search

CSSEGISandData / COVID-19

Novel Coronavirus (COVID-19) Cases, provided by JHU CSSE
https://systems.jhu.edu/research/public-health/ncov/
29.12k stars 18.4k forks source link

No data at all for 3/14 #746

Open marshallstanton opened 4 years ago

marshallstanton commented 4 years ago

Confirmed cases are completely blank for all locations/rows.

Ruegg commented 4 years ago

This was a very poor decision, it broke my tracker which had 6000 users as of the last 24 hours.

Bost commented 4 years ago

:man_facepalming:

loulouOz commented 4 years ago

It is been now 3-4 days they have issues, and they fixed it in the next couple of hours. Just condition your app to keep latest data available and that will solve your problems for now even if this is not an ideal solution. 👍

Bost commented 4 years ago

Just condition your app to keep latest data available and that will solve your problems for now even if this is not an ideal solution. +1

The problem is, here in Europe it's almost 2 a.m. and since about 4 weeks I go to bed at 4 a.m. every day because of all these screw-ups... And I guess I'm not the only one. And you know Europe is the hotspot now. And the people here - pretty everyone really wants to know what's the progress... bloody hell :(

loulouOz commented 4 years ago

I understand, maybe write and automate a little script that will check the last column of the first csv to identify potential errors, if none download all and update your app else wait. You are gonna lose your head if you keep on checking manually... good luck!

shahesam84 commented 4 years ago

Calm down guys. This happens everyday and they fix it afterwards. If you have problem better to fix your code to avoid showing incomplete data. These guys are doing great providing us with such valuable collected information.

Bost commented 4 years ago

You are gonna lose your head if you keep on checking manually... good luck!

I rely on https://github.com/ExpDev07/coronavirus-tracker-api and they've been delivering on promise (except one exception yesterday - Ok nobody is perfect). So no matter what, I really want to see that my https://t.me/corona_cases_bot is working. Call it a sense of responsibility... Hmm. I'm tired.

Ruegg commented 4 years ago

Yeah I fixed mine just by checking whether the last element per line is empty, then removing it.

mmonfort2908 commented 4 years ago

I'm in Australia and I've noticed that the pattern is they initially update the daily datasets first and then the full time series. The full time series has latest dates added but with blank values and then a while later the values are added. It's not a big deal if you set your app to cope with this.

Would be great to get some transparency from them on what the schedule but they are doing an amazing service to keep this up. Have any of you guys read the Lancet article they wrote on this that's in the main page of the repo? Looks like there is a large reliance on manual work and people to get this done.

I hope it continues because I'd rather have this and a few delays/hiccups than nothing at all. Let's be encouraging to them please.

Bost commented 4 years ago

amazing service [..] manual work

The former is unfortunately not true, the later has been obvious since the early days and all in all "amazing service" and "manual work" goes rarely together, when computers are involved.

Computer mouse is the wrong tool for any serious data manipulation and MS Excel is not the data source it's the data sink! :( Don't you see it?

yy commented 4 years ago

I'm building an alternative data repository (with code and workflow to clean and process the dataset): https://github.com/covid19-data/covid19-data Feel free to use it (and help)!

simra commented 4 years ago

As an alternative for country-level data I'm also pulling the WHO numbers from here: https://ourworldindata.org/coronavirus-source-data

analyzewithpower commented 4 years ago

@Bost #agree - the data on their map -https://coronavirus.jhu.edu/map.html -is accurate and error-free, why not share that same data with us even if it means restructuring the file?

An export of the data from their dashboard would suffice,

analyzewithpower commented 4 years ago

@simra - same here, I am currently looking for a source by state.

yy commented 4 years ago

@simra hey I've made a workflow to further clean the Our World in Data team's dataset here: https://github.com/covid19-data/covid19-data#daily-case-data-based-on-who-report-cleaned-by-our-world-in-data

Bost commented 4 years ago

@Bost #agree - the data on their map -https://coronavirus.jhu.edu/map.html -is accurate and error-free, why not share that same data with us even if it means restructuring the file?

An export of the data from their dashboard would suffice,

That's a very good question! I'm bookmarking myself this conversation!!!

mmonfort2908 commented 4 years ago

amazing service [..] manual work

The former is unfortunately not true, the later has been obvious since the early days and all in all "amazing service" and "manual work" goes rarely together, when computers are involved.

Computer mouse is the wrong tool for any serious data manipulation and MS Excel is not the data source it's the data sink! :( Don't you see it?

Better than having nothing out there though right?