Filippo Valsorda :go:

@FiloSottile elsewhere / Cryptogopher / Go crypto maintainer / Professional Open Source maintainer / RC F'13, F2'17 

https://mkcert.dev / https://age-encryption.org / https://filippo.io/newsletter

🕳️ “Gaze not into the abyss, lest you become recognized as an abyss domain expert, and they expect you keep gazing into the damn thing.” —@nickm

Public Key

npub1whzyg92c6fsvpjjcnn504z0a2pfwenctp872sgmedqg2np4drj8qwakn2m

NIP-05 Address

filippo_at_abyssdomain.expert@mostr.pub

Profile Code

nprofile1qqs8t3zyz4vdycxqefvfe6863874q5hveu9snl9gyduksy9fs6k3erspz4mhxue69uhhyetvv9ujumt0wd68ytnsw43qqxeurc

Publishing to

relay.mostr.pub

Author Public Key

npub1whzyg92c6fsvpjjcnn504z0a2pfwenctp872sgmedqg2np4drj8qwakn2m

Show more details

Last Notes

2024-10-28 00:44:27

- reply

Fundamentally I think I’m complaining about the fact that the cache is 1) not a cache and 2) magic and apparently undocumented. What does it store? What are the import semantics with and without cache? (Yes I’ve read the man pages. No I won’t read the source.)

2024-10-28 00:14:02

I can't quite get my head around the zfs import/export concept.
When I replace a drive I like to first resilver the new one as a USB drive, then swap it in. This changes the device name (even using by-id).
If I use the cache it will not find the formerly-USB drive, and immediately get it out of sync.
zfs is perfectly capable of assembling a pool by just looking at disks, but that's a "forced" import of a not-exported pool, which sounds bad?
What am I supposed to do?

2024-09-24 10:17:39

- reply

Ok, looks like "pbpaste -Prefer public.html" lets me get the HTML of what was copied, so this boils down to HTML sanitization, which I do have experience with!
(https://github.com/microcosm-cc/bluemonday is a Geomys project!)

2024-09-24 09:59:17

Is there a tool or website that lets me strip all markup from copy-pasted text EXCEPT bold, italics, and lists?
I want to draft emails in Google Docs, but not carry over the font, paragraph spacing, and all that.

2024-08-07 12:35:31

If you look at a crypto/rsa invocation, you can’t actually know if it’s secure or not. The key size is nowhere in the type system. This is unusual among Go crypto packages.
I propose we fix that in Go 1.24. https://github.com/golang/go/issues/68762
If we break a production application with this, it’s overwhelmingly likely that we are fixing a security issue.
I know it will break some tests, but it’s hard to justify the real world risk, and we have a robust system of GODEBUG flags now.

2024-03-20 18:56:08

Kind of ridiculous that the best paid ebook UX is “purchase from wherever, then copy-paste the ISBN to https://annas-archive.org to download a usable file”…

2024-03-10 12:20:29

- reply

@npub1fdy…h5g8 @npub1acy…p50z I mean, yeah, it does mean that.
Open Core is a way to make money that funds OSS maintenance. It’s incompatible with undermining the money making mechanism.
Working hard on a PR *never* entitles anyone to getting their code merged and maintained upstream.

2024-02-21 12:57:55

- reply

Why is the Mastodon default to allow open registration, anyway? It should be a heavily discouraged setting, with moderated signups being the default.
If the moderation team can't keep up with registration requests surely they can't moderate those accounts either.
We've seen this before with open SMTP relays.

2024-02-17 14:17:19

- reply

@npub13w3…9k68 I don’t remember calls for Threads or Tumblr federation to be opt-in. Did I miss it?

2024-02-17 14:05:41

- reply

Also, harassment-enforced consent is a poor substitute for actual security and privacy controls.
Even if you don’t care about how it impacts its targets, it doesn’t work on anyone operating fully in bad faith.
This is the only place I know where it’s acceptable to tell users they have protections that are not actually enforceable. If a company did it we would call it false sense of security.

2024-02-17 13:53:24

I have so far seen no compelling arguments for why a bridge is fundamentally different from any other instance, and instance federation is opt-out by design.
If an instance is too big, block it. If it’s too unmoderated, block it. If you don’t approve of its implementation, block it. If you don’t like it, block it.
It’s pretty core to how the Fediverse works.
No one opted-in to interact with abyssdomain.expert.

2024-02-13 10:09:13

Interestingly, the .af NIC just suspended inet.af, too. It used to host Go modules. This suggests queer.af maybe wasn't specifically targeted for being LGBTQ+ friendly, but for being unrelated to Afghanistan.
https://bsky.app/profile/bradfitz.com/post/3klbnykibm32j
/cc @npub12fw…xddv

2023-11-16 09:55:00

- reply

@npub1a4c…t2kf Taking negative depreciation into account though feels like allowing financial speculation into the planning. How is it different from saying “buying a car and $GOOG at the same time is cheaper than car sharing, because stock growth makes up for the difference”? (Since if we can make assumptions on the used car market in the future while planning we can also make assumptions on the stock market.)

2023-10-24 15:03:09

With all the discussion around how NIST curves were generated, you might wonder: why don't we let sysadmins generate their own curves like Diffie-Hellman parameters?
Because it's a bad idea! Since the '90s we learned that the marginal advantages don't compensate at all for the added complexity.
Parameters are best fixed and standardized.
https://words.filippo.io/dispatches/parameters/?source=Mastodon

2023-10-05 21:55:59

The seeds of the NIST elliptic curves (P-256, P-386, etc.) were selected by the NSA and are rumored to be like
SHA-1("Jerry and Alice deserve a raise. 1398")
However, Jerry Solinas, who picked them, forgot them and then passed away this year.
We should crack them!
I'm announcing a $8,192 bounty (tripled if donated to charity) for anyone cracking these hashes and solving this mystery.
https://words.filippo.io/dispatches/seeds-bounty/

2023-08-07 13:53:37

- reply

@npub15kj…6rr9 Do I need to learn what Vite is to use it?

2023-08-07 13:22:05

I'm this 🤏 close to announcing age-encryption :age: for the npm ecosystem, but I can't for the life of me make ts-mocha work now.
tsc wants "foo.js" imports, while ts-mocha wants "foo" imports. AFAICT it broke when I added the required target option to tsconfig.json, so I can compile to js and publish to npm.
I'm mailing age stickers to anyone that can make it work, ideally without adding or changing 15 config options.
https://github.com/FiloSottile/age.ts

2023-08-05 19:51:48

Oh hey, file / libmagic 5.45 added support for age! :age:
https://bugs.astron.com/view.php?id=462
https://cdn.masto.host/abyssdomainexpert/media_attachments/files/110/838/748/656/662/599/original/e429985a5c077819.png

2023-06-06 17:24:52

- reply

@npub17tw…7zm4 I want signed pushes not commits, and I want it for every push not just commits generated inside GitHub.

2023-06-06 11:31:45

It would be really nice if GitHub automatically signed git pushes (maybe as an opt-in thing, and maybe tags and tarballs too) so we could prove that a certain build was made from what was pushed to a certain GitHub branch at a certain time.
IMHO would replace 80% of the use cases for git push or commit signing, with zero DX overhead: developers would still use what they use now to auth to GitHub, and then downstreams would have signed releases or point-in-time to build from.