Wikipedia talk:Improving referencing efforts/2007

Comments[edit]

Very interesting and I generally concur. However, I would suggest this should be tagged {{essay}}. Addhoc 15:54, 1 April 2007 (UTC)[reply]

Could you elaborate on why you think that. Essays are to share your views on Wikipedia, while this is a proposal that is trying change a process without altering policy. This change will only occur with large community input because I can't do it all myself. I have a lack of technical bot knowledge to start with. Mgm|^(talk) 16:50, 1 April 2007 (UTC)[reply]

Ok, firstly, let me clarify that I'm only suggesting changing the tag. Basically you are saying "we should increase the efforts to provide references for all the material in Wikipedia and use technical tools to our advantage to reach this goal" and I'm not certain that constitutes a proposal for a guideline. Possibly a new wikiproject? Addhoc 17:16, 1 April 2007 (UTC)[reply]

Stupid me. I forgot to read the template in detail to see if it applied to the situation. I've "subst'd" it and altered the text. I do not want to create a new project. I want to improve the effectiveness of the existing one. - Mgm|^(talk) 17:54, 1 April 2007 (UTC)[reply]

A very similar focus[edit]

I started up a discussion about verifying sources here and did not end with a "thumbs-up" to continue my goal of a sub-project on verifying and attributing articles. I wonder how the opinions in this discussion will affect this proposed WIkiProject. Sr13 (T|C) 02:19, 2 April 2007 (UTC)[reply]

Add this and this to the list. Sr13 (T|C) 02:27, 2 April 2007 (UTC)[reply]

No RefBot for you[edit]

User:RefBot has been blocked. You'll get no help from that. (SEWilco 03:51, 2 April 2007 (UTC))[reply]

List of Bad Sources[edit]

Would it be feasible to begin to create a list of simply bad sources? I have noticed that articles are being referenced more, but the references themselves are to partisan websites and self-published material. Such a list should err on the side of caution, but to create a forum for the discussion of the scholarly merits of material and then to use this list to weed out bad sources may provide a little automated support for an overworked community. It's hard enough trying to reference facts, trying to find out if references are actually legitimate and not just to some little propaganda site that no one has heard seems even worse. Elijahmeeks 08:05, 8 April 2007 (UTC)[reply]

I'm not sure that a list would be helpful - it would be practically infinite. I think that WP:RS needs to be more explicit about what constitutes a reliable source. I have been working on a "source" essay that I was going to attach to my user page because I am always explaining why autobiographies are unreliable or why popular histories are less reliable than scholarly histories, etc. There needs to be a page explaining why scholarly sources are more reliable in detail; it should also explain what kinds of information can be taken from what kinds of sources (for example, why relying on a corporation for information about itself or its products is naive at best). Awadewit 07:43, 9 April 2007 (UTC)[reply]
- Wikipedia, though, is practically infinite, and it seems to be lurching along. The problem we have--and this is not a Wikipedia problem but a social epistemological problem--is that the number of scholarly sources is also near-infinite. My worry is not with the people who don't understand reliable sources, but rather the people who ideologically oppose them, for whom education will not stop their poor sourcing. Also, given the vast size and activity on Wikipedia, we need to start implementing automated procedures to prevent editors from appearing to create a well-sourced and reliable article that actually links to several dozen crackpot-fringe websites. Elijahmeeks 13:56, 9 April 2007 (UTC)[reply]
- Wikipedia:Autobiography and User:Uncle G/On sources and content#Evaluating sources already cover some of the ground that you want to cover. Uncle G 15:17, 9 April 2007 (UTC)[reply]
  - Thank you for those links; the autobiography page covers a slightly different issue - it focuses on writing an autobiography of oneself. I ran into trouble with editors using autobiographies as the primary source for a historical personage (see the archived FAC of Ronald Reagan). Also, I find the "Evaluating sources" section of the other page clear but not detailed enough. The editors that I end up having source debates with would need much more explanation than is on that page. The distinction between primary and secondary needs to be made much clearer, apparently, for some people (see Wikipedia:Featured article candidates/Grand Duchess Maria Nikolaevna of Russia, for example) as well as the idea of checking book reviews of popular histories and biographies in academic journals (see Talk:Cicero, for example). I don't think that any of these examples are malicious misusing of sources or crackpots; I just think that they are people who honestly don't quite understand how to do research. Awadewit 18:43, 9 April 2007 (UTC)[reply]

The quality of a source is relative. an article on a national event sourced entirely from local newspapers would look odd; an article on a major politician sourced entirely on even national newspaper articles also would be. An article on Reagan based primarily on his autobio would likely by POV, but for less notable figures an autobiography might be the main source of information for details. We use the best we have in view of what the subject is.

In the present state of the world, the number of WP editors actually able to check in academic sources is probably less than half, and the number who will go the the trouble on most topics is much smaller. Correcting this implies that those of us who can and do will spend a major amount of time helping out the others, and I think it simply isn't practical. an encyclopedia written by amateurs will have amateur sourcing. Now, I have a POV: I am an academic librarian and think that everyone ought to try to work to a much higher standard; but I have been an academic librarian for quite a while now, and I know the likelihood of that goal. We may be able to improve things a little, but only if we ask for a small and reasonable amount of improvement. DGG 01:54, 12 April 2007 (UTC)[reply]

Well, I think that at least FAs should require scholarly sources (or the closest available for their respective topics); if these are the "best" wikipedia has to offer, their research should reflect that standard as well. Also, I would argue that the sourcing policies should encourage people to look for the best sources. Currently, they do not really outline what those are in any great detail, so someone who doesn't know the difference between a peer-reviewed work and a non-peer-reviewed work will not learn much from the policies. I do not intend to read sources for other articles - I intend to educate others on how to find and evaluate sources. In the end, this is the only way to fix this problem. The few of us who know how to do real research have to teach others - we cannot do all of the research ourselves! (By the way, I feel that if one is writing an article of an obscure figure based entirely on an autobiography, this should be prominently announced somewhere on the page. This is what I have done on Sarah Trimmer - I have placed a note at the head of the "Bibliography" section explaining that all of the details of Trimmer's life stem from her journal and the biography of her attached to it (written by who knows who). Unfortunately, not everyone is this careful.) Awadewit 02:47, 12 April 2007 (UTC)[reply]

I like your suggestions. It would be a very good idea to start with the limited number of FAs. obviously some will be on topics without academic sources, for for the ones that are, they should be added. This will teach people by example, the most effective way--especially if this is pointed out when adding them. This will, of course, take people who can do the necessary working watching over FA candidates and challenging those that are lacking in this respect. It will also probably take several such people together to insist that articles not be promoted otherwise. (I have been mainly working on sourcing at the opposite end of the scale, the AfD nominations, but I'll give FA a try.) :Indicating the limits of sourcing seems such an obvious step, and I never thought of it. I indicate the limits of a search in the real world, but I never thought of it here. A good idea, and not just for autobio. eg., "only web sources have been used, and others may not be practical for this topic" vs. "only web sources have been used, and help is needed finding others." , "only abstracts have been examined" , etc. Some good templates might help. I've haven't done that before either. DGG 06:38, 14 April 2007 (UTC)[reply]

We can do more with automated software, though. Rating sources, excluding particularly bad sources (such as generically web-hosted sites like tripod, which by definition carry the personal theories of its owner and not any kind of scholarly work at all) and even giving short summaries of sources (Or simply links to a review on Amazon), so that someone else doesn't have to wade through a website or book to find out the salient details related to its accuracy. Education is valuable, but we shouldn't forget that Wikipedia has some technical structures that you can also use to help implement these kinds of initiatives.Elijahmeeks 13:42, 14 April 2007 (UTC)[reply]

Linking to Amazon would not seem the way to start; their reviews are written by anyone who might happen to drop by. I'd even consider them as one of the bad sources that should not be used. There's been a long fairly successful battle to replace links to amazon with ISBNs, which are set to link to a number of dealers. One of the problems about doing this in an automated way is getting agreement on things like this. There is for example the example of IMDB, which is by far the most convenient source for details of obscure films, but which accepts content from anybody. It's used for lack of better, but should never be used to establish notability. DGG 10:21, 15 April 2007 (UTC)[reply]

I agree. Even the so-called "Editorial Reviews" which are supposed to be written by professional reviewers are often no more than puff pieces for the book written by publishers. Awadewit 15:21, 15 April 2007 (UTC)[reply]

On the other hand, sometimes Amazon might be the source of a particular piece of info no one else has like the name of an old publisher. A source is almost never fully bad, it depends on the kind of info you want to reference with it... - Mgm|^(talk) 11:16, 5 November 2008 (UTC)[reply]

Improved technology for recognizing editors who reference[edit]

It sometimes seems to me as if edits that include references are largely unappreciated. Would a tool that examined change histories finding edits that add <ref> tags or {{cite}} templates and then listed editors with high-percentages be a way to start identifying editors who are making a positive difference in this regard? Sdsds 06:22, 14 April 2007 (UTC)[reply]

Maybe, but at present it's hard to access that information. There is a data dump format which provides all edit history for an article but it is presently disabled for performance reasons. There might be other ways, such as monitoring recent changes to spot editors making such changes, then examining the Contributions of such users. (SEWilco 03:28, 20 April 2007 (UTC))[reply]

An alternative way of working[edit]

In y experience trying to find references to a particular article is a relatively inefficient way of sourcing. We use it because that's what we learned to do in school. it can be much more effective to take an issue of a reliable publication in a field of interest, or a comprehensive interesting useful and reliable book, and see what articles the material there supports. This is already the method used for sourcing news items, and has been effectively used for groups of articles. Adding the appropriate references from an issue of say, the WSJ, can help many articles. (and also suggest new ones) DGG 00:30, 19 April 2007 (UTC)[reply]

I do that when I work through a book which has source material. Sometimes it serves as an additional source, sometimes a whole article falls out. (SEWilco 00:47, 19 April 2007 (UTC))[reply]

Since this would mean the user spends less time searching for information and more time on editing, I think that is a good idea. - Mgm|^(talk) 09:52, 19 April 2007 (UTC)[reply]

It depends on what one is doing. When writing for a particular article, effort goes to finding relevant source material. When starting with a particular source, effort goes to finding articles which can use the material at hand. Then complications ensue. Such as when using one public domain history book, for one topic I found several Wikipedia articles had bits of relevant information but the topic was not woven together. Rather than referencing the source, I ended up copying a section from it to create a new article on the topic, then absorbing misplaced pieces from several existing articles. You pick up a book and you see what you've got. Now I have a history book with a bunch of black and white maps, next is a tree reference book. (SEWilco 03:22, 20 April 2007 (UTC))[reply]

Page numbers[edit]

Inline references to a book are good, but what if the book is 700 pages long and does not have much of an index? I can use ref name= to have multiple references to the same book in a given article, but how can I direct the reader to the correct page? Edison 03:25, 24 April 2007 (UTC)[reply]

Three ways. One is simply to write out the references each time including the appropriate page number instead of using the name= shortcut; this is illustrated in one of the examples for the template. It's not as difficult to do as it sounds, for you can copy and paste, but it looks awful in the wikitext and not too good in the reference list. The best in my opinion is to use Harvard referencing, a method designed for scholarly purposes. However, the formats shouldn't be mixed in an article, so using the Harvard system means redoing all existing references.

An easy practical method is to put the page number in the text after the references as Cite error: There are <ref> tags on this page without content in them (see the help page). (p. 699) .DGG 06:31, 24 April 2007 (UTC)[reply]