User:Sj/Modern primary sources

I recently ran across Ram Loevy, a rather comprehensive bio, which was criticized in its GA discussion for using direct primary sources (physical documents and email with the subject).

Gathering and publishing these sorts of sources in the context of relevant articles is a valuable service - moreso than the service provided by quickly researching a few articles that mention the topic and appear in a web search. Editors of the article may choose to use or not use information from those sources, based on changing criteria for authority and verifiability, but sharing the information is important.

That's true whatever the information is, whether it is
 * the geolocation or subject of a photograph (primary source: the photographer),
 * personal information about a BLP (source: its subject or an associate, often via OTRS; it's worth revisiting what our policies are there re: where we are only willing to remove questioned data, and when we are willing to replace it with 'correct' data as claimed by the person).
 * data about an institution or other group (source: an activist whose job is to expose its failings, or a staff member or participant)
 * data about an event (source: eye-witnesses, whether quoted in a news brief, on a blog, or elsewhere)

There's also other metadata relevant to the sources we use in articles. One can usually find a spectrum of secondary sources on a topic, which sometimes draw conflicting conclusions from the same primary source material. An important meta-discussion about an article -- one that should grow and mature over time, not fade into an archive as a series of unrelated talk-page sections -- is about the quality and balance of these sources, and separating primary sources from analysis.

Ideas for improving the status quo
Every topic about a current event, person, or group, is an opportunity to get those people engaged as contributors and participants, not enemies; but we need better processes for soliciting their input and treating them as part of our global community.

Some suggestions:


 * Create a "/References" subpage for each article, or similar namespace, where discussions about references or primary-source feedback exists. That would be the place for subjects of articles to leave information and suggested references to consider.  It could also be the place to post relevant information received by an editor by email.  [For optimal verifiability, a copy including the sender's email address should be sent to OTRS.]


 * Encourage article authors to send lists of questions to the groups and individuals they write about, and to post replies (in a section such as the above; as Q&A with the subjects) where allowed.


 * Write to people and groups that are notable (by our criteria) but have no article; invite them to submit material

Extending this idea beyond people and groups
I am thinking about how museum can better become part of the wikimedia movement; they have information about hundreds of thousands of notable works of art. Each work is listed on a few artist or collection pages here -- but almost none have their own article. The museums have metadata and authority documents indicating the provenance of each, why it was notable enough for them to curate it, descriptive text, and medium-resolution images. But they feel that our OR policies (their curators descriptions were once 'original descriptive work'!) and self-promotion guidelines (writing about our collections is writing aobut ourselves!) mean they are not allowed to publish such things. This is a shame, and a confusion that we should correct.

Even in areas where there may be some obscure conflict of interest, we should strongly encourage and praise contributions of this sort of information -- whether on an article page or as a very-relevant resource for future article editors.

What else can we do? Can we apply something similar to laws and policy issues as well? There are certainly groups dedicated to nothing but 'informing others of the law and policy development'; how do we ensure they feel comfortable assing what they know best to Wikipedia? –SJ + 00:33, 18 March 2010 (UTC)

Some Comments and a Suggestion
I think you are raising an important topic, especially when it comes to BLPs, a longtime problem in Wikipedia. I would contend that in most--but not all--instances, the subject of the article is an ideal source of information. I also believe that in many instances, the subject of the article is willing and even eager to correct misinformation that may appear in other sources. That was what I tried to do with the Ram Loevy article.

I recognize fully that there are times when the articles' subjects will have a biased viewpoint about their own careers or achievements. On the other hand, there are basic facts that are not subject to bias but simply reported wrongly in other sources. The classic example of this is the f Jimmy Wales's birth. The information out there is wrong, Jimmy knows the truth, and he can have no possible vested interest in changing the date from 7 to 8. His statement should be trusted.

There are many other examples like that, which I can think of. They certainly come in to OTRS regularly. When writing Ram Loevy, I simply thought it would be best to avoid the step of having to mail in corrections, particularly since I had access to him. Here are some examples of corrections he sent me via email (a copy of the original email was given to you). The email is quoted verbatim:

''1. Could you please change the order of my titles from "Writer and Director" to "Director and Writer"? The present order maybe a bit miss leading although I was indeed the writer of all my documentaries, however most of my features were written by others while I am mentioned as "with participation of".'' 2. My Mother's name is Elisa not Elisabeth. 3. My wife's name is spelled Zipora. 4. I was born three months - not two months - after my parents arrival to Palestine. ''4a. I Achmad is a documentary and not a feature. The director was Avshalom Katz.'' ''5. Mitrassim was presented to the BBC as a single documentary - not as a series. Mitrassim, Barricades and Chawageze - are different names of the same film.'' ''6. Hirbet Hiza'a is also known as Khirbet Khize. This is the spelling of the name of the story as translated about two years ago.''

Points 1, 2, 4, 4a, and 5 were corrections of facts that appeared in the one official biography (compiled when he received the Israel Prize). 3 was my mistransliteration of the Hebrew (I originally used two p's and an h at the end). Point 6 is a point of confusion because both appear in the literature (it is an English transliteration of a Hebrew transliteration of the name of an imaginary Arab town). My preference was the first because it more accurately represents the sound of the name, but he is correct that the English book (first translated in 2008) uses the latter. What I would point out here is that none of these corrections alters any of the core facts of his life or portrays him any better or worse than he actually is. He simply wanted to get the record straight. There are many other examples of this. In one email he said that an article claiming that there were hundreds of drafts of the script for Bread was an exaggeration. There were only 16.

This is a common problem. Some people will remember how, a few years ago, Wolf Blitzer pointed out on CNN that the article's explanation of who he was named after was wrong. I believe that these type of instances can be avoided in at least some instances if the subjects are approached.

One solution might be to simply submit the correspondence to OTRS, but I do not think that this is ideal, since it cannot be cited. Two people with whom I have discussed this offered solutions. These are not my own:

1. Create a content area where such correspondence can be displayed and linked to as a reliable source. 2. If possible, interview the subject, possibly through WikiNews, so that the information can be linked to as a reliable source.

In many cases of BLPs, both of these are viable alternatives. Given the size of Wikipedia, people would want their articles to be as accurate as possible. I would, however, add a third point.

3. Establish flexible guidelines determining what information supplied by the subject of an article is considered reliable.

One final note regarding original research. The original idea behind NOR was presented by Jimmy in an email dated 12 July 2003. At the time it was offered as a way to avoid people using Wikipedia as a platform for "crackpot theories" in the sciences. This was soon extended to include history as well. Delirium cited a problem in which "there is a tendency by some Wikipedians to produce novel narratives and historical interpretations with citation to primary sources to back up their interpretation of events" (italics are mine).

I do not believe that contacting the subjects of articles constitutes Original Research under these early understandings. Rather, it is an extension of the fact-checking process, which should be encouraged, particularly when dealing with BLPs.

That's all for now. I am sincerely interested in hearing what others think, since I believe that this can be valuable not only for my particular article on Ram Loevy but for many other articles besides. Danny (talk) 15:23, 18 March 2010 (UTC)