raising as a side-issue (*) ? It might be dangerous ground for the VVAT
people but perhaps some group other than ICTE might benefit from it.

The Minister claims that 'the integrity of the new electronic system has
been vigorously tested by six independent, internationally accredited test
institutes'.

Correct me if I am wrong but each of these tests is for a *sub-component*
of the system, e.g. voting machine, STV back-end etc. The *system* has NOT
been tested by any such institute, correct ? So the Minister's claims are
false or at least misleading and should be retracted, correct ?

Any fule kno that you can test individual components but when you put them
altogether you can have a system that doesn't work. The most publicised of
such cases recently that comes to mind is the Mars Climate Orbiter
disaster. Lots of very expensive, well-tested modules all supposed to link
together except one of them used metric units and another imperial.

--
John P.

(*) Of course, the only way that would work would be with a paper-trail.





More information about the E-voting mailing list