updated the atproto specs to reference dasl.ing, a simpler variation/subset of IPLD.
hope this is more accessible for folks digging in to atproto-from-scratch implementations. the core primitives in IPLD (DAG-CBOR, CIDs, CAR) are great, but the fully flexible/general system can be overwhelming
hope this is more accessible for folks digging in to atproto-from-scratch implementations. the core primitives in IPLD (DAG-CBOR, CIDs, CAR) are great, but the fully flexible/general system can be overwhelming
specs: update many IPLD references to DASL/DRISL references by bnewbold · Pull Request #496 · bluesky-social/atproto-website
The DASL specifications are simpler and align with what we use in atproto.
This updates most references to the IPLD docs to reference the DASL docs instead. It still discusses the lineage in some p...
github.com
February 6, 2026 at 8:14 PM
updated the atproto specs to reference dasl.ing, a simpler variation/subset of IPLD.
hope this is more accessible for folks digging in to atproto-from-scratch implementations. the core primitives in IPLD (DAG-CBOR, CIDs, CAR) are great, but the fully flexible/general system can be overwhelming
hope this is more accessible for folks digging in to atproto-from-scratch implementations. the core primitives in IPLD (DAG-CBOR, CIDs, CAR) are great, but the fully flexible/general system can be overwhelming
this interplanetary shit got hands
![~/projects/pegasus -> main ? ! —
1 ❯ dune test ipld
Testing `cid'.
This run has ID `QLWAUR1T'.
[OK] cid encoding 0 encode.
[OK] cid encoding 1 encode_empty_string.
[OK] cid decoding 0 decode.
[OK] cid decoding 1 decode_empty_string.
[OK] cid creation 0 create.
[OK] cid creation 1 create_empty_string.
Full test results in `~/projects/pegasus/_build/default/ipld/test/_build/_tests/cid'.
Test Successful in 0.001s. 6 tests run.
Testing `dag-cbor'.
This run has ID `1ZLBRR6Y'.
[OK] dag-cbor encoding 0 encode_primitives.
[OK] dag-cbor encoding 1 round_trip.
[OK] dag-cbor encoding 2 atproto_records.
[OK] dag-cbor encoding 3 invalid_numbers.
[OK] dag-cbor encoding 4 invalid_link_bytes.
[OK] dag-cbor encoding 5 decode_multiple.
Full test results in `~/projects/pegasus/_build/default/ipld/test/_build/_tests/dag-cbor'.
Test Successful in 0.001s. 6 tests run.](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:uu5axsmbm2or2dngy4gwchec/bafkreifkftca5glneqonxsfa4khrjk4xcwdqwupifa4jcrwcd44iam55yy)
July 23, 2025 at 12:40 AM
this interplanetary shit got hands
"Does Bluesky use IPFS?"
Kind of, but not exactly... remember, IPFS has many subsystems & is based around content addressing. @bsky.app utilizes content addressing & a subsystem called IPLD which can be implemented by itself!
Learn more about IPLD here:
Kind of, but not exactly... remember, IPFS has many subsystems & is based around content addressing. @bsky.app utilizes content addressing & a subsystem called IPLD which can be implemented by itself!
Learn more about IPLD here:
IPLD ♦ InterPlanetary Ecosystem Overview
ipld.io
May 3, 2023 at 2:00 PM
"Does Bluesky use IPFS?"
Kind of, but not exactly... remember, IPFS has many subsystems & is based around content addressing. @bsky.app utilizes content addressing & a subsystem called IPLD which can be implemented by itself!
Learn more about IPLD here:
Kind of, but not exactly... remember, IPFS has many subsystems & is based around content addressing. @bsky.app utilizes content addressing & a subsystem called IPLD which can be implemented by itself!
Learn more about IPLD here:
forgot to follow up about my Zig IPLD implementation - it works! I just cut v0.1.0 a few weeks ago with the release of Zig 0.14.
it features two parallel APIs for working with either arbitrary heap-allocated values, or encoding to/from statically-typed Zig types. supports dag-cbor and dag-json.
it features two parallel APIs for working with either arbitrary heap-allocated values, or encoding to/from statically-typed Zig types. supports dag-cbor and dag-json.
GitHub - nDimensional/zig-ipld
Contribute to nDimensional/zig-ipld development by creating an account on GitHub.
github.com
April 2, 2025 at 1:25 AM
forgot to follow up about my Zig IPLD implementation - it works! I just cut v0.1.0 a few weeks ago with the release of Zig 0.14.
it features two parallel APIs for working with either arbitrary heap-allocated values, or encoding to/from statically-typed Zig types. supports dag-cbor and dag-json.
it features two parallel APIs for working with either arbitrary heap-allocated values, or encoding to/from statically-typed Zig types. supports dag-cbor and dag-json.
If IPLD is portable data, UCAN is portable auth.
July 19, 2024 at 12:31 AM
If IPLD is portable data, UCAN is portable auth.
yeah I like dune, dune tests on ipld
July 23, 2025 at 9:20 PM
yeah I like dune, dune tests on ipld
@vmx.cx wrote up a quick tour of what happened in IPLD and multiformats this year:
🧱 New (much faster!) modular crates for Rust IPLD
🐦 Migration guide if you're still using libipld
✅ Rust multiformats (cid, multihash, multibase) production-ready
blog.ipfs.tech/2026-01-ipld...
🧱 New (much faster!) modular crates for Rust IPLD
🐦 Migration guide if you're still using libipld
✅ Rust multiformats (cid, multihash, multibase) production-ready
blog.ipfs.tech/2026-01-ipld...
IPLD 2025 Review: From Monoliths to Modules
The year that brought us modular Rust libraries, faster DAG-CBOR, stable multiformats, and a simpler on-ramp with DASL.
blog.ipfs.tech
January 22, 2026 at 2:17 PM
@vmx.cx wrote up a quick tour of what happened in IPLD and multiformats this year:
🧱 New (much faster!) modular crates for Rust IPLD
🐦 Migration guide if you're still using libipld
✅ Rust multiformats (cid, multihash, multibase) production-ready
blog.ipfs.tech/2026-01-ipld...
🧱 New (much faster!) modular crates for Rust IPLD
🐦 Migration guide if you're still using libipld
✅ Rust multiformats (cid, multihash, multibase) production-ready
blog.ipfs.tech/2026-01-ipld...
Hi, i co-invented IPLD! don't worry, IPLD is merkletrees, and bluesky is built on IPLD.
April 24, 2023 at 7:28 AM
Hi, i co-invented IPLD! don't worry, IPLD is merkletrees, and bluesky is built on IPLD.
is it okay to say DASL instead of IPLD now when talking about CID and CBOR usage in atproto
November 27, 2024 at 10:02 PM
is it okay to say DASL instead of IPLD now when talking about CID and CBOR usage in atproto
I think this is legit to call out!
STAR-lite is cool and solves a couple other paper cuts. but I think churning on low-level formats in the protocol spec itself probably isn't worth it?
I do think the ecosystem can+should migrate away from older flexible IPLD libs to stricter DASL/atproto libs
STAR-lite is cool and solves a couple other paper cuts. but I think churning on low-level formats in the protocol spec itself probably isn't worth it?
I do think the ecosystem can+should migrate away from older flexible IPLD libs to stricter DASL/atproto libs
September 21, 2026 at 8:26 PM
I think this is legit to call out!
STAR-lite is cool and solves a couple other paper cuts. but I think churning on low-level formats in the protocol spec itself probably isn't worth it?
I do think the ecosystem can+should migrate away from older flexible IPLD libs to stricter DASL/atproto libs
STAR-lite is cool and solves a couple other paper cuts. but I think churning on low-level formats in the protocol spec itself probably isn't worth it?
I do think the ecosystem can+should migrate away from older flexible IPLD libs to stricter DASL/atproto libs
Are you using IPLD in a cool way? Or do you know a project that is? If so, please tell me about it! It doesn't have to be fancy computer science magic — I'm interested in "plumbing that gets the job done."
For those who don't know IPLD: ipld.io.
For those who don't know IPLD: ipld.io.
IPLD - The data model of the content-addressable web
The data model of the content-addressable web. It allows us to treat all hash-linked data structures as subsets of a unified information space, unifying all data models that link data with hashes as i...
ipld.io
March 14, 2024 at 7:32 PM
Are you using IPLD in a cool way? Or do you know a project that is? If so, please tell me about it! It doesn't have to be fancy computer science magic — I'm interested in "plumbing that gets the job done."
For those who don't know IPLD: ipld.io.
For those who don't know IPLD: ipld.io.
this will encode/decode your data but doesn't do the fs parts. github.com/ipld/rust-ip...
GitHub - ipld/rust-ipld-core: This crate provides core types for interoperating with IPLD.
This crate provides core types for interoperating with IPLD. - ipld/rust-ipld-core
github.com
August 20, 2025 at 8:04 PM
this will encode/decode your data but doesn't do the fs parts. github.com/ipld/rust-ip...
holyshit, why do so many standards bodies and open standards have the worst fucking logos... IPLD, IPFS, Protocol Labs, W3C (until recently), OAuth, IETF... like bro.
why are there no designers in tech that can donate well designed shit?
no logo cloud on this website i guess.
why are there no designers in tech that can donate well designed shit?
no logo cloud on this website i guess.
wanted to take a stab at making my own version of the atmosphere informational website, not happy with the logo as far as its not representing the atmosphere yet. but a good placeholder for now.
(its literally just a hero right now) thoughts welcome, but not much to show yet.
(its literally just a hero right now) thoughts welcome, but not much to show yet.
April 25, 2026 at 11:27 AM
holyshit, why do so many standards bodies and open standards have the worst fucking logos... IPLD, IPFS, Protocol Labs, W3C (until recently), OAuth, IETF... like bro.
why are there no designers in tech that can donate well designed shit?
no logo cloud on this website i guess.
why are there no designers in tech that can donate well designed shit?
no logo cloud on this website i guess.
To everyone working on DAG-CBOR/DRISL implementations. There was a small update to that specs that shouldn't cause much problems, but is newsworthy: negative zero -0.0 is now forbidden.
I recommend serializing it with the bytes from `0.0`.
github.com/ipld/ipld/pu...
github.com/darobin/dasl...
I recommend serializing it with the bytes from `0.0`.
github.com/ipld/ipld/pu...
github.com/darobin/dasl...
fix: reject -0.0 by vmx · Pull Request #65 · ipld/serde_ipld_dagcbor
For IEEE 754 floats -0.0 is equal to 0.0. It makes sense to have only a single representation for the same numbers, hence always encode it as 0.0, i.e. 0x0000000000000000.
github.com
August 3, 2026 at 6:50 PM
To everyone working on DAG-CBOR/DRISL implementations. There was a small update to that specs that shouldn't cause much problems, but is newsworthy: negative zero -0.0 is now forbidden.
I recommend serializing it with the bytes from `0.0`.
github.com/ipld/ipld/pu...
github.com/darobin/dasl...
I recommend serializing it with the bytes from `0.0`.
github.com/ipld/ipld/pu...
github.com/darobin/dasl...
I got you github.com/DavidBuchana...
GitHub - DavidBuchanan314/dag-sqlite: DAG-SQLite is an SQLite-based represention of the IPLD Data Model
DAG-SQLite is an SQLite-based represention of the IPLD Data Model - DavidBuchanan314/dag-sqlite
github.com
October 8, 2025 at 11:32 PM
I got you github.com/DavidBuchana...
= Floats in AT Protocol =
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
Floats in ATProto - HackMD
Note: I was part of the IPLD team that created the specs/information that is linked from this FAQ.
hackmd.io
March 31, 2026 at 8:00 PM
= Floats in AT Protocol =
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
a modest proposal by @vmx.cx
hackmd.io/@vmx/floats-...
please share if you <3 #atprotocol and also numeric data types
#datamodels #atproto #whynofloats #float64 #ieee754 #atgeo
why alternative packages?
- proves that a protocol's specification is good enough for others to build on, a hard-to-implement protocol is not a good protocol
- atproto's reference packages relies on multiformats/ipld libraries, which is a dealbreaker if you intend to ship web apps (they're big!)
- proves that a protocol's specification is good enough for others to build on, a hard-to-implement protocol is not a good protocol
- atproto's reference packages relies on multiformats/ipld libraries, which is a dealbreaker if you intend to ship web apps (they're big!)
October 19, 2024 at 2:26 PM
why alternative packages?
- proves that a protocol's specification is good enough for others to build on, a hard-to-implement protocol is not a good protocol
- atproto's reference packages relies on multiformats/ipld libraries, which is a dealbreaker if you intend to ship web apps (they're big!)
- proves that a protocol's specification is good enough for others to build on, a hard-to-implement protocol is not a good protocol
- atproto's reference packages relies on multiformats/ipld libraries, which is a dealbreaker if you intend to ship web apps (they're big!)
October 21, 2024 at 6:57 AM
True ATProto nerds form IPLD Lexicons in their mind.
September 9, 2024 at 5:24 PM
True ATProto nerds form IPLD Lexicons in their mind.
"the spec" being violated is the IPLD DAG-CBOR spec
August 29, 2024 at 10:20 PM
"the spec" being violated is the IPLD DAG-CBOR spec
#ArtistFave15 - Nile Rodgers productions - alphabetical order
12 - Sheila (and) B. Devotion - Spacer
Another day, another banger and another song I feel like I’ve known all my life. The video, needless to say, is fantastique!
m.youtube.com/watch?v=Ipld...
12 - Sheila (and) B. Devotion - Spacer
Another day, another banger and another song I feel like I’ve known all my life. The video, needless to say, is fantastique!
m.youtube.com/watch?v=Ipld...
Sheila & B. Devotion - Spacer
YouTube video by 100% Chanson Française
m.youtube.com
March 12, 2026 at 2:02 PM
#ArtistFave15 - Nile Rodgers productions - alphabetical order
12 - Sheila (and) B. Devotion - Spacer
Another day, another banger and another song I feel like I’ve known all my life. The video, needless to say, is fantastique!
m.youtube.com/watch?v=Ipld...
12 - Sheila (and) B. Devotion - Spacer
Another day, another banger and another song I feel like I’ve known all my life. The video, needless to say, is fantastique!
m.youtube.com/watch?v=Ipld...
HAMTs & Other IPLD Data Structures - Rod Vagg
YouTube video by IPFS
youtu.be
March 27, 2025 at 9:07 AM
IPLD stands for Investigators Prosecuting Leftists Discoursing.
July 14, 2024 at 1:29 AM
IPLD stands for Investigators Prosecuting Leftists Discoursing.
Content-addressed archive (CAR) files are like ZIP files for content-addressed data. If you work with IPFS or IPLD, CAR files are handy for moving data archives around.
Here's the spec: ipld.io/specs/transp...
and a thread with some handy tools: bsky.app/profile/subo...
Here's the spec: ipld.io/specs/transp...
and a thread with some handy tools: bsky.app/profile/subo...
November 27, 2024 at 11:21 PM
Content-addressed archive (CAR) files are like ZIP files for content-addressed data. If you work with IPFS or IPLD, CAR files are handy for moving data archives around.
Here's the spec: ipld.io/specs/transp...
and a thread with some handy tools: bsky.app/profile/subo...
Here's the spec: ipld.io/specs/transp...
and a thread with some handy tools: bsky.app/profile/subo...