🎧 Django-Fans aufgepasst!
Fühlt euch wie vor Ort bei der #DjangoConEurope2025 Europe mit unserer neuen Live-Episode aus Dublin. Talks, Community & Open-Source-Diskussionen mit Django Fellow @sarahboyce und natürlich @shezi:, @oryon_dominik und @jochen:
https://python-podcast.de/show/live-von-der-djangocon-europe-2025-in-dublin-tag-1/
#Python #Django
Reduce - The Power of a Single Python Function
I wrote ✏️ 🎉 an article about Running Tasks Concurrently in Django Asynchronous Views 
Buckle-up for our async-journey together! 🚗 #django #python #async
https://fly.io/blog/running-tasks-concurrently-in-django-asynchronous-views/
Für alle, die meinen ich würde zu wenig #podcasten - schaltet doch mal heute Abend 20:00 hier ein:
Yesterday I presented at the "Modern Frontend" Meetup in Microsoft Berlin about #AI, #LLMs and what #ChatGPT means for developers, showing off a lot of #Github #COPILOTx features. Here's my edited recording.
https://www.youtube.com/watch?v=Ou8S9ovMtVw
Blog: https://christianheilmann.com/2023/03/31/modern-web-development-centering-divs-in-new-exciting-wrong-ways-with-ai/
NEW POST: Recently I had an enlightening chat with Xu Hao (our Head of Technology in China) about the techniques he's using to prompt ChatGPT to generate useful code
@marcthiele übers podcasten, kann aber genauso aufs bloggen umgelegt werden.
“Letzten Endes muss man sich einfach, also ich zumindest mich davon freimachen, auf die Zahlen zu gucken. Ich hab bei mir auf den Webseiten auch keine Analytics-Tools mehr laufen, gar nichts, weil's einfach nur ablenkt. Mach den Scheiß einfach für dich. Wenn du's für dich machst und es gerne machst, dann passt schon.”
Hier gibts übrigens eine Episode über LLMs mit einem der Co-Autoren des Transformer-Papers und die ist noch weniger technisch als unsere - aber sehr gut 😁
PEP 684: "A Per-Interpreter GIL" has been accepted for Python 3.12!
https://discuss.python.org/t/pep-684-a-per-interpreter-gil/19583/42?u=hugovk
I agree very much with @simon 's message here: "These are incredibly powerful tools. They are far harder to use effectively than they first appear. Invest the effort, but approach with caution: we accidentally invented computers that can lie to us and we can’t figure out how to make them stop."
Also in my post:
"Honestly, at this point using ChatGPT in the way that I do feels like a massively unfair competitive advantage. I’m not worried about AI taking people’s jobs: I’m worried about the impact of AI-enhanced developers like myself.
It genuinely feels unethical for me *not* to help other people learn to use these tools as effectively as possible. I want everyone to be able to do what I can do with them, as safely and responsibly as possible."
Fun GPT-4 use-case: comparing how different languages/frameworks solve common problems
My prompt:
> Show me code examples of different web frameworks in Python and JavaScript and Go illustrating how HTTP routing works - in particular the problem of mapping an incoming HTTP request to some code based on both the URL path and the HTTP verbDon't show installation instructions, just get straight to the code snippets. Don't show installation instructions, just get straight to the code snippets
Some nice progress on PyPy's 3.10 branch: I just merged my implementation of PEP 626 (precise line numbers for debugging and other tools) and our backport (from CPy 3.11) of PEP 657 (Fine-grained error locations in tracebacks). The latter was implemented by Batuhan Taskaya and me.
Recently at work I got the chance to use a couple of #Python features I'd never used before: positional-only arguments, and keyword-only arguments.
If you've never seen this before, the general form is:
def my_function(a, b, /, c, d, *, e, f)
Where a and b *must* be passed as positional (not keyword) arguments, e and f must be passed as keyword (not positional), and c and d can be passed either way.
Still not sure how I feel about them being first-class language features, but they did solve a kind of weird niche problem I was having and that would have needed more code to solve in other ways.
I wrote about how AI-enhanced development makes me more ambitious with my projects https://simonwillison.net/2023/Mar/27/ai-enhanced-development/
TIL from @adamchainz that when doing type annotations in #Python, `object` is preferable to `Any`, since the latter basically tells `mypy` to opt out of type checking.
https://adamj.eu/tech/2021/05/07/python-type-hints-use-object-instead-of-any/
@jaro @hukl hier ist eine sehr schöne Liste https://github.com/ibraheemdev/modern-unix
Someone referred to Podcasting 2.0 the other day.
Oh my god.
There is only one podcasting.
I can tell this is going to be really tiresome.
You know like Web 3 and all that bullshit.
The other day someone told me that the real Web 3 is AI.
Sigh.
The #Mypy project has reached a huge milestone, v1.0.0! 🥳 40% performance boost and a bunch of other improvements:
https://mypy-lang.blogspot.com/2023/02/mypy-10-released.html?m=1 #Python
SQLAlchemy 2.0.0 released! ://www.sqlalchemy.org/blog/2023/0
Neue Episode: @cfbolz, Dominik und @jochen unterhalten sich über PyPy: ://python-podcast.de/show/pypy-ju #pypyproject #pypy #python
@intro Alles klar, muss mich nur noch dran gewöhnen, die Python-Themen von hier aus zu behandeln. Immerhin sollten takahe und ivory jetzt halbwegs zusammenarbeiten 😄.