Hello WikiProject COVID-19!! Please join the Brooklyn based Sure We Can community for our 2nd NYC COVID-19 themed Wikipedia Edit-a-thon / translate-a-thon - ONLINE - this Saturday, Nov 21st, 2020 11am - 1pm. The edit-a-thon is part of Sure We Can's work with NYC Health + Hospitals to stop the spread of Covid-19. We plan to continue to work on translating the COVID-19 pandemic in New York City article into some of the many languages spoken in New York City; as well as, work on other ideas about how wikipedia could slow the spread of Covid-19.
We’ve compiled, with the help of translators, translations of the first four paragraphs of the COVID-19 in NYC article into Swahilli, Malagasy, Yoruba, Hebrew, Hindi, and Tagalog. You can see the translations in the Meetup page. We are looking for help publishing this translated information onto the appropriate wikipedia projects.
Maybe it's time to update the color schemes on COVID-19 infection data
Hello! I am not a member of WP:COVID-19, however I find the current color scheme for infections on the map to be quite lacking. Take a look at the US-county level data. Alot of states are full black (500+), and there is no distinction of the higher levels of infection. I suggest that we update the color scheme so Black is at 750+, and we introduce a few new levels in-between. Not necessarily the biggest concern, but (for example) showing Wisconsin as entirely black, is not really useful, other than showing that there are a heck-ton of cases here. JackFromReedsburg (talk | contribs) 23:40, 11 November 2020 (UTC)
Agreed, although this is just a symptom of the larger problem we have of our maps not being properly automated. There are hundreds of COVID-19 maps needing constant updating, and it would be a great use of our time to automate the process (including choosing scales), but for a variety of reasons it just hasn't happened. For background, feel free to peruse the maps current consensus items above. {{u|Sdkb}}talk23:58, 11 November 2020 (UTC)
I have a problem with a color scheme too. Is it proper to include the whole map here so it is easy to see what I am seeing? Because I don't know how to just link to the map. The darkest color is in use, but it says the darkest color is for 100,000 cases per million. No one has reached that point, or even close!
Graphs are unavailable due to technical issues. Updates on reimplementing the Graph extension, which will be known as the Chart extension, can be found on Phabricator and on MediaWiki.org.
Hi all - with the pandemic stretching on months longer than many anticipated, I'm finding it overwhelmingly difficult to alone update my local version of the {{COVID-19 pandemic data/United States}} chart, especially as old data has been changing daily. What is do-able is a monthly chart - I just created one at COVID-19 pandemic in Columbus, Ohio#Cases; this may be useful for others where the raw data isn't easily grabbable and/or it's just one person updating it. ɱ(talk)22:16, 19 November 2020 (UTC)
Ɱ, that looks not only much easier to maintain, but also easier to digest for readers and more encyclopedic as opposed to newsy. I'd still prefer to see updating done automatically, but it should be modeled more around something like that. {{u|Sdkb}}talk00:03, 20 November 2020 (UTC)
I don't think we'll ever get to automatic updates in my case, I've already tried reaching out to the Ohio Department of Health about more easily usable data; no reply. ɱ(talk)00:14, 20 November 2020 (UTC)
This one is a biggie... the vaccine races will heat up in the following months. We should decide some standards on how to treat them. Should each vaccine candidate have its article? -- {{u|Gtoffoletto}}talk14:52, 9 November 2020 (UTC)
Follow the usual criteria I think. If a vaccine candidate gets enough media attention and it's plausible it will stay relevant then we can create an article. The first few to finish phase III trials and/or get approved will certainly satisfy that. --mfb (talk) 07:18, 12 November 2020 (UTC)
I'm not sure I have any overarching ideas, but just wanted to kick off a discussion. Rooting around through some of the new/lesser-watched pages in this project, I'm finding quite a few that might be better as part of some other page rather than on their own. I'd say it's one of the trickiest parts about editing in this area: there are a lot of topics, particularly in the "impact" area, that might seem notable if you don't realize that there's already another article already exists that covers most of it (a WP:CONTENTFORK).
A few of these could be helped out by excerpts, but they're still in the process of getting off the ground, and most editors creating the mass of pages don't seem aware of or interested in them. The upshot of all this is that there's a massive amount of duplicated/wasted editor effort spent building things that already exist (or should exist) elsewhere, which is a huge issue since every instance of redundancy cuts our productivity literally in half.
Does anyone have thoughts about how we could try to address this at a macro scale? More redirects/integration would help, since it'd at least make people more likely to become aware of a page before they fork it, but I'm not sure how we'd even encourage that more than we already are. {{u|Sdkb}}talk22:18, 18 November 2020 (UTC)
Sdkb, I'm afraid I don't have any concrete ideas. We could try and make a general conventions list for how articles should be titled to prevent similar content from being produced in redundant pages. For the politics example, I think it would be easier to contribute on the article of the aspect rather than the nation, and then excerpt it over.
I totally agree. I think we are loosing grasp of the COVID pages. Even the main articles are severely understaffed and contain significant mistakes. Most of those problems stem from extreme duplication of information across the pages. We need to restructure and determine some rules ASAP. I keep seeing pages with enormous problems and I just don't know where to start fixing it knowing that the content is duplicated in 5 other articles with slight variations.
I see 2 main problems:
lack of structure with the 3 top pages: we need to resolve the discussion regarding the pandemic/disease/virus articles now. We need a clear direction that eliminates duplication and upon which all sub pages will originate. See discussion here: [1]
lack of guidelines relating to sections and sub-articles. An example is Coronavirus disease 2019#Diagnosis) which is just embarrassing at the moment. There is a subpage COVID-19 testing which is huge. I think we need to make a rule where the section in the main page includes content from the subpage (lead translusion?) so that we have a summary-subarticle approach for all those topics. (relevant discussion here: [2])
lack of guidelines with regional/national articles. When is an article appropriate? Next problem to tackle...
I think the first two issues are crucial for this project. If we don't fix them here there there is no way small discussion on individual pages will ever reach a solution. TenryuuSdkb
-- {{u|Gtoffoletto}}talk14:57, 23 November 2020 (UTC)
Agreed. More on the mental health impact of the pandemic overall is needed, but this isn't quite that as it's also about the decline in precautionary measure observance caused by the fatigue. {{u|Sdkb}}talk18:48, 18 November 2020 (UTC)
I noticed Northeastern University COVID-19 Testing Center in a new-article bot report, and it looks like it needs attention. Devoting an article to a single testing facility seemed like an odd move. However, most of the page isn't even about the facility, instead talking about the university's response in a way that might belong on their website but not here. Most of the footnotes are to a university memorandum that doesn't include the information being cited to it; for example, it has no mention of a "Daily Wellness Check form" or a "testing consent form". XOR'easter (talk) 19:17, 24 November 2020 (UTC)
XOR'easter, thanks for bringing it to our attention. I see a lot of copyediting issues to begin with, but most of it doesn't appear to be encyclopedically useful. I might jump onto that article's talk page to see what can be salvaged. —Tenryuu 🐲 ( 💬 • 📝 ) 04:05, 25 November 2020 (UTC)
Update: I moved it to draftspace for the time being. I don't think it necessarily merits a deletion, but it needs a lot of work before it's suitable for articlespace. —Tenryuu 🐲 ( 💬 • 📝 ) 00:36, 27 November 2020 (UTC)
RexxS What is the MEDRS stance regarding vaccine results? Obviously the results will be published by primary sources. Should we await that some independent organisation reviews them somehow? E.g. FDA approval? The official publication of results probably can be considered "general information" and therefore we could use primary sources if we avoid saying "it is 90% effective" and just report the fact that positive results have been published but the vaccine must still be approved. Might be tricky. -- {{u|Gtoffoletto}}talk14:04, 23 November 2020 (UTC)
@Gtoffoletto: my preference is to wait for good secondary sources to review the primary reports, but I'm aware of the pressure to acknowledge important results. I think we should learn to summarise facts accurately, and not rely on news media, who tend to want to emphasise the headlines. For example, once the AZD1222 interim results are published in the Lancet (or similar), we might write in a Research section something like:
The interim results from AZD1222 phase 3 trials found that the group of 2,741 volunteers receiving a half-dose (~2.5 x10^10 viral particles) followed by a full dose (~5 x10^10 viral particles) some weeks later resulted in 90% fewer participants contracting COVID-19 than the control group. The group of 8,895 that received two full doses showed a corresponding 62% reduction in infection. <referencing the scholarly article>
We should not be writing:
AZD1222 is 70% effective at preventing COVID-19. <referencing BBC News>
I hope editors will try to stick with the best sources. In most cases there will be full results published in scholarly journals (primary sources) and we can use them with care, but quality reviews are certain to follow. --RexxS (talk) 14:52, 23 November 2020 (UTC)
@RexxS: I see problems with both proposed texts. The first one is accurate but it is a blatant violation of MEDRS rules. We are clearly referencing a primary study for biomedical info. What if we try to dodge that with something like:
AZD1222 showed positive results during phase 3 trials and has been submitted for review/approval... <referencing the scholarly article OR even better the request for approval or something similar>
This way we are reporting a piece of general information but not a Biomedical claim. In this case MEDRS allows for primary sources. -- {{u|Gtoffoletto}}talk15:18, 23 November 2020 (UTC)
@Gtoffoletto: what are the biomedical claims that my first piece of text is making? – the "contracting COVID-19" and "infection"? It would be clumsy, but I suppose we could use
The interim results from AZD1222 phase 3 trials found that the group of 2,741 volunteers receiving a half-dose (~2.5 x10^10 viral particles) followed by a full dose (~5 x10^10 viral particles) some weeks later produced 90% fewer participants testing COVID-19 positive than the control group. The group of 8,895 that received two full doses showed a corresponding 62% reduction in COVID-19 positive results. <referencing the scholarly article>
Ultimately it's difficult to report any test results from primary sources if we have to hedge even basic issues of measurement that nobody is going to dispute.
I should add that I think your suggested summary is well wide of the mark. Read https://www.astrazeneca.com/media-centre/press-releases/2020/azd1222hlr.html and try to anticipate what the full data will look like – summarising that as "positive results" is editorialising a primary source and does a disservice to our readers who should be making that judgement for themselves. The issue of "has been submitted for review/approval" is a completely different point and I seriously doubt that a journal article discussing the data and preliminary analysis will mention it. --RexxS (talk) 21:33, 23 November 2020 (UTC)
@RexxS: The claim is that the vaccine "works" which is based on a primary source and is against MEDRS. I would like to find a solution that doesn't require an exception to MEDRS if possible. That's why I was proposing relying on a different kind of source (such as the authorisation/approval request maybe?). We can report simply that the vaccine candidate is being evaluated by X after passing the three phases of testing and that would be a piece of general info and not a biomedical claim. Once it is approved the approving authority would certify somehow that the vaccine is safe and effective based on the primary results and would be MEDRS compliant. Or should we just WP:IGNORE MEDRS in this case? Might be a slippery slope from there... -- {{u|Gtoffoletto}}talk19:28, 24 November 2020 (UTC)
@Gtoffoletto: please read the example I gave carefully. There is no claim that the vaccine "works" (whatever that might mean): you are imposing your own analysis on the facts presented. The example is an unadorned report of the interim results of a phase 3 trial. The numbers of volunteers testing positive in two different groups are quite simply matters of fact, not of analysis. Check again what MEDRS requires.
If, at a later stage, a qualified secondary source (such as an approval authority) makes a statement about the effectiveness of the vaccine based on the primary study, we can use it to claim that the vaccine is X% effective, but that inference can't be drawn by Wikipedia editors. --RexxS (talk) 00:39, 25 November 2020 (UTC)
@RexxS: I understand what you are saying. However the same argument could be made of any primary source that we routinely discard per MEDRS. All peer reviewed papers (should) contain facts. However, using those facts to draw useful conclusions is a bit more complex. I might be talking like a MEDRS zealot here but I'm wary of cherrypicking when MEDRS applies. Those are obviously crucial results. So I think as long as we stay away from any direct (or implied) biomedical claim and only report the general information we should be fine with the primary source. I would not give excessive weight to the exact results but would only report the fact that the phase 3 trial has been published/authorisation requested which is the only reliable fact we have. But maybe I'm interpreting too strictly MEDRS -- {{u|Gtoffoletto}}talk09:32, 25 November 2020 (UTC)
@Gtoffoletto: Indeed, the same argument is routinely made about using peer-reviewed sources published in top-quality journals. Normally I do my best to dissuade editors from using them to support biomedical content. However, there is continual pressure on articles surrounding COVID-19 and I've found that It's better to try to educate the super-keen editors who want to report the latest news by attempting to suggest wording that simply extracts factual detail from the sources and avoids any claims of efficacy. etc. I still don't accept that approach for news sources (which regularly get things wrong), but once there is data published in the Lancet, for example, I'd rather suggest acceptable summaries than have to revert unacceptable content, especially if an edit-war starts. WP:MEDPRI outlines the circumstances in which a primary study may be utilised, and it's very important not to draw our own conclusions, even subconsciously.
You state "using those facts to draw useful conclusions is a bit more complex" and I agree – I might even call it an understatement. Drawing useful conclusions from facts is really the job of independent, secondary analysts, which is why we put so much store in systematic reviews. The conclusions of any primary study have to be taken with a pinch of salt, because the authors of the study are clearly subject to the bias of being invested in their own work.
I understand the argument about "excessive weight to the exact results". but in fact, reporting those in the detail I suggest allows a neutral summary of the source and illustrates its importance to our article (when it is important, that is). I don't think we're disagreeing much, but it's interesting to air these differences of interpretation. Cheers --RexxS (talk) 16:55, 25 November 2020 (UTC)
@Gtoffoletto: Yup, and I can't say I'm surprised. Here's a ringing endorsement of our policies from the NYT article:
Menelas Pangalos, the AstraZeneca executive in charge of much of the company's research and development ... responded, "I think the best way of reflecting the results is in a peer-reviewed scientific journal, not in a newspaper."
He might as well have added "... nor in our press release".
It looks like the half-doses were a serendipitous accident, and adds to the concerns about rushing the different phases of trials – we should have had optimal dosage determined in phase 1/2, and ought to be well past that by the time phase 3 interim results are being analysed for efficacy. Of course, when it turns out that all of the volunteers who received the half-dose were under 55, it shows we must have an independent analysis of the full data in order to draw any conclusions about efficacy, and that's not a job for Wikipedia editors. --RexxS (talk) 19:37, 26 November 2020 (UTC)
RexxSGtoffoletto I think WP:MEDPRI probably gives us our best guidance. It is a surprising result worth noting, but phrasing is definitely key to avoid giving undue weight to unproven efficacy. I'd suggest it's best we just leave it less specific and point out the reason why it's notable (it was an error in trial administration, and necessitated an extra study to confirm the result): The interim results from AZD1222 phase 3 trials found that the group of volunteers receiving a half-dose followed by a full dose some weeks later produced a larger reduction in participants testing COVID-19 positive than the group that received two full doses, prompting an additional set of studies to attempt to confirm the result. <referencing the scholarly article>Bakkster Man (talk) 14:37, 30 November 2020 (UTC)
I noticed that SusanLesch recently split off Statistics of the COVID-19 pandemic in the United States from COVID-19 pandemic in the United States to help address technical limits. I'm not sure I can get behind this as a solution. The topic of gathering and reporting statistics about the pandemic at a worldwide level might be notable, but that hypothetical article about the data itself is very different than an article that just presents the data for a single country, which is what we have here. And an article that just presents statistics for a subject of another article isn't a separate topic; it's just looking at the same topic in a different way, which is definitionally a WP:CONTENTFORK. People who come to COVID-19 pandemic in the United States should find all the relevant information about the COVID-19 pandemic in the United States, which includes statistics as needed. Can we find another way to handle this issue? {{u|Sdkb}}talk00:37, 30 November 2020 (UTC)
Sdkb, not unless you're willing to get rid of templates being used elsewhere in the article. If there are citations that aren't cutting it anymore, we can remove those. Alternatively, get rid of the navboxes to save on PEIS.
The major problem is that constant additions to statistics is what's causing PEIS bloat; they generally contribute to roughly at least 25% of the allotted PEIS limit. —Tenryuu 🐲 ( 💬 • 📝 ) 03:06, 30 November 2020 (UTC)
Tenryuu, looking for places to cut, I'd start with the Statistics_of_the_COVID-19_pandemic_in_the_United_States#Number_of_U.S._positive_test_individuals_by_date section. Having lines charting the course of the pandemic for what looks like close to every single state is way overkill; those can be saved for the state articles (the charts are very unwieldy as is, given that they're split into a bunch of different groups, which makes them hard to compare). If there's data available at a regional level (e.g. total cases in the midwest over time), that might be better, and would contribute less to PEIS. {{u|Sdkb}}talk03:14, 30 November 2020 (UTC)
The purpose of those charts is to visually compare the timelines of different states, which is quite useful and appropriate. Yes, having it split between four charts is a bit clumsy, but it's hard to make them readable otherwise. pauli133 (talk) 19:52, 1 December 2020 (UTC)
I'd 'like' to combine the tests taken and test positivity charts, but I don't think we can do two Y-axes in one chart with the tools available. pauli133 (talk) 19:57, 1 December 2020 (UTC)
discussion at 2020 Formula One World Championship
Hello, I hope everybody who is reading this is having a lovely day. There is currently an ongoing discussion at Talk:2020 Formula One World Championship#COVID-19 versus SARS-CoV-2 as to what wording should be used when a notable person tests positive for the SARS-CoV-2 virus. I have tried searching for some broader guidelines elsewhere on Wikipedia, but have not been able to find any. This may be a relevant subject of discussion for this WikiProject, and also Wikipedia as a whole, as there is a potential WP:BLP issue involved. HumanBodyPiloter5 (talk) 07:08, 2 December 2020 (UTC)
Pfizer - BioNTech vaccine
The UK Medicines and Healthcare products Regulatory Agency (MHRA) has issued a press release confirming its approval of the BNT162b2 vaccine candidate for use in the UK. The press release is at https://www.gov.uk/government/news/uk-medicines-regulator-gives-approval-for-first-uk-covid-19-vaccine and I assume it is considered reliable on Wikipedia for its statements, as long as editors are careful not to go beyond what the press release actually says. It's worth noting that this is an approval granted "under Regulation 174 of the Human Medicine Regulations 2012, which enables rapid temporary regulatory approvals to address significant public health issues such as a pandemic."
Can we please try not to use news reports based on the press release? The news media are not qualified to make analyses of medical-related statements by a regulatory agency or similar. --RexxS (talk) 14:50, 2 December 2020 (UTC)
Request for comment
I am requesting comments here concerning what this article, Travel restrictions related to the COVID-19 pandemic, is supposed to include. Velayinosu (talk) 02:03, 3 December 2020 (UTC)
Effects of COVID-19 pandemic on a city, state or country
Greetings. Can anyone here say what a section on the effect of the pandemic on a city should look like? I started one under Economy at Minneapolis, thinking that is what survivors are interested in most. I also asked for best practice at WP:CITIES and WikiProject Minnesota but don't expect there will be an answer. Has any city or state or country done this? Thank you. -SusanLesch (talk) 22:05, 3 December 2020 (UTC)
SusanLesch, I see you've asked for help in several places. Please don't do that to avoid WP:TALKFORKs, and just use {{Please see}} instead.
Why is it too recent for Minneapolis, and not too recent for Minnesota? -SusanLesch (talk)
@SusanLesch: It's not Minneapolis vs. Minnesota so much as a non-COVID page vs. a COVID page. Minneapolis has to cover the entire history of the city, and in that context, despite how big it seems now, COVID-19 is not much more of a blip than the Spanish flu. At COVID-19 pandemic in Minnesota, by contrast, the whole page is about the pandemic, and there's plenty of room for a section on its economic impact (as a more localized version of Economic impact of the COVID-19 pandemic). If COVID-19 pandemic in Minneapolis existed, we'd want to have it there, but that page doesn't exist and shouldn't because it'd be too local. {{u|Sdkb}}talk22:28, 3 December 2020 (UTC)
I can't find it in the archives, but I know there's been some prior discussion about limiting ourselves from creating hyper-local pages. The NYC page is definitely justified, but the Portland page probably should not exist—it's completely unsurprising that it's outdated, which is an indicator it's too specific. {{u|Sdkb}}talk00:48, 4 December 2020 (UTC)
Self-trout. This is what I get for copying only the discussion heading and not the entire page. I've amended the link to point to the appropriate page. Thanks for catching that JackFromReedsburg. —Tenryuu 🐲 ( 💬 • 📝 ) 18:02, 4 December 2020 (UTC)
@Tenryuu: - 'given the change in COVID-19 restrictions between England and Wales'? There has been no change! Or do you mean 'given the differences in COVID-19 restrictions between England and Wales'? There is no change, as the signs have been up since March, and are still up especially by the border. What 'change' are you referring to? Secondly, Tenryuu, it is customary to ping an user when discussing that person's edit. John Jones (talk) 15:56, 5 December 2020 (UTC)
John Jones, I am not particularly concerned personally with the situation over on the page; if I misspoke that's on me. The only thing I was concerned with was that there was possibly an edit war that would not have ended well with either of the offending parties.
[I]t is customary to ping an user when discussing that person's edit. The thread in question has not had any input from you, and the only addition I made to it was to inform anyone keeping up with it that an invitation had been extended to this page, so I am confused as to what prompted you to make such a statement. —Tenryuu 🐲 ( 💬 • 📝 ) 16:07, 5 December 2020 (UTC)
Thanks. In my view, you were discussing my edits. I placed an image in an article. It was then taken down with no good reason. I placed it back again. That in my view is not unreasonable. You then bought the discussion here without informing me on my talk page or pinging here. You added the link 3 hours after the thread was started. Can you please explain which statement of mine you are referring to in your last sentence, which causes you confusion? John Jones (talk) 17:02, 5 December 2020 (UTC)
Use of month categories for Covid location pages
Has there been discussion regarding the categorization as monthly events for the various Covid-by-location pages? For instance, the page for my home state of Illinois (COVID-19 pandemic in Illinois) currently has the following categories (among others):
Even though I am generally a fan of categorization of events by month, this seems like overkill, and it seems like "2020 in Illinois" (or soon, "2021 in Illinois") should suffice. We would not categorize the article France during World War II with "January 1940 events in France", "February 1940 events in France", etc. in an attempt to capture every month of the year. Does anyone know if this has been discussed anywhere thus far? Thanks to all for all their contributions. KConWiki (talk) 17:54, 6 December 2020 (UTC)
Something that spans way more than one month shouldn't be in every monthly category I think. Put all these things in a category that goes in Category:2020 in the United States? Or at least everything in a single category that becomes part of these monthly categories. Otherwise the categories get silly. --mfb (talk) 19:45, 6 December 2020 (UTC)
That does seem concerning. I'm not familiar enough with categories to weigh in, but further discussion seems warranted. Perhaps issue some invites to here on pages category folks are likely to be watching? {{u|Sdkb}}talk22:51, 6 December 2020 (UTC)
This stub was moved to Draft space on or about the 1 November, so that it could improve the number of citations it has. Of course it has done no such thing, though it might have, indeed probably would have in mainspace. I was looking for information on this topic and have created a redirect at Bamlanivimab. Had I not done this, I would not have been aware of the draftified stub.
Could the good members of the project, please look at moving this back to article space and bringing it up to date? It's a rather important topic.
This article, which includes a hoverable OpenStreetMap at county granularity, focuses on Gove County. Whether it's emblematic or no, Wikipedia should have such maps, or access to such maps, for counties/parishes/boroughs.
Furthermore, I was surprised that Impact of the COVID-19 Pandemic on Kansas doesn't exist. Each state has had its own response, or lack of response. COVID-19 websites went up in a certain sequence, etc. This should probably be separate from the Gubernatorial responses as such, but a solid template is needed. I'm not here to write stubs. kencf0618 (talk) 15:43, 12 December 2020 (UTC)
Better, more interactive maps are absolutely needed. We've done some innovative things by our normal standards (e.g. the timeline map, for which kudos to Wugapodes), but there's a ways to go before we catch up to the presentations used by news orgs.
To be frank, we likely will never "catch up" to news orgs because we simply do not have the infrastructure or labor. Unlike MediaWiki, news orgs can have by-page custom JavaScript that allows them to do just about anything, but the only page-specific option we have is mw:Extension:Graph. The software back-end for interactive graphs--Vega 2--is from 2016 and is 3 major versions behind the stable release (currently Vega 5), and the extension itself is pretty much unmaintained. We simply don't have the capabilities of a news website or custom-built webpage, and instead of trying to replicate existing coverage, we should focus on areas outside the US without high level data journalism. — Wug·a·po·des23:05, 12 December 2020 (UTC)
There are plenty of places within the U.S. without high level data journalism, but given that the manifold aspects of of the COVID-19 Pandemic shall be the subject of academic study, God knows how many books, and so on, I think we're well placed to use that corpus of information as it becomes available. Contemporary journalism is the just the first draft of history, after all. kencf0618 (talk) 11:03, 13 December 2020 (UTC)
Medical cases charts - change type
The medical cases chart module offers "p" (percentage) and "a" (absolute) for the display of changes in cases as well as deaths. To me the default "p" only made sense during the first wave, where recoveries and deaths didn't make for much of the cumulated cases, so percentage or even doubling time were good measures for the speed the disease spreads with. But once most of the cumulated cases are recovered (like in a second wave), and thus not contagious anymore, this stops making any sense. You simply compare to a number that's no factor at all for the current epidemic development. Furthermore I can't see percentage used in publications anymore since months, be it press or medical. For these reasons I dared to change both options to "a" in Template:COVID-19 pandemic data/Germany medical cases chart, but while some agreed to the reasoning on the local talk page, others were competely outraged for the one and only reason, that almost all other country charts use percentage. So the question is, should all countries medical cases charts use the same option for cases and deaths differences, and if so, which one? My personal opinion would be: Leave it to the different countries chart templates, but give a recommendation to use percentage until the peak of the first spread (wave) is reached, and absolute afterwards (cases). And for deaths I cannot see much sense for percentage at all, so I'd recommend to use this option only, if it's predominant in current publications, and absolute in all other cases. -- Kohraa Mondel (talk) 13:15, 1 December 2020 (UTC)
@Kohraa Mondel: You mentioned that different countries should be allowed to use what option they want to use, but someone appears to be aggressively changing that on some templates without the necessary agreement of other editors. Edits like this, this, and this are becoming troublesome. The change was recommended but not required, but this user was editing several other templates just to force his indirect opinions on those. Can others discuss this proposal first before implementing the change? LSGH (talk) (contributions) 01:41, 3 December 2020 (UTC)
In the first place this of cause was a proposal to be discussed here. Further on there is no project wide current consensus on that topic, so country's templates are not bound to a certain option, but still an agreement among the maintainers of respective articles is required. Also a more comprehensive style of reasoning, than in the examples you linked, would be highly recommended. But what can I do about it? In my opinion it would be best if some more people take part in the discussion right here, and we hopefully reach some agreement that might lead to a current consensus on that topic. -- Kohraa Mondel (talk) 05:50, 3 December 2020 (UTC)
If an RfC could be held here, then we would need to inform the several other maintainers that there is a certain part of the template which we all need to agree on before implementing or rejecting the change. The proposal is good and well-intentioned, but arguments and concerns from them need to be heard first. A situation where the type used (percentage or absolute) is not always the same will confuse readers. In my opinion, considering the outcome of such an RfC, the template maintainers themselves should be the ones deciding if and when they want to implement the change on each template that they maintain. LSGH (talk) (contributions) 10:13, 3 December 2020 (UTC)
As it looks to me, we share mostly the same view on that topic. Only in one point I'm not quiet clear what you mean: you'd rather demand all articles having the same option used, or leave it with a recommendation? Anyways, is there any procedure or means to RfC the maintainers? -- Kohraa Mondel (talk) 11:45, 3 December 2020 (UTC)
The maintainers would need to be notified of an RfC, which may be held on this talk page, but I do not also know how to do so. Maybe a notice on some other project namespace pages of this WikiProject may suffice, though some might not notice if country-related articles are their primary area of editing. Of course, inputs from other editors and casual readers are also welcome. The objective could be to discuss if using the absolute type is allowed, and the result will be a recommendation. The norm as of this time is to use the percentage type, but because it's not set in stone, an argument similar to "because it's used in all other templates" should not be considered. I won't demand that one option or the other be used everywhere, but the decision should come mainly from the maintainers. LSGH (talk) (contributions) 14:52, 3 December 2020 (UTC)
Ok, not too many editors have been attracted to take part in this discussion, or maybe placing requests for comment in the charts talk pages might not have reached the majority of maintainers. So I'll take an approach by a) posting RfCs at the countries pages themselves (top 20 by cumulated case count) and b) define my own idea of criteria for the decision for discussion. In my personal opinion criteria for new cases metrics should be: 1) what kind of metrics is predominantly used in current publications worldwide, 2) what kind of metrics depicts the infections dynamics best, and 3) what kind of metrics makes ongoing infections development best comparable between countries. There (in my restricted personal view of cause) predominant in publications is absolute daily difference, 7-days-incidence and 14-days-incidence. Infections dynamics would be depicted best by relating new cases to active cases (to those that are conatgiuos, but this might be original research). And finally comarability might be best if we uniformly display 7-days- or 14-days-incidence on all charts, but this would require some augmentation of the medical cases chart module. Anyways, I would be happy to gather some more opinions here, as I am of the decided opinon, that the current use of percentage is of not much value or relevance at all, and thus should be changed. -- Kohraa Mondel (talk) 22:40, 5 December 2020 (UTC)
As far as I can see, many countries already report absolute numbers instead of percentages. I'm not sure about weekly increments; most of the actively maintained templates are updated daily, though some countries choose not to report on weekends, even if the number of cases in those countries keep on rising. LSGH (talk) (contributions) 10:20, 6 December 2020 (UTC)
Some countries national health organisations report last 14 days infections per 100,000 inhabitants (like Spain) or last 7 days per 100.000 (like Germany). This is usually done on a daily basis, even though some countries do not issue reports on weekends. To integrate that kind of metrics into the chart, the lua-module behind the charts would require some augmentation, and an additional parameter providing the number of inhabitants would be required from maintainers. Until (if a all) such programming is done, I'd still vote for (no suprise) recommending absolute differences. -- Kohraa Mondel (talk) 14:59, 6 December 2020 (UTC)
That could become an optional parameter or two, but if it needs too much code, a new template might be needed. Besides, not all countries are publishing such a figure. The percentage option is good for comparing the rate of increase, but there are other places in the country articles that are better suited for that purpose because the bars in the charts do not use the logarithmic scale. LSGH (talk) (contributions) 00:30, 7 December 2020 (UTC)
You'd need no more additional data than before, only population count. So you do not depend on any (health authority) publications, if that's what you mean. It's as simple as subtracting cumulated cases of 7/14 days before from current, and dividing by population. -- Kohraa Mondel (talk) 13:54, 7 December 2020 (UTC)
Just the same way as with percentages or absolute difference. This would be the publication based incidence (New infections published over last 7/14 days). That would slightly differ from from incidences based on registration, as published by some health authorities, but actually less biased (registration based incidence often lacks cases already registered but not yet committed to some national health instance, thus reducing the outcome). -- Kohraa Mondel (talk) 16:28, 10 December 2020 (UTC)
Regarding rate of increase: if you had a heavy amount of infections during first wave, your numbers are way smaller than if you had a lightweight 1st wave infection count. So comparability (between countries) is only given in the rather rare cases of very similar temporal development. -- Kohraa Mondel (talk) 14:10, 7 December 2020 (UTC)
How about countries which have not yet appeared to flatten the first wave or enter a second wave? The rate of increase would be less relevant because it may have already gone down but the absolute number of daily cases still remains constantly high. LSGH (talk) (contributions) 10:16, 10 December 2020 (UTC)
To all I know of exponential growth, the discussed percentage is the reproduction rate (not to be confused with the reproduction number) as long as all, or at least almost all, of the base count takes part in reproduction (and thus growth). Meaning, the discussed percentage is as long a relevant metrics as the amount of active cases makes up the wide majority of the cumulated case count. This is usually the case in a (strongly positive) exponential growth phase of a 1st infection wave, and might also be in such a phase of a 2nd wave where the magnitude outnumbers the 1st wave by far. But in almost all countries this is not the case anymore. So the ideal metrics to depict an epidemics current dynamics would be to relate new cases to pre-existing active cases. And also this would mean comparability between countries as a measure of how agressively the virus is spreading. Only problem is that this is considered original research, you'll find such measures in population growth (like birth per year per living woman) but up to now I have not come across such thing in publications on covid. And also some countries do not register or estimate recoveries. -- Kohraa Mondel (talk) 15:14, 10 December 2020 (UTC)
I edit case counts and charts for numerous US states, and have also realized that the delta columns in the cases chart are no longer quite as relevant. Perhaps these columns could show the change from the 14 days prior value? I agree this is stepping out onto the slippery slope of original research. The current way is however still useful to compare the death rate with the few week's earlier new case rate. As to the original question, my policy is to use (a)bsolute for counts under 1000 and (p)ercent for above 1000. Counts under 1000 have less than three significant digits, thus not quite enough to calculate a percentage that will be shown with up to two significant digits. With larger numbers, an (a)bsolute delta becomes incomprehensible. Example from today's data: +77 or +1.1% increase? EphemeralErrata (talk) 09:10, 13 December 2020 (UTC)
14-days-incidence or 7-days-incidence IMHO wouldn't be original research, such numbers are published by several countries as spain or germany and can be calculated from daily cases reports plus number of inhabitants. Relating new cases to active cases would be a very useful metrics, but is not used with covid, so you might have original research there - you rather see the basic reproduction number that can't be calculated by published data only and itself is not available in many countries' daily publications. About comparabilty of percentages: For me it was the main reason to push the change to absolute when I found that the view get's biased: For todays (dec 14) US cases you get +1,1% that is +177266 and for the 13 days prior +1,1% you actually have +148440. -- Kohraa Mondel (talk) 10:03, 14 December 2020 (UTC)
If I'm not mistaken that would require changes in the underlying medical cases chart module (lua code). Or two different chart templates must be maintained and embedded in the switcher template on the countries main page. Is it that what you mean? -- Kohraa Mondel (talk) 05:21, 9 December 2020 (UTC)
The button there would duplicate what is already displayed. Maybe a toggle that is similar to what is used in Jakarta would do this better? I was looking for the parameters in Template:Infobox settlement that would produce the toggles that allows the template to show any one or all of the maps, but I could not find it on either page. LSGH (talk) (contributions) 10:16, 10 December 2020 (UTC)
The switcher template does the trick (see example, template). Only, as with the maps, you switch between current versions ob two different objects and thus would have to maintain two different medical cases charts for each country. -- Kohraa Mondel (talk) 15:24, 10 December 2020 (UTC)
We've got some more out-of-control forking, folks. National responses to the COVID-19 pandemic in Africa was created a few weeks ago, which is in addition to National responses to the COVID-19 pandemic and COVID-19 pandemic in Africa (which includes sections on a bunch of countries). All of these, unless they're using excerpts, duplicate the leads of the individual country pages. This is not a small problem—every single fork doubles our work, and makes things more confusing for readers. I'm too burnt out to go into battle with whoever created this one, but I'd strongly suggest this project make eliminating forks and nipping them when they're created more of a priority. We have enough work to do building and maintaining all the actually different COVID-19 topics, let alone when we allow multiple pages covering the same thing. {{u|Sdkb}}talk20:48, 14 December 2020 (UTC)
In my view we are here to help organize and facilitate more knowledge... we should not try to suppress the accumulation of more information by telling country projects they should not move or split info - especially when it's causing accessibility concerns in references because of template limits. A split or move is the preferred way to move forward over deletion in many cases WP:SPINOFF. As articles get to big/detailed - especially when causing accessibility problems because of template limits- information should not be deleted or superseded simply because of sizes... but rather moved to retain the information be it to a different article or in some cases a new article WP:SUBARTICLE. Many countries have multiple articles covering different aspects of the pandemic....be it country/provincial/state level articles or thoses covering specific topics like economic impact or government response. I agree as this is done main articles that become summary overviews should do some transcluding if and when possible WP:DETAIL.--Moxy🍁23:25, 14 December 2020 (UTC)
This sounds like an inevitability, though. With the influx of information (from reliable sources) about COVID-19, it seems like that pages that specific will end up being created naturally when pages get split. It sounds like this should be an issue that is addressed by a general COVID-19 pandemic consensus item that should be workshopped before being put up for an RfC. —Tenryuu 🐲 ( 💬 • 📝 ) (🎁 Wishlist! 🎁) 00:26, 15 December 2020 (UTC)
The creation of more detailed sub-articles is natural and warranted. The creation of overview articles with a slightly different scope while existing articles remain underdeveloped and become outdated is also natural and absolutely not warranted. There seems to be something that just makes people to want to start fresh, and it's a huge problem for this project given that our articles don't naturally divide the same way e.g. WikiProject Animal's do. We need to control the forking much more tightly than we are, or the mess is just going to sprawl further, wasting editor effort and confusing readers. From my perspective, the problem is quite obviously out of control, but if we can't even agree it's a problem then we're in for a bad time. {{u|Sdkb}}talk00:50, 15 December 2020 (UTC)
Aside from creating merge/split proposals for individual pages I don't see there being much that individual editors can do without starting an RfC or similar discussion to decide how and when articles should be split off for this WikiProject. —Tenryuu 🐲 ( 💬 • 📝 ) (🎁 Wishlist! 🎁) 02:43, 15 December 2020 (UTC)
As of today, Britain's first guidance on how to diagnose and clinically manage long covid is published (https://www.nice.org.uk/guidance/ng188). As a result, the page Long Covid(edit | talk | history | links | watch | logs) needs overhauling, as do various other articles. The document refers to other materials "for healthcare professionals" available here: here, issued in June and here, issued in November, updated yesterday. I have added some definitions from the new NICE guidance to the lead, but now the details of the guidance should be looked over with an eye to incorporating the information in a way compliant with Wikipedia medical policies with which I am not familiar, so it'd be good to have some experienced help now that authoritative sources are becoming available.
The NICE:
have used the following clinical definitions for the initial illness and long COVID at different times:
Acute COVID-19: signs and symptoms of COVID-19 for up to 4 weeks.
Ongoing symptomatic COVID-19: signs and symptoms of COVID-19 from 4 to 12 weeks.
Post-COVID-19 syndrome: signs and symptoms that develop during or after an infection consistent with COVID-19, continue for more than 12 weeks and are not explained by an alternative diagnosis.
The bottom two bullet points (4 weeks +) are what they define as "long COVID", though they also state that:
Post‑COVID‑19 syndrome may be considered before 12 weeks while the possibility of an alternative underlying disease is also being assessed
They state that infected
People may also develop signs or symptoms of a life-threatening complication at any time and these need to be investigated urgently.
It also notes that:
People need good information after acute COVID-19 so they know what to expect and when to ask for more medical advice. This could help to relieve anxiety if people do not recover in the way they expect. ... Accessibility of information ... is particularly important after acute COVID-19 because people may have cognitive symptoms ('brain fog') or fatigue, making it difficult for them to take in long or complex information.
@Elliot321: Interesting idea! I've never heard of glossaries being hosted on Wikipedia before, which unfortunately probably does say something about how much demand there is for them. My initial thought is that it's a fairly old-school method of organizing information. 90% of the use of glossaries in books is people looking up a specific term that has been mentioned that they want to understand, and that's not really a use case here, mainly because of wikilinks, and in rarer cases because of {{abbr}}. However, I do think it's worth giving some concerted though to how best to handle complicated terms in our COVID-19 articles, much of which is potentially applicable to other areas on Wikipedia. One good strategy is reducing the need for them overall—for the lead of COVID-19 pandemic, I wrote It might also spread via contaminated surfaces, not It might also spread via fomites. Beyond that, there's wikilinks, but we need to be wary of MOS:OVERLINK, and there's a a school of thought that says wikilinks shouldn't be used as explanations at all. When we can't rely on links, that's where {{abbr}} comes in, particularly for table headings, but we might want to consider adopting it in body text as well. It unfortunately doesn't work for mobile, though, which is a big obstacle.If we do end up going with a glossary page, considering the circumstances in which someone might use it should be top of mind. Building it so that it will remain up to date is also important. For that, I notice that the blurb for covidiot is very similar to the definition at Wiktonary. Is there a way to transclude that definition so that it will be synced? I hope all that is fuel for thought. {{u|Sdkb}}talk01:39, 17 December 2020 (UTC)
@Sdkb: yeah, glossaries are indeed somewhat rare. I was originally thinking of doing a "Impact of the COVID-19 pandemic on language" - since many new terms have been coined - but I thought a glossary could potentially be more useful for defining existing terms (and people would be more likely to view it than an "impact on language" article, especially if featured prominently in the infobox. The line between dictionary and encyclopedic glossary is somewhat awkward and is described at the MoS page I linked - particularly this section comments on it. I don't think there's a way to embed content directly from Wiktionary. Elliot321 (talk | contribs) 01:46, 17 December 2020 (UTC)
Elliot321, eh, what should be in a proposed glossary is more appropriate in a separate conversation (or a new task force at this point). In regards to the creation of the glossary, I am still saying yes. —Tenryuu 🐲 ( 💬 • 📝 ) (🎁 Wishlist! 🎁) 07:58, 17 December 2020 (UTC)
Elliot321, no worries, I started the whole thing. I'm going to stand back and let other regular editors comment on this before I start going into details like term inclusion and the glossary's visibility. —Tenryuu 🐲 ( 💬 • 📝 ) (🎁 Wishlist! 🎁) 17:13, 17 December 2020 (UTC)
I have never heard the phrase "social distancing" before this pandemic, although I have no doubt that it was used in certain fields. Like autism, it wasn't a thing! But in any case its provenance shall be tracked down eventually. kencf0618 (talk) 12:30, 19 December 2020 (UTC)
Is our use of the term "recovery" realistic?
Due to factors such as "Long Covid" and the possibility that COVID can cause long-term health problems, I feel that our use of the word "recovery" may be too optimistic. In particular, what is classified as a "recovery" can vary between states and countries; some may require verification, some may automatically declare it a "recovery" if they are not hospitalized within 14 days of a positive test.
I personally think we should try to deprecate or be more careful in use of the word "recovery", either making sure there is specific disclosure/footnotes stating how the region defines a "recovery", or avoiding use in favor of terms such as "no longer considered active"/"inactive", "inactive, non-mortality", etc.
Any thoughts on this topic? It might fall under a similar category of semantics to those arguing over the differences between a "death with COVID" and "death from COVID".... ViperSnake151 Talk 04:26, 22 December 2020 (UTC)
The article needs to be WP:MEDRS compliant. If you check the section on October and November, you'll see reliable secondary sourcing of claims to earlier cases. Fences&Windows22:51, 22 December 2020 (UTC)
This article on a new strain of SARS-Cov-2 was created yesterday. Is there a guideline on articles about coronavirus strains and is the information here is better merged somewhere else? Hemiauchenia (talk) 23:47, 20 December 2020 (UTC)
There will be large number of new articles in relation to vaccination programmes across the world. Like COVID-19 vaccination programme in the United Kingdom, there will be many new country-wise articles. How they should be named and organised? Please discuss and prepare guidelines. I propose COVID-19 vaccination in [country], similar to COVID-19 pandemic in [country], as a naming style. -Nizil (talk) 05:53, 22 December 2020 (UTC)
Nizil Shah, while there could be eventually enough content to warrant a standalone article on vaccination programs by country, it's still too early to think about it. I agree with Alexbrn and if they do get very large we can start thinking about splitting them. —Tenryuu 🐲 ( 💬 • 📝 ) 🎄Happy Holidays!⛄ 06:19, 22 December 2020 (UTC)
Strongly agree. Please advise editors to add the content to the corresponding 'COVID-19 pandimic in COUNTRY' article. If there's substantial content, create a new Vaccine section in that article. - Wikmoz (talk) 22:13, 24 December 2020 (UTC)
Vaccination counts
Now that mass vaccinations have started in English speaking areas, is it time to add vaccination counts to all the other statistics we maintain? Specifically, do infoboxes need a line for vaccinations? (The US state of South Dakota just started publishing a county by county breakdown of vaccinations. I added this data to South Dakota's case count table.) EphemeralErrata (talk) 05:49, 17 December 2020 (UTC)
Wouldn't bother with counts unless some particular figure is news-worthy (e.g.[4]) as we'd be getting into WP:NOT territory. Does Wikipedia track the vaccine count of any other vaccine? Alexbrn (talk) 06:02, 17 December 2020 (UTC)
Oh my, here comes another type of data for us to keep track of. As I've argued before, I think we've failed quite badly at centralizing case/death/recovery counts—they're recorded in tons of different places, and updating in one place doesn't update everywhere. This is an opportunity for us to do better, but it will be difficult. We need to first agree on where to store the information (Wikidata would be my preference, as that's basically what it's designed to do, and it's multilingual) and then we need to make sure that every single place where that information appears, it's sourced from the centralized database. I'm probably going to be too burnt out about this to help much, so good luck. {{u|Sdkb}}talk06:28, 17 December 2020 (UTC)
Yeah, Wikidata's the thing (please discuss there). Mind you, it's a tough job since care will be needed to manage the data to account for different types of vaccines, 1st dose vs 2nd dose (where applicable) and so on. It's really the sort of thing that requires paid statisticians to labour over. But by all means a few enthusiasts can make a start here and then walk away leaving a steaming mess. That is, after all, the way we're dealing with the whole COVID-19 thing! Alexbrn (talk) 06:35, 17 December 2020 (UTC)
COVID-19 vaccine has come a long way over the last month but could use some additional editor attention to help improve content quality and address excessively detailed and redundant statements. - Wikmoz (talk) 22:08, 24 December 2020 (UTC)
This article really needs to be split, possibly three ways. All the information about development and deployment is crowding out information about the actual vaccines themselves. I'll come up with a proposal on the article talk page. John P. Sadowski (NIOSH) (talk) 21:46, 26 December 2020 (UTC)
The vaccines have multiple aspects, as does the pandemic itself. "Manifold" doesn't begin to cover it...! That said, a condensed version would be helpful. kencf0618 (talk) 21:27, 28 December 2020 (UTC)
Current best practice for tables/histograms over time?
Hi, could someone point me to examples of whatever the current best practice is for tables/histograms over time please? I'm looking at updating COVID-19 pandemic in the Canary Islands (outdated since April), but it's going to be quite a bit of work, and I'd rather get the format right from the start. Thanks. Mike Peel (talk) 18:19, 30 December 2020 (UTC)
Mike Peel, some pages are better than others, but I don't really know of any gold standard, as COVID-19 data on Wikipedia as a whole is kind of a mess. I'd check the usual suspects (U.S., UK, India, Australia) and go with whichever page seems to be working best. Ideally, the data should be automatically drawn from a centralized data source (Commons, Wikidata) that is automatically or regularly updated itself. {{u|Sdkb}}talk20:11, 30 December 2020 (UTC)
Resolved
