Tyler Alicea
tyleralicea
Kinja Data Engineering
10/6/18
3:35 PM
Save

Open policy analysis talk was good, although it was pretty much the generic argument for the pros of open source software

10/6/18
3:20 PM
Save

The real disappointing thing, I think though, is that the room cleared out from the previous talk. People don’t want to talk about policy (even if it’s only a policy-adjacent topic!)

10/6/18
3:19 PM
Save

So far interesting talk on policy analysis and why reports would be better if the models were open source

10/6/18
3:08 PM
Save

We’ve now reached the part of the program where we watch our presenter install python packages he forgot to install

10/6/18
2:46 PM
Save

We’re getting some real criticism of that YouTube “Smash that bell and subscribe!!” culture from a talk presumably about writing your own Git client

10/6/18
2:44 PM
Save

Guys I’m not going to lie, this reviewing all the PyGotham cookies is really starting to get to me

10/6/18
2:08 PM
Save

Also ostensibly only being 3.7 is a short-term hurdle since our code isn’t 100% on the latest version of Python

10/6/18
2:07 PM
Save

Yeah that makes sense. I suppose if you’re a “purist” and want to use something out of the box it makes sense. Although my understanding was that the type and value validation wasn’t as good with dataclassses

10/6/18
2:02 PM
Save

@josh this might just be me not being imaginative, but can you think of any solid use cases where we would actually use dataclasses? They seem neat in theory

10/6/18
1:55 PM
Save

Fun fact, dataclasses have been backported to 3.6 and can be installed with pip