Wikipedia talk:WikiProject Medicine

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by Looie496 (talk | contribs) at 16:24, 13 July 2018 (→‎color x-ray: comment). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

    Shortcut

    Welcome to the WikiProject Medicine talk page. If you have comments or believe something can be improved, feel free to post. Also feel free to introduce yourself if you plan on becoming an active editor!

    We do not provide medical advice; please see a health professional.

    List of archives

    ICD-11

    Has been released. https://icd.who.int/ frameless|right Doc James (talk · contribs · email) 15:38, 20 June 2018 (UTC)[reply]

    Is it time to migrate ICD-11 to a separate article? Looie496 (talk) 13:03, 21 June 2018 (UTC)[reply]
    I'm in favour of a early WP:SPINOFF for this. Although the ICD article is "only" ~36kB; it'll be more work to WP:SPLIT later (ie once ICD-11's adoption begins).
    At some point; it'll also need adding to {{Template:Medical resources}}. Little pob (talk) 15:37, 21 June 2018 (UTC)[reply]
    Support both those suggestions. Doc James (talk · contribs · email) 19:34, 21 June 2018 (UTC)[reply]
    ↑Support. I'm not an expert on ICD, but from what I've read, ICD-11 introduces significant changes/developments in how to best conceptualize, organize, identify/label, and code diseases, injuries, etc. So another important reason to create a separate article now. ¶ I have read the psych section; submitted 5 proposals and commented on a couple. It is fantastic that WHO has--and will continue to--solicit input from anyone who has access to the Internet and can support their proposal(s) with scientific evidence. On the other hand, WHO has made a mistake that so many science organizations (and other groups) make. It goes something like this: "We are some of the most intelligent people in the world. Therefore, we don't need some snooty librarian types nitpicking how we write things up." Consequently, at least in the psych section, the median writing quality score is about a "C", which is not good for a classification system that seeks achieve precision, reliability, and validity.   - Mark D Worthen PsyD (talk) 01:51, 29 June 2018 (UTC)[reply]
    agree as well--Ozzie10aaaa (talk) 10:47, 8 July 2018 (UTC)[reply]

    XY female -- redirecting or creating an article

    At XY female (edit | talk | history | protect | delete | links | watch | logs | views), we have an IP edit warring when it comes to "XY female" redirecting to XY gonadal dysgenesis. The IP keeps redirecting it to XY Female, which is not a Wikipedia article. Finally, Natureium created a small article under the "XY female" title, and I changed the redirect back so that it points to XY gonadal dysgenesis because I considered the small XY female article an unnecessary WP:Content fork and that we should consider what Wikipedia:Manual of Style/Medicine-related articles#Article titles states. We also do not have an article titled XX male; it redirects to XX male syndrome. IVORK and Insertcleverphrasehere, who were involved, might also want to weigh in.

    Thoughts? Should "XY female" be a redirect, a disambiguation page (like Insertcleverphrasehere felt it should be), or a small article? Doc James, any idea on what should be done? Flyer22 Reborn (talk) 16:22, 26 June 2018 (UTC)[reply]

    I don't think a separate list of conditions that can result in this phenotype is necessary, but after being reverted by multiple people, I gave in and created a stub. Regardless, a disambiguation article is incorrect. A disambiguation page disambiguates pages with the same name. This is a short list of conditions with different names, and if needed, the page would be a set index. I also don't understand why among all the edit warring, no one bothered to fix the name. Natureium (talk) 16:47, 26 June 2018 (UTC)[reply]
    It's always tough to communicate with IP-hoppers, even though they're trying to work in good faith. In the case of this topic, the hot-button sociological issues only complicate the discussion. Move-warring can be a messy business. Better to start a structured discussion somewhere obvious in article talkspace, such as the redirect target's talkpage. Once the IP engages there, recommend they register an account. LeadSongDog come howl! 20:52, 26 June 2018 (UTC)[reply]
    This ref supports the current redirect.[1] Do we have other refs which say it also means other things? Doc James (talk · contribs · email) 17:57, 26 June 2018 (UTC)[reply]
    Can't see that ref for some reason DocJames. Battleofalma (talk) 11:04, 27 June 2018 (UTC)[reply]
    @Battleofalma and Doc James: That's because Google books limits the availability of the pages that it displays by country. Since James posted a link to books.google.fr, you can either set your vpn to Montreal (as I did before I spotted a simpler solution) and view the book; or you could follow this link https://books.google.com/books?id=wGoj9RtTcVIC&pg=PA507#v=onepage&q&f=false instead. Cheers --RexxS (talk) 20:18, 27 June 2018 (UTC)[reply]
    • Comment Note that XY Female (with a capital F) exists and should either be a redirect to XY female or (if that is redirected) redirect to the same target as that page. I don't have an opinion about which is more desirable. Thryduulf (talk) 12:42, 27 June 2018 (UTC)[reply]
    This I have tried pointing out numerous times. Natureium (talk) 13:38, 27 June 2018 (UTC)[reply]
    I think that we should redirect it to Sex reversal or make a short article. Redirection to XY gonadal dysgenesis is inappropriate. --Sharouser (talk) 13:02, 27 June 2018 (UTC)[reply]
    Sharouser, why do you feel it's inappropriate? Is it because of what Trankuility stated below? If we can demonstrate what the WP:Primary topic is, we can go by that. Flyer22 Reborn (talk) 05:05, 29 June 2018 (UTC)[reply]
    Same as Trankuility stated below --Sharouser (talk) 06:54, 9 July 2018 (UTC)[reply]
    Do you have sources supporting that? Natureium (talk) 23:05, 27 June 2018 (UTC)[reply]
    I’m not here very much these days, but both of these came up using simple, self explanatory Google searches:
    A transgender woman: Cycling Canada, Ontario Cycling Association, the Union Cycliste Internationale (UCI) and Canadian athlete Kristen Worley settle human rights application to promote inclusive sporting environments, 18 July 2017.
    AIS XY female”: XY Female with Complete Androgen Insensitivity Syndrome with Bilateral Inguinal Hernia, Rema V. Nair and S. Bhavana Trankuility (talk) 23:18, 29 June 2018‎ (UTC)[reply]
    NPP here, any further points to be made? An IP just remade the page, and it doesn't look like there is much of a consensus here. Any ideas? Xevus11 (talk) 04:54, 7 July 2018 (UTC)[reply]
    I have reverted the IP hopper and will request semi-protection of the page. There clearly is no consensus for creation, and the IP is impeding on the consensus process. Flyer22 Reborn (talk) 10:23, 7 July 2018 (UTC)[reply]

    Wikipedia:Frequently misinterpreted sourcing policy

    Some here may want to weigh in on this at Wikipedia talk:Frequently misinterpreted sourcing policy. Flyer22 Reborn (talk) 06:02, 1 July 2018 (UTC)[reply]

    commented--Ozzie10aaaa (talk) 11:42, 2 July 2018 (UTC)[reply]
    Fairly daft essay acting as a coatrack for some very daft axe-grinding. We've had this kind of thing before and procedurally it's probably just best to let it sink into well-deserved obscurity: it is just after just all an editor's opinion. Alexbrn (talk) 16:55, 2 July 2018 (UTC)[reply]
    The difference as I see it is that advocacy ducks is less wrong and more just incoherent, while this essay includes a whole bunch of falsehoods and mistakes regarding the distinction between objectivity and neutrality as well as that between original/primary research and secondary sources/reviews. It does however include one very constructive discussion, though it phrases it in a less than useful manner, simply assuming that what is written in that essay will become the de facto methodology for Wikipedia. The wish to include primary sources together with secondary sources is a good one, and acknowledges a problem in academia where citations point to sources in support of statements which in turn cite something else, which in turn cites something that doesn't even support the statement it is later used to support. I've meant to write an essay on WP:Citing citations. However, there are three problems with the idea of simply tacking primary sources to secondary sources:
    • Technical — we need to differentiate between primary and secondary sources clearly in the reference list, and these is no template to tie primary sources to secondary sources.
    • Risks promoting biased results — Even if we require secondary sources next to all primary sources, the inclusion of primary sources will increase their presence on WP, and thus make them seem more important. The selection of which primary source is included is just as likely to be as biased as any primary source included on its own.
    • Time — any introduction of primary sources next to secondary sources will require more work from editors to ensure that bias is not included in the selection of which primary sources are chosen for inclusion — and time is required to vet not only the secondary source, but also that the primary source is the best one of all the primary sources cited.
    So to cut a long story short, I disagree with everything EDIT: much in the essay, as anyone who is interested enough to want a primary source can read the secondary sources we cite and find the original research from there. Carl Fredrik talk 20:18, 2 July 2018 (UTC)[reply]
    I wonder if Hildabast has anything to say about this. Carl Fredrik talk 20:22, 2 July 2018 (UTC)[reply]

    The title "Frequently misinterpreted sourcing policy" is not what the page is about. It claims "This is a list of key points of frequently misinterpreted sourcing policy, guidelines, and community norms at Wikipedia."[2] It is not about that. It is a disagreement with sourcing policy. It is not about misinterpreting policy. If it was about misinterpreting policy then it would be providing specific information on what is being misinterpreting rather than dispute the reliability of sources such as the US FDA. It is about "Frequent disagreements with sourcing policy" or "Frequent disputes with sourcing policy". The title should be changed to reflect what the page is about. QuackGuru (talk) 15:11, 7 July 2018 (UTC)[reply]

    This appears to have originated in a very different context at Wikipedia:Village pump (policy)#Not a democracy?. WhatamIdoing (talk) 18:25, 7 July 2018 (UTC)[reply]

    Postmenopausal hormone therapy article title

    What to call this article? See Talk:Postmenopausal hormone therapy#Strange title - is this "postmenopausal hormone therapy"?. A permalink for the section is here. Flyer22 Reborn (talk) 08:04, 1 July 2018 (UTC)[reply]


    unassessed articles

    any help w/ Medical unassessed articles which is around 500 would be appreciated, thanks--Ozzie10aaaa (talk) 14:44, 3 July 2018 (UTC)[reply]

    It's been in the thousands before. The easiest way to reduce the sheer numbers is to find the BLPs in Category:Unassessed medicine articles and paste {{WPMED|class=Stub|importance=Low|society=yes}} on most of them (on the talk page, to replace whatever WikiProject Medicine template is currently there). (Most of biographies are stubs.) This is easy work that even inexperienced editors can be successful at. (In fact, assessing articles is how I got involved in this group.) WhatamIdoing (talk) 21:20, 3 July 2018 (UTC)[reply]
    I got through about 100 of them. We have some interesting and some strange articles. Natureium (talk) 23:25, 3 July 2018 (UTC)[reply]
    thanks Natureium--Ozzie10aaaa (talk) 00:06, 4 July 2018 (UTC)[reply]
    A number of them are typically also paid for articles created by socks. Remember to mark those as appropriate. Doc James (talk · contribs · email) 11:03, 8 July 2018 (UTC)[reply]

    Conflicts of interest in academic publishing

    Conflicts of interest undermine the reliability of some health- and science-related academic journal articles cited on Wikipedia. The Sponsored Point of View panel discusses this problem in 2012

    The contents of the conflicts of interest in academic publishing article may be of interest to editors here. The article is currently in DYK on the main page, and extensively discusses medical articles (the medical community writes about this subject more than other academic fields). While it's extensively referenced, it is not well-written (I wrote most of it, so I should know), with a rather atomized structure and many statements which are more a summary of the references than a phrase in an article. Edits and criticism would be very welcome, if anyone feels so inclined.

    Thanks to Headbomb for prompting me to write this article (and to both Headbomb and others for contributing to it). HLHJ (talk) 16:26, 3 July 2018 (UTC)[reply]

    I did? I don't remember doing so, but if the end product is better coverage of encyclopedic topics, I'm all for it. Headbomb {t · c · p · b} 16:30, 3 July 2018 (UTC)[reply]

    Links to DAB pages

    Despite this WikiProject's valiant efforts to keep on top of these problems, new medicine-related links to DAB pages which need expert attention keep getting created all the time. This is my latest batch. As always, if you do solve a problem, take off the {{dn}} tag, and post {{done}} here.

    Thanks in advance, Narky Blert (talk) 21:21, 3 July 2018 (UTC)[reply]

    Help wanted

    For a number of years we have been experiencing a steady decline in the number of administrators as a result of attrition and a declining number of editors willing to consider adminship. Things have reached a point where we are starting to experience chronic backlogs in important areas of the project including noticeboards, requests for closure, SPI, CSD & etc. If you are an experienced editor with around two years (or more) of tenure, 10k edits give or take and no record of seriously disruptive behavior, please consider if you might be willing to help out the community by becoming an administrator. The community can only function as well as we all are willing to participate. If you are interested start by reading WP:MOP and WP:RFAADVICE. Then go to WP:ORCP and open a discussion. Over the next few days experienced editors will take a look at your record and let you know what they think your chances are of passing RfA (the three most terrifying letters on Wikipedia) as well as provide you with feedback on areas that might be of concern and how to prepare yourself. Lastly you can find a list of experienced editors who may be willing to nominate you here. Thank you and happy editing... [Note:This page may not be on my watchlist so if you want to reply to me, please either ping me or drop me a line on my talk page.] -Ad Orientem (talk) 01:49, 4 July 2018 (UTC)[reply]

    I'm not qualified to be an admin, but if I were I would think thrice about throwing my name into the hat. Some of the discussions about proposed admins that I've read have seemed brutal. I found myself thinking, "Why would anyone volunteer to be stretched on the rack for a stressful, mostly thankless, unpaid job?" I offer my impression simply as data to consider. I could be projecting or misinterpreting what happens.   - Mark D Worthen PsyD (talk) 05:38, 4 July 2018 (UTC)[reply]
    That reminds me: Another reason to not be an admin is that some hackers target admin accounts. This is partly done to put spam on the main page, but it's also partly because if you want to install malware, then what better way to do it than to hack into the account of someone who can instantly install Javascript, with no pesky, interfering code review processes, on one of the most widely visited and trusted websites in the world? Even five minutes could infect thousands upon thousands of machines. Oh, and once you're in, there's apparently a way to steal other people's accounts, so one weak password on just one admin's account can cause problems for everyone. So for those of you who are already admins, please, please, please use a unique and l-o-n-g password for your accounts (I believe that MediaWiki will accept passwords significantly longer than 200 characters), and then go read m:Help:Two-factor authentication and consider setting it up. Also, the devs (including volunteer devs) have been talking about splitting this risk off from regular admins for about a decade, and I'm hoping they'll finally get that done this year. So things may be getting better soon. But in the meantime, protect your accounts. WhatamIdoing (talk) 19:03, 5 July 2018 (UTC)[reply]
    There's really only one thing that's important in making a strong password: length. A truly random string of just 12 lower case letters will take a couple of hundred years to crack by brute force using today's technologies. All the rest of the stuff that so-called "security experts" shove down our throats is just theatre that they conjure up to justify their existence. If you think I'm being too optimistic, use 16 letters instead. That would probably keep you safe for about 90 million years. The electric costs for the hacker would exceed the sum of the world's worth, which ought to discourage most of them. --RexxS (talk) 20:42, 6 July 2018 (UTC)[reply]
    RexxS, From your description above it appears that randomness is also a factor. How much is the strength compromised by being reasonably easy to memorise? · · · Peter (Southwood) (talk): 07:09, 7 July 2018 (UTC)[reply]
    WhatamIdoing, does Mediawiki accept passwords with spaces, or does it assume that the first space indicates the end of the password? · · · Peter (Southwood) (talk): 07:09, 7 July 2018 (UTC)[reply]
    @Peter: A lack of true randomness is only weaker if the hackers know the way in which the randomness is deficient. Against a simple brute-force attack, it's not very important. For Wikipedia I use a pass-phrase – that is, a series of words, including spaces and some uncommon words. It is naturally easier to remember, but has to have many more characters than a random sequence. A phrase with 10 words would take about the same time to crack via a 4000-word dictionary attack as 25 random letters would take using brute force (about 500 million million million years and the hackers won't have any electricity by then). Of course it feels like it takes almost that long to type it in. --RexxS (talk) 11:52, 7 July 2018 (UTC)[reply]
    I don't think that randomness is important, but "anti-non-randomness" probably is. Common passwords (e.g., qwerty, password, google, 123456) should be strictly avoided. Anomie can probably tell us how spaces are processed in MediaWiki. (I know that MediaWiki won't complain if you use a password that contains spaces, but I don't know if they're ignored in the calculations.)
    What you need to know: Special:ChangeCredentials is your friend (and will instantly ask for your current password). Some web browsers (including Firefox, but not Safari) will show the date when you saved your password there; that can give you an idea of how long yours has been around.
    Also, please don't look at RexxS's numbers and think it means a 12-character password is unassailable. It's probably "good enough" – and it is dramatically better than the average 5-character-long password – but you'll want to keep in mind that those 200-year estimates are usually based on a single consumer-grade computer, which is a rather quaint approach in a world with 20 billion internet-connected computers, many of which can be "borrowed" by any determined hacker. A modest cracking array will be able to crack a 12-character, all-lower-case password in a year or less, and a mid-sized botnet will be much faster. A dozen characters is "good enough" because almost none of us are going to have someone who's going to dedicate thousands of dollars' worth of effort to discovering our passwords, but cracking a 12-character, all-lowercase password is technologically feasible. Most of us only need to defend against the more casual attackers – the guy who's manually testing all the accounts to see who's used popular bit of profanity as their password, or someone who's got a list of passwords from a different cracked website and is seeing whether anyone has the same username and password here.
    If this subject interests you, then you might also be interested in https://www.grc.com/haystack.htm This interactive tool shows how long it would take to check every possible password of a particular type (e.g., all possible combinations of 12 lowercase characters: centuries for a single computer, but mere minutes for a massive cracking array). Statistically speaking, you can safely assume that your password is going to be towards the middle of this range (i.e., neither the first one attempted nor the last). Finally, the good news here is that the Technology folks will notice if a major hacking attempt is underway, and they can take steps to discourage it. But you really can help by setting up a decently long password (everyone) and adding two-factor authentication (if you're an admin). WhatamIdoing (talk) 18:22, 7 July 2018 (UTC)[reply]
    Thanks WAID for putting my figures into context. Yes, a massive parallel setup will crack the 12 letter passwords in much less time than I indicated, but folks who spend that much money on super-powerful rigs won't be using their investment on trying to get access to your Wiki account, so 12 is probably "good enough". I have my generator set to 20 characters (because I can), if that helps the context. The other thing I really should have said is that the most insecure thing you can do is re-use a password over multiple sites. It's reassuring that the Tech department will keep an eye out for attacks, but that is no defence if you use the same name and password here as you use on the website for your local grocery shop. If the grocery shop has its password file stolen, it can be cracked at the hacker's leisure, and then they will only need one attempt to log into your account here (unless you have 2FA). HTH --RexxS (talk) 21:32, 7 July 2018 (UTC)[reply]
    @WhatamIdoing: MediaWiki doesn't have any restrictions on what you use as your password, beyond certain policy restrictions like a minimum length. It takes whatever you submit as-is and runs it through a PBKDF2 key-derivation function to get the password hash stored in the database. Anomie⚔ 19:43, 8 July 2018 (UTC)[reply]

    Request for input

    This discussion about scope and naming on List of questionable diseases could probably use the input of folks who "know stuff" about medicine. GMGtalk 11:24, 4 July 2018 (UTC)[reply]

    Help still needed. The AfD decided keep but tasked us with renaming the article. So far, it has proven difficult to get consensus on what the article is about and thus on what a new name should be. The article has been extensively edited since the AfD result, but not all those edits have been welcomed. And debate continues over the applicability of WP:MEDRS. Bondegezou (talk) 15:22, 8 July 2018 (UTC)[reply]

    Comorbidities

    Are not mentioned in MEDMOS for diseases. Where do you all think they should go? Am looking at this edit to Diabetes... Jytdog (talk) 02:42, 6 July 2018 (UTC)[reply]

    Hmm. The ideal approach might depend upon the details. Here's some ideas off the top of my head: ==Epidemiology== if they co-exist but don't really interact. ==Pathophysiology== if the causes are related. ==Symptoms== if the obvious presence of one means you should investigate the existence of the other. ==Treatment== if it affects treatment (e.g., "Many people with Disease also have Scary, in which case they can't take this Scary-exacerbating drug").
    Other people may have other/better ideas. WhatamIdoing (talk) 16:30, 6 July 2018 (UTC)[reply]

    One alternative is to introduce the following division into MEDMOS:

    ==Related diseases==
    ===Differential diagnosis===
    ===Comorbidities===
    Just thinking about it now makes sense to me, and it's odd that it hasn't been raised before. It would of course require splitting out Differential diagnosis from the ==Diagnosis== sections, but this isn't a formal issue and will be resolved over time.
    Another alternative is to place it as follows:
    ==Diagnosis==
    ===Differential diagnosis===
    ===Comorbidities===
    Though this feels less natural for me. Carl Fredrik talk 16:40, 6 July 2018 (UTC)[reply]

    Use of the actual word may be a bridge too far for a general audience. Most people will cope with the word "diagnosis", or even "differential" thanks to medical shows... I doubt "comorbidity" will slide down quite so easily. Basie (talk) 07:29, 7 July 2018 (UTC)[reply]
    It would require some short clarification of what a comorbidity is on each article; to which an alternative is ===Associated diseases=== (To further make myself clear I am in favor of splitting comborbities from other sections, because a treatment section would be overly detailed with such information, and it is more likely to be read if it is under an independent header. We know that most people only read the sections they believe are relevant to them.) Carl Fredrik talk 10:23, 7 July 2018 (UTC)[reply]
    If we want people to read it, then we should use a plain-language title.
    I think that the separate vs same section question depends upon how many there are. In other conditions, there might only be one sentence to write.
    Are you thinking about a strict definition of comorbidities (technically unrelated diseases that happen to occur at the same time (e.g., common cold + broken arm = comorbidity) or a broader one (diabetes causes kidney failure)? If the latter is more important, then a ===Complications=== section (or something like that; for all the things that are commonly present together because they have related causes/processes) might be more appropriate. WhatamIdoing (talk) 17:25, 7 July 2018 (UTC)[reply]
    When is it appropriate to include very technical information, well beyond the comprehension of 95% of visitors, e.g., drugs (medications) and the chemical information in the sidebar? Only in sidebars? What about the fact that many medical & other health professionals use Wikipedia to learn about all sorts of topics related to their studies/profession? Are they not an important part of our audience? (I understand that Wikipedia is not a medical textbook.) I don't mean to come across as harshly challenging or the like. These are simply questions I've had for a while that this discussion brought to mind.   - Mark D Worthen PsyD (talk) 23:17, 7 July 2018 (UTC)[reply]
    Markworthen, I took the liberty of moving your comment out of the middle of WAID's. I agree, it's a tricky needle to thread but I don't think the fact that health professionals use Wikipedia means we should allow the content to drift further away from something a general audience will understand. Cheers, Basie (talk) 23:26, 7 July 2018 (UTC)\[reply]

    Medical outsourcing

    Just created that. it is a bit crappy. Also created Category:Medical outsourcing... Jytdog (talk) 06:05, 6 July 2018 (UTC)[reply]

    Added Travel nursing. There's probably a few other candidates scattered around. Basie (talk) 08:13, 6 July 2018 (UTC)[reply]

    Request for more information in article

    Checking breathing devices

    At Talk:Breathing_gas#Improve_medical_usage. I don't know whether any of you think this article should be expanded to include the requested information. Cheers, · · · Peter (Southwood) (talk): 16:04, 6 July 2018 (UTC)[reply]

    commented--Ozzie10aaaa (talk) 21:02, 6 July 2018 (UTC)[reply]

    Comment on draft

    Your comments on Draft:TRIORAL rehydration salts are welcomed. Please use either Yet Another Articles for Creation Helper Script by enabling Preferences → Gadgets → Editing → check Yet Another AFC Helper Script, or use {{afc comment|Your comment here. ~~~~}} directly in the draft. Thank you. Sam Sailor 09:37, 9 July 2018 (UTC)[reply]

    Commented. And user needs to disclose... Doc James (talk · contribs · email) 13:41, 9 July 2018 (UTC)[reply]

    Level of reading dispute

    Hi, I am having a dispute with User:Sbelknap about reading levels in articles about drugs. See the version of Clonidine here, compared with this one. There is some discussion at my talk page here.

    There has been a discussion about reading level at WT:MEDMOS (the only thing there now). Am also placing a note of this at WT:PHARM since this was pharmacology content.

    What are folks' thoughts? Jytdog (talk) 13:25, 9 July 2018 (UTC)[reply]

    Yes simplification is important. Doc James (talk · contribs · email) 14:10, 9 July 2018 (UTC)[reply]
    "Everything Should Be Made as Simple as Possible, But Not Simpler" - Albert Einstein
    I think their version of the lead would be great in a section on mechanism. The lead is for random people who just want to know a simple summary of the drug and google it. The rest of the article can dive into more depth. Natureium (talk) 14:16, 9 July 2018 (UTC)[reply]
    Many ledes for pharmacology articles are wrong, trivial or both. I strongly agree with Doc James that simplicity is important. However, the root of simplicity for drugs should be in the conceptual articles, not the drug articles. The drug articles should use technical terms that are defined in simple terms in the conceptual articles. For example, This is the first sentence I proposed for the lede for the clonidine article: Clonidine (trade names Catapres, Kapvay, Nexiclon, Clophelin, and others) is a centrally-acting sympatholytic α2 adrenergic agonistCite error: The opening <ref> tag is malformed or has a bad name (see the help page). My intent was to provide hyperlinks to articles that would explain "centrally-acting" "sympatholytic", "alpha-2 adrenergic receptor", and "agonist." Alternatively, this could be expressed using multiple sentences. Either way, these concepts are the essence of what clonidine is in a pharmacology context. Thus, these should be in the lede. Here is the lede for adrenergic drug from Encyclopedia Britannica: "Adrenergic drug, any of various drugs that mimic or interfere with the functioning of the sympathetic nervous system by affecting the release or action of norepinephrine and epinephrine. These hormones, which are also known as noradrenaline and adrenaline, are secreted by the adrenal gland, hence their association with the term adrenergic. The primary actions of norepinephrine and epinephrine are to mediate the “fight-or-flight response.” Thus, they constrict blood vessels (vasoconstriction), which increases blood pressure, and accelerate the rate and force of contractions of the heart. Adrenergic drugs that produce or inhibit these effects are known as sympathomimetic agents and sympatholytic agents, respectively."— Preceding unsigned comment added by Sbelknap (talk • contribs)
    I disagree strongly with the assertion that "the root of simplicity for drugs should be in the conceptual articles, not the drug articles". I think it is the other way round.
    An average reader coming across an article about a drug is less likely to understand pharmacological and physiological concepts than those finding their way to articles on drug classes. A patient is not necessarily aware that his drug is of a certain type, and that simvastatin is a statin isn't even immediately obvious, if you do not know that statins exist. That said, we have ample room for detailed description in the body of articles, and Jytdog's lede summary is better for this reason.
    Also, arguing that Encyclopedia Brittanica does things one way is not an especially strong argument around here. Carl Fredrik talk 23:15, 9 July 2018 (UTC)[reply]
    Like Doc James, I am of the mind that articles about drugs should be as simple as possible (without eroding the validity of the content). Generally, we should strive to simplify or define jargon (or provide both) when applicable (e.g. using the term "dry mouth" in place of "xerostomia"). The purpose of these efforts is to improve the accessibility of medical content for a lay audience. I believe that this should be true even for more "complex" sections, such as the "Mechanism of Action" subsection under "Pharmacology."
    In reference to the clonidine article specifically, I think it would be a good idea to break up the MoA § by indication/effect. I may put some work into that to display my meaning. I concur with Carl Fredrik that drug articles should be the focus of simplification, since I imagine that these receive greater lay-person traffic than technical, "conceptual" topics in pharmacology. ―Biochemistry🙴❤ 03:07, 10 July 2018 (UTC)[reply]
    This "centrally-acting sympatholytic α2 adrenergic agonist" should never occur within the first sentence of any article. You only learn this stuff near the end of medical school or into residency.
    One should not need a professional degree in the subject matter in question to find a Wikipedia article useful. Doc James (talk · contribs · email) 10:18, 10 July 2018 (UTC)[reply]
    I am more familiar with science education in the US and to a lesser extent, in the UK and the EU, and so can't speak to the situation in Canada. In the US concepts of "adrenergic" "cholinergic" and "dopaminergic" are introduced to elementary school students age 11 - 13, are covered again in Freshman Biology in High school (age 14), are covered in more detail in AP Biology during the Junior or Senior year (age 16 - 18), in college during Introductory Biology (age 18 - 19), and for biology majors during Neurobiology (age 19 - 21), then in medical school physiology, again in medical school pharmacology, and during clinical rotations in pediatrics, medicine, Emergency medicine. The concept agonist and antagonist are introduced in Freshman High School Biology. The concept of alpha 1 and alpha 2 adrenergic receptors, beta 1 and beta 2 adrenergic receptors, dopamine receptors, and acetylcholine receptors are first introduced in AP Biology in High School. The concept of centrally-acting vs. peripherally-acting are covered in AP Biology. I served as a Science Olympiad coach for elementary school students aged 11 - 13 for several years. I do have some idea what an intelligent laymen knows and what they can learn. Our team finished second in the nation. I have also taught pharmacology and clinical pharmacology to medical students for 34 years. For 10 years, I was on the National Board of Medical Examiners Pharmacology Committee that created the exams for the USMLE Step 1 Exams. Perhaps Doc James misremembers his own educational experience. Regardless, his comments about education on neuroscience do not accurately desccribe the experience of American students.Sbelknap (talk) 19:00, 10 July 2018 (UTC)[reply]
    That's surprising. It hasn't been that many years since I was in school, but I didn't learn about the concepts of "adrenergic", "cholinergic", and "dopaminergic" until I was an undergrad. I probably learned about antagonists/agonists in high school, but I took a lot more science classes in high school than were required. I think the rest of the concepts you talk about I learned during undergrad. Natureium (talk) 19:04, 10 July 2018 (UTC)[reply]
    Perhaps not so surprising. There is a tendency to have amnesia for initial experience with material that is also covered later. Both you and Doc James appear to be victims of this phenomenon. It might be best to move to some more objective source of evidence than our own idiosyncratic experience. Here is the Khan Academy lesson for High School AP Biology on neurotransmitters: https://www.khanacademy.org/science/biology/ap-biology/ap-human-biology/modal/a/neurotransmitters-their-receptors

    I am beginning to understand the reasons why the wikipedia pharmacology articles are so terribly bad. The community of editors interested in these pages is completely wrong-headed as to what an intelligent laymen knows (or knew in high school). You all simply do not know your audience.Sbelknap (talk) 19:43, 10 July 2018 (UTC)[reply]

    Or... you could be wrong. Jytdog (talk) 21:47, 10 July 2018 (UTC)[reply]
    "In the US concepts of "adrenergic" "cholinergic" and "dopaminergic" are introduced to elementary school students age 11 - 13". I got no dog in this fight, but I can tell you that this isn't part of the Canadian curriculum at any point. I'm a 33 year old man with a Master's degree in physics who took all sciences possible in K-12, and 2 biology classes for my undergrad, and I couldn't tell you what those things are without looking them up. This is the first I hear of them. Headbomb {t · c · p · b} 22:18, 10 July 2018 (UTC)[reply]
    Which is another matter entirely. Your assertion that you don't remember these things does not mean that you never learned these things. Sbelknap
    I can't remember what I never learned in the first place. Those are specialist terms. I know what dopamine and adrenaline is, but this not K-12 material, and the word "adrenergic" does not evoke adrenaline to the non-expert. And absolutely no one would have any clue what acetylcholine. I recognize 'acetyl' because I have some background in chemistry, but hell if I know what choline is, or that "cholinergic" would refer to it. Headbomb {t · c · p · b} 01:49, 11 July 2018 (UTC)[reply]
    Well, I tried to find evidence to back up your claim, and I've failed. Here are two searchable links to 7th-grade biology textbooks, and I have been unable to find anything in them that suggests that these concepts are even named at this level. Not even simpler concepts, like "serotonin" and "dopamine", are turning up for me. The focus seems to be on anatomy ("optic nerve" and "dura") rather than function. Perhaps someone else wants to have a go at seeing whether they can find these concepts in a seventh-grade biology textbook? I don't doubt that a dedicated child with special extra coaching could understand this; I'm just unable to find any evidence that it's presented to in the regular biology curriculum at this age. Only 5% of American kids are going to be taking AP Biology. Unless we're writing Wikipedia articles for the educationally privileged 5%, then "let's put it in the first sentence, because they cover this in AP biology" doesn't seem like an appropriate plan. WhatamIdoing (talk) 01:15, 11 July 2018 (UTC)[reply]
    For whatever it's worth, I got through a major in biochemistry and taught various biology courses to first- and second-year science and medicine students in Australia for several years before I ever learned "adrenergic" or "cholinergic". Neither term appears in the >1300 pages of our general biology textbook for first-years. Adrian J. Hunter(talk•contribs) 03:06, 12 July 2018 (UTC)[reply]
    Lets take the lede for clonidine as an example:
    "Clonidine is a medication that is used to treat high blood pressure, attention deficit hyperactivity disorder, anxiety disorders, tic disorders, withdrawal (from either alcohol, opioids, or smoking), migraine, menopausal flushing, diarrhea, and certain pain conditions."
    A definition ought to first say what a thing is, then (maybe) say what it is used for; e.g., Hammer n. A hand tool consisting of a handle with a head of metal or other heavy rigid material that is attached at a right angle, used for striking or pounding.
    The key facts about clonidine are that it is 1. sympatholytic 2. alpha 2 adrenergic agonist 3 centrally-acting
    It is marketed under many brand names.
    False. It is hardly "marketed" at all. Instead, it is sold as a generic drug. Even the patch has a generic version.
    The adverse effects include sedation, dry mouth, and low blood pressure.[1]
    The Adverse drug effects paragraph in the lead ought to have a sentence for "most frequent" and a clause for "most serious." The essential information is that most adverse effects are mild and tend to diminish with continued therapy. The most frequent (which appear to be dose-related) are dry mouth, occurring in about 40 of 100 patients; drowsiness, about 33 in 100; dizziness, about 16 in 100; constipation and sedation, each about 10 in 100. The most serious adverse effects are low blood pressure, depression, and slow heart rate.
    Clonidine treats high blood pressure by stimulating α2 receptors in the brain stem, which decreases peripheral vascular resistance, lowering blood pressure.
    No. The blood pressure lowering effects of clonidine are caused, at least in part, by activating alpha 2 receptors in the lower brainstem.
    Clonidine also may cause bradycardia, probably due to a temporary increase in vascular resistance caused by clonidine stimulating α2 receptors in smooth muscles in blood vessels.
    No. Clonidine increases parasympathetic outflow, which may slow the heart rate.
    This blood pressure raising effect is seen when the drug is given intravenously, and is not usual when clonidine is given by mouth or by the transdermal route.[5]:201-203
    To what antecedent is "This" referring?
    This sort of awfulness occurs throughout the pharmacology article ledes. When one looks through the revision history, one finds attempts to correct the errors, but these are shouted down by wikilawyers. Pharmacology is a science. If an editor does not respect that, then that editor is more of a vandal than an editor. Sbelknap (talk) 22:58, 10 July 2018 (UTC)[reply]
    I don't think I agree with this analysis. From my POV, the "key fact" sounds a lot more like "your grandfather is probably taking the stuff in that bottle to treat high blood pressure" than "biochemists classify it in these ways".
    On a tangent, a few years back, we collected comments from readers about Wikipedia articles. The comments on rare-disease articles have stuck with me. They didn't ask for more information about how to classify it or what the etiology was, or anything like that. They asked whether their friend's baby was going to die. I learned from this that the stuff that interests our intellectual, ivory-tower contributor base is very different from what our readers are actually looking for. IMO, if your target is the rare person who is going to look up all of those classification terms, then you can count on that person to read past the first sentence. You can, in fact, count on them reading the entire article, which is something that almost nobody actually does. But the one who just needs to learn "Scary Disease doesn't kill kids" or "Those are Grandpa's high blood pressure pills"? They don't need that. WhatamIdoing (talk) 01:39, 11 July 2018 (UTC)[reply]
    The wikipedia editors for rare diseases don't seem to agree with your perspective. Here is the beginning of the lede for Maple syrup urine disease (MSUD), also called branched-chain ketoaciduria, is an autosomal recessive[1] metabolic disorder affecting branched-chain amino acids. It is one type of organic acidemia.[2] The condition gets its name from the distinctive sweet odor of affected infants' urine, particularly prior to diagnosis, and during times of acute illness.[3] — Preceding unsigned comment added by Sbelknap (talk • contribs) 03:17, 11 July 2018 (UTC)[reply]
    That kind of introductory sentence is exactly why we got those complaints about the articles' contents.  :-/ WhatamIdoing (talk) 23:30, 11 July 2018 (UTC)[reply]

    Here is a good example from the first paragraph of a technical wikipedia article:

    Esterel is a synchronous programming language for the development of complex reactive systems. The imperative programming style of Esterel allows the simple expression of parallelism and preemption. As a consequence, it is well suited for control-dominated model designs.

    Notice that the terms "synchronous programming language", "complex reactive systems", "imperative programming style", "parallelism", "preemption", and "control-dominated model designs" are used in this first paragraph of the lede. If one doesn't understand these terms, then most of them have hyperlinks to decent articles that explain them. That is the model that the pharmacology articles ought to use.

    Each technical discipline has its own core vocabulary. There are about 400 core technical words in pharmacology. About 100 or so of these words cover 90% of the need for technical vocabulary. This attempt to eschew technical words in a technical article is sheer folly. It is not what technical articles for other disciplines do. Here is another example:

    Staggered tuning is a technique used in the design of multi-stage tuned amplifiers whereby each stage is tuned to a slightly different frequency. In comparison to synchronous tuning (where each stage is tuned identically) it produces a wider bandwidth at the expense of reduced gain. It also produces a sharper transition from the passband to the stopband. Both staggered tuning and synchronous tuning circuits are easier to tune and manufacture than many other filter types.

    The function of stagger-tuned circuits can be expressed as a rational function and hence they can be designed to any of the major filter responses such as Butterworth and Chebyshev. The poles of the circuit are easy to manipulate to achieve the desired response because of the amplifier buffering between stages.

    Applications include television IF amplifiers (mostly 20th century receivers) and wireless LAN. Sbelknap (talk) 23:52, 10 July 2018 (UTC)[reply]

    Basically, say it twice in the lead - first simply, then with the technical terms. What about the unintelligent layman, which using your standards would seem to include most of us? Johnbod (talk) 01:24, 11 July 2018 (UTC)[reply]
    On the Esterel example: I happen to have the advantage of knowing what those "vocabulary words" are, but that text didn't really tell me anything about that programming language. So if you know as little about Esterel as I did, here's the answer: it's an old programming language that would be useful for programming things like stoplights, banks of elevators, vending machines, and (in the realm of biology) tools to measure reflexes. It looks like the main commercial use might be audio processing equipment. That class of programming languages is better at handling simultaneous input (e.g., two vehicles arriving at an intersection at exactly the same time) than normal async programming languages. If I were working on that article, I'd probably start with a sentence such as "Esterel is an uncommon programming language that is often used to control machines and computers that react to real-time events. It may be used to process audio or to control hardware, such as elevators and traffic lights". I'd bet that the typical person will get a lot more out of my description than out of the list of classifications. WhatamIdoing (talk) 01:55, 11 July 2018 (UTC)[reply]
    A technical article in wikipedia ought to use the core technical vocabulary of the relevant discipline. Otherwise, the article is shorn from its intellectual base. Instead of redefining the core technical words across hundreds of articles, use hyperlinks. That is their purpose. There is actual research on hyperlinks: they help the average reader more than the highly-motivated reader.[1]
    I read the Esterel lede differently. The first paragraph of the lede for the Esterel article accurately and precisely specifies what the language is. The lede is accessible to the non-expert, providing the hyperlinks so as to understand the descriptors in the lede.
    Esterel 5.0 is not an "old" language. It is in active use in industry and in academia. In fact, I use Esterel in my research. The Esterel family of languages includes Lustre, which was used to build SCADE, the system that runs the Airbus 380. I expect that your characterization of Esterel as a stoplight language would make Gerard Berry laugh. Sbelknap (talk) 02:59, 11 July 2018 (UTC)[reply]
    Berry's first publication about Esterel was in 1983. That's the same year that the Apple Lisa (selling for "just" the equivalent of $30576), MS-DOS version 2, Microsoft Word version 1, and Lotus 1-2-3 v1 were released. 3.5" floppies were brand-new, and available in single-sided only, and most consumers were still swapping 5.25" ones. Almost nobody had hard drives at home, and when they did, they often held barely 5 or 10 MB. Consumers were coveting Intel's 286 chip, and larger businesses were looking at fancy "minicomputers" that were "only" as big as a washing machine, rather than the size of a truck, not counting the tape drive or expensive, washing-machine-sized hard drives you needed for your data. The fastest computer in the world was the Cray, and despite being big enough to double as furniture, it had less computing power and less memory than the iPhones being sold today. In 1983, "the internet" had 1,000 hosts, "the backbone" was critical infrastructure (DNS was brand-new), and it would have zero websites for another two decades. Borland's C++ and MS Windows were still dreams, and PostScript was still in John Warnock's garage. So, yeah: by most people's standards, 1983 really is "old" for computing, even if some of us are old enough to personally remember it.
    Oh, and if Berry's amused that Esterel can be used for traffic control, then he's been amused at least since he published that idea himself in 1999 (e.g., page 10, section 1.2). WhatamIdoing (talk) 23:28, 11 July 2018 (UTC)[reply]

    If this approach you all are using to writing pharmacology articles is so great, then why are the articles so terrible? Have you all considered that using folksy language instead of technical terms renders these articles uneditable? Isn't this the real reason why a 14 year old article with hundreds of edits is still filled with errors? For example, here is one edit:

    (Changed Sympathoolytic to the correct Sympathomimetic in line 1. Clonidine is an alpha 2 agonist, therefore a sympathomimetic. (medical student)) (undo)

    Let us imagine that this well-meaning medical student saw this first sentence in the lede instead of the pile of rubble that was present at the time: clonidine is a centrally-acting, sympatholytic alpha-2 adrenergic agonist. This student would then stop and think and realize that his own understanding was close to being correct, but not quite. He would not have then introduced yet another error into the article. You all are thinking about some lowest common denominator reader that will read the article. But editors are also readers, and they will unintentionally vandalize complex descriptions written in folksy language. They are more likely to respect technical language, reviewing it for accuracy, and perhaps improving it. I suggest that the wikipedia pharmacology articles are uneditable in their current form. A new approach is needed. Lets use the approach that other technical articles use. Sbelknap (talk) 03:43, 11 July 2018 (UTC)[reply]

    See WP:OTHERCRAP and WP:NOTFINISHED. Headbomb {t · c · p · b} 03:33, 11 July 2018 (UTC)[reply]
    On the contrary, Sbelknap, I believe that the use of jargon is what easily contributes to mistakes—both in interpretation of sources, and in a reader's understanding. Wikipedia is meant for the lay public, so let us write it in such a way that it is accessible to them. However, let us do so in a way that is stimulating and useful to those with more advanced knowledge. I am not convinced that it is somehow impossible to write both clearly and informatively. I do not pretend myself to be an expert in such writing, but it is something I strive for. What say you re: the ADHD sub§ under the MoA§? ―Biochemistry🙴❤ 03:56, 11 July 2018 (UTC)[reply]
    There is some poor-quality evidence that clonidine may have a modest effect in ADHD in adolescents. I notice that the sub for this article does not cite Hirota's meta-analysis [2], which perhaps it should. AFAIK, there is no evidence that clonidine acts via the central alpha-2 adrenergic mechanism in ADHD (and no cite is provided in the sub you point to), nor is there evidence that alpha-2 adrenergic receptors are of some central importance in ADHD. There were some studies in an animal model of ADHD, but my understanding is that this model is not considered relevant to the human ADHD phenomenon. Perhaps if there were studies cited on this, we could see. Sbelknap (talk) 11:04, 11 July 2018 (UTC)[reply]
    Steve, above you wrote that you work with kids in science olympiads who are in grade school and AP biology classes. When you write "intelligent layperson" you appear to be talking about people way, way over on the flat part of the bell curve. Super-smart people who are actually studying this stuff. That is horribly skewed and absolutely not who we write for.
    No one here will argue that our articles do not need improvement. They do.
    But as long as you keep generating content aimed at people who are a) super smart and b) educated in biology/pharmacology (these are not the same sets of people, necessarily) you are going to keep wasting your own time and the time of experienced editors. That is not building the encyclopedia but rather squandering the most valuable resource we all have -- time. Jytdog (talk) 20:05, 11 July 2018 (UTC)[reply]
    @Sbelknap: I think you missed the point of my writing entirely, and are trying to argue for the incorporation of primary research here. The cited article is a review that describes the connection between alpha-2A ARs in ADHD. My point was to get your feedback on whether you thought the section, in terms of its readability (which is the topic we are discussing at present, yes?) was acceptably comprehensive while also accessible to a lay person. My intention was to demonstrate to you that it is indeed possible to accomplish both; we don't need to overwhelm our readers with jargon. Could I have summed the section up with a one-liner like, "Clonidine's alpha-2A AR agnoism induces a sympatholytic effect on downstream prefrontal cortex neurons."? Probably, but that would be practically useless to a layperson.―Biochemistry🙴❤ 03:01, 12 July 2018 (UTC)[reply]

    References

    1. ^ https://pdfs.semanticscholar.org/63e0/f751d8fa17d565a7b8c98fde6625f1773727.pdf
    2. ^ Hirota, Tomoya, Shimon Schwartz, and Christoph U. Correll. "Alpha-2 agonists for attention-deficit/hyperactivity disorder in youth: a systematic review and meta-analysis of monotherapy and add-on trials to stimulant therapy." Journal of the American Academy of Child & Adolescent Psychiatry 53, no. 2 (2014): 153-173.
    Jytdog, (and other interested editors) I challenge you to provide one example of a wikipedia article on a medical drug that meets the wikipedia criteria for a wikipedia featured article. Those criteria are here: [[3]] One article.
    The content I have attempted to create is for intelligent laypersons, which is a large fraction of any human population, anywhere. I am suggesting that as long as editors of wikipedia pharmacology articles are prohibited from using the core technical vocabulary of the discipline, the quality of those articles will remain poor. In my opinion, the *cause* of the poor quality is, in part, the policies and culture that are unique to medical editing on wikipedia. Many of these pharmacology articles have existed for more than a decade, yet remain rife with error. If you review the revision history for the pharmacology articles you will notice that many attempts at correction of errors are thwarted. The pharmacology articles lag behind other technical articles in wikipedia, which have steadily improved. I am suggesting that the prohibition on the use of the core technical vocabulary of pharmacology in these articles is the *cause* of the poor quality of the pharmacology articles relative to those for other STEM disciplines. Editors in other areas of wikipedia apparently accept/encourage the use of core technical vocabularies.
    Doc James stated, "One should not need a professional degree in the subject matter in question to find a Wikipedia article useful." I provided evidence that contradicted this idea that a professional degree was required to understand the core vocabulary of pharmacology. Here is that video Khan Academy lesson for High School AP Biology on neurotransmitters: https://www.khanacademy.org/science/biology/ap-biology/ap-human-biology/modal/a/neurotransmitters-their-receptors
    Perhaps this is progress, as you now concede that some high school students achieve mastery of some of the technical vocabulary of pharmacology. Many less accomplished high school students also are familiar with the meaning of "adrenergic", "cholinergic", "dopaminergic", "serotonergic", "agonist", "antagonist", etc. I note that my use of the word "agonist" has been removed from wikipedia articles that contain the word "antagonist" a few sentences away.
    The wikipedia guidelines and culture do not respect pharmacology as a science. There are good reasons for using technical words, as they facilitate clear communication. Technical words are widely used in wikipedia, with hyperlinks to articles that explain them, often very well. Pharmacology is an outlier in wikipedia on this issue.
    Jytdog, you are certainly wise to not waste the time of experienced editors. However, under the current wikipedia culture, the efforts of editors of pharmacology articles *is* being wasted. The product of these efforts is quite poor. The pharmacology articles are condescending to their audience. I frankly don't see utility of most of these articles for any audience. The quality problems could quickly be addressed by using the same approach to technical vocabulary that is used in other STEM articles.
    I propose that we compile a core vocabulary of pharmacology terms (agonist, half-life, alpha adrenergic, etc), with broad input from interested wikipedia editors. I envision eventually using about 100 terms. We would first change wikipedia pharm policies to permit use of these core technical terms in wikipedia drug articles. We would then want to expend some effort to improve the pharmacology vocabulary articles. Many articles for these core terms exist, though some improvements would be advisable if we are going to use them to build the articles for specific drugs. Is there a mechanism on wikipedia for doing modified Delphi trials to build such a list? Or is there some other suitable mechanism we could use? Perhaps just a wikipedia page for core terms of pharmacology? What do you suggest? Sbelknap (talk) 21:44, 11 July 2018 (UTC)[reply]
    You complain fiercely about the poor quality of articles but you make edits like this, where you persist in writing crappy citations that other people need to work over to make them usable in an online encyclopedia, even as you put tons of weight on clickable wikilinks. Doc James and I have both showed you the tool that allows you to create an elegant citation with links in seconds. You blow that off.
    You also quote something without saying who or what you are quoting (I learned not to do that in junior high school).
    You do not understand our mission and are so busy being contrarian that you are screwing up very basic editing details and wasting everyone's time.
    This place survives in the tension between individual boldness and minding the community (what you dismiss as "wikilawyering" is actually the "constitution" that holds this place together). You are all the way over on the "bold" side. It is not a sustainable approach to editing. Jytdog (talk) 23:30, 11 July 2018 (UTC)[reply]
    Strongly against any proposal that would create a mandated list of vocabulary terms to use whilst writing WP:PHARM articles. I'll be frank Sbelknap: even if such a list existed, I would not use it. It would be an overwhelming task to create, let alone memorize, such a list, and any attempt to apply it would present an insurmountable, bureaucratic burden—especially for new editors.―Biochemistry🙴❤ 02:49, 12 July 2018 (UTC)[reply]
    I accept Jytdog's criticism that I did not use the correct tool to edit in this case. I apologize for this, but did not know (still do not understand) that I was doing anything incorrectly. I will stand by the *content* of what I wrote, which was a big improvement over what was there. The net effect is that this article is worse than it could have been. Perhaps rather than fierce attacks on process errors, gentle reminders and requests to repair an editing error could be used? Or is that not part of the etiquette of the wikipedia community?
    I note that Jytdog does not address my above proposal, which was exactly the attempt to engage the community that Jytdog seems to be asking for. Perhaps Jytdog could share his opinion on my proposal? I openly solicited Jytdog's opinion in my comments directly above. Jytdog invited me to this page to engage in this discussion and I agreed to participate. I concede that process matters. Will Jytdog concede that content matters? Perhaps Jytdog can provide us with a *single* pharmacology article that meets the criteria for a wikipedia featured article? Sbelknap (talk) 00:02, 12 July 2018 (UTC)[reply]
    You are asking the wrong person about FAs; I find that whole structure to have gone wrong and to have become an engine of drama-mongering, and I stay far, far away from it.
    I already said that much of WP, including our health content, needs work.
    That is why this new crusade of yours is such a tragedy. Please just try to learn the basics. Formatting citations is explained in WP:MEDHOW. Please actually read WP:MEDMOS and try to follow it. Learn what we actually do here and how we do it and why before you try to reform everything. I try to explain these very basic things in User:Jytdog/How. I want our articles to be high quality and accessible to every day people. That takes work. Letting your academic scientist flag fly when you write here is not OK. You need to condescend (one of my favorite theological terms -- to come down to be with) and write for our target audience without giving up rigor. This is hard.
    I have said that you could be immensely valuable here. Instead you keep wasting our time and yours with these crusades. Please just learn what we do and how and why and try to follow it, and work. Once you show that you "get it" -- that you are what we call clueful -- you will be able to persuade people here. Jytdog (talk) 01:42, 12 July 2018 (UTC)[reply]
    The tragedy is that thousands of editors have contributed to wikipedia drug articles and nearly all of them are flawed. I have spent many months learning what you all do and why. I have pored through revision histories for dozens of drug articles. There appears to be a serious problem, I intend to fix this problem. If the current approach works, then please point me towards one medical drug article that meets the criteria for a wikipedia featured article. One article. I will use it as a learning device for my edits. Sbelknap (talk) 03:33, 12 July 2018 (UTC)[reply]
    I spend a lot of time looking at pages and how they got to their present state as well. This place is truly insane. Anybody can come here and write anything.... and you know, bell curves being what they are, most edits are bad or mediocre, some are good and a few are very good. But even the very good ones fall out of date unless somebody new who is competent comes along and updates what that person did. It just lifts my heart when I see somebody make a good edit and I whoop when I see a great one.
    There are a lot of ways that people make bad edits. Some of them are vandalism or spam, some are just wrong, some remove good, well sourced content, some are "news" that are not "accepted knowledge"... some put UNDUE weight on some aspect, some are hyper-technical. There are many ways to go wrong. There is content i would love to build but I spend most of my time fixing bad content. Just doing maintenance.
    And it kills me when I see somebody who could be making great, mission-oriented edits making bad ones. You really could do great things here Steve. But when I see edits like this, I will fix them. They are not good edits here. Here. You could do so much better.
    Again, I do not care about FA. You are asking the wrong person. If you want to ask me to point you to an article with pharmacology content that I think is really good and on-mission, I will try to find some. Jytdog (talk) 04:03, 12 July 2018 (UTC)[reply]
    Sbelknap, if you want to see an FA – a process that was basically "tuned" to fit history and the other humanities, and which IMO does not fit technical subjects perfectly – then see Bupropion or any of the other articles listed at Wikipedia:WikiProject Pharmacology#Featured content. You should be able to find the exact revision that was approved by looking in the boxes at the top of the talk page. WhatamIdoing (talk) 17:34, 12 July 2018 (UTC)[reply]
    The current version of the bupropion article is quite similar to the FA version. The lede (current version) begins:
    Bupropion, sold under the brand names Wellbutrin and Zyban among others, is a medication primarily used as an antidepressant and smoking cessation aid.[7][8][9] It is an effective antidepressant on its own, but is also used as an add-on medication in cases of incomplete response to first-line SSRI antidepressants.[10]
    There are only a few published reports describing clinical trials of bupropion. Based on what little evidence is available, bupropion is a weak antidepressant, having almost no effect on depression. There is a high-quality meta-analysis published in 2018 in a good journal that synthesizes the available evidence on bupropion as an antidepressant.[1]
    On the single most important issue, antidepressant efficacy, the wikipedia bupropion article gets it spectacularly wrong. There are many other errors in the bupropion article. I counted 10 serious errors after reading carefully through the entire article once (~45 minutes of reading). I would rate the quality of the wikipedia bupropion article as very poor. I would agree that the wikipedia bupropion article is better than nearly all the other wikipedia drug articles I've read. I expect that many readers of the wikipedia bupropion article will want to know whether or not bupropion is effective, and the article misinforms them on this point.
    So what are we to do? Sbelknap (talk) 21:37, 12 July 2018 (UTC)[reply]
    Well, that claim is supported by a source from 2005, so I'd start by wondering whether recent review articles held a different view. A quick search of reviews in core clinical journals didn't turn up anything recent, but perhaps a more thorough search will turn up a couple of useful sources. WhatamIdoing (talk) 06:42, 13 July 2018 (UTC)[reply]

    For what it's worth, I don't think the bupropion article has been compliant with the WP:FA criteria since I first read through it 3–4 years ago. As for the clonidine article, it's fine to be as detailed/technical as one wishes when describing a drug's pharmacodynamics in the body of the article, but article leads are supposed to be pretty dumbed down. Normally, I'd dedicate 1 of 4 lead paragraphs to a drug's pharmacology for a fairly lengthy drug article, using really basic/simplistic language – while still maintaining accuracy – to describe its pharmacodynamics.

    In any event, bupropion probably needs to undergo a WP:featured article review. Seppi333 (Insert 2¢) 09:49, 13 July 2018 (UTC)[reply]

    Lets be very clear here. There has never been evidence that bupropion was particularly effective for depression. The serious flaws in this article were present at the time it was first recognized as an FA. The article has been of very poor quality since it was a stub. It has gained length but never improved. I've gone through the revision history, which is fascinating. I note that Jytdog, Doc James, and others who are high-frequency editors have made many contributions to the bupropion article. Somehow they all missed the fact that there is very little evidence available about bupropion and what is available shows that thre the drug is ineffective. Numerous other errors remain uncorrected.
    Something is very wrong here. If one were to deliberately create drug articles that obscure the truth, one could hardly have done worse. There are many high quality articles in other technical areas of wikipedia but none of them are about drugs. Those who are knowledgeable about drugs are greeted with hectoring wikilawyers who attack them until they stop editing. Look at the revision histories and then look at the editing history of knowledgeable drug article editors who are attacked. Please explain how I am wrong here. I am struggling mightily to assume good intentions by those who police the wikipedia drug articles. The evidence speaks for itself.
    I again ask knowledgeable wikipedia editors to kindly provide links to wikipedia drug articles that they consider to be of high quality. I have not found any. If I am to achieve the "clueful" state that Jytdog mentions, I will need at least one example of how this community and these processes (which are unique to this subculture of wikipedia) have produced at least one wikipedia drug article that is high quality. Please advise. Sbelknap (talk) 11:54, 13 July 2018 (UTC)[reply]
    Even if every single fact is completely backwards, those articles are still good models for writing style. The FAC regulars are better at "brilliant prose" (a requirement for FA status) than at evaluating scientific evidence (which isn't even mentioned in the FA criteria). WhatamIdoing (talk) 15:55, 13 July 2018 (UTC)[reply]
    @Sbelknap: There are only a few entries at , and most of those could use a review. That said, Linezolid does seem to have a lede that we could afford to emulate elsewhere. Serpin's lede is rather more technical than necessary, but is still encouraging. LeadSongDog come howl! 16:21, 13 July 2018 (UTC)[reply]

    ref

    1. ^ Monden, Rei, Annelieke M. Roest, Don van Ravenzwaaij, Eric-Jan Wagenmakers, Richard Morey, Klaas J. Wardenaar, and Peter de Jonge. "The comparative evidence basis for the efficacy of second-generation antidepressants in the treatment of depression in the US: A Bayesian meta-analysis of Food and Drug Administration reviews." Journal of affective disorders 235 (2018): 393-398.

    Requested move

    Input is requested at Talk:Vocal_cords#Requested_move_7_July_2018. Natureium (talk) 14:22, 9 July 2018 (UTC)[reply]

    Natureium — Please post at WP:ANAT as well. Carl Fredrik talk 16:57, 9 July 2018 (UTC)[reply]

    Introducing ScienceSource

    The ScienceSource project, funded by a WMF grant to work on medical referencing, was started in June and is now launching its first participatory activity, its "focus list" of biomedical articles, hosted on Wikidata.

    In the story so far, work has gone on here on Wikipedia:WikiProjectMedicine/Referencing case studies for MEDRS, and on Wikidata relating to Medical Subject Headings (diseases). There is a wikiversity:Wikiversity:ScienceSource mentoring page under development for those who want broader background. We are using lightweight videos for that.

    The focus list, information at d:Wikidata:ScienceSource focus list which is WD:SSFL on Wikidata, is the first pass at which (open access, biomedical) articles ScienceSource should look to download, in future months. We want those with expertise to work with us on compiling an excellent medical bibliography. Once the list is set up, it will be possible to query and display it in multiple ways, so slice 'n' dice will be on the cards.

    It is really not so troublesome to add a single Wikidata statement, once you have the relevant item about the article. And there is a tool to translate a DOI into the item number. Just about every relevant DOI used here in a reference is likely to be present on an item on Wikidata, since recent work has gone on in this area. (For lists, there is technical support available, so do ask.)

    There has been discussion of topics around the project in recent issues of Facto Post (back numbers), which by convention is not delivered here. You can sign up for this mass message to be delivered on your User talk page. I'm always glad to hear from people on my own talk page. Charles Matthews (talk) 09:54, 10 July 2018 (UTC)[reply]

    I followed several of those links and, sorry, I don't understand what this means or what you want people to do? Bondegezou (talk) 10:01, 10 July 2018 (UTC)[reply]
    Would be useful to have all journal articles used within Wikipedia tagged by whether or not they are a "|type= review" versus not. This of course does not related to what is proposed above.
    Would also be useful to have a bot run to add missing pmids / dois if only one is present.
    This looks like a Wikidata project rather than a Wikipedia one. Doc James (talk · contribs · email) 10:13, 10 July 2018 (UTC)[reply]
    @Doc James: Isn't the missing pmid/doi infill on WP still being contributions done by User:Citation bot? I long ago suggested that the PublicationType field on pubmed should be harvested and mapped into our citations' |type= parameter, using values such as "Systematic Review", "Meta-analysis", or just "Review". There was resistance to the idea, based on concerns over data quality at the time, IIRC. LeadSongDog come howl! 16:09, 10 July 2018 (UTC)[reply]
    @LeadSongDog: Thanks, that's interesting history and I wasn't aware of it. The objection can be met by storing the PublicationType field on Wikidata, on the item for the article, and referencing it to PubMed (so implying "for what it's worth"). Charles Matthews (talk) 09:39, 11 July 2018 (UTC)[reply]

    Thanks for the responses. I have just added the three introductory videos (60 seconds) to the Wikiversity page. These are designed to give people some overall idea of the project.

    In terms participants in this WikiProject would understand, the big problem ScienceSource addresses is the one discussed a couple of years ago (see Wikipedia talk:WikiProject Medicine/Archive 85, #24 on the ToC). We are not working on that yet, though it would be good to discuss how we see a new approach. We shall be downloading articles to our own wiki, which is being configured as we speak. Those articles are going to be "good for referencing", on the face of it. The algorithm will check if those articles actually "pass MEDRS". Other software will help locate useful factual content in the articles. Putting it all together, reliably sourced medical facts will be obtained.

    One month in, we are still "in the foothills". The focus list project is to get a preliminary bibliography together. That is what I'm proposing today, for participation.

    Yes, we have to start with data. Whether something is a review or not of course is very much on topic for the project, and I was discussing it on Saturday. There is an issue of scale: it would theoretically possible to find out (for what it is worth) for each article now on Wikidata and from PubMed, whether PubMed reckons it is a "review article". Right now that would require searching 17.5 million articles. If we have a list of candidates of (say) 10K suggestions, we have a more manageable problem. We'd still need to understand better what "review" means. Charles Matthews (talk) 10:26, 10 July 2018 (UTC)[reply]

    Work like doi:10.1186/2046-4053-4-5 and related seems highly relevant to the proposals. Automated selection of papers for systematic reviews is becoming commoner, and there is also increasing interest in automated data extraction from papers. That said, I am dubious about the challenges of obtaining "reliably sourced medical facts" at the end. Bondegezou (talk) 10:36, 10 July 2018 (UTC)[reply]
    OK, the grant process took over three months to complete, through half-a-dozen stages, and much of what was said then can be read on m:Grants:Project/ScienceSource and its talk page. I don't want to rehash that here, for obvious reasons. We said that we'd be trying to engage the medical editors and others in what we are doing, and getting a good baseline for papers is where it starts. (By the way, the company has been involved in work around systematic reviews.)
    The idea of getting people to add some small numbers of papers to a list is actually consciously modelled on m:The Wikipedia Library/1Lib1Ref, and we took advice from someone involved in that effort. Charles Matthews (talk) 12:40, 10 July 2018 (UTC)[reply]
    I just read Wikipedia:WikiProjectMedicine/Referencing case studies for MEDRS and find all that fairly clueful. The hardest thing in all of this, in my view, is identifying "high quality sources". When I go through pubmed looking for reviews, I don't necessarily take the newest, non-predatory one. I also look for "signs" that the piece is really an effort to summarize accepted knowledge in the field vs someone making an argument that might be not really mainstream in the field. One sign of that, is to the extent to which the authors are citing their own papers or are sampling widely from the field. Another is pretty subjective, but seeing the extent to which people are buying woo (see the discussion about PMC 5438513 in this archived discussion here, from last summer; I guess you could look for woo-buying keywords like "guarantee" or the like to filter those out. I almost never cite Frontiers journals because many issues are guest edited, and while articles in those issues are interesting you can bet that people are making arguments rather than really giving the state of play. Unfortunately some Cochrane reviews are the product of advocacy; see for example this discussion here about the HepC review, here about Cochrane's alt med group; here on Psych drugs; see also Tom Jefferson (epidemiologist) for his views on what kind of evidence should "count" for vaccines, which is way, way out there.
    I think the judgement that goes into selecting sources is as difficult and important, as what information is extracted.
    Then there is the whole issue of giving WEIGHT to various findings. You can see the content at Oseltamivir#Medical_use where clinical guidelines and Cochrane reviews directly contradict each other, and you can see the negotiations on the talk page where we worked out how to handle that.
    These are challenges; I am not sure how your project can address them... Jytdog (talk) 19:46, 10 July 2018 (UTC)[reply]

    Thanks for the comments. This is a 12 month project, and we have eight or nine months now to tune what we do with MEDRS. You make some reasonable points about the actual practice. I expect an approach of salami-slicing the difficulties, rather than one big advance. The starting level is publication date/publisher/type of publication. We can gather good data on those.

    After that it will get somewhat messier, from a data point of view. We can address "stale or not" considerations for past-best-before-date reviews, with a list of exemptions. We can also address the "pecking order" issue of some reviews, e.g. Cochrane, being considered ace of trumps in the particular area: Cochrane reviews are at least well documented here, and I have been doing some preliminary work about that.

    Then, we shall set off writing an algorithm in SPARQL, which (in contrast I think to previous proposals) is quite well adapted to problems where the available data is disparate. A SPARQL query that looks at least quite like what MEDRS is trying to say is at least a formal representation of a purported solution. In other words, fixing it up would look like debugging, while trying to get shades of meaning in the hard cases from MEDRS might descend into "semantics". Actually arguing about semantics is perfectly fine in its place: my comment on "review" above was intended to say that. (The ontology of "review" has been studied formally, and I expect to get to that area in due course.)

    As for machines making good "judicial" decisions, I doubt we'll crack that as an issue of principle. Actually NPOV here admits multiple points of view under some kinds of RS criteria. Where the machine says "yes" and there is a consensus here that the answer should be "no", we do have a debugging situation, and the rationale would be very interesting input into our work.

    Anyway, thanks too for the kind words about the case studies page. It is obviously a first pass, only, and to illustrate how we'd try to apply some of the MEDRS discussions here. Charles Matthews (talk) 08:45, 11 July 2018 (UTC)[reply]

    So, is this a rule-based approach rather than using machine learning? Bondegezou (talk) 10:17, 11 July 2018 (UTC)[reply]

    Not machine learning. I know that "algorithm" often does mean machine learning these days, but we mean it in the Computer Science 101 sense, Turing machines ... When I say SPARQL, it is partly because it is the query language I understand, and partly because it is a natural way to draw in facts from Wikidata. We will be using human checking of the facts (relation extraction), by the way, and combining it with algorithmic processing of the publication metadata. Charles Matthews (talk) 10:43, 11 July 2018 (UTC)[reply]

    Thanks for the details. Machine learning techniques using the full text of papers are coming on in leaps and bounds, as per doi:10.1186/2046-4053-4-5 I mentioned previously. Generally this is in a decision support role, e.g. the use of machine learning to prioritise what papers a human should look at for a review, or using the machine learning answer as a second reviewer. I would have thought that sort of approach would be valuable in a Wikipedia context, but this time selecting reviews that Wikipedians should be looking at. However, these are perhaps ideas for another project. Bondegezou (talk) 10:57, 11 July 2018 (UTC)[reply]

    I'm aware of these ideas: in fact one of the authors in the paper you cite is in the group at UCL with which my company has worked. Here, though, we are working on a problem rather directly related to Wikipedia editing. Charles Matthews (talk) 13:05, 11 July 2018 (UTC)[reply]

    I'd like to echo the concerns about treating Cochrane reviews as the (singular) ideal source. Although not perfect, Cochrane is the best... for what it does... which isn't everything. You might a Cochrane review for ==Treatment==, ==Screening==, ==Diagnosis==, and ==Prevention==, but you usually want a textbook for ==Classification==, ==Causes==, ==Pathophysiology== and ==Symptoms==, and a plain narrative review for ==Research directions==. WhatamIdoing (talk) 17:28, 11 July 2018 (UTC)[reply]

    What kind of help is being requested?

    User:Charles Matthews what kind of help is ContentMine seeking from members of this WikiProject, concretely? Would you please describe the kinds of tasks you are asking us to do? Jytdog (talk) 17:16, 11 July 2018 (UTC)[reply]

    At this point, we are asking for bibliographical help: we need to find around 30K "suitable" open access titles to download and process, and the chosen approach is, firstly, tagging on Wikidata.
    Later on, we'll need co-operation on our own wiki, at sciencesource.wmflabs.org, in the fact-checking process that will be carried out via an annotation system (technically similar to the one now on Europe PMC). We'll also need to understand the "expert" process of source criticism, that we are trying to mechanise. Charles Matthews (talk) 17:53, 11 July 2018 (UTC)[reply]

    Intended use of the resulting tool by the community

    User:Charles Matthews ....three questions:

    What exactly will this tool do - will it just analyze sources and report, or do you intend it to actually edit articles, or to suggest content and sourcing?

    Do you imagine this as a WP:BOT or Extension or WP:Gadget or WP:Scripts or something else that will reside on WMF servers, or is this something else, that would reside somewhere else?

    How do you expect it to be used and by what kind of WP editor; what kind of skills would be needed to use it? Thanks. Jytdog (talk) 17:16, 11 July 2018 (UTC)[reply]

    Not a tool as such. It will be a wiki in its own right, where work goes on as described as above. The outputs from the project will be referenced statements on Wikidata. The real task for the ScienceSource project is to build a community there. Once there is a community, it will behave in a way antipodal to a bot: it will raise issues and think of other purposes for the system. We really hope.
    There will be an algorithm, that is being refined, that evaluates sources. It will be pretty trivial to adapt to checking existing sourcing across Wikipedia, for example. Where there are gaps in referencing, it would be rather more work to locate a decent secondary source that might serve as a reference: but that is a known other problem, getting machine-readable topics associated to articles, and we'd have done some of the work by the initial narrowing of our "corpus".
    Given that the wiki is a Wikibase site, i.e. is like Wikidata in look-and-feel, I think Wikidatans may comprise a proportion of editors. The annotation system will resemble other ones that are popular. I realise medical editors may initially be a hard sell for participation, given that they have plenty to do here. The work will be "upstream" of what goes on in daily Wikipedia life. As they said back in the Renaissance, ad fontes. The scientific sources of clinical medicine – we really are trying to shine some daylight on them. As for UX, we have a deal with the WMF involving consulting their people, so that "skills" don't become the limiting factor. Charles Matthews (talk) 18:06, 11 July 2018 (UTC)[reply]

    Intended use by the company

    User:Charles Matthews would you please clarify what value this project and the grant brings to the company, and how the company intends to use the tool and any data/know how it gathers while building it, to increase the value of the company or to provide new ways to serve its paying customers?

    You are asking editors to volunteer to help you; please describe the value that our help would bring to the company and its efforts to remain viable as a commercial enterprise. Thanks. Jytdog (talk) 17:16, 11 July 2018 (UTC)[reply]

    Obviously getting such a grant is a boost to any company in a similar position, given that (a) Wikipedia is a household name, and (b) Wikimedia is a vastly influential player in the open sector. The data will be open, and openly licensed. The software will be open source and posted to GitHub or equivalent.
    ContentMine applies for both grants and contract work, and cannot do otherwise. These issues were hashed over on m:Grants talk:Project/ScienceSource#Questions about applicant and alternative funding sources early on in the grant application. I can add a bit to that, now. The wiki, we hope, can be a platform for further projects on the scientific literature. There is really no reason it needs to be a one-trick pony, but of course if the community doesn't like a new direction it won't hang around. The annotation technology is innovative in the Wikimedia context, and seems to have potential to contribute also in the development of "structured data on Commons" and edutech. We'll have to see, but Wikibase sites (Wikidata-like wikis) are a coming thing, and our version may have legs. Technical innovation is like that, and one can't really know whether the technology is going to have a "killer app".
    To get back to community. I'm a Wikisource person, and the basic idea of downloading open access papers there has not taken off. The grant, for me, allows Wikimedia to have a good look at text-mining, a major technique of contemporary tech, in a setting where people can see what it is and isn't. The added value of volunteer work will be basically verification that facts said to be stated in papers are actually there when you go carefully through the language. Any third party will be able to harvest those verifications just as much as ContentMine. Charles Matthews (talk) 18:28, 11 July 2018 (UTC)[reply]
    Thanks for your replies here. Thanks for linking to the questions about the company side at the grant page, where you wrote It defines itself as not-for-profit... however the listing at CompanyHouse is unambiguous that this is a business. One of the things that went very wrong with the company doing medical videos is that there was some weak signalling that it was thinking about becoming a nonprofit; it never was and it never did. ContentMine is a for profit entity; please don't fuzzify this.
    It is immensely valuable to the movement, in my view, that the company intends to make the tool and the data available to all comers for all purposes. I am hearing that.
    But the answers above were not clear about the value to the company outside of getting the grant itself. Can you please answer that clearly? Thx — Preceding unsigned comment added by Jytdog (talk • contribs) 19:09, 11 July 2018 (UTC)[reply]
    (You need to sign.) The founder in the old thread gave the "Shuttleworth" answer to being a non-profit. In the UK, the non-profit does not exist in the same way as in the USA. You are a charity, which is a much more restricted kind of entity, or you are a business of some sort. For example, I was the first hire of Wikimedia UK in 2010. At that point they were a type of company (limited by guarantee) and were aspiring to become a charity. They got there in the end, but there were real and serious questions whether they should qualify under English&Welsh law.
    So, being a business in terms of registration is what ContentMine is.
    I shall do my best to find a further answer to your question, since this is clearly a matter that concerns you. Perhaps you want my take on the "business case", from ContentMine's point of view, for applying for the ScienceSource grant in January of this year. It all started back in August, when there was a strategy gathering over a weekend in Cambridge. I brought up annotations, about which I'd been talking with Zazpot, and which had been mentioned as an approach to text-mining by Peter Murray-Rust. At that point we were brainstorming about "revolutionising literature search", about how the company had technology in two directions, one being ElasticSearch run on new papers on Europe PMC, and how it needed to find a way into medicine and education because these affect basically everyone.
    The MEDRS idea came up when I was on the phone to RexxS, months later. It ticked a couple of boxes, medicine clearly, and greater participation in text-mining, which was the Achilles heel of WikiFactMine, the previous project of the company. We talked in December to the WMF, and on the call community participation came up again, from the Foundation's side, as one thing they'd really like to see in a tech project.
    So the business case was that: MEDRS was a clear target that would interest many people; internally ElasticSearch probably was something of a tech dead end, and had been a bit of a stretch in WFM; and there was an algorithmic slant that fitted rather well for a Cambridge company that had made one tech hit (the fatameh tool by T Arrow). I won't give you more blow-by-blow, but we had encouraging noises from Foundation people, qualified by the need to draw conclusions and "learnings" from the previous grant. Well, I wrote the learnings, frank comments I'd made were thrown around during the grant process, as you'd expect, and in the end we made it through the process. From the point of view of the company, this largely open debate (plus two interviews) has allowed for a fresh start.
    I'm not that surprised to be asked again about the kind of issue you raise, and indeed I know rather more about the commercial position than I did at the start of the year. In terms of ContentMine, I don't see what to add, though. It is certainly not the average start-up. Charles Matthews (talk) 20:02, 11 July 2018 (UTC)[reply]
    The concerns I am raising here consumed this board in an extremely ugly way not very long ago. Please read Wikipedia_talk:WikiProject_Medicine/Archive_110, which is only one place where the matter was discussed. I am raising the kinds of questions that people who were very, very upset in that discussion were raising, so that they are raised now, at the beginnning. I am trying to avoid us going through that again.Jytdog (talk) 20:12, 11 July 2018 (UTC)[reply]
    OK, I can see you are arguing from a precedent. But not the exact relevance of that precedent to anything within my control.
    You might assume my good faith, and appreciate that volunteer engagement in the project I'm introducing has been the constant theme of ContentMine's dealings with the WMF throughout. Indeed the reporting at m:Grants:Project/ScienceSource/Timeline is dominated by a need to report in detail on all the many efforts ScienceSource has said it will make to engage volunteers.
    We were offered the full grant we applied for, in the middle of May. but conditional in particular on submitting a detailed publicity plan to the WMF. The metrics in that (submitted) plan are as given in the long second table. (Full disclosure: the grant-runner has not yet signed off on the plan.)
    Perhaps this will have some clarifying value for you. If the precedent you see is of commercial promotion on Wikipedia, and you see an analogy, there is this: the Foundation has made the grant to ContentMine, in their kind of phrasing, while trusting the company on tech, but regarding its track record as weak on self-promotion. In short, the driver of something that may be bothering you is the WMF. Charles Matthews (talk) 20:45, 11 July 2018 (UTC)[reply]
    Nothing is bothering me and I am not arguing anything. I am trying to help get a full disclosure on the table. That full disclosure would have been a useful starting point for discussion to get further buy-in from the editing community.
    You do not understand where I am coming from with regard to companies doing work in Wikipedia generally, nor with respect to the issues here.
    I think people from companies have a place in WP, but their presence and activity have to be managed very carefully, with very clear disclosure and full buy-in from the editing community. I was not one of those, who was very upset over the Osmosis thing. I listened to those who were.
    Based on your approach to this discussion, I will not be participating in this project.
    I recommend that folks here steer very, very clear of this project. Jytdog (talk) 21:05, 11 July 2018 (UTC)[reply]
    I'm naturally sorry to hear that. You may not, in turn, understand where I come from on COI. I was deeply involved in drafting WP:COI in 2006, and commented at the time that the issue was going to be of increasing importance to Wikipedia. Charles Matthews (talk) 21:12, 11 July 2018 (UTC)[reply]
    I don't care where you are coming from nor what you have done in the past. Your approach to this discussion is all I needed to turn and walk away. Jytdog (talk) 21:37, 11 July 2018 (UTC)[reply]

    In the cold light of early morning now, it seems to me that I can do little good in returning to this thread; but on the other hand that I need to. I should at least note that my reply just above, mentioning COI, was not in fact to the text immediately above it, where the language on "full disclosure" now stands: "full disclosure" was introduced in one of the series of edits marked "ce" that can be seen in the page history, and my reply was to a previous version of what Jytdog wrote. I take it that from full disclosure we should be reading this as "full disclosure, the acknowledgement of possible conflicts of interest in one's work", but applied to the company. Yes, it could be a problem if the company's "intended use" of the grant had some conflict of interest with the missions of Wikimedia (WMF or movement), or Wikipedia. Companies can be led to behave in ways that "increase the value of the company", and so on, in the terms in which the question was first posed.

    It has been noted before that the project has little to do with Wikipedia editing, and in Wikimedia terms is centred round Wikidata. I frankly don't see that ContentMine has any conflict of interest in what it intends to do in relation to Wikidata.

    It is ironic to me, at least, that this is a tale of two guidelines, running from MEDRS, where I'm actually just learning the ropes, to COI, where the distinction between potential and actual COI in Wikipedia terms is burned on my Wikipedian heart. Charles Matthews (talk) 04:22, 12 July 2018 (UTC)[reply]

    Nope. If you don't understand something you should ask.
    If you had taken the time to read the Osmosis discussions (which you obviously have not) you would have seen that one of the things that upset people here the most, was the way that the company was leveraging the relationship and the work for its own benefit in ways that were not fully explained up front. that is one of the things that was upsetting people.
    That is what this section was intended to get stated now. I was asking you to "fully disclose" how the company would benefit from this relationship and how it would use the products, such that those kinds of questions would not arise only after the fact.
    The response only makes it more clear to me that I have no desire to be involved with your project, and intensifies my warnings to others to steer clear of this. Jytdog (talk) 18:01, 12 July 2018 (UTC)[reply]
    I'm confused Jytdog as to what is not being disclosed. What exactly is full disclosure to you? Expecting a perfect summary of each possible way an entity may benefit from a specific action is naïve, and I find Charles Matthews answers to be exemplary. Even if we demanded a review of each possible way something may benefit the entity — we could never get a satisfactory answer, because giving such an answer is impossible as it is not possible to know what the potential benefits may be. The fact that he acknowledges COI is enough, and per our guidelines (WP:MEDCOI) it does not bar him from making contributions.
    I think it is totally unreasonable to suggest anyone read about the Osmosis case, and it would be nothing but a massive waste of time. That whole debacle was a tempest in a teapot — and the amount of text is enough to keep someone occupied for days straight just reading back-and-forth from a heated and sometimes quite irrational debate. There were some potential issues, but nothing near what it was turned into — and even then — the Osmosis case is so dissimilar to this case as to be entirely irrelevant.
    However seeing as you seem to have identified some problems, would you please opine on what you think is "leveraging work of the community too far in favour of a company", and whether there are any specific responses that are problematic? Saying that someone hasn't taken time to read about every possible past case or that disclosure is not "full" from the get-go — as well as saying that someone "should ask", when they are engaging in discussion — is not constructive. Carl Fredrik talk 19:16, 12 July 2018 (UTC)[reply]
    You are wanting rehearse the Osmosis thing and I am not going there. You (and everyone else) will of course react to this proposal and to these responses how ever you see fit. Jytdog (talk) 19:33, 12 July 2018 (UTC)[reply]
    Points from me, and thanks to Carl Fredrik for giving a third opinion:
    • I'll take up the logo thing, since at very least it is possibly misleading.
    • I reject the implication made of an oppressive form of argumentation, which seems odd given the way I linked to the related issues raised by John Cummings.
    • Some of what is said above is hurtful, but I'm content just to move on now.
    • The Osmosis reasoning seems to me guilt by association, and while I didn't know about it as a matter of background, and I don't mind frank discussion about my own actions, I'm not happy with the way it has been used here as a slur against ContentMine. Charles Matthews (talk) 03:47, 13 July 2018 (UTC)[reply]

    Note about the WMF vs en-WP vs WP:MED

    User:Charles Matthews I see that the company is already citing the WMF wikimedia movmement logo on its partners page. Please be aware (if you are not) that the WMF is not the same as the English Wikipedia (or any other WMF project) and that each project governs itself. Please be careful that the company avoids saying things like it is "partnering with Wikipedia" or "partnering with WikiProject Medicine in Wikipedia" unless you get absolutely clear buy-in from the editing community. (something like an RfC perhaps) We just went through a very ugly thing (which I am not trying to revive, everybody!) where a company that was uploading open-licensed medical videos to WP was... overly aggressive in claiming a relationship with Wikipedia and this with this wikiproject in its promotional materials, which was part of what led to a severe backlash here (which is still recent and somewhat hot). OK? (I am posting this here, so it is very public, and very unambiguous) Jytdog (talk) 17:16, 11 July 2018 (UTC) (fix per comment below Jytdog (talk) 23:37, 11 July 2018 (UTC))[reply]

    That's the Wikimedia movement m:logo, not the foundation's logo. The foundation's logo has a single, uniform color (usually 100% black, not multi-colored) and says "Wikimedia Foundation" underneath it. WhatamIdoing (talk) 18:11, 11 July 2018 (UTC)[reply]
    Jytdog, thanks for the heads-up anyway. I'm not management, I am a 15-year Wikipedian, I'll try to keep them straight. Charles Matthews (talk) 18:30, 11 July 2018 (UTC)[reply]
    Not a 15-year-old Wikipedian, otherwise you're really not wearing well, Charles, --RexxS (talk) 19:06, 11 July 2018 (UTC)[reply]
    If it's ageist jokes now, you dinosaur, I'm going to ask whether you were suited up in that Thai cave. Charles Matthews (talk) 20:51, 11 July 2018 (UTC)[reply]
    User:Charles Matthews, away above you wrote my company; just above here you wrote I'm not management, I am a 15-year Wikipedian. The lack of clarity and hat-switching is very unhelpful and is the kind of thing that will cause major problems later. Jytdog (talk) 18:06, 12 July 2018 (UTC)[reply]
    Surely not. "My company" = "the company I work for", not "the company I own", certainly. My position is defined as Wikimedian in Residence. Roles in the team were defined at m:Grants talk:Project/ScienceSource#Comment 6 and hold with the slight revision that the "work package owner" parts of job titles are not being used. Cesar Gomez is the manager. Charles Matthews (talk) 09:00, 13 July 2018 (UTC)[reply]

    Commons Picture of the Year

    c:Commons:Picture of the Year/2017/R2/Gallery is in the second round of voting now; if you're looking for a little break, then head over and take a look. While you're there, you might want to look at c:Category:Featured pictures on Wikimedia Commons and relevant categories to see whether there are any images that might be useful in your favorite articles. WhatamIdoing (talk) 21:29, 10 July 2018 (UTC)[reply]

    HgenetInfoDB

    The gene database HgenetInfoDB does not have its own article (and apparently never has) but it is still linked from over a dozen articles. However the database itself appears to be offline [4], suggesting that all wikilinks to it and links to urls in the database should be removed (typically they are included in infoboxes like the one on 5-HTTLPR). IntoThinAir (formerly Everymorning) talk 16:33, 12 July 2018 (UTC)[reply]

    All the infobox ones can be removed by altering the Template:Infobox single nucleotide polymorphism. Would be good to mention their. Doc James (talk · contribs · email) 17:17, 12 July 2018 (UTC)[reply]
    It looks like it was added in 2008 by a logged-out editor from Denmark. I don't know who to ask/who might know about this. Maybe a question for Wikipedia:WikiProject Molecular and Cell Biology? WhatamIdoing (talk) 20:59, 12 July 2018 (UTC)[reply]

    Wiki Project Med Foundation is Applying to be a Thematic Organization

    Please support here meta:Wiki_Project_Med/Application#Support_Us! Doc James (talk · contribs · email) 17:15, 12 July 2018 (UTC)[reply]

    great endeavor--Ozzie10aaaa (talk) 14:58, 13 July 2018 (UTC)[reply]

    Research funding by JDRF

    We have a conflicted edit request from an intern at one non-profit (a watchdog of other nonprofits) about another non-profit's level of research funding, sourced to the watchdog's website. See Talk:JDRF#Request_edit_on_12_July_2018.

    Not strictly medical, but related to stuff we do here. Am interested in others' thoughts. Am unsure what is good, in this situation. Jytdog (talk) 17:43, 12 July 2018 (UTC)[reply]

    commented--Ozzie10aaaa (talk) 20:27, 12 July 2018 (UTC)[reply]

    color x-ray

    should be placed in an article or made standalone article[5][6]...IMO--Ozzie10aaaa (talk) 20:43, 12 July 2018 (UTC)[reply]

    It's a press release -- not even a primary source. Without better sourcing, there is no justification for doing anything with it. Looie496 (talk) 00:17, 13 July 2018 (UTC)[reply]
    We already have an article on medipix, the underlying CERN technology. If/when RS become available for the "color" x-ray device, they could be added to the medical imaging subsection there. --{{u|Mark viking}} {Talk} 00:29, 13 July 2018 (UTC)[reply]
    Picture looks like a CT image not a plain X ray. Agree will need better sources. Can go here Medipix#X-ray_imaging Doc James (talk · contribs · email) 13:42, 13 July 2018 (UTC)[reply]
    University of Canterbury--Ozzie10aaaa (talk) 13:56, 13 July 2018 (UTC)[reply]
    As far as I can make out, what they are doing is CT scanning using a mixture of X-ray energies, and using differential attenuation of different frequencies to assign colors to voxels. Looie496 (talk) 16:23, 13 July 2018 (UTC)[reply]

    Feedback request

    I'm requesting feedback on article content in nootropic. The issue is described at Talk:Nootropic#Coverage of CNS stimulants. Seppi333 (Insert 2¢) 07:52, 13 July 2018 (UTC)[reply]

    Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:WikiProject_Medicine&oldid=850101796"