Simon Howard fragglet

python3 -m venv venv
./venv/bin/pip install --upgrade pip
./venv/bin/pip install py-bcrypt==0.3
export CRYPTOGRAPHY_DONT_BUILD_RUST=1
#export CFLAGS="$(pkg-config --cflags eopenssl33)"    # on OpenBSD
#export LDFLAGS="$(pkg-config --libs eopenssl33)"
./venv/bin/pip install cryptography==3.4.8
./venv/bin/pip install autobahn==21.2.1 buildbot-worker

Instructions for installing OpenDarwin 8.0 in VirtualBox

You want the file darwinx86-801.iso

Set machine type to Windows NT 4 (you want to avoid an EFI machine)
PIIX3 chipset, 256MB of RAM is okay. Set ethernet controller to Intel 82540EM (Darwin 8.x has a driver for it)
Enable ACPI with VBoxManage modifyvm OpenDarwin --acpi on
Set up storage: PIIX3 controller with host i/o cache enabled.

Virtual hard disk should be first disk. I created an 80GB drive.

I'm publishing this as a public gist so that it's googleable for anyone else looking for replacement rubber feet for their Apple IIc.

Rubber tends to perish over time and when restoring an old Apple IIc I found that the rubber feet on mine were in a bad condition. A good (perfect?) replacement seems to be 3M Bumpons, specifically part number SJ5003 (11.1mm in white). I believe these may even be the originals that were fitted on the IIc as shipped, because this story on folklore.org describes Apple using Bumpons in 1983.

This is a response to the draft design for Go 2 error handling.

I think that the main thing I dislike about the proposed check/handle blocks is that they feel too much like a kind of "magic goto" that breaks the normal program flow. I worry that it won't be straightforward to follow the logic of what happens when a check fails. In the simple handle err { return err; } it isn't so bad but the examples with chained handle blocks seem quite counterintuitive.

I'm in agreement with Alessandro Arzilli in that I think chaining is the main source of my objection. It seems to me that the

Keybase proof

I hereby claim:

I am fragglet on github.
I am fraggle (https://keybase.io/fraggle) on keybase.
I have a public key whose fingerprint is 6D2C 117E 0310 6644 97AA 9546 F6C2 EE9C 2335 4344

To claim this, I am signing this object:


	_update_auth_sock() {
	if ! ls ~/.ssh/auth_socks \| grep -q ""; then
	# No sockets
	return
	fi

	local sockfile

	for sockfile in $(ls -t ~/.ssh/auth_socks/*); do

	"""Extracts media URLs from a tweet archive."""
	import glob
	import json

	for filename in glob.glob("tweets/*.js"):
	with open(filename, "r") as f:
	data = f.read()
	_, data = data.split("=", 1)
	tweets = json.loads(data)
	for tweet in tweets:

	doom2map/map01.dwd
	doom2map/map03.dwd
	doom2map/map06.dwd
	doom2map/map07.dwd
	doom2map/map08.dwd
	doom2map/map09.dwd
	doom2map/map10.dwd
	doom2map/map12.dwd
	doom2map/map13.dwd
	doom2map/map16.dwd

	Corel Draw has a bug in its SVG parsing: it treats all movement commands as absolute. By default, inkscape will use both absolute and relative movement commands.

	There are two ways to deal with this:

	1. Convert the SVG to PDF in inkscape. Corel's PDF import is less buggy.

	2. Tell inkscape to output SVGs with only absolute movement commands:
	2a. Open the Inkscape Preferences window (via Shift+Ctrl+P or File \| Inkscape Preferences).
	2b. Go to "SVG Output".
	2c. Uncheck "Allow relative coordinates".