sigwinch28
Joined 8 years ago
Karma
867
Submissions
229
Member since
Jul 20
Recent submissions
Comment · 1 month ago
I will now refer to senior engineers with a bone to pick with a codebase as “hungry goats”.
Comment · 1 month ago
I think bit-aligned encodings (i.e. schemes where the output alphabets have a size that is a power of two) are the sweet-spot for encoding schemes due to simplicity of encoding and decoding compared to non-bit-aligned schemes like base84.
Base16 is verbose, but each symbol in the alphabet carries exactly 4 bits (hence why a single byte in hex is two symbols). This is nice for encoding bytes. Base64 requires two non-alphanumeric symbols such as + and / or - and _ because A-Za-z0-9 only provides 62 characters. Each output symbol encodes exactly 6 bits of the original input. But when we're encoding whole bytes we sometimes need to use padding (when the input is not a multiple of 3 bytes and that ambiguity is harmful in the context that decoding occurs in). Then there's Base32, which is what currently seems like a sweet spot to me and is what I'm considering using for identifiers in my own systems. It only requires 32 characters, so can easily take a range of A-Z0-9 or a-z0-9 excluding visually-similar characters. Like Base64 it can also require padding in some settings.
I quite like the general scheme used in the Bech32 and Bech32m framing idea, which allows for a human-readable prefix before a `1` and then the base32 data follows. This can be done because `1` is excluded from the Bech32 alphabet. This prefix can be used to differentiate between kinds of identifier, for example:
https://github.com/bitcoin/bips/blob/master/bip-0350.mediawi...
But to address the article directly, I'm not sure the complexity is worth the gains in the table in https://github.com/jedisct1/zig-base84#encoding, which states that we save ~6.5% of characters, and for 128 input bytes, the resulting Base64 string is 171 characters, while the Base84 string is 161-166 characters.
I also dislike that the chosen alphabet breaks text selection; GitHub very carefully picked their current token formats so that the whole token is selected with a double-click:
> One other neat thing about _ is it will reliably select the whole token when you double click on it. Other characters we considered are sometimes included in application word separators and thus will stop highlighting at that character. Try out double clicking this-random-text versus this_random_text!
https://github.blog/engineering/platform-security/behind-git...
For example, compare double-clicking on the article's Base84 alphabet:
> ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789!#$%&'()+,-;=@[]^_`{}~
with the Bech32 Base32 alphabet:
> qpzry9x8gf2tvdw0s3jn54khce6mua7l
Regardless though we still have the fundamental issue that bytes encoded as Base16, Base32, Base64, or even Base91/Base84 will be _longer_ when encoded. Base32 can encode up to 159 bytes before hitting a filename limit of 255 bytes. I can't remember whether filenames are 16-bit on Windows (UCS-2? UTF-16?), but I speculate on average that maybe you could save a lot of bytes by first ensuring that filenames are UTF-8 before encrypting them, since filenames on a lot of computeres, especially ones running in the west, likely contain a lot of latin characters. You could even switch between encodings to get the best bit-packing in filenames.
At the moment it looks like turbocrypt forbids encrypted filenames that are too long: https://github.com/jedisct1/turbocrypt/blob/4905241d271e84a0...
I think turbocrypt could switch to using a surrogate file when an encrypted filename exceeds 255 bytes to allow for encrypting any filename permitted by the filesystem, while still preserving the property that the same filename in a different directory has the same encrypted filename. If an encrypted filename is too long to fit on the filesystem, maybe its hash could be stored as the filename instead (which is extremely unlikely to collide). Then we could store the full encrypted filename in a `.name` file in the same directory. We could do similar for directory names that are too long. We could then use a less dense but bit-aligned encoding scheme like Base32 with an all-lowercase alphabet without any punctuation to make filename text selection straightfoward. We would avoid all case sensitivity traps that can occur with things like Base64 and Base84. Let's pick two filename prefixes, say "tcf1" for "turbocrypt filename" and "tch1" for "turbocrypt filename hash". Then we could have a directory layout like this:
tch1q9x8gf2tvdw0 # a file whose encrypted filename exceeds 255 bytes; this filename is a hash of the encrypted filename, e.g. SHA-256.
tch1q9x8gf2tvdw0.name # a file whose contents are the full encrypted filename of tch1q9x8gf2tvdw0.
tcf1tvdhc6mua7l9x8s3qprz # a file whose encrypted filename does not exceed 255 bytes.
Once a system to support long filenames is implemented, the size of the alphabet used for encoding the filenames (like Base84) becomes less important; Base16, Base32, Base64, or another bit-aligned encoding scheme could be used.
As a very small added bonus, you could even implement a bit-aligned codec such as base16 or base64 using SIMD via a bunch of swizzling, shifting, bitmasking, and AND/XORing, making it very fast should the need arise.
Comment · 4 months ago
Conversely, as a hobbyist photographer, I want to do the exact opposite for most photos I take.
I would like my camera info, especially the body, lens, focal length, and settings in the image. I recently discovered that software like Darktable can even take a gpx file and photo timestamps to add coordinates to photos taken on a camera without a GNSS receiver.
Comment · 9 months ago
Measurement trains filled with cameras and LIDAR
For example, in the U.K.:
Comment · 10 months ago
A writable file closing itself when it goes out of scope is usually not great, since errors can occur when closing the file, especially when using networked file systems.
Comment · 1 year ago
Or it’s simply an indicator of a schema that has not been excessively normalised (why create an addresses_cities table just to ensure no duplicate cities are ever written to the addresses table?)
Comment · 1 year ago
Via Wikipedia:
> The developer offered full refunds to the game for macOS and Linux owners regardless of how long they had the game.
https://en.wikipedia.org/wiki/Rocket_League#Free-to-play_tra...
https://www.rockpapershotgun.com/rocket-league-ending-mac-an...
Comment · 1 year ago
I’m from the U.K. and I consider the government’s actions around digital privacy to be somewhere between incompetent and malicious.
Comment · 2 years ago
>waiting for 2 different lights just to get to the opposite corner.
A solution sometimes seen in London is a “Pedestrian Scramble”, where pedestrians are explicitly given full (and even diagonal) access to a junction with all other traffic stopped.
Comment · 2 years ago
With SSO, the party running the SSO decides what the authentication policy is.
For example, where the authentication request is coming from (on-site, managed device), what methods are being used (hardware second factor, Authenticator app).
These are all things that the SSO can check at time of authentication, before a token or session key gets issued to the user. Also, all of these things can be checked again when doing any auth flows for the various linked services.
So with stolen SSO credentials, they might be worth diddly squat to you if you didn’t think to also be on-site or on a managed company device (physically or virtually).
Comment · 2 years ago
Reuse of identifiers seems to be a theme in aviation https://news.ycombinator.com/item?id=37401864
Comment · 2 years ago
If we replace “YAML” with “JSON” and then talk about naive text-based templating, it seems wild.
That’s because it is. Then we go back to YAML and add whitespace sensitivity and suddenly it’s the state-of-the-art for declaring infrastructure.
Comment · 2 years ago
A lot government funding stipulates open access publication of some form.
Comment · 2 years ago
A smartwatch for kids could be so good if it was designed in a way to be educational, but most importantly, which respects a child’s privacy utmost, even from their parents in terms of tracking.
For example, a maps app, to always get the kid home if they’re lost. Medication reminders. Fitness tracking. Emergency SOS. A calendar to remind them about family birthdays and upcoming holidays. School timetables. Medical ID. Payment cards or passes for travel (in Western Europe a lot of schoolchildren commute by themselves, especially on public transport) and spending their allowance. Let the kid choose to notify their family of their location as and when they want to. Empower them to use tech to their advantage but put their privacy first.
Children are going to end up as adults in this world regardless of whether we teach them, so we should be teaching them the benefits and warning them of the many bad actors. We should be teaching our children the skills they need to navigate the modern world. This includes technology and abusive/controlling relationships.
I believe a good responsible smartwatch for kids can exist. Alas, this is Google and helicopter parents exist, so this product is not it.
Don't bother with the broken plaque-ridden keyservers. Use WKD instead: