[game_preservation] The role of comment/opinion on referencing game history articles/things

[Andrew Armstrong]

On 12/02/2010 08:36, Martin Goldberg wrote:

> Commentary should probably not be subjective in nature in a scholarly

> database.  I.E. if it's known for something, or the author has a

> conflict of interest, etc., you'd want to reference (i.e. create a

> link to) a reliable/verifiable third party source.  Things in regards

> to incorrect info can certainly be noted via an errata tag of some

> sort.  But with some sources you're going to be dealing a rather large

> listing of them (such as Steve Kent's book History of Video Games).

>    

Martin, I'm not too sure where the boundaries lie (thus the original 
post!) between basically what amounts to a library database (where to 
get a text named "XXX" which is in library category "YYY"), and a 
completely user-submitted system (not sure of a great comparison here, 
the old school "lists of links" on search engines might be one). You 
imply scholarly to be in the middle? I presume you'd want such a 
database to have more then ISBN and ISSN numbers or URL's with titles, 
these additional comments/notes on accuracy are part of that I guess - 
so perhaps it'd not be a wholly scholarly database (at some point it 
might well have original articles - if SIG members want to write some in 
the future and don't have anywhere to release them, why not?).

Basically, what if the information was submitted by good sources? Are we 
a reliable source to basically say "This is wrong about this". I'd like 
to think that was possible - only because we'd reference who wrote the 
comment (so they can be tracked to explain it), and explain in some 
detail, yes, with additional references if needed, why it was the case 
this is wrong.

Your Steve Kent example was the one I was thinking of from previous 
discussions - since I hadn't read it (still need to actually) I don't 
know of and probably still will miss references to incorrect facts and 
so forth. I wonder if anyone has done a proper article on all the 
inaccuracies of any kind of book on videogames, but I've not seen them 
if they have! (thus nothing to reference easily as you said).

Also, this brings up a good question of how reliable does a reference 
have to be to be a reference in proving something was right or wrong? Do 
we simply have to have X number of places repeat a fact for it to be 
true? (and not simple things like dates, but more complex facts like 
people's creative involvement in projects). I hope the database will 
hold things like sales figures - if possible to obtain original sources 
- and other simple facts if it was obtainable, but I don't think I know 
anywhere which lists such data en masse yet, which is a bit of a pain if 
you are sure such sale figures are wrong, or dates of release are off.

Urg, I'm getting a headache thinking of the circular nature of 
references now, hahaha :)

If nothing else, eventually I'll get a prototype working and by that 
point I hope it'd be easier to point out how to perform such tasks 
practically rather then theoretically. Since it'd be online it allows it 
to change in any case - perhaps worth trying one method (your suggestion 
of using total references for inaccuracies) then trying or thinking 
about other more complex ones later.

Thanks for the consideration, made me think a lot :)

Andrew