home.social

#datomic — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #datomic, aggregated by home.social.

fetched live
  1. Почему мы не написали ещё один Bad CaRMa

    «Bad CaRMa» — глава из Dreaming in Code Скотта Розенберга (каламбур на CRM и «карму») про CRM-систему Vision в компании Upstart. Архитектор задумал предельно гибкую схему: одна-единственная таблица DATA , куда сложили все 150+ бизнес-сущностей — 240+ колонок с именами вроде string82 и numeric31 , метаданные и данные вперемешку. Схему ведь больше «никогда не придётся менять». Практики на грани

    habr.com/ru/articles/1062550/

    #EAV #модель_данных #Bad_CaRMa #PostgreSQL #Datomic #память_агента #ИИагенты

  2. Почему мы не написали ещё один Bad CaRMa

    «Bad CaRMa» — глава из Dreaming in Code Скотта Розенберга (каламбур на CRM и «карму») про CRM-систему Vision в компании Upstart. Архитектор задумал предельно гибкую схему: одна-единственная таблица DATA , куда сложили все 150+ бизнес-сущностей — 240+ колонок с именами вроде string82 и numeric31 , метаданные и данные вперемешку. Схему ведь больше «никогда не придётся менять». Практики на грани

    habr.com/ru/articles/1062550/

    #EAV #модель_данных #Bad_CaRMa #PostgreSQL #Datomic #память_агента #ИИагенты

  3. Почему мы не написали ещё один Bad CaRMa

    «Bad CaRMa» — глава из Dreaming in Code Скотта Розенберга (каламбур на CRM и «карму») про CRM-систему Vision в компании Upstart. Архитектор задумал предельно гибкую схему: одна-единственная таблица DATA , куда сложили все 150+ бизнес-сущностей — 240+ колонок с именами вроде string82 и numeric31 , метаданные и данные вперемешку. Схему ведь больше «никогда не придётся менять». Практики на грани

    habr.com/ru/articles/1062550/

    #EAV #модель_данных #Bad_CaRMa #PostgreSQL #Datomic #память_агента #ИИагенты

  4. Has anyone done #bitemporal stuff with postgres 16? I understand SQL:2011 support has landed in postgres 18 but I'm hesitant about upgrading (being very conservative). I'm aware of github.com/scalegenius/pg_bite and github.com/xocolatl/periods but I'm no database expert and I feel like a baby engineer dipping my toes in. What do y'all recommend?

    (I don't want to use #xtdb or #datomic because I know even less about these and I want to be able to maintain what I build)

  5. Has anyone done #bitemporal stuff with postgres 16? I understand SQL:2011 support has landed in postgres 18 but I'm hesitant about upgrading (being very conservative). I'm aware of github.com/scalegenius/pg_bite and github.com/xocolatl/periods but I'm no database expert and I feel like a baby engineer dipping my toes in. What do y'all recommend?

    (I don't want to use #xtdb or #datomic because I know even less about these and I want to be able to maintain what I build)

  6. Has anyone done #bitemporal stuff with postgres 16? I understand SQL:2011 support has landed in postgres 18 but I'm hesitant about upgrading (being very conservative). I'm aware of github.com/scalegenius/pg_bite and github.com/xocolatl/periods but I'm no database expert and I feel like a baby engineer dipping my toes in. What do y'all recommend?

    (I don't want to use #xtdb or #datomic because I know even less about these and I want to be able to maintain what I build)

  7. Has anyone done #bitemporal stuff with postgres 16? I understand SQL:2011 support has landed in postgres 18 but I'm hesitant about upgrading (being very conservative). I'm aware of github.com/scalegenius/pg_bite and github.com/xocolatl/periods but I'm no database expert and I feel like a baby engineer dipping my toes in. What do y'all recommend?

    (I don't want to use #xtdb or #datomic because I know even less about these and I want to be able to maintain what I build)

  8. Has anyone done #bitemporal stuff with postgres 16? I understand SQL:2011 support has landed in postgres 18 but I'm hesitant about upgrading (being very conservative). I'm aware of github.com/scalegenius/pg_bite and github.com/xocolatl/periods but I'm no database expert and I feel like a baby engineer dipping my toes in. What do y'all recommend?

    (I don't want to use #xtdb or #datomic because I know even less about these and I want to be able to maintain what I build)

  9. Even though I mostly use #Datalog databases these days (mostly #Datomic), many #PostgreSQL tidbits make me (unreasonably?) happy. Like this one: “Aggregate first - join later”

    cybertec-postgresql.com/en/sup

  10. Even though I mostly use #Datalog databases these days (mostly #Datomic), many #PostgreSQL tidbits make me (unreasonably?) happy. Like this one: “Aggregate first - join later”

    cybertec-postgresql.com/en/sup

  11. Even though I mostly use databases these days (mostly ), many tidbits make me (unreasonably?) happy. Like this one: “Aggregate first - join later”

    cybertec-postgresql.com/en/sup

  12. Even though I mostly use #Datalog databases these days (mostly #Datomic), many #PostgreSQL tidbits make me (unreasonably?) happy. Like this one: “Aggregate first - join later”

    cybertec-postgresql.com/en/sup

  13. Even though I mostly use #Datalog databases these days (mostly #Datomic), many #PostgreSQL tidbits make me (unreasonably?) happy. Like this one: “Aggregate first - join later”

    cybertec-postgresql.com/en/sup

  14. Awesome to hear Davis Shepherd and Jonathan Indig from Netflix talk about how stable the Clojure / Datomic ecosystem has been over the past 10 years, especially in comparison to the number of breaking changes that other open-source infrastructure projects have gone through in the same amount of time.

    The talk is called ”A Decade on Datomic” for when videos are published

    #Clojure #Datomic #ClojureConj #ClojureConj2025

  15. Awesome to hear Davis Shepherd and Jonathan Indig from Netflix talk about how stable the Clojure / Datomic ecosystem has been over the past 10 years, especially in comparison to the number of breaking changes that other open-source infrastructure projects have gone through in the same amount of time.

    The talk is called ”A Decade on Datomic” for when videos are published

    #Clojure #Datomic #ClojureConj #ClojureConj2025

  16. Awesome to hear Davis Shepherd and Jonathan Indig from Netflix talk about how stable the Clojure / Datomic ecosystem has been over the past 10 years, especially in comparison to the number of breaking changes that other open-source infrastructure projects have gone through in the same amount of time.

    The talk is called ”A Decade on Datomic” for when videos are published

  17. Awesome to hear Davis Shepherd and Jonathan Indig from Netflix talk about how stable the Clojure / Datomic ecosystem has been over the past 10 years, especially in comparison to the number of breaking changes that other open-source infrastructure projects have gone through in the same amount of time.

    The talk is called ”A Decade on Datomic” for when videos are published

    #Clojure #Datomic #ClojureConj #ClojureConj2025

  18. Awesome to hear Davis Shepherd and Jonathan Indig from Netflix talk about how stable the Clojure / Datomic ecosystem has been over the past 10 years, especially in comparison to the number of breaking changes that other open-source infrastructure projects have gone through in the same amount of time.

    The talk is called ”A Decade on Datomic” for when videos are published

    #Clojure #Datomic #ClojureConj #ClojureConj2025

  19. Today's Google "AI overview" hallucination: "#Datomic between".

    No, there is no `:between` operator in Datomic. As a Google user, I need to waste time figuring out that this is a hallucination, and then scroll down to results that actually make sense.

    I'm increasingly fed up. I guess I'll just bite the bullet and switch to Kagi.

  20. Today's Google "AI overview" hallucination: "#Datomic between".

    No, there is no `:between` operator in Datomic. As a Google user, I need to waste time figuring out that this is a hallucination, and then scroll down to results that actually make sense.

    I'm increasingly fed up. I guess I'll just bite the bullet and switch to Kagi.

  21. Today's Google "AI overview" hallucination: "#Datomic between".

    No, there is no `:between` operator in Datomic. As a Google user, I need to waste time figuring out that this is a hallucination, and then scroll down to results that actually make sense.

    I'm increasingly fed up. I guess I'll just bite the bullet and switch to Kagi.

  22. Today's Google "AI overview" hallucination: "#Datomic between".

    No, there is no `:between` operator in Datomic. As a Google user, I need to waste time figuring out that this is a hallucination, and then scroll down to results that actually make sense.

    I'm increasingly fed up. I guess I'll just bite the bullet and switch to Kagi.

  23. Today's Google "AI overview" hallucination: "#Datomic between".

    No, there is no `:between` operator in Datomic. As a Google user, I need to waste time figuring out that this is a hallucination, and then scroll down to results that actually make sense.

    I'm increasingly fed up. I guess I'll just bite the bullet and switch to Kagi.

  24. New blog post! „No, really, you can’t branch #Datomic from the past (and what you can do instead)”

    #Clojure

    blog.danieljanus.pl/2025/04/22

  25. New blog post! „No, really, you can’t branch #Datomic from the past (and what you can do instead)”

    #Clojure

    blog.danieljanus.pl/2025/04/22

  26. New blog post! „No, really, you can’t branch #Datomic from the past (and what you can do instead)”

    #Clojure

    blog.danieljanus.pl/2025/04/22

  27. New blog post! „No, really, you can’t branch #Datomic from the past (and what you can do instead)”

    #Clojure

    blog.danieljanus.pl/2025/04/22

  28. New blog post! „No, really, you can’t branch #Datomic from the past (and what you can do instead)”

    #Clojure

    blog.danieljanus.pl/2025/04/22

  29. There are hornets in #Datomic that may sting you. Beware!

  30. There are hornets in #Datomic that may sting you. Beware!

  31. There are hornets in #Datomic that may sting you. Beware!

  32. There are hornets in #Datomic that may sting you. Beware!

  33. There are hornets in #Datomic that may sting you. Beware!

  34. Is it possible to make a re-frame #localfirst app, syncing #datomic on the back-end and #datascript on the front-end? How would you do that?

    #clojure

  35. Is it possible to make a re-frame app, syncing on the back-end and on the front-end? How would you do that?

  36. Is it possible to make a re-frame #localfirst app, syncing #datomic on the back-end and #datascript on the front-end? How would you do that?

    #clojure

  37. Is it possible to make a re-frame #localfirst app, syncing #datomic on the back-end and #datascript on the front-end? How would you do that?

    #clojure

  38. Me finally deciding I should try #datomic, going all the way through helloworld only to realize that the "datomic local" flavor doesn’t do stored functions.

    (╯°□°)╯︵ ┻━┻

    So to even try those I'd have to spin up a local postgres, deploy datomic transactor, and figure the pg schema.

    #xtdb looks like fun again.

  39. Me finally deciding I should try #datomic, going all the way through helloworld only to realize that the "datomic local" flavor doesn’t do stored functions.

    (╯°□°)╯︵ ┻━┻

    So to even try those I'd have to spin up a local postgres, deploy datomic transactor, and figure the pg schema.

    #xtdb looks like fun again.

  40. Me finally deciding I should try #datomic, going all the way through helloworld only to realize that the "datomic local" flavor doesn’t do stored functions.

    (╯°□°)╯︵ ┻━┻

    So to even try those I'd have to spin up a local postgres, deploy datomic transactor, and figure the pg schema.

    #xtdb looks like fun again.

  41. Me finally deciding I should try #datomic, going all the way through helloworld only to realize that the "datomic local" flavor doesn’t do stored functions.

    (╯°□°)╯︵ ┻━┻

    So to even try those I'd have to spin up a local postgres, deploy datomic transactor, and figure the pg schema.

    #xtdb looks like fun again.

  42. Me finally deciding I should try #datomic, going all the way through helloworld only to realize that the "datomic local" flavor doesn’t do stored functions.

    (╯°□°)╯︵ ┻━┻

    So to even try those I'd have to spin up a local postgres, deploy datomic transactor, and figure the pg schema.

    #xtdb looks like fun again.

  43. I have been playing with #datomic in #clojure, but it leaves me wondering how schema migrations (adding new entities) is done in production. What do you use?

  44. I have been playing with in , but it leaves me wondering how schema migrations (adding new entities) is done in production. What do you use?

  45. I have been playing with #datomic in #clojure, but it leaves me wondering how schema migrations (adding new entities) is done in production. What do you use?

  46. I have been playing with #datomic in #clojure, but it leaves me wondering how schema migrations (adding new entities) is done in production. What do you use?

  47. I have been playing with #datomic in #clojure, but it leaves me wondering how schema migrations (adding new entities) is done in production. What do you use?

  48. soooo close to getting my evil #datomic on #sqlite+#litestream idea running and I’m now running into an undebuggable fly.io error 😵‍💫😭

  49. soooo close to getting my evil #datomic on #sqlite+#litestream idea running and I’m now running into an undebuggable fly.io error 😵‍💫😭

  50. soooo close to getting my evil #datomic on #sqlite+#litestream idea running and I’m now running into an undebuggable fly.io error 😵‍💫😭

  51. soooo close to getting my evil #datomic on #sqlite+#litestream idea running and I’m now running into an undebuggable fly.io error 😵‍💫😭

  52. Jepsen (a small consultancy focused on #distributed #systems safety) released their analysis of #Datomic safety and correctness - and concluded:
    We found no behavior which violated Datomic’s core safety claims ... Indeed, we believe Datomic’s inter-transaction safety properties are stronger than promised. jepsen.io/analyses/datomic-pro

  53. Jepsen (a small consultancy focused on #distributed #systems safety) released their analysis of #Datomic safety and correctness - and concluded:
    We found no behavior which violated Datomic’s core safety claims ... Indeed, we believe Datomic’s inter-transaction safety properties are stronger than promised. jepsen.io/analyses/datomic-pro

  54. Jepsen (a small consultancy focused on #distributed #systems safety) released their analysis of #Datomic safety and correctness - and concluded:
    We found no behavior which violated Datomic’s core safety claims ... Indeed, we believe Datomic’s inter-transaction safety properties are stronger than promised. jepsen.io/analyses/datomic-pro

  55. For those of you who are Datomic users, and database users in general, I'd love to hear what you think about #Datomic's concurrent intra-transaction semantics. Do you struggle with traditional databases that let transactions observe intermediate state from earlier in the same transaction? Have you accidentally written transaction functions that assumed serial semantics? How has Datomic's model changed the way you design transactions?