On Mon, Sep 9, 2019 at 8:48 AM Neil Van Dyke <n...@neilvandyke.org> wrote:
> If someone is up to doing such a startup: find a > startup CEO who can get funding, start coding with the copious > information and community support available for Racket... and then feed > back your success story, as well as feed back useful generic open source > modules you'll probably have to write to get to launch. (And if you can > get enough funding for that startup, I could help with everything except > the funding&sales schmoozing.) > One other similar, but maybe easier to achieve idea, is that Racketeers find opportunities to create internal tools in their companies, prototype one, and try to convince others that Racket can save plenty of cost for that specific task. If this is done successfully, then the next step is to present the tool at related conferences and how it made everything easy and saved time and cost. I currently work at a team which creates a distributed SQL database ( https://github.com/citusdata/citus). Ideas that I have: - Improve the way we write tests. We have different test suites, but the main one is just to run a plain SQL file and compare it with the expected output file. This is not a very good to write tests, e.g. (1) What we are testing for is not clear (command output? not erroring out? rolling back successfully? not getting involved in distributed deadlock?) (2) There is no flow control, so if we want to try combination of few things, we need to write all of the combinations manually, ... - Our HA tool at https://github.com/citusdata/pg_auto_failover is heavily shell based, and writing shell scripts is not very joyful. A racket/... wrapper on top of the CLI might help with writing tools that use the HA tool. - Maybe we can find opportunities to use Cosette ( http://cosette.cs.washington.edu/, a Racket project) to find bugs in or prove correctness of PostgreSQL. ... Two barriers I have: 1. Convincing others that Racket is a better tool for these tasks than Python, Ruby, ... 2. I am currently full-time writing C code for our database engine, so doing something useful will need good planning. If for example we can make an outstanding testing framework, one can present it at postgres conferences, and then other postgres companies might start using Racket. As a side note, we have a Lisper in our team who used CL to create a widely used postgres tool: https://github.com/dimitri/pgloader, and another team member who used Haskell for another widely used tool: https://github.com/PostgREST/postgrest. None of these projects is being sponsored by our company, and they were personal projects AFAIK. -- Hadi -- You received this message because you are subscribed to the Google Groups "Racket Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to racket-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/racket-users/CAK%3D1%3DWrqL%2BO%3DSRe3O0joA7CjQh%3D3t99CkLXyp0j-q%2BQgLcJanQ%40mail.gmail.com.
