3 ms·

> In general, you don't really get to compact tombstones meaningfully without consensus so you really are pushing at least remnants of the entire log around to

by josephg 8mo ago

> In general, you don't really get to compact tombstones meaningfully without consensus so you really are pushing at least remnants of the entire log around to each client indefinitely. This is often much less of a problem in practice than people think. Git repositories also grow without bound, but nobody seems to really notice or care. For diamond types (my library), I lz4 compress the text content itself within the .dt files. The metadata is so small that in many cases the space savings from lz4 compression makes the resulting .dt file - including full change history - smaller than the document on disk. If anyone really cares about this problem, there are a couple other approaches: 1. In our Eg-walker algorithm, we don't use stable guids to identify insert positions. Thats where other algorithms go wrong, because they need to store these guids indefinitely in case someone references them. For eg-walker, we just store relative positions. This makes the algorithm work more like git, where you can do shallow clones. And everything works, until you want to merge a branch which was split off earlier than your clone point. Then you should be able to download the earlier edits back to the common branch point and merge. To support merging, you also only need to store the metadata of earlier edits. You don't need the inserted content. The metadata is usually tiny (~1-4 bits per keystroke for text is common with compression.) 2. Mike Toomim's Antimatter algorithm is a distributed consensus protocol which can detect when its safe to purge the metadata for old changes. It works even in the presence of partitions in the network. (Its conservative by default - if a partition happens, the network will keep metadata around until that peer comes back, or until some timeout.) > The same few guys (Martin Kleppman, Kevin Jahns, Joseph Gentle, probably others) pop up all over the more recent optimisations. Alex Good is another. He's behind a lot of the huge improvements to automerge over the last few years. He's wicked smart.

aguacaterojo 8mo ago
The man himself. Yeah agreed you guys have solved it. It is more a misfired crud dev instinct for me that sees it as wasteful. Just a different paradigm and not big in practice. I've got eg-walker & Diamond Types in my reading/youtube backlog. Diamond Types went further down the backlog because of "wip" on the repo! I will look into Antimatter too.
- josephg 8mo ago
  > It is more a misfired crud dev instinct for me that sees it as wasteful. You're not alone. Lots of people bring this up. And it can be a real problem for some applications, if they store ephemeral data (like mouse cursor positions) within the CRDT itself.
rudi-c 8mo ago
re: git repositories That's partly because repositories rarely need to be cloned in their entirety. As such, even when you need to do it and it's a couple hundreds mb taking a few minutes, it's tolerated. In situations where a document needs to be cold loaded often, the size of the document is felt more acutely. Figma has a notion of GC-ing tombstones. But the tombstones in questions aren't even something that get created in regular editing, it happens in a much more narrow case having to do with local copies of shared components. Even that caused problems for a small portion of files -- but if a file got that large, it was also likely to be important.
- josephg 8mo ago
  > even when you need to do it and it's a couple hundreds mb taking a few minutes, it's tolerated. Well written CRDTs should grow slower than git repositories. > In situations where a document needs to be cold loaded often, the size of the document is felt more acutely. With eg-walker (and similar algorithms) you can usually just load and store the native document snapshot at the current point-in-time, and work with that. And by that I mean, just the actual json or text or whatever that the application is interacting with. Many current apps will first download an entire automerge document (with full history), then using the automerge API to fetch the data. Instead, using eg-walker and similar approaches, you can just download 2 things: - Current state (eg a single string for a text file, or raw JSON for other kinds of data) alongside the current version - History. This can be as detailed & go as far back as you want. If you download the full history (with data), you can reconstruct the document at any point in time. If you only download the metadata, you can't go back in time. But you can merge changes. You need history to merge changes. But you only need access to history metadata, and only as far back as the fork point with what you're merging. If you're working online (eg figma), you can just download history lazily. For client/server editing, in most cases you don't need any history at all. You can just fetch the current document snapshot (eg a text or json object). And users can start editing immediately. It only gets fancy when you try to merge concurrent changes. But thats quite rare in practice.
  - rudi-c 8mo ago
    > If you're working online (eg figma), you can just download history lazily. You can download the history lazily, that's a special case of incrementally loading the document. If the history is only history, then sure. But I was understanding that we were talking about something that may need to be referenced but can eventually be GCed (e.g. tombstones-style). Then lazy loading makes user operations that need to reference that data go from synchronous operations to asynchronous operations. This is a massive jump in complexity. Incrementally loading data is not the hard part. The hard part is how product features need to be built on the assumption that the data is incrementally loaded. Something that not all collaborative apps will care about, but Figma certainly did, is the concept that the client may go offline with unsynced changes for an indefinite amount of time. So the tombstones may need to be referenced by new-looking old changes, which increases the likelihood of hitting "almost-never" edge cases by quite a bit.

by josephg 8mo ago

aguacaterojo 8mo ago

The man himself. Yeah agreed you guys have solved it. It is more a misfired crud dev instinct for me that sees it as wasteful. Just a different paradigm and not big in practice. I've got eg-walker & Diamond Types in my reading/youtube backlog. Diamond Types went further down the backlog because of "wip" on the repo! I will look into Antimatter too.

josephg 8mo ago
> It is more a misfired crud dev instinct for me that sees it as wasteful. You're not alone. Lots of people bring this up. And it can be a real problem for some applications, if they store ephemeral data (like mouse cursor positions) within the CRDT itself.

rudi-c 8mo ago

re: git repositories That's partly because repositories rarely need to be cloned in their entirety. As such, even when you need to do it and it's a couple hundreds mb taking a few minutes, it's tolerated. In situations where a document needs to be cold loaded often, the size of the document is felt more acutely. Figma has a notion of GC-ing tombstones. But the tombstones in questions aren't even something that get created in regular editing, it happens in a much more narrow case having to do with local copies of shared components. Even that caused problems for a small portion of files -- but if a file got that large, it was also likely to be important.

josephg 8mo ago
> even when you need to do it and it's a couple hundreds mb taking a few minutes, it's tolerated. Well written CRDTs should grow slower than git repositories. > In situations where a document needs to be cold loaded often, the size of the document is felt more acutely. With eg-walker (and similar algorithms) you can usually just load and store the native document snapshot at the current point-in-time, and work with that. And by that I mean, just the actual json or text or whatever that the application is interacting with. Many current apps will first download an entire automerge document (with full history), then using the automerge API to fetch the data. Instead, using eg-walker and similar approaches, you can just download 2 things: - Current state (eg a single string for a text file, or raw JSON for other kinds of data) alongside the current version - History. This can be as detailed & go as far back as you want. If you download the full history (with data), you can reconstruct the document at any point in time. If you only download the metadata, you can't go back in time. But you can merge changes. You need history to merge changes. But you only need access to history metadata, and only as far back as the fork point with what you're merging. If you're working online (eg figma), you can just download history lazily. For client/server editing, in most cases you don't need any history at all. You can just fetch the current document snapshot (eg a text or json object). And users can start editing immediately. It only gets fancy when you try to merge concurrent changes. But thats quite rare in practice.
- rudi-c 8mo ago
  > If you're working online (eg figma), you can just download history lazily. You can download the history lazily, that's a special case of incrementally loading the document. If the history is only history, then sure. But I was understanding that we were talking about something that may need to be referenced but can eventually be GCed (e.g. tombstones-style). Then lazy loading makes user operations that need to reference that data go from synchronous operations to asynchronous operations. This is a massive jump in complexity. Incrementally loading data is not the hard part. The hard part is how product features need to be built on the assumption that the data is incrementally loaded. Something that not all collaborative apps will care about, but Figma certainly did, is the concept that the client may go offline with unsynced changes for an indefinite amount of time. So the tombstones may need to be referenced by new-looking old changes, which increases the likelihood of hitting "almost-never" edge cases by quite a bit.