Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

github.com

Post reply on HN
Print view View original

71–80 of 236 posts

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#71

Post by oleganza »

pavlov wrote: post #12
ajkjk wrote: post #4

Unrelated, but this links to "Crockford's alphabet", https://www.crockford.com/base32.html , which is a base-32 system that includes all alphanumeric characters except I and L (which are confusable with 1), O (which is confusable with 0), and U (????). The page says the reason for excluding U is "accidental obscenity'. What the heck is it talking about?

True Latinists find the letter U vulgar to the point of obscenity because it didn’t exist in Cicero’s time.

Trve Latinists wovld appreciate yovr point.

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#72

Post by oleganza »

stronglikedan wrote: post #16
Racing0461 wrote: post #8

Earlier quoted context omitted.

yep, youtube video ids has/had? same issue where it would have things like fag/f4g etc in it. eg: google "allinurl:fag site:youtube.com"

You can prevent any obscenity, O and 0 confusion, and I and L confusion, just by excluding vowels. If someone interprets "f4g" in an offensive way, then they have bigger issues than can be dealt with in software.

The problem is not being offended per se, but having your user id accidentally become "user_123fuck567" — that's akin to having a vulgar license plate on your car's forehead. People don't appreciate how lucky they sometimes are.

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#73

Post by hinkley »

hinkley wrote: post #27
ajkjk wrote: post #4

Unrelated, but this links to "Crockford's alphabet", https://www.crockford.com/base32.html , which is a base-32 system that includes all alphanumeric characters except I and L (which are confusable with 1), O (which is confusable with 0), and U (????). The page says the reason for excluding U is "accidental obscenity'. What the heck is it talking about?

A coworker and I came up with basically this same set about 4 years before Crockford. We were trying to solve the url slug problem, and they were long enough that we felt 5 bits per byte would reduce transcription annoyances. In the end I think we had a couple of characters to spare, and so, sitting by ourselves because everyone else had gone home for the day, we ranked swear words by how offensive they were to prior…

Other comments are jogging my memory. I think we went case sensitive (62 characters -> 30 spares), eliminated aA4, eE3, iI1l oO0 (maybe Q), uU, which is 16 characters, 14 to go. Remove the remaining 7 numbers (once you remove most for leetspeak what's the point of the rest?), nN, yY. That leaves 2 left and I can't recall what we did with those. Maybe kK or rR.

Y is pretty versatile for pissing people off.

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#74

Post by hfkwer »

pphysch wrote: post #34
Daegalus wrote: post #31

Earlier quoted context omitted.

Ideally, you should version that as UUIDv8. Since that is a more custom implementation, but I guess changing the random bits with type info works fine for UUIDv7, they jsut arent random anymore.

Not every bit of UUID is required to be random. The goals are smallness, uniqueness, monotonicity, resistance to enumeration attacks, etc. Not randomness for randomness sake. My UUIDv7+ can be consumed as a standard UUIDv7. It is not intended to be v8. A program can treat the last 16 bits as random noise if it wants.

> It is not intended to be v8.

There is already a UUIDv8. It's defined as vendor-specific UUID. https://www.ietf.org/archive/id/draft-peabody-dispatch-new-u...

> Some example situations in which UUIDv8 usage could occur:

> An implementation would like to embed extra information within the UUID other than what is defined in this document.

Isn't that exactly what you are doing?

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#75

Post by nickjj »

tomnipotent wrote: post #53
changoplatanero wrote: post #43

what does k-sortable mean?

That if a hundred servers are generating (uuid, timestamp) tuples that are subsequently merged on a single machine, and sorted by uuid, it would have almost the same order as if sorted by timestamp. This property is useful for RDBMS writes, when the UUID is used as a primary key and this locality ensures that fewer slotted pages need to be modified to write the same amount of data.

> it would have almost the same order as if sorted by timestamp.

Is there documentation covering the scenarios on how the order can become out of sync and what the odds are? There's a big difference between "almost" and "always" if we're talking about using this as a database PK.

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#76

Post by rbera »

titanomachy wrote: post #70
codeulike wrote: post #48

Earlier quoted context omitted.

If I and O are already excluded and you also exclude U that removes a lot of potential rude looking three letter combinations like *** and *** and *** and also the four letter ones like **** and **** and the dreaded ****. Of course because you have A then **** is still a possibility but very very unlikely

Wow I didn't know HN even had obscenity filters, and I've been here for many years. Guess that's a credit to the general civility of the community. EDIT: It appears that other people in this thread are freely using profanity, so either your comment was targeted by automation due to the unusual density of banned words, or it's a joke that went over my head :)

That explains it, I was very confused by what I assumed was self-censoring, since the comment didn’t actually clarify anything. I wish there was an accepted way to disambiguate asterisks from server side filters.

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#77

Post by Nelkins »

Daegalus wrote: post #29
wood_spirit wrote: post #2

UUIDv7 has been taking HN by storm for years now! When is it going to become a proper standard, and when are libraries and databases and all the rest going to natively support it?

Its been going through drafts and improvements. It's very close to being standardized, and many libraries are supporting it already, or new offerings are being added. For example I maintain the Dart UUID library, and my latest beta major release has v6, v7 and a custom v8. There is a list of them somewhere, I know I get pinged on every new draft by the authors because I am listed as a library maintainer on one of the…

How much does it change between drafts? Close enough to where I could use it in production?

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#78

Post by carlsverre »

kortex wrote: post #65

Does the prefix ("user_") get recorded in the DB (so every string in the column starts with the same "user_"), or does are there constraints and other clever chicanery to save those bytes in every record? Or do modern DB engines even care? Is this premature optimization?

The authors have created a specialisation for Postgres that leverages a custom type which is a tuple of type and uuidv7: https://github.com/jetpack-io/typeid-sql/blob/main/sql/typei...

This is more optimal for Postgres while making it slightly more difficult to interop between the db and the language (db driver needs to handle custom types, and you need to inject a custom type converter).

And while there are hacks you can do to make storing uuid-alikes as strings less terrible for db engines, if you want the best performance and smallest space consumption (compressed or not) make sure to use native ID types or convert to BINARY/numeric types.

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#79

Post by macintux »

rbera wrote: post #76
titanomachy wrote: post #70

Earlier quoted context omitted.

Wow I didn't know HN even had obscenity filters, and I've been here for many years. Guess that's a credit to the general civility of the community. EDIT: It appears that other people in this thread are freely using profanity, so either your comment was targeted by automation due to the unusual density of banned words, or it's a joke that went over my head :)

That explains it, I was very confused by what I assumed was self-censoring, since the comment didn’t actually clarify anything. I wish there was an accepted way to disambiguate asterisks from server side filters.

I assumed this was a riff on the classic bash.org transcript.

http://www.bash.org/?244321

Re: Type-safe, K-sortable, globally unique identifier inspired by Stripe IDs

#80

Post by AtNightWeCode »

changoplatanero wrote: post #43

what does k-sortable mean?

That it is a nearly sorted sequence. Typically the first part of the ID is a timestamp. So you may sort the IDs down to the second it was created. But for IDs created at the same time the order is random. This can be used for caching, database performance and so on.
Post reply on HN