# Database Poll

**URL:** <https://discourse.julialang.org/t/database-poll/24715>\
**Category:** Community\
**Created:** [May 29, 2019, 5:47am UTC](https://discourse.julialang.org/t/database-poll/24715 "2019-05-29T05:47:24Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![tk3369](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/tk3369/32/2824_2.png) [@tk3369](https://discourse.julialang.org/u/tk3369)\
**Post date:** [May 29, 2019, 5:47am UTC](https://discourse.julialang.org/t/database-poll/24715/1 "2019-05-29T05:47:24Z")

</div>

Given this [high traffic topic about the state of database libraries](https://discourse.julialang.org/t/highly-concerned-with-the-state-of-database-libraries/11877/106), I think it may be time to run a quick poll.

1. Are you interested in accessing database from Julia programs?

_Poll ([view on site](https://discourse.julialang.org/t/database-poll/24715/1))_

1. What kind of databases do you need to access?

_Poll ([view on site](https://discourse.julialang.org/t/database-poll/24715/1))_

---

<div class="post-metadata">

**Author:** ![fborda](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/fborda/32/8654_2.png) [@fborda](https://discourse.julialang.org/u/fborda)\
**Post date:** [May 29, 2019, 12:49pm UTC](https://discourse.julialang.org/t/database-poll/24715/2 "2019-05-29T12:49:04Z")

</div>

Good and easy database support is really important for expanding into the services (Web Services, ETLs, random microservice stuff) which is one of the main areas people use dynamic languages (and one that uses what Julia is good at, such as ML and Statistics) and one of the areas that suffers as well with the two-language problem. Elixir for example created an Erlang (another previously highly focused tool) resurgence by focusing on a comprehensive and easy to use database support (Ecto) and Web Framework (Phoenix).

Plus, having a way to pull data from a database to a dataframe (with any kind of preprocessing through SQL), processing the data using any tool and inserting/updating the result back to the database just as easy as you would now with a CSV may end up convincing people of this workflow (even if you can indeed do that without a database).

Though I agree with the old topic that it’s up to people who want it now to start building, and hopefully reach a critical mass eventually when support will sustain itself.