Fork me on GitHub

@rhansen Here's an example of something that breaks if you expose internal ids:

👍 8

Two questions: 1) Is excision planned for datomic-cloud? Need to handle GDPR 😕 2) Is gc-storage performed automatically in datomic-cloud?


(Btw, I've been playing with datomic-cloud for about a week now and I love it. It's simplified a lot of my code, and I don't look forward to work on projects that use other dbs. A big high five to cognitect for this product)


1) we hear you! and 2) yes

👍 4

I’m in a situation where I’m looking at databases for storing a lot of academic publishing-related information. Metadata about articles, journals, books, reports, publishers, authors and so on. There will be blobs of PDFs with fulltexts of the articles, books, etc. as well, and XML versions of those same PDFs. Is Datomic a good fit for this type of data? (I expect the blobs will end up in storage outside of the database)


@henrik I think Datomic is a great fit for the metadata; that sort of multi-source integration is a very good use case; I would definitely store the blobs out of band (i.e. S3 or elsewhere) and keep a reference to them in Datomic (i.e. a URI or other content address)


Thank you @U05120CBV! The amount of data is likely to grow to approach the amount of digitalized academic publishing data available in the world (i.e., all data from the top 10 publishers to begin with, then the top 20 and so on). Still good?


probably, but i’d want to know more about scale in terms of # of datom


Would it be possible to do something like a Skype chat with yourself or someone else at Cognitect and talk through the specifics?


Sure. I’m traveling next week, but would the week after work? You can email me at <mailto:[email protected]|[email protected]> and we can set something up


when i pull a datom from :tx-data and read its attribute via :a, i get back a number rather than a keyword. why is that?


@sekao all schema definitions are just datoms themselves


try pulling the :db/ident of the number you got back (it’s an entity ID)


i.e. (d/pull (d/db conn) [:db/ident] (:a my-datom))


makes sense. and also you just blew my mind. thanks!

Daniel Hines16:07:45

Is there an easy way to write reports/dashboards against a datomic database?

Daniel Hines16:07:58

For a non-developer, that is.

Daniel Hines16:07:38

To clarify, my business analyst and database programmer colleagues are running into a bunch of problems which I think Datomic solves beautifully, but they depend on tools like SSRS and PowerBI to write lots and lots of reports.


We’ve used another project from the Clojure landscape: Unfortunately, they don’t support Datomic (yet). I would love for them to, though. I’m sure Datomic can be set up to export continuously to one of the supported databases, however.


On a couple of my projects we just use transaction log/onyx to stream stuff in to RedShift, etc. There’s not much direct support from your typical BI tools for datomic


anyone know if it's possible to get the datomic ions tutorial working on windows (and thus without the clojure command line tools)?


You will need at least Java installed and Clojure’s jar…. clojure cli is just a script that calls Java with a well-crafted classpath