On 15/05/07, David Gerard <dgerard(a)gmail.com> wrote: Oh, I left a pretty pointy message on both his talk page and that of the editor he reverted. Reference stripping is unacceptable behaviour and attracts strong penalties for good reason. - d.

The objection of it being a subscription site is of course wrong, but it seems a little more complicated: Some or all of the links reverted were to a dead site. This is a different problem. I really do not see a how a link to a site that cannot be reached by anyone at all is a reference. The only way to go would be to find it in a old file somewhere--or to have made a permanent link by one of the available methods in the first place, or to have an equivalent print link. This is one of the known hazards of using purely online references. DGG On 5/15/07, John Lee <johnleemk(a)gmail.com> wrote: -- David Goodman, Ph.D, M.L.S.

But I could put in a link to a dead site and use it to reference anything at all, true or false, and nobody would ever be able to dispute it. I think the reasonable thing to do is to leave the link in the wikitext, and comment it out with an explanation. Out of print books, even the most esoteric, are still in libraries. Unless dead links are in an internet archive somewhere, they are nowhere at all. DGG On 5/15/07, John Lee <johnleemk(a)gmail.com> wrote: -- David Goodman, Ph.D, M.L.S.

Ah, but you can;t make up book references safely. I can go to a research library and find a copy, or -- if it really makes a difference-- I can ask you for a scan. You can pick the most obscure book, and someone at WP will be there. ( This question has come up before in the context of using unique manuscripts as sources, and I think the potential availability was given as a justification for using them if actually necessary. Where this has also come up is in the use of master's theses for local history--typically they are not on the web, but as a unique copy in the library of the university where the degree was given. But there is likely to be another person at whatever state university it is, and these can if it is actually important enough to go the trouble be ordered as microfilm from the university, if not available in any other way. DGG On 5/15/07, David Gerard <dgerard(a)gmail.com> wrote: -- David Goodman, Ph.D, M.L.S.

On 5/16/07, David Goodman <dgoodmanny(a)gmail.com> wrote: Do we agree that: a) A non-subscription reference is better than an equivalent subscription reference b) A subscription reference is better than no reference Also that: c) A dead link is better than no reference unless there is no hope whatsoever that the link could be used to retrieve the source in the future. The problem arose at [[Smile (band)]] where an excellent source just disappeared off the web. Archive.org had some of the text, but was missing the scan of a crucial letter-to-the-editor. I wanted to keep the link on the basis that it might eventually come back, or serve as some kind of trail to find the thing again. Not sure whether someone else killed it. Steve

On 5/16/07, MacGyverMagic/Mgm <macgyvermagic(a)gmail.com> wrote: Well, as I put it on the talk page after I reverted, meatspace newspapers require subscriptions, but we don't ban them as references. Treat a web source which requires subscription as a meatspace source and you'll see how ludicrous it is to assume that subscription web sources are unverifiable and inaccessible. (I just remembered that {{cite news}} even has a specific option for mentioning that the source is subscription-only.) This reminds me a bit of the time when a Singaporean editor said my Malaysian newspapers weren't verifiable for him because they are censored in Singapore. Johnleemk