[Foundation-l] Ensuring veracity of articles based on print sources

[Jonathan Leybovich]

Andrew Gray wrote:
> 
> 
> My position is that the proposal originally suggested in this thread -
> of confirming the existence of books so as to deal with bad-faith fake
> sources - just won't work, because it means a good deal of work but is
> trivially easy for the people who we assume are trying to fool us to
> keep fooling us. We need fact checking. But having a system that
> sounds like fact checking and looks like fact checking but doesn't
> work is a net detriment.
> 

Almost every system of security relies on countermeasures that are- not 
invulnerable- but only cost more for the attacker to breach than are 
worth the effort of breaking.  Outrageous claims that contradict common 
knowledge or sense are going to be scrutinized and weeded out anyway, 
source or no source.  And so what we really have here is the possibility 
of bad-faith editors introducing fabricated citations to support, at 
worst, moderately false assertions.  There are various levels of 
automatic validation that could be done at the time the citation is 
created using data from open bibliographic databases- from checking 
whether the book is of an appropriate subject and that the page cited 
falls within the book's pagination range to more involved CAPTCHA-like 
challenges such as: Where was the book published? and: What is the 
highest Roman numeral used in the preface?  This is only moderately 
easier for an owner of the source to verify (and indeed, bad-faith 
owners could own the source in question anyway), but it quickly becomes 
for trouble than it's worth for the citation creator if their only 
motivation is mischief.  Before introducing countermeasures we need to 
determine where that threshold actually lies and so avoid introducing 
more hoops for good-faith editors to jump through than are absolutely 
necessary.