Jump to content

Wikipedia:Manual of Style/Medicine-related articles/RFC on lead guideline for medicine-related articles

From Wikipedia, the free encyclopedia

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.


Is the guideline for leads in medicine-related articles in this version of the Medicine-related articles Manual of Style in agreement with Wikipedia:Manual of Style/Lead section? Should WP:MEDLEAD be in sync with WP:LEAD?

Please avoid threaded responses in the "Survey" section, and start a new sub-section in the "Discussion" section for threaded commentary. 22:10, 22 December 2019 (UTC)

Survey

[edit]
See chart of sample positions in Survey (continued) below. SandyGeorgia (Talk) 23:08, 30 December 2019 (UTC)[reply]
Collapsed and replaced below, SandyGeorgia (Talk) 23:06, 25 December 2019 (UTC)[reply]
  • No, not in sync, but should be. Remove all content from the section, WP:MEDMOS#Lead: there is nothing unique about leads in medicine-related articles, this is a fork of a standard guideline, and there is little in this section that is in agreement with Wikipedia's broader guideline, WP:LEAD. Some of the forked guideline text directly contradicts Wikipedia's broader guidelines, and application of these recommendations has resulted in choppy prose, a loss of clarity, and over-cited leads.

    See the discussion at the talk page of Featured article Schizophrenia for a sample of problems resulting from application of this guideline. The specific issues in this text—that has been disputed for many years—are:

  1. Language can often be simplified by using shorter sentences, having one idea per sentence, and using common rather than technical terms. No such restriction on sentence length is in LEAD. Guidelines on making technical articles accessible are already available (MOS:JARGON and WP:TECHNICAL), and they do not include restrictions on sentence length. The advice about using common rather than technical terms is contained in the general sections of MEDMOS, and is not unique to the lead. Application of this sentence length restriction and oversimplified language has led to a loss of clarity and the precision required in medicine (see Talk:Schizophrenia example cited above).
  2. It is also reasonable to have the lead introduce content in the same order as the body of the text. This is not true for every topic, and forcing the lead to a specific flow causes prose deterioration in articles (particularly Featured articles with carefully written leads) where the flow of information may need to be presented differently than the set structure that has been imposed. There is no such requirement at LEAD, and forcing a set structure can actually cause the lead to be less understandable to a broad audience.
  3. Avoid cluttering the very beginning of the article with pronunciations or unusual alternative names. This is distinctly at odds with the wider guideline, MOS:LEADALT.
  4. Medical statements are much more likely than the average statement to be challenged, thus making citation mandatory. This text is at odds with LEAD (see example at Talk:Schizophrenia). Text that must be cited is clearly discussed in broader guidelines: no evidence that medical content is any different has been presented. That overcited leads aid the reader has been rejected by the broader community, and medicine is not an exception.
  5. The final sentence at WP:MEDLEAD indicates why these deviations from WP:LEAD have been introduced: To facilitate broad coverage of our medical content in other languages, the translation task force often translates only the lead, which then requires citations. Translation to other languages may be a laudable goal, but to facilitate a non-English-Wikipedia project, restrictions in medical articles that go beyond WP:V and LEAD are being added to MEDMOS, resulting in deterioration of leads on en.wikipedia. The chart does not contain information unique to medical content.
Medicine project guidelines must remain consistent with Wikipedia's broader guidelines. If the Wikipedia community wants to force citations and short sentences in a set order into leads for purposes of translation to other languages, that should be accomplished at WP:LEAD, not within the guidelines of one Wikiproject. SandyGeorgia (Talk) 22:10, 22 December 2019 (UTC)[reply]
  • Remove, replace, or rewrite sections in MEDLEAD to better reflect Wikipedia-wide WP:LEAD guideline, as explained in 1–7 below.

    I have collapsed my original response and re-written it to hopefully address concerns about how the RFC is framed. Discussions at the WikiProject level have failed to resolve intractable disputes resulting from application of this local WikiProject guideline (or the items it encompasses even before they were added to the guideline page) for many years now, and broader examination of each of these sentences is needed.

    The last uncontested version of WP:MEDMOS had no Lead section, because there was no information specific to medical articles thought necessary. It said only: "Adding sources to the lead is a reasonable practice but not required as long as the text in question is supported in the body of the article", which was in sync with Wikipedia policy and guideline.

    See the discussion at the talk page of Featured article Schizophrenia for a sample of problems resulting from application of this guideline. I have highlighted in yellow phrases of particular concern, and prioritized from 1–7 the sentences from WP:MEDLEAD that I see as the most problematic, so that these numbers can be referenced in Survey and Discussion below.

1. To facilitate broad coverage of our medical content in other languages, the translation task force often translates only the lead, which then requires citations. Remove entire sentence.

While the goals of translation are laudable, translation is not unique to medicine, and if the broader community wants en.Wikipedia lead guidelines to be rewritten to accommodate translation of leads only, that should be addressed at WP:LEAD.

This is the tail that is wagging the rest of this dog; transparency in this discussion would be aided by disclosure from participants and board members of this Wikimedia project in their responses here. Separate programs or products being advanced by Wikimedia Foundation projects with their own governing boards should not be constraining content in the English-language Wikipedia. Requiring citations—beyond what en.Wikipedia policy requires, for the purposes of a Wikimedia project—should be examined at the level of Wikipedia-wide guidelines, and not in one WikiProject guideline.

2. Medical statements are much more likely than the average statement to be challenged, thus making citation mandatory. Remove this sentence entirely and return to wording similar to the last uncontested version.

No evidence is given that everything in the lead of a medical article is "more likely than the average statement to be challenged". As an example, what in this sentence is likely to be challenged?

Influenza, commonly known as the flu, is an infectious disease caused by an influenza virus.

The word "mandatory" in a guideline or policy page raises red flags; WP:BLP is where we would, for example, most expect wording of this nature, yet the word "mandatory" does not occur there. MOS:LEADCITE says:

Because the lead will usually repeat information that is in the body, editors should balance the desire to avoid redundant citations in the lead with the desire to aid readers in locating sources for challengeable material. Leads are usually written at a greater level of generality than the body, and information in the lead section of non-controversial subjects is less likely to be challenged and less likely to require a source; there is not, however, an exception to citation requirements specific to leads. The necessity for citations in a lead should be determined on a case-by-case basis by editorial consensus. Complex, current, or controversial subjects may require many citations; others, few or none. The presence of citations in the introduction is neither required in every article nor prohibited in any article.

To those who wonder why not fully cite medical leads anyway because extra citations can't hurt and may help: overcitation can and does impact readability and accessibility in leads by constraining the lead writer's ability to easily summarize general information well cited in the body. Precisely because leads are "written at a greater level of generality" to aid in comprehension and accessibility and provide a highly readable overview of the most important parts of the article, forcing citations to each clause of the lead can negatively impact the very concern this addition to MEDLEAD pretends to address. The idea that citations are mandatory anywhere has been well rejected by Wikipedia-wide guidelines.

3. The lead of an article, if not the entire article, should be written as simply as possible without introducing errors or ambiguity. Remove the highlighted yellow and blue text, and replace the blue with wording along the lines of: while taking great care that clarity is retained and error or ambiguity are not introduced.

The clause "if not the entire article" extends beyond the cited quideline, which clarifies that there may be sections that have more technical aspects (pathophysiology comes to mind in medicine) and may not be written as simply as the lead:

Some articles are themselves technical in nature and some articles have technical sections or aspects. Many of these can still be written to be understandable to a wide audience. Some topics are intrinsically complex or require much prior knowledge gained through specialized education or training. It is unreasonable to expect a comprehensive article on such subjects to be understandable to all readers. The effort should still be made to make the article as understandable to as many as possible, with particular emphasis on the lead section.

The blue highlighted wording is not problematic per se, rather its application requires fluency in the article's topic and literature that is not always evidenced by those advocating for this addition to MEDLEAD.

4. It is also reasonable to have the lead introduce content in the same order as the body of the text. Remove.

There is no such language anywhere in Wikipedia's lead guidelines, and no reason to expect medical articles should have a specific order of information in the lead. As in the first and second examples, the application of this clause—across broad swatches of medical articles by an extremely small number of editors who may not be fluent in the topics they are editing—has and does impede, again, the very thing we seek: readability and accessibility. Sometimes the most easily readable and simplest version of a lead has to introduce concepts in a different order than that which advocates prefer. (Why advocates want this done remains unclear.)
5. The British National Formulary for example often uses "by mouth" rather than "oral". Replace with a better example, for reasons covered in the Discussion section.

6. Avoid cluttering the very beginning of the article with pronunciations or unusual alternative names; infoboxes are useful for storing this data. Remove; alternates and pronunciation may be needed at times.

This is much better covered at MOS:FIRST, MOS:LEADALT and MOS:LEADPRON.
7. Statements that are not problematic, but may be better covered in Wikipedia-wide guidelines or more useful there: No issue if these stay or go to broader guidelines, but unsure if the LEAD section is warranted if any of the above is removed.
Language can often be simplified by using shorter sentences, having one idea per sentence, and using common rather than technical terms.
It is useful to include citations in the lead, but they are not obligatory.
Around a third of readers of English Wikipedia, have English as a second language.
Most readers access Wikipedia on mobile devices and want swift access to the subject matter without undue scrolling.
File:2019 Wikipedia reader native language by language.png
SandyGeorgia (Talk) 23:06, 25 December 2019 (UTC)[reply]
Revised positions on points 1–7 provided at Survey (continued) below. SandyGeorgia (Talk) 15:19, 31 December 2019 (UTC)[reply]
  • Reject RfC wording as a leading question where there is only one right answer. However, this is not the website to argue that the rules must be followed without exception. Is there any actual problem apart from pointless arguments about whether complex jargon (because it's an encyclopedia) is preferable to comprehensible wording, or whether the rule about no-citations-in-lead is mandatory? Are Pokemon articles really the same as medical articles? Johnuniq (talk) 22:40, 22 December 2019 (UTC)[reply]
  • Reject RfC Adding a second even more leading question doesn't make the first question less leading. Also "agreement" or "sync" are not the important question but whether something is permitted in which case consensus would prevail. AlmostFrancis (talk) 00:00, 23 December 2019 (UTC)[reply]
    This has become a farce. When the OP did not get their way they changed the RFC completely. When that didn't work the moved the RFC to the talk page and added only people they valued. This is silly at this point.AlmostFrancis (talk) 04:23, 29 December 2019 (UTC)[reply]
    So first you thought that the RfC, in its original form, should be rejected, but then when the OP took your advice about it not being well constructed and tried to make the request clearer, now you're mad about that? (Also, congratulations on reaching your first anniversary as a Wikipedian, on the day of your first post here.) WhatamIdoing (talk) 23:37, 29 December 2019 (UTC)[reply]
    I see the OP's special invitees feel free to ignore the prohibition on threaded replies. At no point has the OP made anything clearer. They are repeatedly changing the form in the hope of getting their way. Sarcasm noted. AlmostFrancis (talk) 23:59, 29 December 2019 (UTC)[reply]
  • Yes in sync right now however we are here to write an encyclopedia We have a guideline that says "The content in articles in Wikipedia should be written as far as possible for the widest possible general audience." which is similar to saying "The lead of an article, if not the entire article, should be written as simply as possible without introducing errors or ambiguity." This was discussed in 2015 when this text was added.[1] Wikipedia has received a lot of criticism, both in the academic[2][3] and lay press, for being overly complicated. We are not talking about writing the entire article to a grade 6 or 8 reading level (or any of the article for that matter). What we are discussing is trying to approach a grade 12 reading level for the leads which we have made progress on over the last 5 years in the area of medicine.[4] Translation is tremendously important. Our mission is to provide "every single person on the planet... free access to the sum of all human knowledge". This helps achieve that. The medical translation effort has resulted in more than 6 million words of text being translated. User:Subas Chandra Rout has translated more than 1,000 of these leads into Odia, a language spoken by about 40 million people, in which machine translate does not exist, and in which basically no medical content previously existed online. Efforts to try to isolate English Wikipedia from collaborating with other language versions is disappointing. Doc James (talk · contribs · email) 01:41, 23 December 2019 (UTC)[reply]
  • I don't see that WP:MEDLEAD contradicts the general principles of MOS:LEAD. It seems to me that MEDLEAD builds upon MOSLEAD in the same sort of way that MEDRS builds upon WP:RS, and does so because we have specific needs within our topic area. We know that our medical articles have impact well beyond the English-speaking world, and that they are accessed by many readers whose first language is not English, as well as being frequently translated into many other languages. We understand the importance of getting clear and simple medical information to readers in countries when disease outbreaks occur: Wikipedia is acknowledged as one of the foremost sources of that information. If you summarise the advice in MEDLEAD as "keep the language simple and cite the content", it doesn't seem to me to contradict the spirit of any of the guidance in MOSLEAD concerning the text of the lead, in particular "The lead should stand on its own as a concise overview of the article's topic." and "The lead must conform to verifiability ... there is not ... an exception to citation requirements specific to leads." I remain convinced that the advice given in MEDLEAD is good advice, and has a good purpose behind it. I don't believe that removing the advice or attempting to make it say only what is in MOSLEAD is beneficial to medical articles or to their readers. --RexxS (talk) 03:03, 23 December 2019 (UTC)[reply]
    Additional positions on points 1–7 provided at Discussion on Survey section below. SandyGeorgia (Talk) 23:08, 30 December 2019 (UTC)[reply]
  • I worked with others many years ago to polish MEDMOS and helped push it to guideline status. We examined featured content at the time, and got input from expert writers. Guidelines should reflect best practice as performed by multiple editors in consensus, and must never be amended in order to support an idiosyncratic approach. There's a clear COI and this is a guideline-fork being used to support non-consensual editing at odds with wider Community guideline and practice. That shouldn't be how Wikipedia works. Why does MEDMOS have a chart of native languages? The tail-wagging-the-dog argument wrt translation is yet another example of where off-wiki projects lead to a sub-optimal kludge that makes Wikipedia worse. Our leads should be a joy to read, not choppy fragmented sequences of incoherent factoids littered with unnecessary citation noise. I don't accept the argument at all that this makes translation harder, but fundamentally, that isn't Wikipedia's primary concern, which is to our readers. There is no good justification for any of the content in WP:MEDLEAD. -- Colin°Talk 10:11, 23 December 2019 (UTC)[reply]
  • RfC comment. I can't figure out whether I want to say "yes" or "no", but I can say what I think about it. I think that it's perfectly acceptable for medical content guidance to have some differences from general guidelines, because medical content risks misleading our readers to make bad decisions about their own health care. "First, do no harm" trumps "first, do not violate MOS". I think it's fine to have lots of inline citations in the leads of medical articles. I think it's reasonable to pay some attention to accessibility in terms of "reading level", but I also think that all writing on Wikipedia should be, well, well-written – and the specifics of wording on a given page can be determined through discussion and consensus at that page. I've looked at the edits that were contested at MEDLEAD, and I don't have any problems with the longer version. --Tryptofish (talk) 23:02, 23 December 2019 (UTC)[reply]
  • Oppose change — While I am tempted to reject the RfC outright on the ground that it is biased, there is I believe also a need to reject that there is any issue with WP:MEDMOS. Rehashing the issue with a less biased question would be irresonsible use of editor time, as I believe the issue can be resolved once and for all.
As MEDMOS is an extension of MOS, it is free to suggest additional scrutiny and higher standards — which is also all that it does. We must reject the idea of guidelines being "in sync", rather refocusing on whether they violate WP:MOS or WP:LEAD. WP:MEDMOS is written in such a way that it mandates nothing, and each example above concerns a suggestion — each of which I also agree are reasonable, relevant, and often necessary. I find it profoundly silly that the suggested additional care detailed in MEDMOS should violate other policies or guidelines.
The only point that actually goes against other guidelines (not simply extending upon them) is SandyGeorgia's point 6 — which states that pronunciation and rare alternate names should not be included in the lede body. This is a matter of deep concensus within the medical community, and is due to the preponderance of alternative names, sometimes upwards of 10 (solved well by using the other names=-parameter in Template:Infobox medical condition (new)). This provision has been included in WP:MEDMOS for years, and overturning it would disrupt readability and searchability of articles — and demand extraordinary efforts to enforce while only resulting in worse articles. I believe this justifies its inclusion despite contradicting suggestions from WP:LEADALT and WP:LEADPRON. Carl Fredrik talk 23:02, 25 December 2019 (UTC)[reply]
Additional positions on points 1–7 provided at Survey (continued) below. SandyGeorgia (Talk) 15:23, 31 December 2019 (UTC)[reply]
  • Comment: I understand where SandyGeorgia is coming from, but I lean toward yes, in sync. Flyer22 Reborn (talk) 03:10, 27 December 2019 (UTC)[reply]
  • Support these changes to bring the page closer to general guidelines. Cas Liber (talk · contribs) 05:40, 27 December 2019 (UTC)[reply]
  • Support these changes, and generally oppose the notion that a primary purpose of enwiki medical article leads should be ease-of-translation. It's more important that we accurately and readably state the information in English than that we make it easy to translate that information into another language. I find the thought that people who speak a certain language are getting their medical information primarily from Wikipedia deeply frightening. We are not competent for that task. We are competent to write a tertiary source that summarizes secondary sources. We shouldn't even be trying to be some kind of WebMD. Levivich 18:21, 27 December 2019 (UTC)[reply]
  • yes in sync, the changes that have been suggested will make medical content more difficult to understand for laypeople like me and especially for people who have English as an additional language. John Cummings (talk) 17:12, 2 January 2020 (UTC)[reply]
  • Comment It would be much easier for people to take part in this conversation if the RFC was reworded to fit within the standard style of RFCs so people can use 'support', 'oppose' etc. Please can it be changed? John Cummings (talk) 17:12, 2 January 2020 (UTC)[reply]
  • Comment I'm finding this RfC to be very confusing so I'll just state my views without trying to label them. I understand and support the concept that community consensus trumps local consensus, and I'm sympathetic to the view that for Wikipedia in general, lead sections don't need citations. However, the role of the English Wikipedia has evolved in the past few years, such that in addition to being an important standalone project, enwiki - and more specifically, its lead sections - is also an important source for translation on issues that are critical for the well-being of humanity. What we're doing for medical articles in terms of having translation-friendly and fully-cited leads is an innovation that we should be talking about spreading to other topics of serious global concern, such as climate change. Writing for translation is a practice that's become normal within the technical communication profession because every large software company produces docs in multiple languages - it's a pain sometimes, but as a writer you get used to it and can take pride in it. Clayoquot (talk | contribs) 18:04, 8 January 2020 (UTC)[reply]
  •  Comment: I agree totally with Doc James because as a native speaker of Arabic language and a medical student, English Wikipedia provided me with simple language that is easy to understand and not too technical which helped me alot during my study. In addition this language being not complicated for someone who is not native English speaker, played an essential role in my contribution to medical content in Arabic Wikipedia in that I found it easy language to translate from and usually when I want to translate article that is not in English I really suffer from that and often I ask friends to translate it to English fist as mutual language then I can translate it to Arabic. Please save that essential role of English Wikipedia as a tool to spread knowledge to other languages and please don't make it just for certain linguistic/ social class of readers. Regards--Avicenno (talk) 08:41, 12 January 2020 (UTC)[reply]
  • Whatever this is, I think I agree with it. Though I see no problem with citing in the lead or the lead matching the body order (though I agree citing probably shouldn't be "obligatory" and that medical statements aren't "much more" likely to be challenged – political statements are). What I do know is that any medical article maintained by Doc James has far too many short sentences in the lead and is incredibly disjointed to read. I wish I could change it to be more in line with basically every other article on the entire Wikipedia. James even states on his user page that he may have some difficulty with spelling or grammar, though I do not know if this affects how he sees appropriate sentence length. All I know is that his current standard is not an optimal universal benchmark to base lead sentence length or accessibility on. I mean, I was looking at the lead for Harlequin-type ichthyosis:

"There is no cure. Early in life constant supportive care is typically required. Treatments may include moisturizing cream, antibiotics, etretinate, or retinoids. It affects about 1 per 300,000 births. There is no difference in rate of occurrence between sexes. Long-term problems are common. Death in the first month is relatively common. The condition was first documented in 1750."

It gave me a headache. So yes, perhaps changing the medical lead guide to specifically include WP:TECHNICAL's statement, "However, using too many short sentences in a row becomes monotonous; vary sentence length to maintain reader interest" would be helpful. SUM1 (talk) 08:51, 19 January 2020 (UTC)[reply]

Discussion

[edit]

Discussion of JohnUniq's comment

[edit]

I added a second question to hopefully make the choice less binary.[5] Pokemon articles are quite different from medical articles; clarity in medical topics is undoubtedly more important than in Pokemon, and community-wide guidelines have been vetted to enhance clarity and accessibility. SandyGeorgia (Talk) 23:04, 22 December 2019 (UTC)[reply]

After the fact, I understood that one question could not address seven different issues, and re-wrote my Survey response to address each issue separately. I regret that I failed to separate the issues at the outset, and apologize for the length because of my faulty framing; framing an RFC is never easy, and I regret that the first responders to this RFC declined to point out to me specifically why it was malformed. SandyGeorgia (Talk) 22:30, 25 December 2019 (UTC)[reply]

For newcomers to the RFC, there are good examples of how to address the seven separate issues in the Discussion on survey section. SandyGeorgia (Talk) 19:51, 26 December 2019 (UTC)[reply]

More discussion

[edit]

SandyGeorgia, it appears that WP:TECHNICAL says to "Use short sentences when possible. Comprehension decreases dramatically when sentence length exceeds 12 words. However, using too many short sentences in a row becomes monotonous; vary sentence length to maintain reader interest." This direction to use short sentences feels more restrictive than merely observing that "Language can often be simplified by using shorter sentences". WhatamIdoing (talk) 05:14, 23 December 2019 (UTC)[reply]

Precisely. And the drive to make them ever shorter, and ever simpler, than Wikipedia-wide guidelines suggest, leads to a loss of clarity and specificity (so essential to making medical content digestible, accessible and relevant). Adding to that the drive to put a citation in the lead on each text fragment means that sometimes the clearest and most accessible wording can't be used, because it can't be cited to (or at least, hasn't been in examples I've seen) the simple sources (not necessarily the highest quality sources) used. I'll continue in RexxS's section below. SandyGeorgia (Talk) 15:43, 23 December 2019 (UTC)[reply]
SandyGeorgia WhatamIdoing, what if we changed the MED guidance from "Language can often be simplified by using shorter sentences, having one idea per sentence, and using common rather than technical terms." to "Use short sentences when possible. Comprehension decreases dramatically when sentence length exceeds 12 words. However, using too many short sentences in a row becomes monotonous; vary sentence length to maintain reader interest, [and use common rather than technical terms]." While referencing may be contributing to the issue of disengaging language in the leads it's a necessary evil against the greater threat of bad information being introduced imo. Ian Furst (talk) 16:34, 23 December 2019 (UTC)[reply]
Thanks for offering a suggestion (if we had seen more of that in talk page discussions, we might not be here with an RFC :)

The "greater evil" is that bad information is being left in articles as the Medicine Project has switched from a focus on writing quality content, to writing only leads for translation. This is a big problem, and we should all be worried about the amount of incorrect information that is in our medical content, as the Medicine Project no longer focuses on the kind of work it did five or ten years ago (weekly collaborations, improving article assessments, etc.)

On the guideline, we want people to actually read the page, rather than have it become bloated by repeating what is in Wikipedia's main guidelines. We should be adding at MEDLEAD only information which is specific to medicine and explains how to interpret the main guideline for medicine articles. If we think LEAD is insufficient, we should be addressing it. None of what you suggest is specific to medicine, and the problem we have seen is that an over-zealous application/interpretation of sentence length is negatively impacting clarity. SandyGeorgia (Talk) 16:58, 23 December 2019 (UTC)[reply]

I can't say that I've ever written a lead for translation (or even had it in mind when I wrote it) but I do agree that it's something to consider. I looked at some articles from 10 years ago, and they had references in the lead but lacked a summary of the main points of the page. I think it's beyond this RfC to reorient what you characterize as a misguided focus by all of wikimedicine (nor do I agree it's needed). I do think we could improve the readability of our leads. Is there a metric for readability? E.g. should we have an RfC that we will build leads to a grade 12(?) level in english and all agree to a specific engine to test our leads? Ian Furst (talk) 19:23, 23 December 2019 (UTC)[reply]
But what about the need to improve readability of our leads is specific to medicine? And not one of the "yes in sync" opinions above have addressed how they believe that specifying a set structure for leads improves readability; having written many leads, IMO that in fact constrains readability. None have addressed how we can write accessible and easily digestible leads if we have to follow a specific order in which to present that information (a requirement not found in any wider guideline), even if that topic does not lend itself well to that order, and add a citation on each little piece, even that which is common knowledge or better covered in detail in the body (a requirement specifically rejected by Wikipedia guideline and policy).

The goals put forward by the text at MEDLEAD are at odds with each other, and it is not surprising that the Medicine Project has not produced a Featured article since the trend set in, when the project once saw good annual growth in good and featured content (sample, which as of 2019 shows a net decline). These (non-standard for Wikipedia) practices force less accessibility and readability, and less well written content in leads; that is why the Wikipedia-wide guideline does not restrict the writer in the ways MEDLEAD is now. IF we want to improve leads, LEAD and other guidelines tell us how. If we want leads only for translation, that project has their own content-forked pages anyway.

Thank you, at least, for engaging the topic, which was not happening when the discussion was local. SandyGeorgia (Talk) 19:36, 23 December 2019 (UTC)[reply]

This is a discussion worth having; happy to engage. I'll expand on 'in sync'. (1) I've written my fair share of LEADS and have never felt constrained by the need to reference. With that said, I'm a surgeon not a copywriter so my sentence structure tends to be shorter, more concrete, and less readable. Articles from a decade ago had plenty of referencing in the LEADS. In short, I personally don't see this as a cause of readability issues (2) How does MEDLEAD help? It prevents hyperbole and downright BS (esp from commercial interests). While this may not help readability it strongly encourages factual statements. If we accept that referencing equates to shorter prose (which I argue against) then this may be a necessary trade-off imo. (3) GA/FA status. I've only taken 3 articles to GA so take this with a grain of salt, but I found them to be an exercise in Wikilawyering rather than an intense copyedit. Having written articles irl that have been subject to extreme copyedit processes, that's what I expected, but not what happened. Add to that the LONG turn-around times, and our group consensus goals to get all top-importance articles at B-class or above as a priority and I can understand why there is a lack of desire to push articles forward. Also, my sense (and maybe someone can add stats to this) is that few of our articles have a single 'champion' rather we have a more of a group effort. Last, translation was/is an important goal imo but that doesn't mean we can't tip the balance back to more complex prose. Many of the top importance articles are already translated. I still believe Wikipedia in general and Wikiproject Medicine specifically has done more to improve access to life-saving information than almost anything in history. It is a project well worth having intense, meaningful debates about. Best. Ian Furst (talk) 01:53, 24 December 2019 (UTC)[reply]
Thanks again for engaging, Ian Furst. I don't want to appear to ignore you or RexxS (section below), but have decided to wait a bit for the temperature to subside in here before adding more thoughts. Generally, my hope is that, now that we are finally discussing, reasonable editors can come to a compromise on different portions of the disputed guideline text (some portions of it remain completely unaddressed by responses here). The "Me, too, per editor so-and-so" declarations that have become commonplace at WPMED have never led us anywhere useful, and dialogue is finally happening, so I am hopeful. More later, best regards, SandyGeorgia (Talk) 02:02, 24 December 2019 (UTC)[reply]
The change to voting rather than discussing is not specific to WPMED. It's gotten so pervasive that I've been contemplating a barnstar-like award that entitles the recipient to actually WP:Ignore all rules when the rules get in the way of improving articles.
As I said at WT:MEDMOS, I don't have strong views about whether we should minimize redundancy or support one-stop shopping. But I don't feel like MEDMOS's factual observation that "Language can often be simplified by using shorter sentences" either contradicts or exceeds the requirements of TECHNICAL. If I were making changes, this would not be one of the changes that I recommended. WhatamIdoing (talk) 20:41, 24 December 2019 (UTC)[reply]
Agree that phrase is one of the least problematic aspects of the new text at MEDLEAD, and I now understand that the RFC is malformed, because there are so many different pieces of MEDLEAD that are problematic, yet the way I phrased it (one question) seems to ask for one answer. I did not foresee that problem (me, again :) and yet, people responding are not looking at the other problems. More later, SandyGeorgia (Talk) 20:49, 24 December 2019 (UTC)[reply]
Coming back now to address the post from Ian Furst at 01:53 24 December.

Regarding your point (1) on leads and readability: having read thousands (literally) of Featured article candidate discussions on leads during my tenure as FAC delegate, it is my opinion (and that of many others, according to LEAD) that overciting a lead can and does constrain writing, impeding the clearest possible narrative written at a higher level of generality. I will role this into separate sample discussions when I answer RexxS in his section below, because the CREEPy enforced order of content in the lead is also part of that problem.

On your point (2), I believe the idea that we need citations in the lead to help avoid quackery and commercial interest content being inserted into leads is a strawman: WP:MEDRS already gives us the sourcing tool to keep quackery and commercial content out of articles, so we don't need to change lead guidelines for that goal.

On your point (3), you will get no argument from me about the usefulness (or lack thereof) of the GA process, but most FAs most assuredly have what you call a "single champion", typically the WP:WBFAN nominator, but at times, someone else who took the article on after a Featured article review. This is so well understood that the FAR process requires notification of the original nominator when submitting an FA for review, and most FAs whose original nominator is no longer active on Wikipedia end up deteriorating. On this score, most of WPMED's FA writers are gone; that there has not been an FA from the Medicine Project since this trend set in, and that most FA writers are gone from the project, is not IMO coincidental. On a personal note, I once watchlisted every WPMED FA, but I removed most of them from my watchlist when their leads were negatively impacted by the trends evidenced in this guideline; to continue to watchlist them would mean I would need to take them to FAR, so I washed my hands generally. Whenever I do come back to look, I find deterioration (again, see the sample discussion at Schizophrenia; we can't expect Casliber to fix them all).

More significantly, whether we look at B-class or GA/FA level, WPMED in general has moved away from systematically improving content in the body of articles (as once was done in the now-defunct Wikipedia:WikiProject Medicine/Collaboration of the Month/History), to a focus on more exclusively editing leads, to support off-en.Wikipedia projects. Medical collaboration to improve content has been replaced in a project focused almost entirely on off-en.wikipedia ventures, altering leads of even Featured articles, and at a time when the effects of student editing require constance vigilance from competent editors. Gone are the times when we collaborated to bring an article to a higher standard.

As WPMED loses core content writers (including every one of its FA writers), how much has been lost in addressing content issues? Whether we care that no FAs have been produced by WPMED since this trend set in is not the issue: when we lose editors capable of writing at the FA level and replace them with editors who can offer little more than me, too, per-so-and-so in important discussions, where do we end up? This is being done in the belief that it helps "children in sub-Saharan Africa", but at a disservice to English-speaking readers who may read the entire article, or may be looking for information in one section of the article. On that score, we demonstrably have errors being introduced by editors who focus on leads only, and neglect significant errors in the body of articles. This should be a concern. We do not help anyone, in sub-Saharan Africa or otherwise, by focusing on leads to the exclusion of overall content, and relative to years past, that is where we are now. As you say, many of the top articles are already translated, and the translation project has its own content-forked pages for translations, so why should it be imposing guidelines that are not in accordance with Wikipedia-wide guidelines in order to facilitate translation? If an off-Wikipedia venture wants to force requirements into medical content to make translation easier, could they at least not do it in a way that assures that medical content cannot meet the requirements of Featured articles, so that our writers who are capable of producing that content are not chased off ? SandyGeorgia (Talk) 20:36, 26 December 2019 (UTC)[reply]

@WhatamIdoing, SandyGeorgia: Referring to the first two comments, I fully agree. Refer to my comment above in the first section. SUM1 (talk) 08:57, 19 January 2020 (UTC)[reply]

Content fork?

[edit]

Colin suggests that MEDLEAD is a content fork, which of course is a guideline about articles, and not project space. The relevant advice for guidelines is actually neither policy nor guidance and is at WP:GUIDELINEFORK, which recommends not to create a page "that conflicts with or contradicts an existing one". I maintain that MEDLEAD neither conflicts with or contradicts MOSLEAD, but supplements it for medical-related articles. It is common practice on Wikipedia for sub-topics to receive different, sometimes inconsistent guidance, and an obvious example is how WP:NPROF and WP:NSPORT differ from WP:GNG and from each other – the criteria in NPROF replace those in GNG; while those in NSPORT supplement GNG.

We are not a bureaucracy, with hide-bound rules, and the guidance given should reflect best practice, and if necessary, we have an RfC to determine what that is.

Turning to specifics: why shouldn't MEDLEAD have a chart of native languages? It helps editors understand the issues faced in making our content available to its readers. Our vision is "a world in which every single human being can freely share in the sum of all knowledge" and that means we should be doing whatever we can to increase the availability of our content to every single human being, which includes improving its readability and accessibility, as well as making it easier to translate. This not "tail-wagging-dog"; it is a fundamental principle of writing Wikipedia – what Jimbo called "the nearest thing we have to a prime directive".

There is nothing sub-optimal about making the reading level of our leads (medical or otherwise) low enough to be accessible to the broadest possible audience. That means writing shorter, more direct and less complex sentences to aid comprehension. It means using the simplest phrase available, while still preserving meaning. Wikipedia is not the place for editors to show off their vast vocabulary by using "oral" instead of "by mouth" or "renal" instead of "kidney". I'll gladly sacrifice some of the joy of reading a lead of Shakespearean quality for having one that a 15-year old kid in sub-Saharan Africa can grasp immediately. That's the justification for what's in MEDLEAD, which makes it clear that our editors are pragmatists, not elitists, and write content for the whole world, not just the privileged few. --RexxS (talk) 14:39, 23 December 2019 (UTC)[reply]

RexxS, once again, yes, once again, you haven't read carefully what I wrote and spent the first part "correcting" something I didn't write I did not say "content fork". I wrote "guideline fork" both in the text above and the edit summary. I should also point out that we do have actual content forks for translation: Wikipedia:WikiProject Medicine/Translation task force/RTT(Simplified)L lists lots. So the argument that our leads, read by millions of en.wp readers and thousands of times every day by poor Alexa, needs to be corrupted in order to help translation is, well rubbish. We end up with articles saying weird stuff like Hydromorphone: "It may be used by mouth". Indeed at Haloperidol the infobox has "by mouth, IM, IV, depot (as decanoate ester)" as though anyone familiar with the other jargon and abbreviations would stumble over "oral". The irony is:
  • "By mouth" -> oral (Spanish), par la bouche (French), mit dem mund (German), pela boca (Portuguese), per via orale (Italian)
  • "oral" => oral (Spanish), oral (French), oral (German), oral (Portuguese), orale (Italian)
So we've substituted a word "oral" that has a direct equivalent in every major European language, for a word "mouth" that is exclusive to English. We've also failed to help learners see the word "oral" used in a common context, the most important way that we learn and establish words. By going out of our way to avoid "taken orally" we fail to educate our readers with examples of common English they might hope to pick up when reading an encyclopaedia. But hey, apparently the sentence "Neuroleptic malignant syndrome and QT interval prolongation may occur.", is just fine -- a pair of incoherent factoids with no meaning. And "Haloperidol may result in a movement disorder known as tardive dyskinesia which may be permanent" is just grammatical garbage. Haloperidol is a noun can can't "result in" anything, even if "result in" was ever good English. There is more to writing accessible English than inserting baby words here and there and Wikipedia does have role in helping readers see words in context rather than hiding them away. Some basic competence with grammar, and stepping back to see if what has been written is a coherent paragraph that makes any sense. Once again, WP:MED is focusing on the wrong thing, and forgetting the basics. -- Colin°Talk 16:26, 23 December 2019 (UTC)[reply]
@Colin: No, you're wrong as usual. Not only that, but you're offensive as well. You have no rational argument, so choose to make personal attacks on those who disagree with you. Are you incapable of engaging on the issues, as Sandy has so ably done below? Take a leaf out of her book and learn to engage civilly for once.
Of course I read your entire rant. I read your appeal to authority, but you have to understand that your prior contributions give you no special privilege here. It doesn't matter whether you wrote all or none of MEDMOS, you don't own it. I read your attempt to claim the authority of "expert writers" and to smear the current consensus guidance as "an idiosyncratic approach". What utter nonsense. I read your attack on other editors by smearing them with the "COI" label. There is no conflict of interest here, only the conflict between your elitism and the current guidance that seeks to make our content available to as broad an audience as possible. I even read your edit summary (guideline fork) and I understand perfectly that you were trying to make a negative association, rather than debate the real issues. Your contributions here so far have been poisonous to constructive debate.
If you think that It may be used by mouth or by injection into a vein, muscle, or under the skin is "weird stuff", why not simply amend it to It may be taken by mouth or by injection into a vein, muscle, or under the skin or It may be administered by mouth or by injection into a vein, muscle, or under the skin? If it had read It may be used orally or by injection into a vein, muscle, or under the skin, it would still be "weird stuff" requiring improvement to satisfy the English idiom.
There nothing ironic about using "by mouth" which has direct translations into French, German and Portuguese. A Swahili translator is likely to recognise "mouth" as "mdomo" and translate "by mouth" as "kwa mdomo" (which is the Google translation of "orally" into Swahili). A quick check with Google shows, for example, "orally" = "mondelings" in Africaans and "усно" in Macedonian, both of which unrelated languages incorporate the word for mouth, so it's not just English that employs that idiom. Why would you want to reject guidance that suggests using "by mouth" instead of "orally", when it's commonly used in expert sources such as the British National Formulary? What makes you think "taken orally" is any way superior to "taken by mouth", other than fewer folks will understand it? Isn't that the agenda you're pushing?
How will removing the guidance "The lead of an article, if not the entire article, should be written as simply as possible without introducing errors or ambiguity" lead to an improvement in comprehensibility for the sentence "Neuroleptic malignant syndrome and QT interval prolongation may occur.? Your agenda would celebrate that sort of jargon, while the current guidance that you're so keen to do away with is exactly the recipe to fix the problem.
We will all agree that "there is more to writing accessible English than inserting baby words here and there", but that's not what the current guidance asks us to do. It advises editors to write as simply as possible without introducing errors or ambiguity. Just because you can't see any other way of accomplishing that than "inserting baby words here and there" doesn't mean that other editors are so limited in their writing abilities. --RexxS (talk) 20:48, 23 December 2019 (UTC)[reply]
It is a complete myth that the British National Formulary uses "by mouth" instead of "orally". It does have section headings, each of which begin with "By ---" such as "By mouth", "By rectum", "To the eye using eye drop", "By intravitreal injection", "By intravenous infusion", "By intramuscular injection, or by slow intravenous injection, or by intravenous infusion" and also even "By mouth using oral solution". For that pattern of writing, "oral" simply doesn't fit. But elsewhere, such as when listing side-effects or detailing administration procedures, they are happy to have sub-headings such as "With oral use", "With parenteral use", "With intravenous use", "With oral use in children". And elsewhere in the prose, oral is used routinely. Compared to all these other multi-syllable medical terms, "oral" is something any elementary school child knows after their first visit to the dentist: "oral health" being kind of a big issue with dentists. We fail our readers when we write clumsy non-standard English like "It may be used by mouth" and fail them when we are weirdly inconsistent in dropping "oral use" but keeping "intravenous use" or even "IV". -- Colin°Talk 08:48, 24 December 2019 (UTC)[reply]
So it's not actually a myth; it's just that BNF uses the terms interchangeably? I think that makes the point that there's nothing wrong with using "taken by mouth", although I'm sure we all agree that "used by mouth" is not idiomatic English (but then again, neither is "used orally" for a drug).
I don't want us to fail our readers, either, but I still don't see how removing advice to write simply actually helps fix the problem of jargon like "intravenous use" or "IV". Surely the existence of such unnecessary complexity is an argument to keep the advice?
I'm sorry, but there are millions of elementary school children outside of the developed world who never had a first visit to the dentist, and many more, who having seen a dentist, have still never heard the word "oral". Are we now to discount them as part of our audience? Surely it's understood that while English remains the lingua franca of the internet, the English Wikipedia is the first port of call for anyone who hasn't access to medical advice in their native language? --RexxS (talk) 16:59, 24 December 2019 (UTC)[reply]
I don't think this exchange is productive. I think we could do with some fresh voices here. -- Colin°Talk 18:46, 24 December 2019 (UTC)[reply]
I'm still taking deep breaths, and will respond in depth (as I said) when the heat subsides and people start listening as well as talking (that is, reading for understanding as well as writing to win a point).

But from my perspective as a Spanish-language interpreter in a free clinic for migrant farm workers without insurance, and as someone who is also conversant in Italian, and who lives in a household where fluent German exists, I think by focusing on specific words, we're missing the broader points we should be addressing in how to write the best possible content for any reader in any language. In romance languages, the more technical jargon (typically with a Latin root) is much more likely to lead you to directly to the translation. We don't speak of ear, nose and throat in Spanish: we speak of otorinolaryngología. We don't speak of kidney damage; since kidney = riñon, renal is a word commonly used (as in insuficiencia renal). Interpreting in Spanish means I am frequently using the technical jargon.

An appeal to the children in "sub-Saharan Africa" indicates to me we might try harder to hear/read the intent of those who argue a different point than our position.

When medical editors stop writing or improving content because the environment has become so toxic, then we'll end up with nothing left to translate for anyone in Africa or Venezuela, as well as articles that are inaccurate in the lingua franca, English. So, what is the approach here that will help us get better medical content in English, so we have content worthy of translation? Could those who believe that Wikipedia's content is helping that child in sub-Saharan Africa consider the possibility that, in spite of honorable intentions, they might be deceiving themselves, and weigh their concerns for that child in Africa against the possibility that deficiencies in English-language medical content may be impacting that child in inner city Detroit? If not, we will see an accelerated pace of deficient Wikipedia content in the body of articles, with leads written by editors who aren't necessarily topic experts and don't even know what they've gotten wrong in the leads they are putting up for translation. SandyGeorgia (Talk) 19:16, 24 December 2019 (UTC)[reply]

RexxS Thank you for pointing out that an RfC is the next necessary step for resolving this dilemma (after the first responders rejected the RFC).

I maintain that MEDLEAD does not supplement LEAD as a guideline should, rather extends beyond LEAD and even contradicts it (and other guideline pages) in several instances, and in ways that is affecting the precision and clarity so necessary for medical content. I also maintain that, if we are to extend our lead guidance in the way that MEDLEAD has done, that should be done in the main guideline page (LEAD).

Translating articles is not unique to medicine. If the broader community believes we should alter our LEAD guideline to incorporate the needs of a translation project, then we should address that in LEAD. Do we write leads for the English-language Wikipedia, or leads in a Wikipedia in English for the purpose of translators who may not be fluent in English? Further, considering the Translation Task Force implements its own static versions of articles for the purpose of translation (see Wikipedia:WikiProject Medicine/Translation task force/RTT(Simplified)L and example at Wikipedia:WikiProject Medicine/Translation task force/RTT/Simple Dementia with Lewy bodies), why is it necessary to alter the leads of every medical article? (A separate issue is that experienced medical editors should not be altering leads without also adjusting the body of the article.)

Some examples of how over-simplified wording, constrained by citations on each fragment and sentence length, are sub-optimal.

  1. Schizophrenia is a mental illness characterized by ... strange speech ... "Strange" seems to have been chosen as a word that translates easily, but it does nothing to describe the kind of speech characteristic of schizophrenia. Mork had strange speech with Mindy.
  2. Males are more often affected and onset is on average earlier in age. Here is a sentence constrained to 12 words. Does it tell the average reader what they need to know?
  3. About 20% of people eventually do well, and a few recover completely. Again, a sentence constrained to 12 words. What does "do well" mean here? Don't end up hospitalized? Don't commit suicide? Don't need medication? Don't have impairment in certain realms of functioning? The reader is given no idea (and the reasoning given for that on talk page is related to over-citation: the desire to stick with what one simple source states).
  4. This is the result of increased physical health problems and a higher suicide rate (about 5%). The suicide rate is 5% higher than the non-affected population or 5% overall ? A lack of clarity and specificity results from keeping sentences too short and overly simple, rather than when possible as suggested by the broader guideline.
These kinds of examples occur when editors not thoroughly familiar with the full body of literature on a given topic edit the lead only, to over-simplify for translation and impose a specific order on the text presented. SandyGeorgia (Talk) 16:48, 23 December 2019 (UTC)[reply]
Thanks for the constructive response, Sandy. It often helps when we can examine specific issues that illustrate more general concerns.
Am I correct in summarising your first concern as "text that is over-simplified is not as good as it should be"? If that's right, then I can understand why you feel that advice to keep text simple may lead to over-simplification. However, MOSLEAD gives this guidance:

"It is even more important here than in the rest of the article that the text be accessible. Editors should avoid lengthy paragraphs and overly specific descriptions"

That is surely an injunction to keep text simple, even more so than in the body of the article. MEDLEAD gives this corresponding advice:

"The lead of an article, if not the entire article, should be written as simply as possible without introducing errors or ambiguity."

That seems to me to be the correct advice, and I don't think that the argument that editors sometimes fail to avoid "introducing errors or ambiguity" is a reason to remove the guidance. Wikipedia abounds with problems where editors have failed to follow guidance, but the solution is to improve the content to fix the problems, not remove the guidance.
You also seem concerned that citations in the lead cause problems for the reader. I don't see that issue at all. The crux of MOSLEAD's guidance on citations is this:

information in the lead section of non-controversial subjects is less likely to be challenged and less likely to require a source; there is not, however, an exception to citation requirements specific to leads. The necessity for citations in a lead should be determined on a case-by-case basis by editorial consensus. Complex, current, or controversial subjects may require many citations; others, few or none. The presence of citations in the introduction is neither required in every article nor prohibited in any article.

Compare that with MEDLEAD's advice:

It is useful to include citations in the lead, but they are not obligatory. Two reasons for using them are: Medical statements are much more likely than the average statement to be challenged, thus making citation mandatory. To facilitate broad coverage of our medical content in other languages, the translation task force often translates only the lead, which then requires citations.

Now, I believe that there is nothing in MEDLEAD that contradicts MOSLEAD. Medical advice is often far more complex and potentially controversial (consider the problems of fake information like the anti-vaxxers and of traditional cures, which can only be countered by citing good sources) than more mundane topics like pop culture and sport. Medical articles are demonstrably more often translated than the average article and so there is an increased requirement for citations in the lead.
Both of those factors distinguish medical content from many other topics, which then drives MEDLEAD to extend MOSLEAD, but there is nothing in the MEDLEAD guidance I quoted that contradicts MOSLEAD. --RexxS (talk) 19:49, 23 December 2019 (UTC)[reply]
I wonder how much of this is the advice given, and how much is the skill with which the advice is implemented.
It happens that I've gotten a lot more skillful at writing in simple(r) English in the last few years. I have nowhere near the level of automaticity that a couple of my teammates do – writing to minimize the chance that machine translation will mangle your meaning is a complex art – but I can do it fairly well. My first response to "Males are more often affected and onset is on average earlier in age." is that this example is not a simple sentence. If I were trying to write that information simply, it would look something like Men are more likely to have schizophrenia than women. Men develop symptoms earlier than women, too." (assuming that's what the age of onset bit was trying to say). The example sentence fails to apply the "one thought per sentence" advice, and the article is poorer as a result. WhatamIdoing (talk) 06:07, 25 December 2019 (UTC)[reply]
RexxS, on the samples, I hope you will trust me that Schizophrenia is not a random sample, and we see this kind of lead editing repeated across articles by experienced editors driving leads to oversimplification while ignoring important text in the bodies of articles. When competent medical editors start unwatching articles because of the pervasiveness of this problem, the quacks win.

See discussion in the section below (Discussion on survey) on the problem with the word mandatory in a guideline, and my views on whether BLP-style citation requirements should be added to medical content. That is not the argument here, though, which is that they should be "mandated" (not a word found in any policy) in leads only, which does nothing to protect medical content from anti-vaxxers and quacks who add bullroar to the bodies of articles. Fuller discussion of that below.

You point out that Wikipedia abounds with problems where editors have failed to follow guidance, but the solution is to improve the content to fix the problems, not remove the guidance. We are in violent agreement in principle. For example, this removal of guideline text to accommodate student editing errors, rather than correcting the wording to address the problem is not what I am advocating, and is a good indication that competence is required in editing guidelines.

In this case, we have repeatedly seen extreme application of the principles behind the guideline, and edits that impact clarity and accuracy of medical content. If editors aren't able to understand Wikipedia policy, and our guideline wording is facilitating poor editing, then we should work to clarify the wording to reflect policy and best practice. Instead, we see "I like it, per editor-so-and-so" !voting throughout WPMED discussions, and no attempt to refine the wording in the disputed guideline text to help assure the best possible understanding. The wording we have does not reflect policy, and we have experienced medical editors who do not follow policy. We need to make our guidelines clearer to deal with not only the anti-vaxxers, but to also provide clarity to our established editors. Please take a more careful look at the type of wordsmithing that is needed to improve MEDLEAD; that sort of wordsmithing was a key part of the drive that saw MEDRS and MEDMOS written so well a decade ago, and that gave us guidelines that served the project well until around 2015. WAID points out that some people are better than others at simplifying language, but we have less editors doing this than you can count on one hand; our guidelines should be written as well as the Wikipedia-wide guidelines are, to leave no doubt as to what we intend. When we end up with few competent editors willing to even engage medical articles, leaving editing to those who lack nuance, understanding of the topic, or adequate time to make sure edits are sound, the anti-vaxxers and quacks only get the upper hand. SandyGeorgia (Talk) 02:04, 27 December 2019 (UTC)[reply]

@Sandy: I accept that editors of all abilities contribute to Wikipedia, and I don't disagree with your observation that sometimes edits leave much to be desired. Nevertheless, this is a wiki, and I've always advocated that the way to solve those problems is to fix them when we see them. I'm an educator by profession, and I would dearly love to find ways to teach editors to write better. Tangentially, Wikimedia UK recently organised an event where Johnbod explained to editors some of the techniques involved in writing a Featured Article – how cool is that? Anyway, I'm all in favour of improving the advice to editors, rather than removing it, and I think that is one way to cultivate the editors we have. Of course I'm not in favour of using the word "mandatory" in our guidance. There are much better ways of expressing approval of something than suggesting it should be compulsory, so I'm happy to help you explore how we can improve what we currently have. --RexxS (talk) 22:04, 27 December 2019 (UTC)[reply]
It wasn't quite that (and it was 13 months ago); more a survey of FA coverage and processes, encouraging people to join. I mostly riffed off the compiled list of links at User:Johnbod/FAC Skillshare. Johnbod (talk) 22:25, 27 December 2019 (UTC)[reply]
Please pardon my faulty memory, John. As I get older, the year before last seems more and more like the month before last. Skillshare is still a cool idea, and I'm always keen to celebrate ideas that help editors write better. Cheers --RexxS (talk) 22:35, 27 December 2019 (UTC)[reply]

Disclosures reverted

[edit]

CFCF reverted my additions of disclosures for himself and Doc James as members, officers, or founding members of the Wikimedia Project that has an interest in advocating for these changes. There may be other missing disclosures. SandyGeorgia (Talk) 23:20, 25 December 2019 (UTC)[reply]

I did so based on your request SandyGeorgia:

Please avoid threaded responses in the "Survey" section, and start a new sub-section in the "Discussion" section for threaded commentary. 22:10, 22 December 2019 (UTC)

It seems odd that these rules should apply to everyone but you, and additionally odd that you can choose what is a relevant disclosure. I don't find my engagement in Wikimedia Medicine relevant in any way, much as I would not find it relevant to disclose that I am a Wikipedia editor, have been to Wikimania, have recieved scholarships from the WMF or have an MD (I do however voluntarily disclose these facts on my userpage User:CFCF).
If you wish to discuss these affiliations, feel free to do so — but at least follow your own rules regarding the flow of discussion in the RfC. Carl Fredrik talk 23:24, 25 December 2019 (UTC)[reply]
I read what your reasoning was in edit summary so you need not repeat it; while I disagree with it, I let it stand since I don't editwar. Adding disclosures to !votes is commonplace in many !voting forums such as AFD and RFA, and it is done right under the !voters declaration; I don't consider it continuing a threaded discussion. Re: It seems odd that these rules should apply to everyone but you..., I view that as an unhelpful comment of the type that promotes a battleground environment. The disclosures can be done here in this separate section if my adding them under signatures is a concern. SandyGeorgia (Talk) 23:37, 25 December 2019 (UTC)[reply]
Disclosure from RexxS
As the question has been raised, I'm happy to clarify my agenda for editing medical articles on the English Wikipedia. I simply want to advance the vision of making the sum of human knowledge freely available to every person in the world. For me, that not only means writing and curating important content, but also finding ways of making that content available for everyone. That's the reason why I involve myself with medical articles, with accessibility and with technical issues as the major uses of my time. Folks may note from my user page that I am also a trustee of the chapter Wikimedia UK and I am chair of the newly-recognised thematic organisation, Wikimedia Medicine. Both of those are independent affiliates of the Wikimedia Foundation, but I really must make it clear that when I edit a medical article, I edit as a simple editor, not as an agent or representative of either of the affiliates that I have the privilege of contributing to. Both of those bodies will have policies or goals concerning Wikimedia projects, but my role on those bodies is to help make the policies and goals, not to implement them. If anyone would like to discuss further, then may I suggest that my talk page would be a better venue, rather than clog up this already lengthy and entirely unconnected discussion. --RexxS (talk) 22:27, 27 December 2019 (UTC)[reply]

Discussion on survey

[edit]

1. disagree. I feel function over form is more important here and the focus on global SDG's means we need to facilitate the work of the translation task force imo. Ian Furst (talk) 02:56, 26 December 2019 (UTC)[reply]
2. disagree. See previous discussion on the risk of non-MEDRS sourced statements by commercial interests and charlatans. Ian Furst (talk) 02:56, 26 December 2019 (UTC)[reply]
3. neutral. I don't think this change will have much impact. I have no opposition to it. Ian Furst (talk) 02:56, 26 December 2019 (UTC)[reply]
4. agree. Maintaining the same order, while great to ensure content is well summarized, disrupts the create of great prose. I favour form over function here. Ian Furst (talk) 02:56, 26 December 2019 (UTC)[reply]
5. no opinion. I saw the argument but would prefer to here from someone that actually translates. I'm suspect there is a word for mouth in most languages and I wonder how it translates. Ian Furst (talk) 02:56, 26 December 2019 (UTC)[reply]
6. disagree. We've built a section in infoboxes to handle this info. I'm surprised by your stance on the SandyGeorgia, the preambles ruin the flow of some articles. I prefer them in the infoboxes. Ian Furst (talk) 02:56, 26 December 2019 (UTC)[reply]
7. I don't understand the survey question in #7. Ian Furst (talk) 02:56, 26 December 2019 (UTC)[reply]

Thanks for the re-do, Ian Furst and my apologies for the formatting. I added clarification on to point 7. SandyGeorgia (Talk) 03:05, 26 December 2019 (UTC)[reply]

My list:

  1. Whether supporting the translation task force is a site-wide goal is something that should be handled separately. It can be a WPMEDF goal without being a WPMED goal, and it can be a WPMED goal without being a goal of the entire English Wikipedia. This particular guideline belongs to the entire community, not WPMED or WPMEDF, and therefore any such goals should be decided by the entire community, and separately from this more technical discussion.
  2. Even if we stipulate that medical statements are much more likely than the average statement to be challenged, that does not mean that the citation must be located in the lead. The rest of the community seems to be able to manage even statements like "<BLP> is a convicted murderer" without copying the citations to the lead. I therefore doubt that medical content in the lead, much of which ("Vulvar cancer is a cancer of the vulva") is both unlikely to be "mandatory" and unlikely to need separate/special rules compared to the rest of the encylopedia. I think that it is more appropriate to say that "It is useful to include citations in the lead, but they are not obligatory" than to say that they are "mandatory" or "required".
  3. I don't have any problem with telling editors to write "as simply as possible". I hope that editors reviewing articles will remember that "as simply as possible" is in many cases not very simple at all. I'm not overly concerned with introducing ambiguity in the lead. For example, a carcinoma is "a cancer", "a neoplasm", and "a malignancy". An equally deadly malignant non-carcinoma is not "a cancer" – depending upon the definition that you use for cancer. I am concerned about introducing significant errors; I am not concerned about introducing unimportant errors, of the sort that call leukemia "a cancer of the blood" instead of "a hematological malignancy".
  4. I think that prescribing the order of the lead's content is WP:CREEPy.
  5. It's not just BNF that recommends "by mouth" rather than "oral", and in general I think it's okay. But these recommendations are usually aimed at people writing labels on the bottles of prescription drugs, and they may not apply as directly to encyclopedia writing. On the more general subject, if you compare synonyms for everyday words, the "simpler" (shorter) word in English is usually more or less Germanic, and the "more complicated" word often has a cognate in Latin. So if your goal is to make things easier for a native English speaker with limited reading skills, you may make things harder (i.e., "more difficult" – hard and difficult is an example of this) for people who speak Spanish or French fluently.
  6. I thought that pronunciations had mostly been moved to infoboxes (when those exist, and they usually do). We usually give two or three alternate names, and we should resist the temptation to give 10 or 20.
  7. I have no position on the "minimize redundancy" vs "one-stop shopping" debate. WhatamIdoing (talk) 18:23, 26 December 2019 (UTC)[reply]
Back now with more time. Ian Furst, on point 6, I am not in disagreement with the text; as I stated, I think this territory is much better stated already at MOS:FIRST, MOS:LEADALT and MOS:LEADPRON, which do a better job of covering nuance and the possibility that there may be instances when they are needed. See WAID's point just above this on No. 7: it is a matter of whether we do a good job of repeating already available information. Both No. 6 and No. 7 are what WAID called "one-stop" vs. "minimize redundancy". IMO, if we have nothing useful to add, it is better not to spread guideline info across many pages that then have to be kept in sync.

WP:MEDLEAD shows every mark of inexperienced guideline or policy writers, laying down absolutes in inflexible terms, with no consideration for whether the text in a guideline syncs with policy.

For example, mandatory is not a word associated with citation found in any policy or guideline page anywhere else on Wikipedia.

That medical content in general should have BLP-style required inline citation to high-quality sources is something I have argued repeatedly, for years, and that has been rejected. How, then can we impose citations specifically in the lead, when the lead is a summary of the body, and we don't require them in the body? Now, if you want to push (as I have, for years) for a BLP-style policy on medical content within articles, with content then summarized to leads, well I'm on board, but that is not what this lead guideline discussion is about. A BLP-style requirement for citation on medical text has never been endorsed by WPMED or by the community. We can't mandate it in a guideline about LEADs only, without an RFC on whether we can require inline citations to high-quality sources for medical content, as we do for BLPs. For those arguing that point (2) is acceptable, they would have a stronger argument if they argued same for the body of articles, and used the kind of language that is used in guideline and policy writing (eg WP:BLP, WP:V, WP:RS, etc.) For those who want us to have similar requirements on medical content as we do on BLPs, I have been on board for years, and would love to see that RFC. This is not that RFC, and if we have a lead guideline that goes beyond what policy pages on Wikipedia state, we have a problem of being out of sync with policy. SandyGeorgia (Talk) 01:15, 27 December 2019 (UTC)[reply]

My personal views on the seven points raised:
0. Preamble: I don't agree with the suggestion to "Remove, replace, or rewrite sections in MEDLEAD to better reflect Wikipedia-wide WP:LEAD guideline". If MEDLEAD is to merely "reflect" MOS:LEAD, that is, merely mirror it, the there's no point in having distinct advice for leads of medical articles. MEDLEAD should be a refinement, an extension, of MOS:LEAD that takes into account the differences between the average article and medical articles. I believe that has been an agreed principle throughout the existence of MEDMOS, and I don't think we should be attempting to make medical guidance to editors no more than a reflection of what we already have elsewhere.
1. The work done in translating Wikipedia's medical content into many other languages is acknowledged as valuable. It is part of our mission not just to write an encyclopedia, but to ensure that its content can be disseminated to every person on the planet. We should be reminding editors of that fact, and I see no reason to remove guidance that asks them to keep it in mind. Nevertheless, I would have no problem with rewording the advice along the lines of

To facilitate broad coverage of our medical content in other languages, and to respond quickly to the need to make good medical content available in emergencies, the translation task force often translates only the lead. Adding citations to the lead then helps it stand on its own as a concise overview.

I've rephrased the second sentence using wording directly from MOS:LEAD.
2. I've always been led to believe that medical science is not always intuitive. The vast number of folk remedies, coupled with the desire by big pharma to promote their latest money-maker, means that we are continually faced with searching for the sources to refute inaccurate information. "Not in the source cited" must be one of my most used edit summaries. So my experience tells me that medical statements are much more likely than the average statement to be challenged. And it's usually me that's having to do the challenging, in order to preserve the accuracy of our content. Having citations in the lead makes it much easier for an editor to spot a rogue sentence that's been inserted, by the simple expedient of looking at the nearby references to see whether they support the text. Once again, though, I'm no fan of forcing editors to give citations, particularly where there is blue-sky stuff like "Influenza, commonly known as the flu, is an infectious disease" (of course whether it's caused by a virus or some other agent is the sort of claim that we ought to consider sourcing). I believe consensus is that there is greater benefit in adding citations, even in the lead, to medical articles than to many other topics, and I oppose removing useful guidance explaining that to editors. I would be happy to see the guidance rewritten, however, to remove the word "mandatory", as I don't find that helpful. On the other hand, I don't want to see editors removing citations from the lead without a very good reason. Perhaps something like

Medical statements are more likely than the average statement to be challenged, so editors are encouraged to add citations in the lead, particularly for complex or controversial topics.

I've rephrased the second part using wording directly from MOS:LEAD. Overcitation can be problematical, but one well-chosen citation to support each fact stated is unlikely to lead to the issues described in WP:CITEOVERKILL. If the presence of citations in the wikitext makes editing difficult, then I've found that the use of WP:List-defined references goes a long way to keeping the text clear when editing.
3. I don't have a problem with the proposed re-write, other than the phrase "error or ambiguity are not introduced". It may be a en-gb thing, but I'd use singular "is" for the verb there. And I'm not a fan of "great care"; just "care" works well for me.
4. I think it is important to help editors ensure that they have summarised all of the article's salient information in the lead. To that end, it makes sense to me to give advice to work through the content of the article sequentially when writing the lead. That makes sure you've missed nothing out, and I have always assumed that was the reason for the guidance. Of course, if someone wants to re-arrange the lead later to turn it into brilliant prose, there shouldn't be a prohibition on that, but I don't see that as an issue with the current wording. I wouldn't object if we wanted to re-word the guidance to something like

"When writing the lead, editors should ensure that they write a comprehensive summary of all of the main points of the article. One way to achieve this is to follow the order of the content in the body of the article."

Although that's not bad advice for any article, IMHO.
5. I'm always happy to see a better example used if we can find one.
6. I disagree with removing the guidance to "Avoid cluttering the very beginning of the article with pronunciations or unusual alternative names". MOS:LEAD guides us to "briefly summarize the most important points covered in an article in such a way that it can stand on its own as a concise version of the article ... It is even more important here than in the rest of the article that the text be accessible. Editors should avoid lengthy paragraphs and overly specific descriptions". MOS:FIRST guides us to be "wary of cluttering the first sentence with a long parenthesis containing alternative spellings, pronunciations, etc." In a medical article, the pronunciation of a drug is a trivial matter. When I go to the pharmacy to collect my atorvastatin, neither I nor the pharmacy assistant care whether we pronounce it exactly; we both know what it is. That's not the sort of content we need in the lead, especially not in the first sentence. We should be giving guidance to editors not to clog up the start of articles with such trivia. That applies double for alternative or trade names for medicines. The start of the Paracetamol article is not the place to list the dozens of alternative names it is sold under, that that is covered much better elsewhere in the article. The advice in MOS:ALTNAME is fine, generically, but really needs to be more robust for use in medical articles where the problem of alternative names is often much greater.
7. My preference is to give more advice to editors, rather than less, so I would prefer to see the guidance mentioned there retained.
I'll address the issue of disclosure in the relevant section. --RexxS (talk) 21:36, 27 December 2019 (UTC)[reply]
Nice work. I will examine the specifics as I have more time to think about how to address our areas of agreement as constrained by my poor design of the RFC. SandyGeorgia (Talk) 23:11, 27 December 2019 (UTC)[reply]
I have added a section on the talk page here to attempt to summarize the views so far and include areas where views are merging or diverging so that can be brought back to this page in summary form. SandyGeorgia (Talk) 16:54, 29 December 2019 (UTC)[reply]
  • Your feedback above, Clayoquot, is very helpful, and the confusion is entirely my fault for a malformed RFC, followed by attempts to fix it. If I could encourage you to delve past the confusion, in the Survey (continued) section, there are seven different items of MEDLEAD that need consideration (beyond citations in the lead). Some of those positions evolved as editors here worked towards consensus building on talk. One of those is whether we can use the word mandated, and there are other recommendations for wording and positions to view in that section. If I can trouble you to engage further, might you contemplate the seven different points under discussion in that section? SandyGeorgia (Talk) 18:14, 8 January 2020 (UTC)[reply]
    Hi SandyGeorgia. Thanks for asking, but I'll have to leave the details to others as I don't have time to properly think these issues through. Best, Clayoquot (talk | contribs) 06:02, 13 January 2020 (UTC)[reply]

Survey (continued)

[edit]

See discussion at the talk page of this RFC. Sample responses on the RFC have ranged from "no change needed" to "delete the whole MEDLEAD section of MEDMOS", with many partial positions in between those two. This table presents a summary of the range of possibilities raised so far. It may assist those wanting to examine each point of the contested text, and facilitate responses such as Support 3A, or Oppose 5B, for example. SandyGeorgia (Talk) 23:08, 30 December 2019 (UTC)[reply]

Current Position A Position B Position C Position D
1. To facilitate broad coverage of our medical content in other languages, the translation task force often translates only the lead, which then requires citations. Translation is done from simplified forks of our leads, not the leads directly. Citation clutter makes our articles harder to read by our readers. Excess citations may be included in the wikitext inside HTML comments, which are hidden from our readers, and then surfaced in the simplified translation-fork page. To facilitate broad coverage of our medical content in other languages, and to respond quickly to the need to make good medical content available in emergencies, the translation task force often translates only the lead. Adding citations to the lead then helps it stand on its own as a concise overview.
Citations in the lead have no effect on readers, but are valuable to editors maintaining text integrity.
No evidence that translating leads of medical articles improves medical outcomes has been presented. Remove this and related text from MEDLEAD, and conduct a separate RFC on whether LEAD should be rewritten to accommodate article translation for all articles, not just medical. Retain all text in points 1–7.
* No change is needed.
* MEDLEAD is in sync with LEAD.
Translation is tremendously important. Wikipedia's mission is to provide "every single person on the planet... free access to the sum of all human knowledge". Translation helps achieve that. The medical translation effort has resulted in more than 6 million words of text being translated.
2. Medical statements are much more likely than the average statement to be challenged, thus making citation mandatory. No evidence that this is more true than in biographical, economic or political articles, for example. Claiming citations are thus generally mandatory is in direct contradiction to policy, which is case-by-case. Medical statements are more likely than the average statement to be challenged, so editors are encouraged to add citations in the lead, particularly for complex or controversial topics.
Any editor who regularly helps maintain medical articles knows that statements are not only more likely to be challenged, but also require a higher standard of sourcing than the average Wikipedia article. The leads of complex and controversial topics are the prime target for POV-pushing and subtle vandalism, where citations are most needed, cf WP:BLP.
Adding sources to the lead is a reasonable practice but not required in all instances if the text in question is supported in the body of the article. Editors are encouraged to add citations in the lead for direct quotes, hard data, or statements likely to be challenged.
"Vulvar cancer is a cancer of the vulva, the outer portion of the female genitals" is not likely to be challenged. No evidence that medical statements are more likely to be challenged has been presented. Overciting leads constrains the ability to craft an overall summary of the article's content.
While BLP-style citation requirements may be appropriate for all medical content, even BLP does not mandate citations in the lead (see Donald Trump). A separate RFC about applying a BLP-style inline citation policy to medical content would be useful.
It's fine to have lots of citations in the leads of medical articles. It's acceptable for medical content guidance to have some differences from general guidelines, because medical content risks misleading our readers to make bad decisions about their own health care. "First, do no harm" trumps "first, do not violate MOS".
3. The lead of an article, if not the entire article, should be written as simply as possible without introducing errors or ambiguity. No need to repeat existing policy or guideline, which does not say "as simply as possible" but "as understandable as possible". There is a difference, as the latter requires good writing more than it requires baby words. The lead of an article should be written as simply as possible without introducing errors or ambiguity.
I would not have a problem if the guidance were to focus on making the text as accessible as possible or as understandable/comprehensible as possible.
The lead of an article should be as understandable as possible without introducing errors or ambiguity.
The clause "if not the entire article" extends beyond the cited quideline, which clarifies that there may be sections that have more technical aspects (pathophysiology comes to mind in medicine) and may not be written as simply as the lead.
"The content in articles in Wikipedia should be written as far as possible for the widest possible general audience" is similar to saying "The lead of an article, if not the entire article, should be written as simply as possible without introducing errors or ambiguity." Wikipedia has received a lot of criticism, both in the academic[6][7] and lay press, for being overly complicated. Wikipedia should try to approach a grade 12 reading level for the leads and has made progress towards this in the area of medicine. [8]
4. It is also reasonable to have the lead introduce content in the same order as the body of the text. Per WAID there is no justification for this for any article, let alone medical ones. When writing the lead, editors should ensure that they write a comprehensive summary of all of the main points of the article. One way to achieve this is to follow the order of the content in the body of the article.
This is a sensible practice and sufficient justification for the advice.
Editors should ensure that the lead is a comprehensive summary of all of the main points of the article. One way to achieve this is to follow the order of the content in the body of the article, although no specific order is required.
Prescribing the order of the narrative is CREEPy; the additional yellow-highlighted text is needed to prevent the leads of articles, including Featured articles, from being diminished by imposing a particular order.
5. The British National Formulary for example often uses "by mouth" rather than "oral". Per notes on RFC, the BNF does not prefer "by mouth" to "oral" at all. This deceptive statement should be removed. Examples should not be removed, but should be replaced by better ones. The present example is not deceptive, as the BNF actually does use "by mouth" as well as "oral". Most languages in the world translate "by mouth" exactly, including French, Portuguese and Romanian. Remove. Problems have been raised with every example given of words that can be substituted. This information is not needed in the MEDLEAD section, as it is already covered in the same guideline at the sections on Writing for the wrong audience and Careful language.
6. Avoid cluttering the very beginning of the article with pronunciations or unusual alternative names; infoboxes are useful for storing this data. No medical reason for this. Adequately covered by other guidelines. Avoid cluttering the beginning of the article with pronunciations or multiple alternative names. Remove here, this widespread information is already better covered at the guidelines on first sentences, alternate names and lead pronunciation.
Avoid duplicating information across guideline pages that is already covered better elsewhere.
This is a matter of deep concensus within the medical community, and is due to the preponderance of alternative names, sometimes upwards of 10 (solved well by using the other names=-parameter in Template:Infobox medical condition (new)). This provision has been included in WP:MEDMOS for years, and overturning it would disrupt readability and searchability of articles — and demand extraordinary efforts to enforce while only resulting in worse articles. I believe this justifies its inclusion despite contradicting suggestions from WP:LEADALT and WP:LEADPRON.
7. Miscellaneous:
* Language can often be simplified by using shorter sentences, having one idea per sentence, and using common rather than technical terms.
* It is useful to include citations in the lead, but they are not obligatory.
* Around a third of readers of English Wikipedia, have English as a second language.
* Most readers access Wikipedia on mobile devices and want swift access to the subject matter without undue scrolling.
* File:2019 Wikipedia reader native language by language.png
Remove. This is gratuitous agenda-pushing and has no place in any guideline. We have other guidelines to recommend writing style. Retain Indifferent, but support minimizing redundancy in guidelines over "one-stop shopping" that replicates other guidelines, and avoid repeating info here that is better covered elsewhere. Most of this information either is better covered elsewhere, or belongs elsewhere. It is acceptable for medical content guidance to have some differences from general guidelines, because medical content risks misleading our readers to make bad decisions about their own health care. "First, do no harm" trumps "first, do not violate MOS".
  • Position D; No change needed/RfC is disruptive"Oppose change to the guideline on the grounds the current RfC is too broad. There are too many questions raised all at once — causing confusion and making it unlikely that actionable consensus will arise.
    Further points:
    • The question of whether the guideline is out of sync with policy has been answered — as an invalid question.
    • Guidelines clearly can and do act to extend upon other policies and guidelines — with suggestions as to how certain articles are best written.
    • As a guideline, MEDMOS has included the provisions that are up for discussion here for at least the past five years, without evidence of any major issues caused by any of the non-mandatory suggestions detailed in points 1 through 7.
Continually raising this debate is WP:GRIEFING, as is evidenced by WP:FORUMSHOPPING on a number of different pages.
It should also be noted that this extension of the RfC was posted with 9 hours notice for comments [9] on New Years Eve — and most of it was written in the few days between Christmas and New Years, when it was known that many users could not participate. Position D does not accurately represent the position of opposition to the RfC, but it is the best one here.
Carl Fredrik talk 09:05, 31 December 2019 (UTC)[reply]
There is neither griefing nor forum shopping here. Let's have this content discussion without accusing those with whom we disagree of WP:ALLCAPSSHORTCUT misconduct. Levivich 16:13, 31 December 2019 (UTC)[reply]
Arbcom on local consensus: Where there is a global consensus to edit in a certain way, it should be respected and cannot be overruled by a local consensus. WPMED cannot mandate citations in leads. SandyGeorgia (Talk) 22:14, 1 January 2020 (UTC)[reply]
Arbcom can't create or amend policy. The policy at WP:CONLOCAL contains the sentence "... participants in a WikiProject cannot decide that some generally accepted policy or guideline does not apply to articles within its scope", but what generally accepted policy or guideline would not apply if a project decided to require citations for articles within its scope? --RexxS (talk) 09:57, 2 January 2020 (UTC)[reply]
WP:LEADCITE. CONLOCAL says that WPMED cannot overrule the MOS's statement that "The presence of citations in the introduction is neither required in every article nor prohibited in any article" and declare that citations are required in the lead of every article within its scope. However, CONLOCAL is about what WPMED (a group of people) can do; it is not about what the community-wide guideline (MEDMOS) can do. If the whole community says that MEDMOS ought to require citations in the lead of every article about medical conditions, then we should deal with the resulting WP:PGCONFLICT by changing LEADCITE to say "The presence of citations in the introduction is neither required in every article nor prohibited in any article, except for articles primarily about health content, which are required to have a citation after every sentence", or whatever editors decided. WhatamIdoing (talk) 01:27, 8 January 2020 (UTC)[reply]
There is no more PGCONFLICT in MEDMOS requiring citations in the lead that there is for BLP requiring inline citations for any controversial statements in the leads of BLPS. That's because guidance that says "in the general case we don't require or prohibit something" is not contradicted by guidance that requires or prohibits something in particular cases. We haven't changed LEADCITE to say "The presence of citations in the introduction is neither required in every article nor prohibited in any article, except for articles primarily about biographies of living persons, which are required to have inline citations for every statement that is controversial or likely to be challenged". So why would we feel the need to do a similar change in this case? --RexxS (talk) 02:27, 8 January 2020 (UTC)[reply]
I don't think that analysis is sound. The difference between WikiProject WhatamIdoing making up rules, and the whole community making up rules, is the critical point in that sentence of CONLOCAL. WikiProject WhatamIdoing is welcome to type out "rules" until her fingers bleed, but I can't require that my rules be followed, even if I pound on my keyboard and say those articles are within my scope. However, the whole community could type the same rules, and compliance with those absolutely can be enforced. MEDMOS does not belong to WPMED. The whole community can put whatever rules it wants on this page, and they will never violate that line in CONLOCAL.
And if (as happens from time to time), the whole community says on one of its own global-consensus pages (say, LEADCITE) that something is never required for any article, and on another of its global-consensus pages (say, MEDMOS) that the same something is always required for some articles, then the community (not WPMED!) has made self-contradictory rules, and the community needs to adjust one or both of its global-consensus pages so they stop being self-contradictory. WhatamIdoing (talk) 05:55, 8 January 2020 (UTC)[reply]
Well, I'm pretty sure your analysis is not sound. It founders on the first step where you assume that MEDMOS is the creature of WPMED. It is not. It is part of the Manual of Style, and the first line you read states: "This guideline is a part of the English Wikipedia's Manual of Style." where "guideline" links to Wikipedia:Policies and guidelines. That makes it very clear that these guidelines "are developed by the community ...". It doesn't matter how ineffectual WikiProject:WAID or WikiProject:MED will be in enforcing anything; these guidelines carry the force of community consensus. I'm sorry but your argument beyond the first three sentences don't make sense to me. Have you missed the point that this page (MEDMOS) is one of the community's own global-consensus pages, not a subpage of WPMED? You could be forgiven for thinking we are still discussing at WTMED, given the amount of churn that is currently occurring.
Therefore, I stand by my reasoning that a general piece of guidance/policy/rule is not required to mention every specific exemption. All guidelines, including LEADCITE, are subject to the disclaimer "... best treated with common sense, and occasional exceptions may apply." Wikipedia is not a bureaucracy, and there is no genuine conflict to resolve.
Nevertheless, I am on record as stating that I don't believe it is helpful to try to coerce any editor to add citations to a lead if they don't want to. I think that is counter-productive, and I'd like to see our guidance softened, so that we are simply encouraging editors to add citations, for reasons that are well-rehearsed elsewhere. I also think that we must guard against swinging the pendulum too far in the other direction: it would be unacceptable, IMHO, if amended guidance were to be used as a justification for an editor to strip articles of their lead citations, based solely on an ill-founded concept of "readability" for example. Hopefully there is a sensible middle-ground, and editors may find value in moving toward it. --RexxS (talk) 20:34, 8 January 2020 (UTC)[reply]
I think we already agree: It can't technically be WP:CONLOCAL, because MEDMOS isn't one of the hundreds of Wikipedia:WikiProject advice pages, and if there's a conflict between MEDMOS and LEADCITE, it's just a garden-variety WP:PGCONFLICT. WhatamIdoing (talk) 16:16, 9 January 2020 (UTC)[reply]
  • Having worked on the talk page to come towards closer consensus on points 1–7, some of my positions have changed from my first !vote, to the position of Column C.
    With respect to the other sample positions offered so far:
  1. Both A and B are reasonable; there is no evidence for D.
  2. Both A and B are reasonable; D does not encompass policy, adding citations to medical leads has not and does not prevent those leads from being misleading or unhelpful (see Talk:Schizophrenia).
  3. Both A and B are reasonable; D does not allow for more technical sections of articles.
  4. Both A and B are reasonable; no one yet has offered an alternative D.
  5. Both A and B are reasonable; no one yet has offered an alternative D.
  6. and 7, I generally don't care, but find it unhelpful to spread information across multiple guideline pages, and prefer to reference the extant guideline pages when they already cover the territory.

SandyGeorgia (Talk) 15:16, 31 December 2019 (UTC)[reply]

Outcomes

[edit]
Regarding Levivich's statement:

... generally oppose the notion that a primary purpose of enwiki medical article leads should be ease-of-translation. It's more important that we accurately and readably state the information in English than that we make it easy to translate that information into another language. I find the thought that people who speak a certain language are getting their medical information primarily from Wikipedia deeply frightening. We are not competent for that task. We are competent to write a tertiary source that summarizes secondary sources. We shouldn't even be trying to be some kind of WebMD.

It is easy to see that the goal of putting translated medical information in to the hands of people in developing countries is laudable, but do we actually have anything close to an outcome measure that reinforces the idea that giving people access to the lead of, for example, common cold, actually has any effect on overall health? SandyGeorgia (Talk) 23:07, 27 December 2019 (UTC)[reply]

I think it's worth remembering that it's not just a question of translation. A sizeable proportion of the world reads the internet in English by default. That means that even the kid in sub-Saharan Africa has a chance of having a rudimentary knowledge of English. WMF recently published the statistics, so we know that our content is read in many countries that don't have English as their first language. Whatever we can do to make our medical articles understandable, even for a reader who will struggle with the task, is going to be worthwhile. I don't know of any outcome measures that can be reliably applied, because there are going to be too many confounding factors, but anecdotally there is a fair body of evidence that our content is consulted whenever there is an outbreak of a disease in the less developed world. --RexxS (talk) 23:39, 27 December 2019 (UTC)[reply]
So we have data that indicates our content is read: do we have any indication that reading a Wikipedia lead (which is limited information, by definition) improves health outcomes, or is this venture only benefitting the careers of the editors getting their Wikipedia work published in journals? No matter how understandable we make them, I doubt we have any indication that reading Wikipedia leads (in any language) affects medical outcome. SandyGeorgia (Talk) 00:18, 28 December 2019 (UTC)[reply]
We have spikes in the relevant page views and increased readership in affected areas when a disease outbreak occurs. I've always felt that giving advice such as to treat diarrhoea by giving salt and sugar in water would be likely to improve results when compared to folk remedies like starving the affected person. But I can't prove it, and I doubt that any meaningful trial could be constructed. Nonetheless, I still optimistically think that trying to spread accurate information by making it as simple to understand as possible is a worthwhile goal. In many articles, the lead presents a simplified summary of the article – after all, we often tell folks not to put too much detail into the lead, and the proper place for it is in the body of the article. For that reason, I contend that the lead is the best target for us to use the simplest language and to aim for a low reading age, goals that may not be possible in many cases in the body of the article itself. Coincidentally, I'm just about to have paper published which I collaborated on with the Cochrane schizophrenia team at Nottingham. Of course, as I've now been retired for over seven years, I'm not expecting any great benefit to my career, but it may improve my Erdős–Bacon number. --RexxS (talk) 01:41, 28 December 2019 (UTC)[reply]
That's a fair answer :) :) SandyGeorgia (Talk) 01:43, 28 December 2019 (UTC)[reply]
With all the work early this year on Videowiki and the SDG mandate for Wikimania I looked at a lot of this in more detail (plus I do some work with less developed areas of the world). (a) There is excellent proof that non-clinical interventions save lives, (b) Wikipedia is likely the most used medical resource on the planet even for health care providers (although this is regional). Take a look at dengue fever during outbreaks or arabic wikipedia during the last cholera outbreak (FYI we could see a spike before the WHO announced). There is good evidence this is more than just a I-saw-it-in-the-media effect, but instead, real information seeking. Imo we should be proud and do everything possible to ensure our information is accurate (and readable). Ian Furst (talk) 01:20, 29 December 2019 (UTC)[reply]
I have yet to see any evidence that backs up the idea that health outcomes are improved because Wikipedia articles are clicked on during widespread or highly publicized disease outbreaks. SandyGeorgia (Talk) 02:17, 29 December 2019 (UTC)[reply]
We all agree that the primary purpose of enwiki medical article leads is not for ease-of-translation. It is more important for the EN WP leads to be written using easier to understand language for those reading the leads in English. Doc James (talk · contribs · email) 02:40, 31 December 2019 (UTC)[reply]
If we all agree on this, then why does WP:MEDLEAD specifically give ease of translation as a reason for requiring citations in the lead? SandyGeorgia (Talk) 03:02, 31 December 2019 (UTC)[reply]
It is a secondary reason which IMO does not interfere with the primary reason. Doc James (talk · contribs · email) 06:06, 3 January 2020 (UTC)[reply]
User:SandyGeorgia This is both complex and simple. In order to prove what you want we would have to randomize people into groups. This is likely impossible. The only approximation for randomization in this setting is a matched case control cohort study. I have a group in Toronto that could do this, if funded, but I think Wikipedia has better ways to spend 100k. In a study like this we find two groups, one with and one without WP then match them based on age, disease, health access, etc... This is actually is a decent approximation for randomization if done well. Since we have not done this, it leaves us with other more basic questions; 1) do people use Wikipedia during outbreaks - unequivocally yes (regionally) and I can share data with you. We see it in Dengue, flu outbreaks, etc... 2) does knowledge sharing/health literacy improve outcomes? Unequivocally yes depending on disease, think about education campaigns in HIV, diabetes, hypertension, communicable diseases, etc... Given; WP is the most frequently used health resource on the planet and health literacy equates to improved outcomes we should absolutely assume that the information shared is impacting health outcomes until proven otherwise. Ian Furst (talk) 03:06, 31 December 2019 (UTC)[reply]
"Correlation is not causation." Levivich 03:16, 31 December 2019 (UTC)[reply]
Thanks for laying that out for readers who might not understand how controlled studies are conducted (I hope all of us editing medical articles do :) I doubt that the kind of information we put into leads of Wikipedia articles-- and that is being constrained and dictated for the purposes of translation-- affects outcomes. Remember, Wikipedia doesn't give medical advice, and our leads do not include the kind of information that is likely to affect outcomes anyway. There is nothing at dengue fever that would have affected my outcome when I had it if I had read that. People in the developing country where I lived knew more about dengue than we can put in a lead. That's my N=1 response (well, more than 1 since I sure wasn't the only person I knew who had dengue fever in South America). (In fact, the information that would have helped me to a faster diagnosis is not only not in the lead-- it's nowhere in the article, but I digress.) It's a laudable goal, Ian, but the blind pursuit of this goal is compromising our content overall and costing us resources we can ill afford to lose, in editors who could be generating useful content in our mostly inaccurate bodies of articles. We may be generating some nice research papers for editors engaged in these pursuits, but we still have zero evidence we are helping anyone. The most we can get from data is how many people click on an article during an outbreak-- not whether they found anything useful there, much less anything that altered an outcome. SandyGeorgia (Talk) 04:17, 31 December 2019 (UTC)[reply]
User:SandyGeorgia, we definitely have a difference of opinion on the power of information, especially at the scale of Wikipedia. Levivich"Absence of evidence is not evidence of absence" Ian Furst (talk) 14:25, 31 December 2019 (UTC)[reply]
And it's okay to disagree, Ian Furst; I appreciate your good-faith and collegial engagement. I hope you see that my broadest concern is that we have lost competent editors because of the problems throughout WPMED, which are generally evidenced in lead editing. Regardless of how this little (malformed) RFC debacle ends, consider the broader question: will Wikipedia regain competent medical editors, or will we continue to see a focus on formulaic editing of leads by editors who are not topic experts for aims other than those of en.Wikipedia? I see you editing competently in your area of expertise, where you know what you're talking about; kudos to you :) SandyGeorgia (Talk) 14:41, 31 December 2019 (UTC)[reply]

No. 5, "by mouth" vs. "oral"

[edit]

Per the discussions above, the suggestion that we need another example because there is not universal agreement that all of these changes are for the better, can we agree that

  • etiology --> causes

is a better guideline example than

  • oral --> by mouth?

Oral and renal are better word choices for translation to romance languages, and a guideline should not encourage bot-like editing of articles to alter carefully chosen language. Perhaps we can all agree that changing the word etiology in leads to causes is less likely to be disputed, and choose that as an example of wording to consider changing. SandyGeorgia (Talk) 00:27, 28 December 2019 (UTC)[reply]

If memory serves, etiology is a somewhat more complicated and precise concept than cause. Elsewhere on the page, we use the kidney-vs-renal example, which has the virtue of being a sensible synonym in every case I can think of –whereas oral surgeons would probably not consider themselves to be "by mouth surgeons". Of course, if you're looking for a cognate in a Romance language, using kidney doesn't help you, but the fact remains that English word for that body part derived from Proto-Germanic is more common than the English word derived from Latin.
And, since it's mentioned elsewhere in MEDMOS, do we really need to repeat this point in the lead section? WhatamIdoing (talk) 23:46, 29 December 2019 (UTC)[reply]
You raise a good point: it is already covered in two other sections, and there are problems with the etiology --> cause example as well, so I think I'll go with removing it entirely from the LEAD section. SandyGeorgia (Talk) 04:52, 30 December 2019 (UTC)[reply]
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.