Show more

Did you know my !!con West talk from February is online? You should watch me talk about old video games and obsolete text encodings! youtu.be/t37Cr3FDyvU

Also, this is a roundabout way of saying “sorry I haven’t been active on here for a little bit”... haha.

I had my reservations about how useful a smart temperature / humidity sensor would be when Jessica installed it, but it turns out “your apartment flooded and the repair people installed superheating dehumidifying devices” is exactly the thing it’s useful for I didn’t predict.

Misty boosted
#emacs #magit

Tip of the day: you're usually using feature branch but forgot to create the branch and committed your feature to master like an idiot?

magit-branch-spinoff to the rescue.

From http://blog.jenkster.com/2017/07/emacs-feature-branch-tip.html

Great tip, happens to me at least once every two weeks ><

Wow, Apple just released a patch for devices that haven't gotten an OS update in three years. Looks like it's a fix for the GPS epoch bug? At least, Apple says this is for a bug that will start happening November, 2019. support.apple.com/en-us/HT2102

I like how this Docker hub downtime was so famous that Google now links to the one incident for “docker hub status” instead of the main Docker status page.

10x engineers are fake, but did you know 1000x engineers are real?

Engineers who work hard to support their fellow engineers, and help them grow, lift everyone up. One engineer can unblock 1000 engineers.

Misty boosted

“Women working in electronic computing early on were dismissed precisely because they had technical skills, not because they lacked them. As those skills became more highly valued, women were often forced out of the field instead of being promoted …“

📄 Marie Hicks, Against Meritocracy in the History of Computing, 2016 → s3data.computerhistory.org/cor

Once I get this fixed and I’ve confirmed (or disconfirmed) my investigation, I’ll put up a blog post that goes into all this in more detail!

At any rate, the “bad” version of the MPEG file is always having the “data” bit set, while the “good” version varies between “video”, “audio”, and “data”. Feels like Lunar is probably skipping over every non-video sector.

There’s also a format-dependent “coding” byte, which the “bad” tool is setting to 0. It has information about resolution and coding type; Lunar may be depending on it.

I’ll continue investigating, but I think I have all I need to attempt a fix now.

Think I’ve about figured out my problems with the MPEG version of Lunar.

The CD XA format has some metadata that goes along with the actual data in each sector, signalling things like “what kind of data is in this sector?” I guess the intent is that primitive systems can use the sector metadata without needing to actually identify the file type using other means?

Misty boosted

I've written something about the Andaluh and the andaluh-rs crate that I've been working on: danigm.net/andaluh.html

I can’t believe the CD-i standard being made free was actually useful to me. Turns out it encompasses the CDXA and VCD standards? So instead of reading those specs that cost $hundreds I can just read the spec for a failed multimedia console and get what I need.

Huh, new Toot! posting screen. Seems like a handy update!

Probably a weird question but is there any way to stub time calls for C programs so they always return the same value? Debugging the binary output of a program where part of its output depends on dates.

Anyone out there have the White Book (Video CD) standard?

Misty boosted

The WAV audio file format is an application of the Reykjavík International Film Festival.

Misty boosted

So there's a debatable regression in bzip2-1.0.7. It cannot uncompress some files from lbzip2, which worked fine with bzip2-1.0.6, because of the fix for CVE-2019-12900.

Bzip2-1.0.6 and before had a bug with unvalidated input, which a fuzzer found to lead to a buffer overflow.

However, those version seemed to work for lbzip2 files, which (arguably wrongly) were writing the input value larger than bzip2 expected.

gitlab.com/federicomenaquinter - people who want to do code digging appreciated!

Show more
digipres.club

digipres.club is a space for folks interested in productive conversations about, well, digital preservation! If you enjoy talking about how to do memory work with computers, or even with cardboard boxes of old photos, you belong with us on digipres.club. Many of us are/were Twitter users looking for an inclusive and community supported approach to social media. If any of these things sound good to you, consider joining us now.