1
1
mirror of https://github.com/github/semantic.git synced 2024-11-29 02:44:36 +03:00
semantic/weekly/2016-08-08.md
2016-08-08 10:26:35 -05:00

2.0 KiB

What went well?

@robrix: Interpreter stuff was effective in cleaning up / clarifying some previously messy code. diffing algorithm is now a little scripting language.

@joshvera: Javascript syntaxes went well. Self assessments were more streamlined.

@rewinfrey: Self assessments were simple. Pairing with Rob went well (first time really driving).

What didn't go well?

@robrix: Ancillary tasks not going so well (patent process).

@joshvera: Self assessments took time away from working on minor diff summary tasks. Getting C into syntax is going to be difficult.

@rewinfrey: Recognizing that there is a better way to handle the effects in the test generator in the types, but not quite there skillwise to implement it (but also recognize I will get there :) ).

What did you learn?

@robrix: Cemented understanding of free monad and interpreters. learned about type families and type classes for derivative parsers. keep things in single module and break out as needed to avoid orphan instances.

@joshvera: Learned about tradeoffs of performance of free monads we currently use, and learned more about effects. over the weekend read about comonads as pointed spaces.

@rewinfrey: Learned about how to model effects in the type system (via free and cofree), and learned foundational Yesod (routing, controllers, models / migrations, etc.).

Other things?

@robrix: Re: Staff shipping in two weeks -- how do we feel? Our performance issues are still a concern, but responsiveness is being looked at outside the team.

@joshvera: Feeling okay with staff shipping, but performance should be addressed before staff shipping. This seems possible.

@robrix: Please keep in mind the following:

  • If we can't meet the deadline, can we punt?
  • No death marches please.

@joshvera: Let's reconvene later this week and update status of deliverables.

@robrix: Goal is to get to smaller, more regular releases. Right now that is hard, because the initial staffship of Diff Summaries frontloads a lot of functionality that future releases will benefit from.