"Thomas Dalton" wrote
But Cunc dealt with this. 'Verifiable' is an in-principle thing. It is distinct from saying everyone can do homebrew fact-checking on anything mentioned. Verifiability on Wikipedia can't simply be a sceptics' charter: that really would be a problem.
Verifiable means we can actually verify it to be true. Not in principle, but in practise. The whole point of verifiable sources is that we can be sure we don't have things stated on Wikipedia that aren't true - some vague kind of hypothetical verifiability doesn't help that.
No, that's not what we mean, and has never been what we mean. And you seem to be conflating things. We don't talk (much) about "verifiable sources": we base everything on "reliable sources". We talk about verifiable edits, really: we want to constrain editors into only adding material that is verifiable, from reliable sources, accessible to them. Some of the most reliable sources, in the scholarly sense, are some of the least accessible to the general public. (And, frankly, reliability of newspaper reports can vary inversely with circulation. And scholarly monographs with the best information on particular matters are apparently now printed in runs as low as 300.)
So when you say "Verifiable means we can actually verify it to be true", that is not the kind of statement on which so much can be built.
Charles
----------------------------------------- Email sent from www.ntlworld.com Virus-checked using McAfee(R) Software Visit www.ntlworld.com/security for more information
We talk about both verifiability and reliability.
You are not the end authority on Wikipedia.
On 12/8/06, charles.r.matthews@ntlworld.com charles.r.matthews@ntlworld.com wrote:
"Thomas Dalton" wrote
But Cunc dealt with this. 'Verifiable' is an in-principle thing. It is
distinct from saying everyone can do homebrew fact-checking on anything mentioned. Verifiability on Wikipedia can't simply be a sceptics' charter: that really would be a problem.
Verifiable means we can actually verify it to be true. Not in principle, but in practise. The whole point of verifiable sources is that we can be sure we don't have things stated on Wikipedia that aren't true - some vague kind of hypothetical verifiability doesn't help that.
No, that's not what we mean, and has never been what we mean. And you seem to be conflating things. We don't talk (much) about "verifiable sources": we base everything on "reliable sources". We talk about verifiable edits, really: we want to constrain editors into only adding material that is verifiable, from reliable sources, accessible to them. Some of the most reliable sources, in the scholarly sense, are some of the least accessible to the general public. (And, frankly, reliability of newspaper reports can vary inversely with circulation. And scholarly monographs with the best information on particular matters are apparently now printed in runs as low as 300.)
So when you say "Verifiable means we can actually verify it to be true", that is not the kind of statement on which so much can be built.
Charles
Email sent from www.ntlworld.com Virus-checked using McAfee(R) Software Visit www.ntlworld.com/security for more information
WikiEN-l mailing list WikiEN-l@Wikipedia.org To unsubscribe from this mailing list, visit: http://mail.wikipedia.org/mailman/listinfo/wikien-l
No, that's not what we mean, and has never been what we mean. And you seem to be conflating things. We don't talk (much) about "verifiable sources": we base everything on "reliable sources". We talk about verifiable edits, really: we want to constrain editors into only adding material that is verifiable, from reliable sources, accessible to them. Some of the most reliable sources, in the scholarly sense, are some of the least accessible to the general public. (And, frankly, reliability of newspaper reports can vary inversely with circulation. And scholarly monographs with the best information on particular matters are apparently now printed in runs as low as 300.)
So when you say "Verifiable means we can actually verify it to be true", that is not the kind of statement on which so much can be built.
For an edit to be verifiable it has to be verifiable by someone other than the person that made the edit. We're not talking about reliability of sources - it's obvious that a TV show is a perfectly reliable primary source - we're talking about whether someone else can come along and check that what the original editor said is true (assuming the source is right - a wikipedia article can never be more reliable than the sources it uses). It's not necessary that everyone be able to verify it, but a reasonable number of reasonably unconnected people should be able to, otherwise we're open to any number of hoaxes.
For an edit to be verifiable it has to be verifiable by someone other than the person that made the edit. We're not talking about reliability of sources - it's obvious that a TV show is a perfectly reliable primary source - we're talking about whether someone else can come along and check that what the original editor said is true (assuming the source is right - a wikipedia article can never be more reliable than the sources it uses). It's not necessary that everyone be able to verify it, but a reasonable number of reasonably unconnected people should be able to, otherwise we're open to any number of hoaxes.
In fact, the policy is even stronger. [[Wikipedia:Verifiability]] says:
""Verifiable" in this context means that ***any reader*** should be able to check that material added to Wikipedia has already been published by a reliable source." (emphasis mine)
I'm not sure if I agree with that policy, but that's what the policy says.
"Any reader" cannot verify that Cmdr. Pickett gets eaten by a Florgbernian Rumpox in Stargate episode 23 if said Stargate episode isn't publicly available for viewing.
Thomas Dalton wrote:
In fact, the policy is even stronger. [[Wikipedia:Verifiability]] says:
""Verifiable" in this context means that ***any reader*** should be able to check that material added to Wikipedia has already been published by a reliable source." (emphasis mine)
I've always taken that just to mean that classified documents and the like are out of bounds. We don't require that our references be available as books on tape for the benefit of illiterate Wikipedians, or that references are only English-language because not every reader knows French - it might take some effort, but every reader can learn a foreign language or hire a translator.
Stan
On 12/8/06, Thomas Dalton thomas.dalton@gmail.com wrote:
For an edit to be verifiable it has to be verifiable by someone other than the person that made the edit. We're not talking about reliability of sources - it's obvious that a TV show is a perfectly reliable primary source - we're talking about whether someone else can come along and check that what the original editor said is true (assuming the source is right - a wikipedia article can never be more reliable than the sources it uses). It's not necessary that everyone be able to verify it, but a reasonable number of reasonably unconnected people should be able to, otherwise we're open to any number of hoaxes.
In fact, the policy is even stronger. [[Wikipedia:Verifiability]] says:
""Verifiable" in this context means that ***any reader*** should be able to check that material added to Wikipedia has already been published by a reliable source." (emphasis mine)
I'm not sure if I agree with that policy, but that's what the policy says.
I'm sure I don't agree with it. I worked on the verifiability policy a while ago to get it to a rational place. Clearly *any reader* can't check every source used on Wikipedia -- some people just don't have access to the Widener Library, for example, or read French.
"Any reader" cannot verify that Cmdr. Pickett gets eaten by a Florgbernian
Rumpox in Stargate episode 23 if said Stargate episode isn't publicly available for viewing.
We had been talking about commercially available, not publicly available. Those are two different things.
I'm sure I don't agree with it. I worked on the verifiability policy a while ago to get it to a rational place. Clearly *any reader* can't check every source used on Wikipedia -- some people just don't have access to the Widener Library, for example, or read French.
An excellent example. I think it basically boils down to how much work you need to put in to verifying it. As someone else said, it doesn't need to be easy to verify, but it does need to be possible. For some people it will take an unreasonable amount of effort, so it's not really verifiable for them, but as long as some people can verify it with a reasonable amount of effort, that's fine.
We should probably move this conversation to the policy talk page if we want to gather consensus to change that sentence.
"Any reader" cannot verify that Cmdr. Pickett gets eaten by a Florgbernian
Rumpox in Stargate episode 23 if said Stargate episode isn't publicly available for viewing.
We had been talking about commercially available, not publicly available. Those are two different things.
An unreleased episode is neither commercially available or publicly available. It's only available to those people that recorded it. Copyright law allows you to record things off the TV for personal viewing, it doesn't allow you to distribute it to others.
Thomas Dalton wrote:
No, that's not what we mean, and has never been what we mean. And you seem to be conflating things. We don't talk (much) about "verifiable sources": we base everything on "reliable sources". We talk about verifiable edits, really: we want to constrain editors into only adding material that is verifiable, from reliable sources, accessible to them. Some of the most reliable sources, in the scholarly sense, are some of the least accessible to the general public. (And, frankly, reliability of newspaper reports can vary inversely with circulation. And scholarly monographs with the best information on particular matters are apparently now printed in runs as low as 300.)
For an edit to be verifiable it has to be verifiable by someone other than the person that made the edit. We're not talking about reliability of sources - it's obvious that a TV show is a perfectly reliable primary source
I've seen interesting documentaries on sasquatch sightings. Area 51 and perpetual motion machines. :-)
I would venture to say that more than 300 people have recorded each and every Stark Trek episode. i.o.w. Stark Trek episodes are more verifiable than academic monographs. ;-)
Ec
Ray Saintonge wrote:
Thomas Dalton wrote:
For an edit to be verifiable it has to be verifiable by someone other than the person that made the edit. We're not talking about reliability of sources - it's obvious that a TV show is a perfectly reliable primary source
I've seen interesting documentaries on sasquatch sightings. Area 51 and perpetual motion machines. :-)
And they're perfectly reliable primary sources *about the documentaries*. About sasquatch and Area 51, on the other hand, they're secondary sources of perhaps more questionable reliability.
The "reliability" of a source is highly context-sensitive. I don't think anyone's proposing to use Star Trek episodes as sources for articles about real-world physics. :)
I've seen interesting documentaries on sasquatch sightings. Area 51 and perpetual motion machines. :-)
I said primary source - a documentary about area 51 is not a primary source about area 51. It is a primary source about the documentary itself, however, and in that respect is completely reliable. If the documentary says "XYZ" then you can be absolutely sure the documentary says "XYZ" - reliability is generally trivial for primary sources.