no coronavirus thread ?

[Don Carlos]

Big day!

[Eraser]

My country is doing really shitty right now. The government is fucking up in many areas.

For instance, there is a crushing shortage in testing capacity. I have been having a sore throat and runny nose the past couple of days, but getting tested for COVID is pretty much impossible right now. Yesterday when I checked, the only available spot would be next Tuesday, at 6pm in a city that's about 80 minutes driving from where I live. Today I checked again, and there were no available spots at all. It's ridiculous. I don't really have to be anywhere, so I'm just staying at home, but something needs to change here.

[seremtan]

look on the bright side though: no one outside your country cares what goes on there, so you avoid all the bad publicity

[Eraser]

Yes, otherwise you get things like these :olo:
[tweet]https://twitter.com/maxcroser/status/1313046638915706880[/tweet]

[Κracus]

lol, who needs to pay someone to manage a database when we can just use excel!?

[Don Carlos]

:olo:

Fuck me...I live in a country run by absolute mongos

[Eraser]

Best thing is, the solution was not to migrate to a decent database, no, they moved part of the data to a new excel sheet so they've got plenty of rows columns left :tard:

[xer0s]

:olo:

Fuck me...I live in a country run by absolute mongos

Better than a country run by mangos...

[lvlshot]https://ih1.redbubble.net/image.1184011365.6735/flat,750x,075,f-pad,750x1000,f8f8f8.u1.jpg[/lvlshot]

[Eraser]

:olo:

Fuck me...I live in a country run by absolute mongos

lol, at my company, we actually run the software we build on a database named MongoDB, which is a very high performance database system

[xer0s]

[lvlshot]https://pbs.twimg.com/profile_images/886310716579303425/fJqKWCK8_400x400.jpg[/lvlshot]

[seremtan]

Yes, otherwise you get things like these :olo:
*tweet*

on a related note, the csv used to power the Johns Hopkins dashboard is structured with dates (at the day level) as column headers, with countries in rows (and to make matters worse, the daily case values are running totals, not daily counts). at end of sept, the righthand-most column in the dataset was IV :dork: this means that to do any viz oneself (i.e. in Tableau or PowerBI or whatever) you have to pivot the date columns, then add new date columns to the pivot every time you update your own copy from Github (and fix the date column because they didn't use the right format )

as someone who works in data and data viz, this imho this is really shit practice. they only need three columns: date, country, cases. Excel has way more rows than columns available

[Whiskey 7]

Where are we at I wonder?

Cases worldwide

19 September 30,697,734

28 September 33,297,042

10 October 37,121,450

Total deaths
• World 1,072,852
• USA 218,648
• Brazil 149,692
• India 107,450

Source #1 Source #2

[Don Carlos]

Strong numbers

[seremtan]

Yes, otherwise you get things like these :olo:

also, Excel maxes out at 16,384 columns. even with a column per day since the first case there should still be plenty left over

were they using a column per hour? minute? wtf?

[Eraser]

Yes, otherwise you get things like these :olo:

also, Excel maxes out at 16,384 columns. even with a column per day since the first case there should still be plenty left over

were they using a column per hour? minute? wtf?

Famous (only slightly) mathematician Matt Parker explains it very well in this video:
https://www.youtube.com/watch?v=zUp8pkoeMss

For some reason he poured it into some faux news reporting shape for comedic purposes, which distracts a bit from the issue at hand, so I'll give you the quick rundown:

- The claim that columns maxed out is unsubstantiated. The BBC reported this, but hasn't been verified.
- What is known is that an older version of the Excel file format was used (the pre-Office 2007 .xls format), which caps its number of rows at 65,536 (versus 1,048,576 for Office 2007's .xlsx).
- It turns out one case of COVID used more than one row. Official reports say that there was room for about 1400 cases which means there were about 46 or 47 rows per case.
- Because each case had a variable number of rows, it took a long time before someone noticed something was off.

[seremtan]

the part where someone was using columns as rows was especially lolworthy

that said, general understanding of how data ought to be structured is thin on the ground even among people who you'd expect to know better. i've lost count of the number of times i've been asked to build visualisations of datasets in Tableau only to find i have to restructure their shitty clusterfuck of a dataset before i can even start

[HM-PuFFNSTuFF]

https://www.theatlantic.com/health/arch ... ic/616548/

[xer0s]

Interesting. These super spreader events are the real problem. I can now envision all these Italians sitting down for family meals, screaming at each other from across the table as meatballs crumbs and Covid particles fly through the air...

[scared?]

Racist pos... I'm half Italian... Southern poverty law center called...

[HM-PuFFNSTuFF]

Racist pos... I'm half Italian... Southern poverty law center called...

you're half balled :olo:

[scared?]

Lol nope...

[seremtan]

Racist pos... I'm half Italian... Southern poverty law center called...

"hello, SPLC? i'd like you to call someone an extremist then get sued for defamation and lose millions of $$$. thanks bye"

[dekayd]

Racist pos... I'm half Italian... Southern poverty law center called...

So you are half Italian half Moron, got it :olo:

[Captain]

So full Italian then :olo:

[Pext]

So full Italian then :olo:

Mama mia! :olo: