A few years ago I made a scheme whereby you could use sequential numeric IDs in your database, but expose them as short random strings (length 4–20 step 2, depending on numeric value and sparsity configuration). It used some custom instances of the Speck cipher family, and I think it’s robust and rather neat.
Although I finished it, I never quite published it properly for some reason, probably partly because I shelved the projects where I had been going to use it (I might unshelve one of them next year).
Well, I might as well share it, because it’s quite relevant here and interesting:
(Ah, it’s fun reading through that document a bit again. A few things I’d need to update now, like the Hashids name, or in the UUID section how UUIDv7 is no longer a draft, and of sidenote 12 I moved to India and got married and so took a phone number ending in 65536, replacing my Australian 32768. :-) )
bflesch · 1h ago
Hey Chris, that's a really nice blogpost. Not only the content but also the design / sidenotes. What kind of software stack do you run your block with?
It’s lasted for three years of use and three years of disuse, and I hope to replace it with something utterly different (stylistically and technically) by the end of this year, though it may slip to next year. The replacement will be based on handwriting.
bflesch · 44m ago
Thanks. I like it very much, perfect dark mode. The serif font could be a tiny bit bigger for readability. Not a fan of handwriting fonts but you do you :-)
chrismorgan · 41m ago
Who said handwriting fonts?
(I’m not a fan of handwriting fonts either. They’re never truly satisfying, though some with quite a few variants for each character get past the point of feeling transparently inauthentic. But when you can write and draw what you choose, where you choose, that’s liberating.)
bflesch · 1h ago
Oh thanks for sharing this. Many years ago I was asked to code such a thing during an interview and I totally screwed it up, and of course I forgot the name of this technique.
I wanted to use it many times in project for non-iteratable IDs but never found it again.
aabbdev · 3h ago
Hi, I’m the author of uuidv47. The idea is simple: keep UUIDv7 internally for database indexing and sortability, but emit UUIDv4-looking façades externally so clients don’t see timing patterns.
How it works: the 48-bit timestamp is XOR-masked with a keyed SipHash-2-4 stream derived from the UUID’s random field. The random bits are preserved, the version flips between 7 (inside) and 4 (outside), and the RFC variant is kept. The mapping is injective: (ts, rand) → (encTS, rand). Decode is just encTS ⊕ mask, so round-trip is exact.
Security: SipHash is a PRF, so observing façades doesn’t leak the key. Wrong key = wrong timestamp. Rotation can be done with a key-ID outside the UUID.
Performance: one SipHash over 10 bytes + a couple of 48-bit loads/stores. Nanosecond overhead, header-only C11, no deps, allocation-free.
Tests: SipHash reference vectors, round-trip encode/decode, and version/variant invariants.
Curious to hear feedback!
JimDabell · 1h ago
I like the idea.
UUIDs are often generated client-side. Am I right in thinking that this isn’t possible with this approach? Even if you let clients give you UUIDs and they gave them back the masked versions, wouldn't you be vulnerable to a client providing two UUIDs with different ts and the same rand? So this is only designed for when you are generating the UUIDv7s yourself?
move-on-by · 23m ago
Any version of UUID except v4 on the client side would be a mistake- as you are relying on it to provide extra information such as a timestamp which might be manipulated.
Of course, UUIDv4 on the client side is not without risk either- needing to validate uniqueness and not re-use of some other ID. For the UUIDv7 on client side- you could add some sanity validation- but really I think it’s best avoided.
the_mitsuhiko · 1h ago
Two pieces of feedback here:
1. You implicitly take away someone else's hypothetical benefit of leveraging UUID v7, which is disappointing for any consumer of your API.
2. By storing the UUIDs differently on your API service from internally, you're going to make your life just a tiny bit harder because now you have to go through this indirection of conversion, and I'm not sure if this is worth it.
whatevaa · 1h ago
1. Unless API explicitly guarantees that property, relying on that is bad idea. I wouldn't.
the_mitsuhiko · 50m ago
Sure, but that's not really the point is it? If you get a UUID you can store it as a UUID. If the UUID happens to come around as a v7 you get some better behavior in your database, and if it does not, then it does not but there is nothing you can do about.
hnav · 22m ago
depends on the database, famously DynamoDB used to suffer from hotspotting when dealing with monotonically increasing keys
the_mitsuhiko · 14m ago
You're missing the point here. You can always go from ordered to randomness. You cannot go from randomness to ordered. So by intentionally removing the useful properties of UUIDv7, you're taking away some external API consumers' hypothetical possibility to leverage benefits. If I know (as an API consumer) that I have a database that for whatever reason prefers evenly distributed primary keys or something similar, I can always accomplish that by hashing. I just can never go the other way.
aabbdev · 1h ago
You can always treat IDs as UUIDv4, while actually storing them as UUIDv7—combining the benefits of both. From your perspective, they’re just UUIDv4
kevlened · 1h ago
One impact of the_mitsuhiko's second point is during debugging.
Usually if you see an id in your http logs you can simply search your database for that id. The v4 to v7 indirection creates a small inconvenience.
The mismatch may be resolved if this was available as a fully transparent database optimization.
thunderfork · 1h ago
This seems like the kind of tool you would only use where you have the following needs:
1. Not leaking timestamp data (security/regulations)
2. Having easily time-sortable primary keys (DB performance/etc.)
If you don't have both of these needs, the tool is an unnecessary indirection, as you've identified in (2).
However, where you do have both needs, some indirection is necessary. Whether this is the correct one is a different question.
Similarly, if you _must not_ leak timestamps for some real-world reason, (1) is an intrinsic requirement, consumers be damned.
the_mitsuhiko · 49m ago
If you must not leak timestamps then you also cannot really have timestamp ordering internally because you will happen to start leak that out in other ways through collection based endpoints.
JimDabell · 30m ago
Not necessarily. For instance, in situations where unprivileged users can only see single items but privileged users can see collections. But yeah, time-ordering leaks information to people who can see the collection.
miningape · 2h ago
This is interesting, but is almost something I'd rather have the DB handle for me - i.e. I can cast a UUIDv7 to "UUIDv4" (and vice versa) and I could use both in queries (with explicit syntax to annotate which kind is being used / expected)
chuckadams · 1h ago
I remember doing something similar, but I just used two columns, a public uuid, and a bigint primary key that wasn't exposed to the api (this was long before uuidv7). Lacked a lot of the conveniences of using uuid everywhere, but it still handled the use case of merging different DB dumps as long as PKs were stripped out first.
And maybe I misunderstand how the hashing works, but it seems if you're looking things up by the hashed uuid, you're still going to want two columns anyway.
My biggest issue w/ UUIDv7 is how challenging they are to visually diff when looking at a list. Having some sort of visual translation layer in psql that would render them with the random bits first while maintaining the time sorting underneath would be a major UX boost...
nine_k · 1h ago
Write a function that does that, use it in your queries. E.g. simple hex representation + string reversal should help. Or a reversed base64 representation for shorter output.
tracker1 · 1h ago
Interesting project... just out of curiosity, could you give something resembling a couple practical examples of the risk of exposing the time portion of a v7 UUID?
NortySpock · 1h ago
Suppose it's something where the user may be accused of doing something nefarious if a sequence or pattern of behavior is exposed.
- "Ex-spouse: I looked you up on a dating website, and your userID indicates it was created while you were at Tom's party where you swear nothing happened."
- "You say you are in XYZ timezone, but all your imageIDs (that are unique to the image upon creation) are timestamped at what would be 3am in your timezone)"
Granted, for individual messages that are near-real-time, or for transactions that need to be timestamped anyway, it's probably fine, but for user-account-creation or "evergreen" asset-creation, it could leak the time to a sufficiently curious individual (or an organized group that is doing data-trawling and cross-correlation)
thunderfork · 1h ago
Let's say you've got a system that collects medical data - like "store the results of the MRI right after it happens".
For analysis reasons, you want to share this dataset (e.g. for diagnostics on the machine) but first must strip it of potentially identifying information.
The uuidv7 timestamp could be used to re-identify the data through correlation - "I know this person got an MRI on this day, there's only one record with a matching datestamp, thus I know it's their MRI."
tracker1 · 48m ago
Fair enough, thanks... I've got more experience in education/elearning, banking and elections, all of which are likely to have separate timestamp records required anyway, so this kind of scenario didn't really jump out at me.
i'm curious, if you're doing single header, why not also do the stb-style IMPL block + definitions block such that you avoid the issues from accidentally including the header multiple times?
LeicaLatte · 58m ago
Mobile apps often sort by creation time in the UI (chat messages, activity feeds). Since clients only see the masked version, there might be a need to expose a separate timestamp field.
Although I finished it, I never quite published it properly for some reason, probably partly because I shelved the projects where I had been going to use it (I might unshelve one of them next year).
Well, I might as well share it, because it’s quite relevant here and interesting:
https://temp.chrismorgan.info/2025-09-17-tesid/
My notes on its construction, pros and cons are fairly detailed.
Maybe I’ll go back and publish it properly next year.
https://sqids.org/
(Ah, it’s fun reading through that document a bit again. A few things I’d need to update now, like the Hashids name, or in the UUID section how UUIDv7 is no longer a draft, and of sidenote 12 I moved to India and got married and so took a phone number ending in 65536, replacing my Australian 32768. :-) )
It’s lasted for three years of use and three years of disuse, and I hope to replace it with something utterly different (stylistically and technically) by the end of this year, though it may slip to next year. The replacement will be based on handwriting.
(I’m not a fan of handwriting fonts either. They’re never truly satisfying, though some with quite a few variants for each character get past the point of feeling transparently inauthentic. But when you can write and draw what you choose, where you choose, that’s liberating.)
I wanted to use it many times in project for non-iteratable IDs but never found it again.
How it works: the 48-bit timestamp is XOR-masked with a keyed SipHash-2-4 stream derived from the UUID’s random field. The random bits are preserved, the version flips between 7 (inside) and 4 (outside), and the RFC variant is kept. The mapping is injective: (ts, rand) → (encTS, rand). Decode is just encTS ⊕ mask, so round-trip is exact.
Security: SipHash is a PRF, so observing façades doesn’t leak the key. Wrong key = wrong timestamp. Rotation can be done with a key-ID outside the UUID.
Performance: one SipHash over 10 bytes + a couple of 48-bit loads/stores. Nanosecond overhead, header-only C11, no deps, allocation-free.
Tests: SipHash reference vectors, round-trip encode/decode, and version/variant invariants.
Curious to hear feedback!
UUIDs are often generated client-side. Am I right in thinking that this isn’t possible with this approach? Even if you let clients give you UUIDs and they gave them back the masked versions, wouldn't you be vulnerable to a client providing two UUIDs with different ts and the same rand? So this is only designed for when you are generating the UUIDv7s yourself?
Of course, UUIDv4 on the client side is not without risk either- needing to validate uniqueness and not re-use of some other ID. For the UUIDv7 on client side- you could add some sanity validation- but really I think it’s best avoided.
1. You implicitly take away someone else's hypothetical benefit of leveraging UUID v7, which is disappointing for any consumer of your API.
2. By storing the UUIDs differently on your API service from internally, you're going to make your life just a tiny bit harder because now you have to go through this indirection of conversion, and I'm not sure if this is worth it.
Usually if you see an id in your http logs you can simply search your database for that id. The v4 to v7 indirection creates a small inconvenience.
The mismatch may be resolved if this was available as a fully transparent database optimization.
1. Not leaking timestamp data (security/regulations)
2. Having easily time-sortable primary keys (DB performance/etc.)
If you don't have both of these needs, the tool is an unnecessary indirection, as you've identified in (2).
However, where you do have both needs, some indirection is necessary. Whether this is the correct one is a different question.
Similarly, if you _must not_ leak timestamps for some real-world reason, (1) is an intrinsic requirement, consumers be damned.
And maybe I misunderstand how the hashing works, but it seems if you're looking things up by the hashed uuid, you're still going to want two columns anyway.
timestamp + readability
- "Ex-spouse: I looked you up on a dating website, and your userID indicates it was created while you were at Tom's party where you swear nothing happened."
- "You say you are in XYZ timezone, but all your imageIDs (that are unique to the image upon creation) are timestamped at what would be 3am in your timezone)"
Granted, for individual messages that are near-real-time, or for transactions that need to be timestamped anyway, it's probably fine, but for user-account-creation or "evergreen" asset-creation, it could leak the time to a sufficiently curious individual (or an organized group that is doing data-trawling and cross-correlation)
For analysis reasons, you want to share this dataset (e.g. for diagnostics on the machine) but first must strip it of potentially identifying information.
The uuidv7 timestamp could be used to re-identify the data through correlation - "I know this person got an MRI on this day, there's only one record with a matching datestamp, thus I know it's their MRI."