Skip to content

Latest commit

 

History

History
111 lines (67 loc) · 4 KB

File metadata and controls

111 lines (67 loc) · 4 KB

Lecture 4


  • Dashboard
  • Ensure you can load data

Browser Developer Tools

Showing Chrome, but it's similar in other browsers.


HTTP


Remember when we were working with APIs?

requests.get(url)


Visit http://127.0.0.1:5000/echo.


curl -I http://127.0.0.1:5000/echo
curl -X POST http://127.0.0.1:5000/echo
curl -X POST -d "hello" http://127.0.0.1:5000/echo
curl -i http://127.0.0.1:5000/echo > response.txt

curl 'http://127.0.0.1:5000/divide?numerator=5&denominator=2'
curl 'http://127.0.0.1:5000/divide?denominator=2&numerator=5'
curl 'http://127.0.0.1:5000/divide?numerator=5&denominator=0'
curl 'http://127.0.0.1:5000/divide?numerator=5&denominator=abc'
curl 'http://127.0.0.1:5000/divide?numerator=5'

Like cURL, but for downloading files.



Reading response questions

Design responsibility

  • It also raises an important question that the readings only briefly touch on: if users are largely unaware of how files actually work, who bears responsibility for data loss, privacy breaches, or misuse---users, designers, or institutions?

AI & the modern relevance of CLI

  • Elements of the command line: the prompt, followed by a command, and an option reminds me of talking to an AI chatbot. Is this the right way of understanding how it works, or are there differences between the two? I guess one key difference is perhaps that your computer can't "talk back to you", but instead, runs your command, unlike a chatbot.
  • The video mentioned that early interfaces prioritized making things easy for computers, not humans. Will AI change this? When we can talk to computers in natural language, will command lines become like Latin---something only specialists need?
  • Does the efficiency of the CLI still hold true for professionals whose main focus is policy analysis rather than software engineering?

Week 4

Virtual environments

One thing that I'm still unsure of is when to use the global environment versus activating my virtual environment. I understand that virtual environments should be created and activated for each project - but how many projects is too many? What if I forget which virtual environment is for which project? Do I need to keep track of my Notes app?


Jupyter Notebook vs. Streamlit

The power of Streamlit, which transforms python scripts into interatctive apps in a second, is amazing [...] However, since Jupyter notebook has similar functions, which is Python-friendly and shows output immediately, I wonder which tools we should choose, what kind of differences are between these two.


Internet architecture

  • Overall, I would benefit from a quick review of the whole architecture laid out in chapters 3 and 5, even though well illustrated it was a decent amount of new information.
  • I am unsure what it is referring to by "packets".