2012-12-23

This is a miscellaneous collection of notes about teaching programming to non-programmers.

Existing Resources

The first link is well-regarded, despite its intimidating title.

• Zed Shaw's Learn Python the Hard Way
• Software Carpentry
• Khan Academy

Introduction

• Computers: phones, netbooks, tablets, microwaves!

Building blocks

• Variables
• Loops
• Conditionals
• Math
• Functions

Variables

age = 23
name = 'David'

Loops

• For-each
• While
• C-style for: for (i = 0; i < 5; i++) {}
• Do-while (aside)

Conditional expressions

• If
• Switch-case

Data types

• int vs. float vs. char
• Arrays, lists, and strings
• C: terminate with \0, instead of passing around a count
• Character sets and encodings: ASCII, Unicode

Bits

• 0 and 1
• 32-bit and 64-bit computers (architectures)
• Year 2038 problem, vs. Y2K bug

Languages

• Natural languages vs. formal languages
• Aside: Turing completeness, and equivalent power of expression
• Aside: Noam Chomsky
• Machine code, vs. assembly, vs. higher-level languages
• Tradeoffs of different languages
• Larry Wall's State of the Onion: some things are convenient to say in a language
• "Scripting languages" vs. "programming languages"
• "Coding" vs. "programming"
• Web vs. desktop vs. embedded: languages and programming (tradeoffs)
• Standard libraries: "batteries included"
• Compilers and interpreters: translation between languages

Don't Repeat Yourself

• Functions, vs. copy-pasting code
• Classes: associate functions with structured data and/or state
• Syntactic sugar

Security

• SSL and HTTPS
• Crypto: RSA encryption and big prime numbers

• Vulnerabilities and exploits

  • Debian OpenSSL + uninitialized memory (aside: open source security)
• 0-days, and responsible disclosure
• What is a bug?
• Web browsers have limited filesystem access

Portability

• Virtual machines
• Operating systems
• Assembly languages, architectures, and compilers
• The web: abstract away the OS, but variation between browsers

Standards

• Formal vs. informal conventions
• ANSI, ISO
• Language definitions: HTML, Python
• Separation of "language specification" from "language implementation".
• CPython2 and PyPy; Python 3; Jython; etc.

Servers

• What is a server?
• World wide web: HTTP requests/responses as plain text
• Networks and the internet
• DNS: the numbers underneath the internet
• Protocols: HTTP, SMTP, FTP, XMPP, etc.

Graphics

• GUIs and toolkits
• Color encodings; e.g., RGBA
• Pixels: drawing to the screen
• GPUs: 3D graphics, scientific computing, password cracking

Text vs. binary

• File formats, magic numbers
• .txt, .doc, .docx, .exe, XML, JSON, PDF, …
• Comparison: Unix vs. Windows
• Linux kernel binary blobs (e.g., wireless drivers)

Data interchange, file formats

• Binary compatibility
• Windows vs. Linux vs. Mac
• 32-bit vs. 64-bit for a single OS

Coding style

• Unix Philosophy: combine small, simple tools (pipes and input/output)
• Monolithic, universal: Windows registry

Databases

• What is a database?
• Data integrity, normalization, foreign keys (Don't repeat yourself)
• Relational, document, and graph databases: model your data
• Flat file databases
• Index to avoid linear searches: data structures
• Store data sets that are too big for RAM
• JSTOR
• Database servers like MySQL and SQL Server: SQL standard and implementations

Code quality

• Readibility
• Maintainability
• Refactoring vs. rewriting
• Technical Debt

Scientific computing

• Emphasis on get-things-done scripts, vs. "30-year enterprise"
• Python, C, C++, Perl, shell, Java?

Multi-tasking computers

• Multiple cores and processors
• Threading and multiprocessing
• Context switching for CPUs
• Vector and array operations: fast, optimized in hardware

Different kinds of memory

• RAM is faster, but more expensive, than hard disks
• Flash drives vs. hard disks
• Storing and representing objects that are too big to fit into RAM (or HD)

Algorithms, Data Structures

• The math part of computer science
• Real consequences for code: reduce CPU time or memory usage
• Big-O notation
• Often used heavily in interviews for developer jobs

Computational complexity => actual performance

• Premature optimization, and diminishing returns
• "Developer time" is a resource like CPU time, RAM, and hard drive space
• Like Costco's obsessive optimization of toilet paper production costs.
• Right tool for the job: when you have an optimization hammer, everything looks like a nail.

How to learn on your own

• Know how to ask search engines like Google
  • site:google.com
  • this -that
  • "Once upon a time"
• Stack Overflow
• Choose wisely when entering a rabbit hole

Piracy and privacy

• Cute title, don't remember what my point was going to be

The Cloud: buzzwords

Ads pay for free services like Google and Facebook

• Targeted advertisements
• Tracking cookies
• Linking user profile information across services: Google Drive, Google Search, Gmail?

Abstractions

• Easy to understand, and to use
• Hide details that people don't want to know
• Perl saying: make easy things easy, and hard things possible.

Impedance mismatch; i.e., square peg in a round hole

• MATLAB for GUIs
• Strings in Fortran?

iPhone and Android apps, and app development

Making your own website

• Hosting
• HTML/CSS files vs. WYSIWYG editors

LaTeX: nice-looking typography

• Resumes
• Scientific papers

Domain-Specific Languages: slang, jargon, and concise representations

Embedded languages: Microsoft Excel's VBA (Visual Basic for Applications)

Filesystems

• Windows C drive, D drive
• Unix mount points
• Directory structure, and filesystem hierarchy/tree
• NTFS, FAT, ext3, ext4, ZFS

Computer literacy

• opening files
• Command line arguments
• Running out of RAM or hard drive space; "pegging" CPU
• Task manager
  • Monitor resource usage
  • Kill processes
  • Start/stop services
• Windows fundamentals
• Web/browser fundamentals
• Email: clients and servers
• Microsoft Word and Excel
  • The basics: cut-copy-paste, spellcheck, test formatting
  • Intermediate: outlines, keyboard shortcuts
• Networking: connect to wifi

Intellectual property and software patents: technical and legal issues

• "Downloading"
• Torrenting
• MegaUpload
• Napster and Limewire

OS services, daemons, and local servers

Software, hardware, firmware

Back up data to external media: DropBox, Google Drive?

User permissions

• read/write/execute
• administrator privileges
• authentication vs. authorization
• Unix vs. Windows

Multi-user vs. multi-process

Automation: re-run experiments "for free" (simulations)

• Double-check complicated computations with one click

Separation of concerns

• Model-view-controller
• Separate content from presentation (HTML/CSS, vs. JS)

Programming paradigms

• Procedural programming
  • Linear instructions (recipes)
  • Step 1. Do this; Step 2. Do that; …
• Object-oriented programming: associate functions with data, and call them "methods"
• Functional programming: combine little pieces together - lambda calculus?

Automated tests: unit, integration, usability tests

Learning and using legacy tools, and legacy code

Web browsers

• HTTP status codes: 200, 301, 302, 400, 401, 403, 404, 500, etc.

  • Magic numbers, conventions, and abstraction
• What information does your HTTP client send?
  • User Agent string
  • HTTP Cookies
  • Your IP iddress, and MAC address on smartphones?
• Browser cache: save bandwidth, HTTP 304

Routers: packet sniffing, wireless passphrases, MAC filters

The network stack

• Wires and switches
• TCP/IP
• WWW, compare web to internet
• DNS
• HTTP text requests/responses
• Servers
• Browsers as specialized HTTP clients

Operating systems, kernels, and userland

• Unix, UNIX, Linux, GNU/Linux
• Windows, Mac OS X, BSD, Solaris
• System calls
• Funny anecdote: Unix creat

What happens when you press the "Enter" key?

• Keycodes
• Keyboard shortcuts
• Different things seem to happen at a terminal prompt and in an HTML text field

Cracking

• OWASP Top 10
• XSS, CSRF
• Database injection
• Buffer overflows
• Metasploit; john the ripper; LOIC; MAC spoofing
• Anonymous and Lulzsec
• War games
• Arms race: address-space randomization
• White, gray, and black hats
• 0-days and money

Security updates and the software lifecycle

Sharing software

• Code: FTP, GitHub, Bitbucket, and Sourceforge
• Pre-packaged: package managers, "app stores"

Version control systems

• Diffs and patches, vs. full copies
• Microsoft Word's "track changes"
• CVS, SVN
• git, mercurial

Software developer literacy

• Understand the historical context in which a decision was made
• Algorithms vs. heuristics, for humans and computers
• Traditional approach to Intro CS: teach a variety of sorting algorithms
  • Most modern languages give you at least one by default
  • I'd rather focus on fundamental low-level constructs, and general principles

Licensing

• Free (as in freedom/beer): GPL, BSD/MIT/etc.
• Proprietary code
• Creative commons, for content
• Anecdote: copyrighting /bin/true (an empty script)

Miscellaneous

• Windows shortcut anecdote: .lnk challenge