Notices by Foone🏳️‍⚧️ (foone@digipres.club)

my mother in law wants to know what kind of external drive she should get to store ancestry documents long-term.

boy, an archivist is the last person you want to ask that.

anger canceled, I was just misreading curl. it did resume.

so again, why doesn't microsoft's official tool do this?

IT'S 2026 BUY A REAL COMPUTER

anyway I have the file now. There's a metadata JSON file that I don't have, but it looks very fakable. Then I can use the official recovery tool to build me a bootable USB drive, like it's supposed to be able to do

yeah no it just failed verification. So I'm just going to try to download it, again, and again, and again, until I get lucky

FUN FACT: the Surface IT Tool doesn't seem to validate any of the windows version info you give it in the JSON file!

it already hung

slightly tempted to automate it. watch the screen for the visible download amount, and if it hasn't changed in Xty seconds, hit cancel, restart.

currently on attempt #8

#7 made it to 17gb (of 22gb)!

cross your fingers, we're at 19gb and still climbing at 76 Mbps!

OH MY GOD ARE YOU SERIOUSLY DOING THIS, MICROSOFT

21gb!!!!

bad idea: I already MITM'd this once.

I download the file with curl and stick it on a local fast server. Then I set up mitmproxy to silently rewrite requests to their shitty server to my local one, which will be an actual server that works and doesn't randomly drop connections once out of EVERY FUCKING TIME

my first attempt and curling it stalled at 16mb.

not gigabytes, megabytes.

hey microsoft could I mail you some blank floppies and you just return 'em with the file on it? that might be easier at this point

annoyingly I already deleted the file I downloaded earlier.

(I'm juggling laserdisc archival files right now, so my laptop has a VERY full hard drive, and I thought I was done with that file when it failed verification)

my best guess for what is happening: I'm getting randomly loadbalanced onto a bunch of very overloaded servers.

okay the files all moved locally so I can just make mitmproxy point it at the different URL. but I think I have been screaming at this problem enough for one day, so I'm going to stop for tonight.

the surface hub has not defeated me yet, I fight on

I have downloaded the file and I'm now copying it onto my local server.

why didn't I just download it on my local server in the first place, so I wouldn't have to copy it across my house's network?

good question.

I lied. mitmproxy is now redirected to my local server, and Surface IT Tool is downloading from it.

annoyingly slowly, actually. Only 51Mbps? this is 22gb!

(it's probably because mitmproxy is handling all the bytes instead of letting nginx do it)

it boots! it's now recovering

this may finally get us incrementally closer to a version of windows that actually works