[ExI] Essential Upload Data

[spike at rainier66.com]

> On Behalf Of Re Rose via extropy-chat
Subject: [ExI] Essential Upload Data

 

>…I agree with Darin, the Harvard study is highly unlikely to be correct, GIGO and all that. Nothing that any "esteemed scientist" has said so far has had any acquaintance with the reality of this pandemic.

 

>…Models are highly malleable, by adjusting parameters within the model you can take any model and make it say whatever you wish. AI is subject to this as well. Since the parameters are hidden and it is often unclear to non-users what physical property the parameters control, this is an easy way to dupe the public. 

…

 

 

Hi Regina, ja, this is the nature of exponential models: they have an enormous span in their predictions.

 

This looks like a really good site to me:

 

https://covid19.healthdata.org/united-states-of-america

 

>…So - models are difficult and subject to meddling, and bigly caution is advised before trusting them IMHO. 

 

--Regina

------------------------------



No one really knows what the heck is going to happen.  It is easy to envision having the virus slosh around at a low level over the summer then really come back in the fall, November and December, the usual flu season.  

 

I can envision we will handle it differently the second time if it comes back: most governments will have concluded that a shutdown isn’t really a viable option.  It polarizes their societies too much as it has in the US: it creates a great divide between those still getting their paychecks and those who do not.

 

spike

 





-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20200522/8b7bd19b/attachment.htm>