Because I randomly had a conversation about this some weeks ago:
That account is not a bot.
It says in their profile they are not a bot, just neurodiverse.
They just … eithery manually, or via use of an unoptimized regex code, or something, remove most characters other than ATCG from a post…
…and copy paste it into a nearly 20 year old freely available (but woefully underpowered and incomplete) API for genome matching.
Then they copy paste the output manually.
The account also posts non genome matching comments.
Its just a person with a unique hobby.
Unfortunately they are using such an old and underpowered API… and not even formatting their input into this API properly… that it is likely the equivalent of a random species output.
EDIT: Not calling you out personally, as you admit your memory is hazy, but 99% of the people that talk about this account seem to do the most tumblr thing possible:
Do absolutely 0 research or investigation, make up a bunch of personal headcannon (falsehoods, baseless speculation), and then run with it as fact.
I was honestly just curious – it just seemed so out of place. Even in the full post (the source – not the version I posted) I couldn’t figure out how it got from werewolf periods to the moth string.
Ah ok now I remember lol, I forgot it was a person behind the account, thanks for correcting me. Next time I’ll try to be a bit more thorough before posting, I just got kinda excited cause I knew about it lol.
Like i guess i did this: https://imgs.xkcd.com/comics/ten_thousand.png
To what? As in - what in that thread is it relating the moth to? (it’s fine if you don’t remember, but if anyone else does, I need my confusion eased lol)
The ‘idea’ is that a post … of any length … is stripped of all characters other than ATCG, and that what remains constitutes a possible genome sequence.
bark moth couple dingo
becomes
atcg
Then you run this ‘sequence’ against a database of genomic data, and get a nearest match to some species.
The actual problems with this are many.
Actual genomic sequences are orders of magnitude larger than even a lengthy, stripped down tumblr post.
Actually running a nearest match search of a short string / sequence will likely match many, many different species.
Running an exhaustive, ie, accurate, nearest match search against an actually comprehensive database would require using a supercomputer, or at the very least, a lot of powerful, networked computers.
…
EDIT: There’s no real relation to the content or context of the post or thread to the ‘genomic match’ species. None.
Its about as legitimate or useful as trying to decipher the ‘Bible Code’ by running any number of pattern matching algorithms on a modern, English Bible.
There is/was a tumblr bot that would reply with a closest DNA/genome string and a picture iirc (my memory on this is really fuzzy)
Because I randomly had a conversation about this some weeks ago:
That account is not a bot.
It says in their profile they are not a bot, just neurodiverse.
They just … eithery manually, or via use of an unoptimized regex code, or something, remove most characters other than ATCG from a post…
…and copy paste it into a nearly 20 year old freely available (but woefully underpowered and incomplete) API for genome matching.
Then they copy paste the output manually.
The account also posts non genome matching comments.
Its just a person with a unique hobby.
Unfortunately they are using such an old and underpowered API… and not even formatting their input into this API properly… that it is likely the equivalent of a random species output.
EDIT: Not calling you out personally, as you admit your memory is hazy, but 99% of the people that talk about this account seem to do the most tumblr thing possible:
Do absolutely 0 research or investigation, make up a bunch of personal headcannon (falsehoods, baseless speculation), and then run with it as fact.
I was honestly just curious – it just seemed so out of place. Even in the full post (the source – not the version I posted) I couldn’t figure out how it got from werewolf periods to the moth string.
Thanks :) :) :)
Ah ok now I remember lol, I forgot it was a person behind the account, thanks for correcting me. Next time I’ll try to be a bit more thorough before posting, I just got kinda excited cause I knew about it lol. Like i guess i did this: https://imgs.xkcd.com/comics/ten_thousand.png
That’s just humans in general
To what? As in - what in that thread is it relating the moth to? (it’s fine if you don’t remember, but if anyone else does, I need my confusion eased lol)
The ‘idea’ is that a post … of any length … is stripped of all characters other than ATCG, and that what remains constitutes a possible genome sequence.
becomes
Then you run this ‘sequence’ against a database of genomic data, and get a nearest match to some species.
The actual problems with this are many.
Actual genomic sequences are orders of magnitude larger than even a lengthy, stripped down tumblr post.
Actually running a nearest match search of a short string / sequence will likely match many, many different species.
Running an exhaustive, ie, accurate, nearest match search against an actually comprehensive database would require using a supercomputer, or at the very least, a lot of powerful, networked computers.
…
EDIT: There’s no real relation to the content or context of the post or thread to the ‘genomic match’ species. None.
Its about as legitimate or useful as trying to decipher the ‘Bible Code’ by running any number of pattern matching algorithms on a modern, English Bible.
Thanks for the explanation, I would have never figured out the stripping of the post down to those letters part lol