CS 101

Brian Warner - Magic Wormhole: Simple Secure File Transfer - PyCon 2016

Van Lindberg - Structured Data from Unstructured Text - PyCon 2016

Manuel Ebert - Putting 1 million new words into the dictionary - PyCon 2016

Jake Vanderplas - Statistics for Hackers - PyCon 2016

Wendy Grus - When is it good to be bad? Web scraping and data analysis of NHL penalties - PyCon 2016

IPython Notebook in Data Intensive Communities: Accelerating the process of Discovery - PyCon 2016

Sev Leonard - The Fellowship of the Data - PyCon 2016

They even put up their 3-hour paid tutorials for free online:

Tracy Teal - Data Carpentry: An Introduction to Python for Data Analysis and Visualization

PyData 101: Essential data science skills for every programmer, from data to model to visualization

Here’s an assignment for you, one that reminds me plenty of the system that Jer Thorpe et al built to arrange names on the WTC memorial fountain.

A museum living history program has a couple of different programs that they offer, most of which include 3 or 4 “characters”.

Staff are all trained into at least one character. You must be trained in a character to work it. Some characters are on overnight so they can’t realistically work the next day.

Staff have certain preferences. Some do not want 4 hour shifts. Some do not want overnight shifts. Everyone provides a calendar with their availability for the month.

The widget should use an algorithm to go through the calendar and assign shifts. The education director should be able to move people around and have the whole thing reset, setting up some kind of alert when other shifts he’s pinned are in conflict with some rule.

> I’m taking an actual Django Girls workshop this weekend and hoping I can clear some of my weird intellectual Django hurdles.  > > The first one being … how does virtualenv actually work?

I know (knew) that virtualenv creates an isolated python environment. But it feels weirdly inefficient. So what does it actually do? When I run python -m venv flask (Actually ... it gives me an error: "No moduel named venv". So I try python3 -m because a scan of my history suggests that has worked in the past. I get a new error:

> Error: Command '['/home/amanda/Public/flask/flask/bin/python3', '-Im', 'ensurepip', '--upgrade', '--default-pip']' returned non-zero exit status 1

This is what we call "yak shaving" except that I'm pretty sure I actually need this yak hair if I'm going to get anything done. I don't fully understand what is going on here, but Stack Overflow says to try adding a --without-pip flag, which works.

That creates a set of local directories (there's probably a better word for "drpopped them right where I ran the command") although ... I think ... something just clicked for me. Here's what clicked for me:

Every tutorial that I look at for Django or Flask starts by having you create a virtual environment and install a bunch of packages to it. And that means that all over your machine you've got these little venvs hanging outfull of duplicate packages. But the right way to go about this is probably somehting more like ... Create one venv, and then use it over and over for all kinds of applications that require it. Right?

My old strategy was to approve all cookies. Or reject most of them. What I want is to allow temporary session cookies and “has_js” tests, and to stay logged in to the things I’d like to stay logged into, but clear the rest out between sessions.

> git push dreamhost mona

On dreamhost, I have master checked out. I merge mona into that with

> git merge mona

The magic of git is supposed to be that you don’t need a central repository. It’s totally decentralized. That’s pretty cool, even though 90% of git projects are probably based around a central repository on Github. 

I have a couple of repos hosted on Gitorious, which is shutting down, which seems like a great opportunity to figure out this whole “decentralized network of remotes” thing. 

Starting with my collection of decks.

But ... I’m stuck. 

I can’t push to my remote because it isn’t bare and I don’t know what the Right Way™ to do this is.

I could create a separate, new, bare remote. And use that as my central repository. But then I have a central repository again. I could create branches. That seems like how you’re supposed to do, but the honest truth is that I don’t really understand branching.