Python Bytes: #290 Sentient AI? If so, then what?

<p><strong>Watch the live stream:</strong></p> <a href='https://www.youtube.com/watch?v=E-yIQgq08jI' style='font-weight: bold;'>Watch on YouTube</a><br> <br> <p><strong>About the show</strong></p> <p>Sponsored by us! Support our work through:</p> <ul> <li>Our <a href="https://training.talkpython.fm/"><strong>courses at Talk Python Training</strong></a></li> <li><a href="https://testandcode.com/"><strong>Test &amp; Code</strong></a> Podcast</li> <li><a href="https://www.patreon.com/pythonbytes"><strong>Patreon Supporters</strong></a></li> </ul> <p>Special guest: <a href="https://twitter.com/Spirix3"><strong>Nick Muoh</strong></a></p> <p><strong>Brian #1:</strong> <a href="https://github.com/microsoft/picologging"><strong>picologging</strong></a></p> <ul> <li>From a <a href="https://twitter.com/anthonypjshaw/status/1539735691890163714?s=20&amp;t=AofRg_xUHlAmMMhG4uMHlA">tweet by Anthony Shaw</a></li> <li>From README.md</li> <li>“early-alpha” stage project with some incomplete features. (cool to be so up front about that) <ul> <li>“Picologging is a high-performance logging library for Python. picologging is 4-10x faster than the <code>logging</code> module in the standard library.”</li> <li>“Picologging is designed to be used as a <em>drop-in</em> replacement for applications which already use logging, and supports the same API as the <code>logging</code> module.”</li> </ul></li> <li>Now you’ve definitely got my attention.</li> <li>For many common use cases, it’s just way faster. </li> <li>Sounds great, why not use it? A few limitations listed: <ul> <li>process and thread name not captured. </li> <li>Some logging globals not observed: <code>logging.logThreads</code>, <code>logging.logMultiprocessing</code>, <code>logging.logProcesses</code></li> <li>Logger will always default to the Sys.stderr and not observe (emittedNoHandlerWarning).</li> </ul></li> </ul> <p><strong>Michael #2:</strong> <a href="https://github.com/everythingishacked/CheekyKeys"><strong>CheekyKeys</strong></a></p> <ul> <li>via Prayson Daniel</li> <li>What if you could silently talk to your computer?</li> <li>CheekyKeys uses OpenCV and MediaPipe's Face Mesh to perform real-time detection of facial landmarks from video input.</li> <li>The primary input is to "type" letters, digits, and symbols via Morse code by opening and closing your mouth quickly for <code>.</code> and slightly longer for <code>-</code>.</li> <li>Most of the rest of the keyboard and other helpful actions are included as modifier gestures, such as: <ul> <li><code>shift</code>: close right eye</li> <li><code>command</code>: close left eye</li> <li><code>arrow up/down</code>: raise left/right eyebrow</li> <li>…</li> </ul></li> <li>Watch <a href="https://www.youtube.com/watch?v=rZ0DBi1avMM">the video</a> where he does a coding interview for a big tech company using no keyboard.</li> </ul> <p><strong>Nick</strong> <strong>#3:</strong> <a href="https://www.theguardian.com/technology/2022/jun/12/google-engineer-ai-bot-sentient-blake-lemoine"><strong>Is Google’s LaMDA Model Sentient?</strong></a></p> <ul> <li>authored by <a href="https://www.theguardian.com/profile/richardluscombe"><strong>Richard Luscombe</strong></a> (The Guardian)</li> <li>The Google engineer who thinks the company’s AI has come to life</li> <li><a href="https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917">Transcript of conversation</a></li> </ul> <p><strong>Brian #4:</strong> <a href="https://github.com/tonybaloney/rich-bench"><strong>richbench</strong></a></p> <ul> <li>Also from Anthony</li> <li>“A little Python benchmarking tool.”</li> <li>Give it a list of (first_func, second_func, “label”), and it times them and prints out a comparison.</li> <li>Simple and awesome. <pre><code>def sort_seven(): """Sort a list of seven items""" for _ in range(10_000): sorted([3,2,4,5,1,5,3]) def sort_three(): """Sort a list of three items""" for _ in range(10_000): sorted([3,2,4]) __benchmarks__ = [ (sort_seven, sort_three, "Sorting 3 items instead of 7") ] </code></pre></li> </ul> <p><strong>Michael #5:</strong> <a href="https://github.com/agronholm/typeguard"><strong>typeguard</strong></a></p> <ul> <li>A run-time type checker for Python</li> <li>Three principal ways to do type checking are provided, each with its pros and cons: <ul> <li>Manually with function calls</li> <li><code>@typechecked</code> decorator</li> <li>import hook (<code>typeguard.importhook.install_import_hook()</code>)</li> </ul></li> <li>Example: <pre><code>@typechecked def some_function(a: int, b: float, c: str, *args: str) -&gt; bool: ... return retval </code></pre></li> </ul> <p><strong>Nick</strong> <strong>#6:</strong> <a href="https://github.com/TomSchimansky/CustomTkinter"><strong>CustomTkinter</strong></a></p> <ul> <li>A modern and customizable python UI-library based on Tkinter. </li> </ul> <p><strong>Extras</strong> </p> <p>Michael:</p> <ul> <li><a href="https://openssf.org/blog/2022/06/20/openssf-funds-python-and-eclipse-foundations-and-acquires-sos-dev-through-alpha-omega-project/"><strong>OpenSSF Funds Python and Eclipse Foundations</strong></a> - OpenSSF’s Alpha-Omega Project has committed $400K to the Python Software Foundation (PSF), in order to create a new role which will provide security expertise for Python, the Python Package Index (PyPI), and the rest of the Python ecosystem, as well as funding a security audit. (via Python Weekly)</li> </ul> <p>Nick: </p> <ul> <li><a href="https://tosdr.org/">Terms of Service Didn’t Read</a> - Terms of Service; Didn't Read” (short: ToS;DR) is a young project started in June 2012 to help fix the “biggest lie on the web”: almost no one really reads the terms of service we agree to all the time.</li> </ul> <p><strong>Joke:</strong> </p> <ul> <li><a href="https://twitter.com/pr0grammerhum0r/status/1536362784144908288?s=21&amp;t=XQLdkT0BwFv4s15eHbBmPQ"><strong>Serverless</strong></a></li> <li><a href="https://twitter.com/the_unix_guru/status/1541084815927164933"><strong>A DevOps approach to COVID-19</strong></a></li> </ul>