Talk:Duke Mathematical Journal

Impact factor, half-life, etc.

I removed the following:

Of the more than 600 mathematics journals published worldwide, only 170 reach the level of impact required to be included in the rankings of the Institute for Scientific Information. Among these, the Duke Mathematical Journal is ranked at 12, with an impact factor of 1.118 and a cited half-life of >10, the highest score given in this category.

There are at least half a dozen math journals with greater impact factors as measured by ISI. I checked a few years to make sure, but for example, in 2005, only two journals had an IF > 2, Annals and JAMS [1]. In 2001, Duke doesn't even show up in the top 10 for IF of pure math journals [2]. Perhaps I'm mistaking what is meant by "in this category", but certainly Duke is usually evaluated in the same category as these other journals, and it's normal to have half-life > 10 for a major pure math journal.

Besides, I wonder what the point of this is anyway. Numerous people have complained, e.g. in the pages of the AMS Notices, of the various flaws in using IF for math journals. IF just isn't as important in pure mathematics.

...Hm, I see that the lifted text is a copyvio from the Duke Mathematical Journal website [3]. Sigh, well, we don't need to engage in this kind of promotion on Wikipedia. --C S (Talk) 03:17, 27 April 2007 (UTC)

I don't mean to reinstitute apparent self-promotion already rejected by wikipedians, but I think the material is appropriate. The "lifted" material may be copyright violation, but I merely cite it; and the fact that "at least a half dozen journals have more impact" does not contradict the claim that DMJ is ranked twelfth. It's not easy to judge quality, but independent metrics at least speak to notability, and the use of these metrics for funding and job hunting, etc, should not be dismissed. In fact, the effort to rank technical publications, particularly math in peer review journals, was an early influence on methods used today by Google to rank web pages.Pete St.John 22:31, 13 June 2007 (UTC)

I disagree with the re-inclusion of this material. It was initially included only as part of a promotion effort by the publisher of DMJ (cf edit history of article). I don't see the relevance of most of your comments. IF is universally decried as a measurement of worth of mathematical journals, and is not seen as a useful or relevant metric by the general mathematical community. Publishers do like to mention this stuff however. So the only reason to include is to bolster the reputation of DMJ. I'm not saying you are purposefully engaging in promotion, but I think you ought to be careful and consider that your Duke affiliation is causing a possible conflict of interest. --C S (Talk) 23:57, 10 July 2007 (UTC)

Chan-ho, your response has several parts which I will try and address individually.

• "...initially included as part of a promotion...". We don't want to confuse 'apparent intention' with 'effect'. Lots of Wikipedians edit to increase their edit counts (which is deplorable) but if the edits are good, the work is good, even if the motive is petty. It is a logical error to confuse motive with misfeasance.
• "... I don't see the relevance of most of your comments...". To me the relevance of my comments seemed clear; please pick one (or more) specifics and I'll try to amplify.
• "...universally decried as a measurement...". That's very interesting. It doesn't follow from "...numerous people [in citable references, e.g. AMS Notices] have complained of ...flaws" as most things of this world are imperfect. Wikipedia is concerned with notability, however; Mariah_Carey has a subpage just for the list of her Grammy Awards, although lots of musicians have made good music without winning any prizes at all. If there are more useful metrics for math journals you may want to write about that, of if the problems with a particular metric make it unusable you may want to cite references for that.
• "...possible conflict of interest...". Point taken. I won't revert further edits on this point in the article.

The Columbus_Dispatch has (according to it's wiki entry) a circulation of about 350K. The New_York_Times has about 1,100K. To advertisers, a better metric of impact might be sales, or average income of readers, or even advertising revenue itself. Nevertheless the articles give some metrics, cite sources, and readers are generally interested; it helps them choose a paper to read, or to advertise in, or to write for, or whatever.

The significance of particular metrics can certainly be overblown, misinterpreted, or trivialized. But it's OK with me if such metrics are cited. You may like to compose a critique of that particular metric (citing the AMS letters mentioned, perhaps) and put it in an article on the metric itself, linked from here.

Spot checking other journals, I see no other reference to this metric. It's ok with me if we conform to a common standard, although there does not seem to be one. Annals_of_Mathematics says "...It ranks amongst the most prestigious mathematics journals in the world by criteria such as refereeing intensity...". I don't know what "refereeing intensity" is (referees per submission?) and the sentence seems a bit subjective.

Just an historical note: this whole metric thing was novel when I was a kid (at Duke). The idea of weighting references to papers, by the references of the papers that reference the paper (etc), as a more sophisticated and useful metric than just article count or citation count, came from mathematics, and led to Google. Pete St.John 14:52, 11 July 2007 (UTC)

COI?

User C S put a Conflict of Interest tag on the article, with the comment to "see talk", but doesn't seem to have explained it here. I don't want to jump to conclusions, so I left a note at his talk asking him to explain here. Pete St.John (talk) 23:39, 25 January 2008 (UTC) (Duke alumnus, '78)

I don't know what conclusions you would jump to other than the obvious: COI tags are used to invite uninvolved parties to come take a look. I think that's the best course of action here. "See talk" was to explain to anybody that wants to get involved to take a look at the discussion first. --C S (talk) 00:07, 26 January 2008 (UTC)

I do not want other editors to get the wrong idea from this. COI tags are not the first-line in requesting third-party attention to an article or section. That is best served by a "Request for comment". Edit Centric (talk) 04:38, 26 January 2008 (UTC)

I have taken the liberty of removing the COI tagging from the article. C S, what I would suggest is that you research and implement the Request for Comment process instead, therefore inviting third-party investigation and involvement. Edit Centric (talk) 06:00, 26 January 2008 (UTC)

Wikiquette mediation

Okay guys, let's start fresh here. As long as both editors can embrace the following suggestions, I think things can be reasoned out, as it seems that the parties concerned are well-educated;

• No revert warring or edit warring.
• Please be courteous, or at least civil.
• Be clear and concise on what you would like to see for the article.
• Approach edits and content from a neutral point of view.

I will take a "fly-on-the-wall" position on this, unless talks seem to be going in circles or stall. Edit Centric (talk) 06:27, 26 January 2008 (UTC)

One item that I would like to pint out is the reference given for this statement: "Currently DMJ is ranked 10th in impact, among 170 mathematics research journals ranked by the Institute for Scientific Information." The source citation is invalid, in that it needs to point to a third-party site or documentation, not another unlinked statement. The DMJ page is not a valid reference for this statement either, as this would be a direct COI, NPOV and WP:RELY issue. Edit Centric (talk) 07:04, 26 January 2008 (UTC)

• Comment The only way of documenting it properly is a link to JCR, because the search result cannot be linked in a usable manner. The current guidelines for journal articles suggest including this information as basic content. But i strongly deprecate making a comparison with a specific other journal. See Wikipedia:WikiProject Academic Journals for current practice and recommendations. DGG (talk) 09:30, 27 January 2008 (UTC)

I agree with Edit Centric that the Duke Mathematical Journal's own web site is not a reliable source for its own impact factor. The impact factor could be cited from another reliable source, even a paper publication if necessary, but not to anything published in the DMJ. The mention of the impact factor, though legitimate if properly sourced, seems to overbalance the current article, which has little other content. If impact factors are discussed over at WP:AJ as mentioned by DGG, a more specific reference to the discussion may be needed, since I didn't spot it in a quick look. EdJohnston (talk) 22:39, 27 January 2008 (UTC)

If we're "starting fresh" then my objections are irrelevant. Pete St.John (talk) 18:23, 28 January 2008 (UTC)

Oh come on, Pete, your objections are not irrelevant, as long as your position fits within Wiki guidelines. If it's outside the guideline or policy, then you're between the proverbial "rock and a hard place". At any rate, as long as the input is constructive and for the benefit of the article, then we definitely want to hear (read) it. Edit Centric (talk) 20:54, 28 January 2008 (UTC)

My objections weren't about the content, but the unethical editting practices employed. Starting from fresh, addressing the content, doesn't address my concern about ethical editting practices. That said, I happened to read your reply at the other talk page, and generally that was OK (a little left of center? um, OK). There's no relief for my being tagged as COI here but presumably active editors get that alot and it won't matter in the long run. As for the content, I don't give a rattus rattus's posterior for the impact schemes, as DMJ is editted by a Tar Heel now and it's basketball season. Pete St.John (talk) 21:08, 28 January 2008 (UTC)

(Indent reset) So, are we calling this mediation a "done deal" then? As long as all editors are on the same page as far as guidelines and such, sounds like a plan. However, if there is still "unfinished business" on this, I've moved the entire mediation here for further perusal and comment(s). (The idea came when Peter cross-posted some items to my talk space, struck me as a good idea, so I created a dedicated space under my talk to address it!) Edit Centric (talk) 06:16, 29 January 2008 (UTC)

• Let's split the mediation into two parts, 1. The Content and 2. My Objections, which have nothing to do with the content. Is (1) a "done deal"? I'm willing to pass on that judgement. Is (2) a "done deal"? certainly not, so far as I've seen the objection I raised in bringing the Wikiquette item hasn't been concluded, but it has to some extent been addressed, on talk pages, and I'll move to the indicated spot for whatever further. Pete St.John (talk) 18:13, 29 January 2008 (UTC)