PyCon Talks

There are tons of great PyCon talks on YouTube and I haven't even scratched the surface of what's out there but here are a few that struck me.

Development of Open-Source Communities

Keynote - EuroSciPy 2015 - Pieter Hintjens

Key points:

  1. Wikipedia is the model
  2. It has no competition!
  3. The model clearly works
  4. Build a good set of rules
  5. How to Organize Open-Source Projects
  6. Accept pull-requests that pass tests immediately
  7. License is important: GNU GPLv3 is a good choice
  8. Lets people go off and "fork" the code, it can be reincorporated later

Emacs + org-mode + Python

Wouldn't it be nice if you could see the exact code you used to generate a plot in a paper you published two years ago? Wouldn't it be nice to see the LaTeX you used as well? Wouldn't it be nice to see the Python code too? Wouldn't it be even nicer for all of this code to be in one place?

Emacs + org-mode + python in reproducible research; SciPy 2013 Presentation

My take

Jupyter offers similar functionality: markdown, LaTeX, Python, and your code is there for all to see. Emacs is a barrier to entry that I don't want to scale right now.

Also, did you see the presenter is running Windows? It can be done!

Raymond Hettinger - Beyond PEP 8

Raymond Hettinger - Beyond PEP 8 -- Best practices for beautiful intelligible code - PyCon 2015

Ten Points:

  1. PEP 8 - maximum line length of 79 characters
  2. The line width requirement in PEP 8 is annoying
  3. Rather go a couple characters over than "commit an atrocity"
  4. Recommend 4 space indent with 90ish line length
  5. "A foolish consistency is the hobgoblin of little minds"
  6. People skip right over this to the "rules"
  7. Don't put form over substance
  8. P vs. NP -- Pythonic vs Non-Pythonic
  9. Recognize non-pythonic code
  10. Don't be distracted by PEP 8

Keynote - Jacob Kaplan-Moss - Pycon 2015

Keynote - Jacob Kaplan-Moss - Pycon 2015

Ten Points:

  1. "I'm at best an average programmer"

  2. Most people are average and most things

  3. Talent Myth: There are two types of programmers: amazing programmers and crappy programmers

  4. The talent myth drives people out of tech

  5. What does "10x programmer" look like? Introduces dangerous bias!

  6. Programming isn't a "passion" or a "talent" but a collection of acquired skills

  7. Even for people who are in tech the talent myth drives people out

  8. "41% of women leave tech after 10 years, compared to 17% of men"

  9. "In 2003, nearly half of all women who held a computer science B.S. were not employed in a science, engineering, or technical (SET) job two years after graduation."

  10. "We'll know we've been really successful when there are a bunch of average women programmers." -- Lynn Root

  11. "We need to dismantle this myth and build a community that acknowledges that average is actually pretty awesome"

My take

Hi, I'm Rory, and I'm a mediocre programmer.

links

social