Stanford Seminar - Optional Static Typing for Python

EE380: Computer Systems Colloquium Seminar
Optional Static Typing for Python
Speaker: Guido van Rossum, Dropbox & Python Software Foundation
Python is a dynamically typed language, and some of its appeal derives from this. Nevertheless, especially for large code bases, it would be nice if a compiler could find type errors before the code is even run. Optional static type checking promises exactly this, and over the past four years we have successfully introduced this feature into Python 3. This talk introduces the type system we've adopted and the syntax used for type annotations, some tips on how to get started with a large existing code base, and our experience using the 'mypy' type checker at Dropbox. The entire system is open source, and has also been adopted by other companies such as Lyft, Quora and Facebook.
About the Speaker:
Guido van Rossum is the creator of Python. Born in the Netherlands, he moved to the US following Python's success. He has worked at various startups and Google, and is currently a Principal Engineer at Dropbox.
For more information about this seminar and its speaker, you can visit ee380.stanford.edu/Abstracts/...
Support for the Stanford Colloquium on Computer Systems Seminar Series provided by the Stanford Computer Forum.
Colloquium on Computer Systems Seminar Series (EE380) presents the current research in design, implementation, analysis, and use of computer systems. Topics range from integrated circuits to operating systems and programming languages. It is free and open to the public, with new lectures each week.
Learn more: bit.ly/WinYX5
0:00 Introduction
0:33 Python's popularity
3:24 Python's type system
10:31 Why dynamic typing
11:42 Why static typing
13:26 Gradual typing
17:30 The mypy project
22:24 Other static type checkers for Python
23:46 A static type system for Python
27:51 Example in Python 2
28:55 Type inference
31:42 How to spell types
36:40 Escape hooks (unsafe)
39:44 Generics
40:55 Generic class example
43:36 Analyzing runtime type checks
48:03 Protocols (PEP 544) (2)
54:05 Stub files and typeshed
55:34 How to start on a large code base
59:06 Standardization effort

[@skepticalmind2260]

Great move, gradual or static typing does in fact reduce debugging time and improves code readability.

[@MichaelScharf]

Please, if your record speakers with slides, put the focus on slides: at 35:51 Guido shows a slide and talks about it, but it is not shown in the video, which is very annoying. The slide is shown at 36:38 for a fraction of a second.

[@kandhankuhan973]

Didn't know Protocol. Awesome!

[@stefmeden2820]

Great intro to optional static typing for Python, thx

[@AnthonyMcqueen1987]

This a man i want to meet in person

[@ManuelBTC21]

Protocols look very similar to Interfaces in golang. Very good choice!

[@meditationmusicvideos8924]

Thanks for your sir

[@haohuynhnhat3881]

It's number one now

[@MagnusAnand]

MyPy is a great!

[@shehab976]

23:05 laughed at facebook....only legends will understand🔥🤣

[@ManuelBTC21]

In response to 1:13:00 : python is not a chainsaw. If you have a bug in your homework.py, running it will not cut off your foot. It will throw an error at runtime, or it might even just work, because you didn't go down that codepath. The fact is that a type system is extra work. The payoff is worth it even for medium sized codebases, but there are plenty of use cases where "safety above all else" is the wrong approach.

[@AlexRodriguez-gb9ez]

I just saw Erlang's Joe Armstrong talk, and I think that way of handling errors may be better than static types hmm...

[@qwertykesthemwel7360]

Why not python in dutch?

[@dimitriosdesmos4699]

qwerty kesthemwel oh good god, the amount of phlegm one would need just to read 200 line of code in Dutch,......does that answer your question?

[@debvdo]

Too bad he is leaving...