Hacker News new | past | comments | ask | show | jobs | submit | yamafaktory's comments login

What triggers the initial willingness to implement something with hypergraphs was indeed the Wolfram physics project! https://github.com/yamafaktory/hypergraph/discussions/11#dis...


Social media for example.


Maybe companies like Facebook modeled their "social graph" with hypergraphs?


Great, thanks! It does seem like they might be using something like this for social graph. Somebody gave neat explanation on Quora [1]

[1] https://www.quora.com/How-has-Facebook-modeled-its-social-gr...


Thanks a ton for the feedback!

You're right maybe the wording is wrong: query more than manipulate. By the way, querying like jq might be an interesting new feature. Don't know exactly in which direction it should go / what it should do / how exactly it should look like (grammar)

N.B.: I have another project called jql which is actually a simpler - and faster - jq made with Rust https://github.com/yamafaktory/jql


Wow that’s awesome you actually have a jq like implementation. Hmmm since it’s a graph maybe you could go with something like openCypher [0] or GQL [1] (possibly gremlin too) or maybe you could even query it with GraphQL (although GraphQL only traverses down in a tree fashion)? Although some sort of simpler jq like syntax might be able to make more sense. It would need to be able to branch out somehow. I think even an emmet [2] like syntax might work well.

[0]: http://www.jcc.com/resources/jcc-blogs-menu/blog-database-dr...

[1]: https://www.gqlstandards.org/gql-blogs

[2]: https://docs.emmet.io/cheat-sheet/


Thanks for your answer, that's exactly the vision I have for jql: keep it simple and easy to use. jq is super powerful but most of its features are way beyond what should be - IMHO - a small CLI tool to query JSON when what you need is just making a selection / filtering the data. Both tools also have slightly different selection grammars.


How big companies are dealing with such a case? At my company, we have a mandatory audit step in the CI. Of course, one can still bypass it by dropping the step in order to deploy but I'm curious to know how others deal with that.


Hi, awesome work! Love the idea of using S-expression for the query syntax! Funny fact: we share the same project name - even though the core language and the query syntax are different - see https://github.com/yamafaktory/jql.


Hey, looks really nice! I see we had similar goals of simplicity in mind, though with very different approaches.

Sorry for the name duplication! I thought I'm sharing the name only with the jira API jql.

EDIT: I'd love to see some benchmarks. I expect your jql to greatly outperform mine and jq, considering it's written in Rust!


No worries regarding the naming, it's what makes open-source friendly! You can see some benches here e.g. https://travis-ci.org/yamafaktory/jql/jobs/618374730#fold-st... but those actually more like regression tests that are triggered for pull requests to compare them against master (and I should use Github actions probably here now instead of Travis). Thanks for your reply!


Link to the Github project: https://github.com/mit-pdos/noria.


Starting March 15th, 2019 Slack will drop it https://get.slack.help/hc/en-us/articles/115002037526-Suppor...


Visually very pleasant! Full width + responsive


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: