[FRIAM] Big data forensics

[uǝlƃ ☤>$]

It's a wonderful example of careful science. I only have 1 criticism. "There is no plausible scientific reason for the deletion: the sequences are perfectly concordant with the samples described in Wang et al.(2020a,b), there are no corrections to the paper, the paper states human subjects approval was obtained, and the sequencing shows no evidence of plasmid or sample-to-sample contamination."

There's never *any* scientific reason to delete anything. So, the 1st clause in the sentence is *merely* an attempt to rouse the rabble. 8^D Otherwise known as "trolling". But buried under all the excellent, and excellently hygienic, sentences in the paper, it makes that trawl more poignant and well done.

Writ large, though, the phrase "systematic forensis" seems like a paradox. The approach I take, inspired by systems engineering, is to *log* absolutely everything, under version control, persistently. Rather than being a part of systematic forensis, it *facilitates* forensis. In light of our conversation on the myth of the objective, forensics imputes causality into a mesh of events ... hunts down *the* criminal, *the* offending "$ shed -u" command. Nothing brings that to the public forum quite like the gumshoe's pavement-pounding response to her *hunch*.

It doesn't sound quite right to talk of systematic forensics. It sounds more right to say systematic bookkeeping for the sake of more publicizing to the forum.

On 6/23/21 9:42 PM, David Eric Smith wrote:
> Speaking of big data forensics (which no-one was):
> https://www.biorxiv.org/content/10.1101/2021.06.18.449051v1.full.pdf <https://www.biorxiv.org/content/10.1101/2021.06.18.449051v1.full.pdf>
> 
> [...]
> I post because (apart from general interest), in the last paragraph of his introduction, he makes a call for data forensics to be done more systematically.


-- 
☤>$ uǝlƃ