Jump to content

Talk:COVID-19 pandemic by country and territory/Archive 7

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1Archive 5Archive 6Archive 7Archive 8Archive 9

Table appears reordered

This <https://en.wikipedia.org/wiki/COVID-19_pandemic_by_country#Total_cases,_deaths,_and_death_rates_by_country_(Our_World_in_Data)> now orders country by deaths per million, which is IMO much less useful than its previous format of overall cases. Why the change? 79.66.213.100 (talk) 07:47, 4 October 2021 (UTC)

It was agreed that deaths per million is a better indication of how badly affected a country is. You can sort it by total cases by clicking on "Cases" twice. Tol (talk | contribs) @ 14:25, 4 October 2021 (UTC)
I'm suspicious because it looks like it was done to hide embarrassing numbers. I believe I've seen that done before for covid coverage on wiki. Doing as you suggest (clicking on cases) doesn't work for me as I always disable JS, but thanks for letting me know it's possible. Can we revert the format for this to what it was? 92.6.96.220 (talk) 07:54, 5 October 2021 (UTC)
Ah; my apologies. As I said, it was done because there was consensus that deaths per million is a better indicator of COVID-19's effect than total cases, which does not account for a country's size. For a table sorted by total cases, you can see Template:COVID-19 pandemic data. Tol (talk | contribs) @ 18:52, 5 October 2021 (UTC)

@User:92.6.96.220: I agree with you. I didn’t want to mention this as my opposition to this change is mainly about the inaccuracy of the information in the new data table and the new data table not being referenced in any usable way so that figures at the moment cannot be cross checked with reliable sources but what you say is what it looks like to me. I can’t see any other logical reason for it. Please see my comments above in the ‘Concerns’ section. Originally this table was about numbers of coronavirus cases, then deaths around the world. Now it looks like a information distortion table to shuffle information. I’ve always viewed Wikipedia as trusted information.

@Tol ‘It was agreed that deaths per million is a better indication of how badly affected a country is.’ Agreed by who? This table is not about displaying how badly affected a country is or hiding how badly affected a country is. This table is about displaying coronavirus cases and death numbers around the world. It is Not about displaying deaths/million.

‘because there was consensus ‘ Consensus from who? I don’t see any consensus saying, ‘Yes, that’s a really great idea. That’s a real improvement to the table. Let’s do that.’ It looks to me that a perfectly good table which was updated daily with the latest daily accurate and referenced information from official governmental figures around the world has been replaced with a bot supplying inaccurate information as against the ‘accurate’ ( I’ve already gone into this in a small way above ) daily official government figures and is also supplying information which is now two days out of date in some cases ( I haven’t checked them all yet but the bot has only been in operation a few days so more time is needed for accuracy ) compared to data that used to be updated daily with the latest daily figures?

Why are my comments now in a ‘Concerns’ section when I’m sure I posted them in the ‘Remove old table. Keep the table updated daily by bot’ section where @Wikmoz has listed the missing countries? I didn’t open up a new 'Concerns' section heading. — Preceding unsigned comment added by BringBackTheStats (talkcontribs) 03:45, 6 October 2021 (UTC) BringBackTheStats (talk) 03:51, 6 October 2021 (UTC)

@BringBackTheStats: Timeshifter gave you the link with all of the data sources for cases. You posted your initial comment in the subsection "The table is updated daily by a bot", which was to discuss that wording in the article. Because your comment was unrelated to that subsection, and it was very long, I split it into another subsection of the section "Remove old table. Keep the table updated daily by bot". In that section (at the top) it was agreed, by LSGH, Timeshifter, Wikmoz, and me, to use a table with deaths per million. Again, the numbers are sourced from official government sources; see here for a list of sources. The bot has been operating in its current form since 25 July 2021 and in a limited form since 25 May 2021, so it hasn't only been in operation for a few days. Tol (talk | contribs) @ 13:54, 6 October 2021 (UTC)
@Tol: Yes I have looked at the pages before. I spent so much time trying to look for the John Hopkins references that I was frustrated and gave up. The closest I could find was the references on this page https://github.com/CSSEGISandData/COVID-19/blob/master/csse_covid_19_data/README.md but a lot of the links I tried were redundant. I also found this page https://coronavirus.jhu.edu/about/how-to-use-our-data . I tried the first link under the heading ‘About Our Data Sources’ which takes you to yet another covid-19 portal designed for a phone. I’m dreaming Covid portals. I gave up at that point and looked elsewhere unsuccessfully. If I’d clicked on the second link ‘CSSE GitHub’ https://github.com/CSSEGISandData/COVID-19 I’d have been in but I’d had enough of trawling around Github by that point.
Anyway to get back to my point. The references look all good. They all work, they’re credible, reliable sources for the countries that I’m familiar with but there’s a problem and it’s the same problem that I’ve been banging on about. You haven’t checked any of the data for the countries I mentioned, have you?, or you have and are saying nothing. No-one has answered my concerns about the data that appears on the Wikipedia page. The data in those references are as officially correct as we can get but that’s not the data that appears on the Wikipedia table. A good quick example, and you’ve read this yourself, was that OWID posted 137,295 UK deaths within 28 days of a covid test by date reported on the 2nd of October as well researched by @Timeshifter in the sections above. The official figure today for the 6th of October was …. yep, 137,295. Now either OWID has some inside medical information or maybe it’s just very good at predicting the figures or could it really be that their figures are a load of crud … surely not. As I said at the very beginning I’ve been following this very closely since the beginning and they just don’t feel right. The John Hopkins references are fine but the figures that appear on Wikipedia are definitely not. I asked you and @Timeshifter to look at the UK, Ireland ( Ireland is now half right, right cases but wrong deaths which is difficult to do today as Ireland’s deaths are only updated weekly and today’s the day it happens. They’re released together but only the cumulative case number had updated. ), France and Singapore I think, Add Switzerland, Liechtenstein ( they seem to release their figures together but even if you add on Liechtenstein’s 60 deaths it’s still nowhere close ) and Ukraine to that list. I’m just dipping at the moment, I know they’re not right. Maybe I’ve been extremely lucky but what are the chances of my picking up six sets of bad figures ( I’ll give Ireland a pass ) in the Wikipedia data with my first seven picks from the John Hopkins’ references??
All I know is that the Wikipedia old table was updated daily for most countries ( for the larger data sets, it’s a bit difficult to update your figures daily if no-one dies in the smaller territories or like China you come down on Covid like a ton of bricks but it pays off, so far. Things could still change there.) and not only was it being updated daily but it was accurately cross-checking with official figures for the most affected countries whenever I had to check.
My original comments may have been posted in the wrong section but now they’re in an even worse position and are still in the wrong section. They are not in Concerns under the ‘Remove old table. Keep the table updated daily by bot’ section. They’re in ‘Concerns’ under the ‘The table is updated daily by a bot’ which is still the wrong section as you say. Can I have my concerns posted in the correct section please or can I do that myself?
' it was agreed, by LSGH, Timeshifter, Wikmoz, and me, to use a table with deaths per million.”
You keep saying that this was agreed, that there was consensus for that etc but where is it? Am I missing something? I see opposition to this. Where is this deaths per million discussion between you all? Can you post me a link to the death per million discussion or explain how I get to it?
‘Again, the numbers are sourced from official government sources; see here for a list of sources.’
All would be well and good if those numbers from those sources appeared in the Wikipedia table which they do not. As I said before there must be someone in Wikiworld who can build a bot that does the job correctly.
‘The bot has been operating in its current form since 25 July 2021 and in a limited form since 25 May 2021, so it hasn't only been in operation for a few days.’
I thought the bot had only been on Wikipedia since the table changed. I must have misunderstood. If it’s been running since the 25th July then why have the figures in the Wikipedia table only started not to match official data since the table was changed? Is there fault with the bot, the table or is it my Excel that’s stopped working correctly since the table was changed??
BringBackTheStats (talk) 04:34, 7 October 2021 (UTC)
@BringBackTheStats: I'm not sure you understand subsections. Yes, the section with your comments is under "The table is updated daily by a bot.", but it is not part of that section. It is part of the section "Remove old table. Keep the table updated daily by bot" ("The table is updated daily by a bot." is also part of that section). It's in the right section. The discussion to implement the table is right on this page at #Remove old table. Keep the table updated daily by bot. Those government sources are not in the table because we're not getting data from those sources, we're getting data from OWID, which uses JHU, which uses those sources. The bot updates a centralised repository of data, which is then used on other templates, such as Template:COVID-19 vaccination data (the first table that it updated). This data was recently implemented for the templates used on this page. Much of the time, if the numbers are higher, it is because OWID is summing up data reported from more local governments, which tend to report the data before regional or country governments do. Tol (talk | contribs) @ 05:12, 7 October 2021 (UTC)
@Tol: As I said the figures on this Wikipedia table are not the figures in the sources you posted.
In your post above Tol (talk | contribs) @ 13:54, 6 October 2021 (UTC) you say,
‘Again, the numbers are sourced from official government sources; see here for a list of sources.’with the word ‘here’ linking to https://github.com/CSSEGISandData/COVID-19/blob/master/README.md which are the sources that John Hopkins uses. Now you tell me that the figures that appear on the Wikipedia table are not from those sources. Now you’re telling me that the figures originate from those sources but that OWID then run through their own number processing process to produce their own figures which don’t have references. All I can glean is that there are no actual individual references for the numbers that appear on that table.
Before at the beginning of this whole discussion you posted the link
https://github.com/owid/covid-19-data/tree/master/public/data
At the top of the second section, ‘Data on COVID-19 (coronavirus) by Our World in Data’, OWID say that they use
Metrics Source Updated Countries
Vaccinations Official data collated by the Our World in Data team Daily 217
(Official Data with no actual references, sounds good.)
Confirmed cases JHU CSSE COVID-19 Data Daily 194
Confirmed deaths JHU CSSE COVID-19 Data Daily 194
So I checked the JHU UK Death Data here and after a bit of flaffing about with the .csv file, it does match up correctly with the UK .gov site. There are some naming errors and the omission of the individual country data for England, Scotland, Wales and NI which I won’t go into here but the UK totals are correct.
https://github.com/CSSEGISandData/COVID-19/tree/master/csse_covid_19_data/csse_covid_19_time_series
https://github.com/CSSEGISandData/COVID-19/blob/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_deaths_global.csv
They’re not correct on the John Hopkins website however, on this link with the John Hopkins figures about 400 above the UK official figures.
https://coronavirus.jhu.edu/region/united-kingdom
I checked the JHU UK Vaccination Data here and after a bit of flaffing about with the .csv file, it does match up correctly with the UK .gov site. @Timeshifter was asking about this above.
https://github.com/govex/COVID-19/tree/master/data_tables/vaccine_data/global_data
https://github.com/govex/COVID-19/blob/master/data_tables/vaccine_data/global_data/time_series_covid19_vaccine_global.csv
Recoveries are available here.
https://github.com/CSSEGISandData/COVID-19/tree/master/csse_covid_19_data/csse_covid_19_time_series
https://github.com/CSSEGISandData/COVID-19/blob/master/csse_covid_19_data/csse_covid_19_time_series/time_series_covid19_recovered_global.csv
I haven’t checked cases or dipped other countries in the JHU Github files yet but on first attempt it all looks above board with their figures matching up with official references.
So to sum up, the OWID figures are not the official figures that they say that they are using from the JHU data and the numbers have no individual references and so cannot be checked against any verifiable official data base but we’re going to post these unverifiable figures every day on Wikipedia anyway and try to dress it up as if they’re okay.
As you say yourself at the top, ‘There is no way the above-linked table can be as accurate, or as regularly and fully updated, as this table updated daily by a bot: ‘
I keep saying that I’ve been doing this a while now and I know that the OWID figures are not official numbers because I used to check them but they were incorrect and nothing improved over time, so I don’t bother with them any more as they’re nonsense. They were impossible to check then and this exercise shows me that nothing has changed. The WorldOmeter figures that has the Wikipedia disclaimer at the top of this page are better.
I’m glad to have got there in the end.
I note that you have not posted the consensus about this overall change or a link to the deaths per million discussion where it was agreed to do this as well. I have tried to find it but was unfortunately unsuccessful.
BringBackTheStats (talk) 09:27, 8 October 2021 (UTC)
BringBackTheStats. Your posts are very confusing. Is English your native language?
Can you give a simple numbers comparison? For example, like I did for UK deaths.
Go to the OWID sourced templates:
Template:COVID-19 pandemic death rates - or
Template:COVID-19 pandemic death rates by country
Pick one number for one country. Tell us what it is.
Then find the JHU number for that country. Tell us what it is. Tell us exactly where you found it. Link to your JHU source.
--Timeshifter (talk) 11:32, 8 October 2021 (UTC)
@Timeshifter: Cheeky. Your not the first person to have said that lol. But, English is my first language, not my native language but then that applies to most English first language speakers around the world.
This whole thing is confusing. I’ve checked the Wikpedia (OWID) table numbers with official government sites and they don’t match up for deaths or cases for the UK, Ireland (cases yes, deaths no but they probably will have updated by now I hope), France, Switzerland, Liechtenstein, Singapore and Ukraine. I’ve only looked at those so far. OWID says it data is from JHU for deaths, cases and recoveries it doesn’t post, but JHU do collate recovery data.
The JHU data in their Github .csv files does match UK, England, Scotland, Wales and NI data for deaths and vaccinations. I didn’t check cases yesterday but at this point I’m assuming that it does as well.
The data on the JHU website UK page doesn’t match official UK data even though the data in the JHU Github .csv files do match the UK official data.
https://coronavirus.jhu.edu/region/united-kingdom
I’d love to post a table of some sort but I tried to copy in a piece of an Excel page yesterday and this Talk page wouldn’t do it or else I’m not doing it right.
I find Github confusing but I have found that if you click on ‘Ok’ when you get the LibreOffice Calc message box, "The data could not be loaded completely because the maximum number of columns per sheet was exceeded.", then the Github .csv files seem to load up ok or at least they did for me. How much is missing, I don’t know. The .csv files I loaded were confusing with a load of code on them (but by hiding the LHS columns and top rows and then Freezing the page in the right place I was able to make it a manageable spreadsheet. Thanks for mentioning Libre as I always found it a bit unreliable but it has improved a lot.
Yes, I see what you mean if you try to open the .csv links I posted from this page you just get numbers and commas. I’m going to try to upload the JHU Global Deaths .csv file that I’ve fiddled about with so that it’s set up in a useable way when it’s opened. If I can upload and download it in the way that I saved it then I can upload a spreadsheet with countries and numbers and that would be the best way. I’ll try and include the reference links and date and time columns for maybe 10 – 15 countries depending how long it takes. The figures do change during the day and I tend up update my files after midnight to 1:00 am when things have settled down.
I’ve had a quick look. Just France and the UK. I’ve had to post this as a Libre Calc file as Microsoft Office won’t recognise the reformatts in the .csv file saved in Libre. Have a look at France at line 1603. I’ve added in the links after the names on column B, lines 1602 – 1604. Confusing? The French Public Health website used to match the Wikipedia figures. (Not now posted)
Have a look at the UK figures on line 2161. I hope you can follow me. I’ve highlighted the official figures in green and where they match up with French official and UK official. (Not now posted)
‘An automated filter has identified this edit as potentially unconstructive, and it has been disallowed. If this edit is constructive, please report this error.’
I reported, ‘I was trying to post Covid-19 figures with references from various sources added to a Johns Hopkins Github .csv file that had been saved in Libre as a .ods file. As that wouldn't work I took a screenprint and saved as a .jpg file and that wouldn't work either. I checked the Wikipedia copyright page beforehand and as Johns Hopkins public Github files have been uploaded in the past it looked okay.’
No files, no screenprints. Does anyone know how to post these figures?
BringBackTheStats (talk) 15:31, 8 October 2021 (UTC)
@BringBackTheStats: You're replying to Timeshifter, who is a different user. I did not question whether English was your first language, Timeshifter did. I would hypothesise that the data on the website is updated more frequently than the CSV, which is updated daily. Tol (talk | contribs) @ 16:14, 8 October 2021 (UTC)
@Tol: Thanks. Yes, you are right. Lol. Sorry. I’ll change it now although it showing at my end that the page doesn't exist. I'm fairly new to this. It wouldn’t bother me if you did. I was being lazy and copying from a text file. I didn’t check enough. Not like me, eh? I’ve lost my 1,0&8 and a few letters for some reason and a Windows 10 onscreen keyboard that keeps collapsing programs as you’re about to type into them doesn’t help either. You don’t happen to know the best way to post these figures or if there’s any way to post them?
BringBackTheStats (talk) 17:01, 8 October 2021 (UTC)
@BringBackTheStats: You're welcome. As for linking, you're linking to articles such as Tol; I think you mean to link to user pages such as User:Tol. You can make a link to one page but have the link display something else by using a piped link (such as [[User:Tol|Tol]]). For uploading files, you can try the File Upload Wizard. Tol (talk | contribs) @ 17:41, 8 October 2021 (UTC)

BringBackTheStats. Sounds like you need a new keyboard. If you are using a laptop you can plug a keyboard into it too. Same as for a PC.

See Help:Table to learn how to create and edit tables on Wikipedia.

Experiment in user sandboxes. From Help:Table:

Sandboxes help a lot. For example; your user page: Special:MyPage. Create and bookmark some personal sandboxes too. Visual Editor will load very fast in empty sandboxes: Special:MyPage/Sandbox, Special:MyPage/Sandbox2, Special:MyPage/Sandbox3. As many as you want. Share the link when asking for help. To find all your sandboxes: Special:PrefixIndex/User: – click link, add user name to the spot labeled "Display pages with prefix:".

--Timeshifter (talk) 23:46, 8 October 2021 (UTC)

@Timeshifter: Thanks for your help. I’ll try some of that if I get the time. I need a new pc more than a new kepboard. I’ve figured out that I’ve got some sort of weird keyboard with acutes and ‘not’ signs etc. This is the 4th rescued keyboard in this laptop so that could be part of the problem and it’s reached the end of life. I promised myself I’d build a good pc back in the Windows 8 preview days, then Windows 10 came along and things got worse. I meant to post this yesterday but I couldn’t figure out how to reply to your post as it not showing in preview.
Back to the subject in question I noticed yesterday evening that the OWID / Wikipedia table had updated the UK figure but was showing 137,818. The Johns Hopkins Website UK page https://coronavirus.jhu.edu/region/united-kingdom was posting 137,818 deaths from the night before and was still showing that figure. Neither being the official figure.
At present,
Johns Hopkins Dashboard is showing 8,192,541 UK cases and 138,139 UK deaths on https://www.arcgis.com/apps/dashboards/bda7594740fd40299423467b48e9ecf6
Johns Hopkins UK Overview at https://coronavirus.jhu.edu/region/united-kingdom is showing 8,158,935 cases and 138,101 UK deaths as is the Wikipedia / OWID table.
The official UK figures for today are 8,154,306 cases and 137,735 28 day deaths on https://coronavirus.data.gov.uk/details/cases and https://coronavirus.data.gov.uk/details/deaths.
BringBackTheStats (talk) 21:17, 10 October 2021 (UTC)
BringBackTheStats. I got my current PC used from ebay in 2016 for $159 plus $20 shipping from a dealer with a 100% rating and hundreds of sales. HP Pro 6200 MT PC Desktop quad core i5-2400 3.1GHz. It was 5 years old then. So 10 years old now. It works great. I got screwed once on ebay when I tried to buy a new videocard this year from an ebay dealer with a good rating but few sales. He had a good scam going. He sold a few cheap items and got good ratings. Then claimed he had some videocards for sale at a good reasonable price. Perfect scam since videocards are so hard to find now. Screwed a few people before his rating tanked. --Timeshifter (talk) 21:43, 10 October 2021 (UTC)

Table column order

Unfortunately the table format has once again (!!!) changed. Now the current listing on the right side, next to the country, is:

   (1) Deaths per million (2) Deaths (3) Cases

Some weeks ago - and I clearly remember that because I used to visit that table frequently - it was:

   (1) Cases (2) Deaths (3) Deaths per million

So in other words, we could first see the number of total cases, and then the deaths. This was easier to compare than the deaths-per-million ratio first. It's annoying to have to adjust to whoever randomly changes these tables. Would it not be better to simply stick to a variant? Ideally the one that was using cases, deaths and deaths-per-million. Now I have to re-adjust mentally to the changed (and to me) worse variant ... technically a reversion to the old format may cause people to then have to re-adjust as well ... so seriously folks, it would be so much simpler to just stick to something consistently from the get go. Who was making these ad-hoc willy-nilly changes anyway? I guess it may be useful if we could just move the whole column to the left in the browser, that way we don't have to depend on whoever randomly changes these things ...

I found cases-first to be more useful because it gave a "total" comparison situation, as-is, whereas with the deaths-per-million we can have very skewed results, in particular for smaller countries. If you have 1000 people then 50 infected is evidently easier to fluctuate with up and down then when you have a large country, sparsely populated and thus fewer people. So the deaths-per-million is not as useful to me in direct 1:1 comparisons. That is why now that it is listed first annoys me, since it it made that table less useful than it used to be - but even more importantly it is annoying to have to adjust to something new now suddenly ... 2A02:8388:1604:F600:3E29:3607:2855:A3EF (talk) 09:59, 9 October 2021 (UTC)

I think deaths per million is much more useful when comparing countries. It is much more useful than cumulative deaths. That says nothing about the relative impact of Covid on a country.
Cumulative cases per country does not accurately compare countries. In fact it tells nothing about the relative situation in each country. Plus the numbers for cases are less accurate than the number of deaths. Case numbers depend on testing capability, frequency, funding, etc.. Deaths are more obvious, though those numbers are too low too. As proven by the excess death studies. --Timeshifter (talk) 20:56, 9 October 2021 (UTC)
> I think deaths per million is much more useful when comparing countries.
Perhaps, but it's not what I'm after. I much preferred the original (and I said so in the first comment here <https://en.wikipedia.org/wiki/Talk:COVID-19_pandemic_by_country_and_territory#Table_appears_reordered>)
> In fact it tells nothing about the relative situation in each country
It tells us the covid-attributed death total rankings in each country. As a UK brit I want to know where I stand *relative* to other countries. The original table did this well. I would like it back. 79.66.198.78 (talk) 17:11, 10 October 2021 (UTC)
Total cases and total deaths are still in the table. And you can sort them. Unless you have Javascript disabled. I think that is what disables sorting. Not sure. But if that is what you are doing, then you can enable it for Wikipedia. You have nothing to fear from Wikipedia. You are not going to get attacked by malware coming from Wikipedia. --Timeshifter (talk) 17:50, 10 October 2021 (UTC)
> Total cases and total deaths are still in the table - Then the original ordering was equally fine, so why change it? Because you felt a different default ordering was preferable. I would like the old ordering back, and it seems others agree.
> You are not going to get attacked by malware coming from Wikipedia - put that in writing with money your behind it and I might consider it. In the meantime, the only unbreakable window is the one that isn't there. Also irrelevant to the preferability (or not) of the original ordering.
If 4 people ("it was agreed, by LSGH, Timeshifter, Wikmoz, and me, to use a table with deaths per million") then please accept there would be some disagreement and 4 people's decision is a very small base and some people are disagreeing with you. Again, I ask you revert the table. 79.66.198.78 (talk) 18:29, 10 October 2021 (UTC)
The previous much better table in my opinion for a variety of reasons (see above), had ‘Confirmed’, ‘Deaths’, ‘Recoveries’, ‘Death Rate’ columns as per the photo I’ve hopefully uploaded. It also had fully traceable references to the information contained within the table and that information was the correct official government figures. Sorry about the quality of the photo but as you can see from my comments in the ongoing discussion above I’d had a bit of difficulty in uploading screenshots or files to support my arguments against changing this table, when using the Wikipedia Upload Wizard.
I’ve added in the ‘Death Increase From Previous’ day and ‘Case Increase From Previous’ day columns on a spreadsheet. I’ve been recording this page since back in the day when the US had 0 deaths and less than 100 cases and the UK government was still banging on about Brexit with a looming pandemic on the way that nobody appeared to be taking the blind bit of notice of and making no preparations for.
I’ve looked for this deaths/million discussion where ‘it was agreed’ but I can’t find it. Let’s put aside that the information is incorrect and untraceable. Does anyone know where the photo appears? The wizard says that there's a file with that name already saved as if it has uploaded but it's not showing in the preview and the wizard just appears to go around in circles.
File:Wikipedia - Photo Of My Coronavirus Table Photo 20211010 1801.jpg
A photo of the columns that were on the old coronavirus table, Confirmed, Deaths and Recoveries.
BringBackTheStats (talk) 18:58, 10 October 2021 (UTC)
@BringBackTheStats: We've already shown you where to find the list of government sources, and that we get data from OWID rather than directly from them, so citing OWID is correct. I can see the image. The previous table did not have death rates; it only had total cases, total deaths, and total recoveries. Tol (talk | contribs) @ 20:14, 10 October 2021 (UTC)
@Timeshifter: Okay, I thought the Death Rate column might have been my own addition but I didn’t want to take credit. I just check the figures and when there’s a problem, I cross check it with the source reference. I copy, scan for errors and move on.
‘We've already shown you where to find the list of government sources.’ Not for Our World In Data you haven’t. I have no idea where to find those. They’re like a Facebook algorithm or the Coca Cola recipe. I know where to find references for official figures, the only problem is that OWID don’t supply official figures. They begin with official figures sometimes and then appear to manipulate them using other data into something of their own. There are references and names of those who do the number manipulation but very little about what they do to the figures, what they mix the official data with or where they get those other numbers from. Employee credits are not data references. I think that governmental manipulation of the figures is one level too many, nevermind churning the figures through some more. I am completely unable to reference any of their numbers to a credible degree and if you’ll try the same, I doubt if you will either. For the sake of argument let’s assume that I’ve no idea what I’m talking about, can you please post the link to the OWID references for me again and for anyone else who’s interested about where their data comes from?
If you have the knowledge to program a bot to do this, then please take the time to do it properly. That’s what this data deserves. I honestly have no idea what it entails or how much work it is. It may be an awful lot of work. I have no idea. I would be very willing to take some of the load if there’s any way in which I can. The last programming I did was on MS-DOS before Windows was available but if I could help, I will. The old table had the correct figures and was updated on a daily basis although you seem not to think so. I could easily show you that it was.
I do wonder if overhauling a bot is more difficult than trying to figure out how to go about reconfiguring an eighty tab spreadsheet containing at a rough calculation 1,300 sub-columns. I wonder if many school projects around the world have come to a juddering halt because of this change if they were using it as their source. Even changing the order of the columns is a problem. That one simple minor change costs so much work, never mind the accuracy of the numbers. I’ve said it before that major changes like this need to be clearly flagged up on the main table on the main article a few weeks before any changes are made.
BringBackTheStats (talk) 22:36, 10 October 2021 (UTC)
BringBackTheStats. OWID clearly states at the link below that JHU is their source for numbers on cases and deaths:
Coronavirus Source Data
JHU sourcing is covered in the notes section:
COVID-19 pandemic by country and territory#Notes - it says this:
"CSSEGISandData/COVID-19". GitHub. COVID-19 Data Repository by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. See How to Use our Data for more info and links. See: Pandemic Data Initiative. See more sourcing history and info.
Our World in Data (OWID). See Coronavirus Source Data for OWID sourcing info. Excerpt: "Deaths and cases: our data source. Our World in Data relies on data from Johns Hopkins University. ... JHU updates its data multiple times each day. This data is sourced from governments, national and subnational agencies across the world — a full list of data sources for each country is published on Johns Hopkins GitHub site. It also makes its data publicly available there."
--Timeshifter (talk) 23:28, 10 October 2021 (UTC)
If you would like a table sorted by total cases, you can use Template:COVID-19 pandemic data. Tol (talk | contribs) @ 20:08, 10 October 2021 (UTC)
@Tol, so you are saying you're ok with the table format being reverted? If so I'd like other people's input on it before I do. 79.66.198.78 (talk) 20:59, 10 October 2021 (UTC)
I've sorted the table by cases. Does it look good to you? Tol (talk | contribs) @ 21:01, 10 October 2021 (UTC)
It looks great, thank you (edit: and sorry for sounding grumpy).79.66.198.78 (talk) 21:04, 10 October 2021 (UTC)
Just so you know, my edit was reverted, so it's sorted by deaths per million again. If you want to view a table sorted by total cases, you can click here and use that table. Tol (talk | contribs) @ 21:53, 10 October 2021 (UTC)
@Tol - last relevant edit, which was after my thank-you to you was 22:18 'Revert div to limit width (problem fixed)' apparently under your name. I'm not sure what is going on. Did you revert? 79.66.198.78 (talk) 22:12, 10 October 2021 (UTC)
I agree with Dudley Miles in keeping the deaths/million column as the first column, and making it the initial sort order. Dudley maintained the scrolling version of that table for a long time. I have worked on most of the scrolling Covid tables in various ways for a long time. @79.66.198.78: You have a template with cases in the first column here: Template:COVID-19 pandemic data. It is used here: Portal:COVID-19. Its initial sort order is by total cumulative cases.
--Timeshifter (talk) 22:34, 10 October 2021 (UTC)
Who is Dudley Miles (I can't see the name on this page)? Where is that discussion? Why do you feel it's better when others have said they prefer the old way? If it's worth changing here because "it's better" why is it not "better" in the portal you indicate the old template is being used? I don't mind this stuff happening, it's the arbitrariness of it that's bothering me - "we changed it, suck it up" seems to be your attitude. 79.66.198.78 (talk) 22:57, 10 October 2021 (UTC)
See Special:Diff/1049270297. --Timeshifter (talk) 23:08, 10 October 2021 (UTC)
I don't think you understand — I gave you a link to the reversion. The table is in a template, which is transcluded onto this page, so edits to the table are on their own page. Tol (talk | contribs) @ 03:14, 11 October 2021 (UTC)
And again, if you click here you can have a table sorted by cases. Tol (talk | contribs) @ 03:15, 11 October 2021 (UTC)
And again you ignore everything I'm saying. Very well, I've got other things to do and I guess you've put more work into this page than most people so perhaps you have a right to dictate things. I just wish you hadn't done it in such a high-handed screw-you way over a valuable resource. Good luck. 92.6.111.131 (talk) 21:34, 11 October 2021 (UTC)

We have been discussing 3 automated tables:

For the next 2 templates the initial sort order is for deaths per million.

The bottom 2 templates are the same, except one is scrolling and one is not.

The last version of the template with individual references per country is here:

--Timeshifter (talk) 21:23, 10 October 2021 (UTC)