davepeck

14 ms·

151.
by davepeck 3y ago
This is great and showcases Django's talent for basic CRUD apps with handy admin interfaces. One catch: I like to deploy static sites to GitHub Pages. It turns out the Django ecosystem has an answer for that in the form of Django Disti
152.
by davepeck 3y ago
This is neat, thanks. I was looking at your HN OPML file on GitHub [0] and noticed that the `xmlURL` and `htmlURL` attributes are the same for each entry; the `htmlURL` currently points to the feed rather than the site. Do you happen to hav
153.
by davepeck 3y ago
See also https://www.typewolf.com -- a long-running and popular curated look at trends in web typography
154.
Llama: Add grammar-based sampling(github.com)
417 points by davepeck 3y ago|105 comments
155.
The looming demise of the 10x developer(blog.testdouble.com)
8 points by davepeck 3y ago|0 comments
156.
by davepeck 3y ago
At this rate, in its first 24 hours Threads will probably accumulate more user accounts than Mastodon has in its 7 year history. (~13.2M at the moment: https://mastodon.social/@mastodonusercount )
157.
by davepeck 3y ago
https://davepeck.org/ and you can confirm I’m an old by the first post date
158.
by davepeck 3y ago
My impression is that when fly.io has been discussed in the recent past on HN, the conversation tends to be less about databases and more about Heroku alternatives. (?) Out of curiosity, how _does_ fly.io see itself in relation to Heroku?
159.
by davepeck 3y ago
If anyone's curious, here's an old article about working with handles and the Macintosh memory manager: http://preserve.mactech.com/articles/develop/issue_02/Mem_Mg... (The article's examples a
160.
by davepeck 3y ago
Yup. Is there a good/forgiving "drunken JSON parser" library that people like to use? Feels like it would be a useful (and separable) piece?
161.
by davepeck 3y ago
I’ve found it preferable to build directly on top of OpenAI’s API. (I’ve also written a simple API wrapper for llama.cpp hosted LLMs.) Over time I’ve built a small library of utilities, including for summarization. It’s not that much code.
162.
by davepeck 3y ago
There are a few strategies in use today. All involve splitting the content to be summarized into chunks smaller than the context size, summarizing each, and building a full final summary from there (potentially in multiple steps). I wouldn’
163.
by davepeck 3y ago
Azure ($75.3B in 2022 revenue) is only slightly smaller than AWS ($80B). It’s GCP ($26.28B) that’s way behind. Microsoft has always known how to sell through its channels.
164.
by davepeck 3y ago
There are no viable self-hostable alternatives to GPT-4 or even to GPT3.5. The “best” self-hostable model is a moving target. As of this writing it’s probably one of Vicuña 13B, Wizard 30B, or maybe Guanaco 65B. I’d like to say that Guanaco
165.
by davepeck 3y ago
I don't know about the Push 3, but my impression with the Push 2 was that python code bundled with Ableton Live rendered the actual screen contents seen on the Push. Does that seem correct to you? And if so, with some amount of reverse
166.
by davepeck 3y ago
I assume this is showing up in part because it was discussed recently in Matt Levine's Money Stuff [0]; it's an amusing (if short) discussion. [0] https://www.bloomberg.com/opinion/articles/2023-05-18&#x2
167.
by davepeck 4y ago
Seattle Public Schools participates in code.org’s “hour of code” program, which introduces students to programming in general and to the (delightfully good) free resources and programming environments on both code.org and scratch.mit.edu. M
168.
by davepeck 4y ago
Given me great joy over the past 23.5 years (!). Given me a place to say things that matter to me and maybe nobody else. Oh, and kept my mom and dad up-to-date.
169.
by davepeck 4y ago
I happened to be listening to Fazil Say's variations when I read your comment. I can't recommend them highly enough; they're a delight.
170.
by davepeck 4y ago
Hah! This was pretty much my response when I encountered Tailwind for the first time. This post by Tailwind’s author gives some more perspective: https://adamwathan.me/css-utility-classes-and-separation-of-...
171.
by davepeck 4y ago
No offense taken. As a fellow old person (whose first computer as a young kid was a TI-99/4A) I definitely feel this pain. But because I am also an old engineer who works with a lot of other old engineers we are lucky to have a shockin
172.
by davepeck 4y ago
These are excellent, thank you. I maintain similar Django and Flask + compose stacks on behalf of the startup studio I work for so it’s fun to compare notes. For our Django stack, for instance, we have also settled on Postgres, on celery+re
173.
by davepeck 4y ago
Fun. Always love playing with toys like this. A little d'n'b loop: https://peel.fm/c275e61 The banana tapping in rhythm is a great little touch. Fruity!
174.
by davepeck 4y ago
Some potential reasons: 1. The data layer, if you want a high-quality ORM and migrations SQLAlchemy + Alembic (or Django ORM + its built-in migrations) are battle-tested systems that I trust will scale to large teams and not break my data
175.
by davepeck 4y ago
Yes. I moved a very large/mature Django project from use of MyPy to use of Pyright. My experience was that pyright flagged a bunch of new conditions that MyPy seemed to have missed. The cost was that I needed to add (many!) explicit an
176.
by davepeck 4y ago
My experience is much the same: pyright is fast, reliable, and (when it complains) correct. MyPy is… not always those things. The teams I know that use Python generally have all moved to pyright if they started with MyPy. The lack of plug-i
177.
by davepeck 4y ago
In addition to being impressive technically, this is just plain fun. I downloaded a game my friends and I wrote way back in the day [1] and... it worked in my browser. Such a nice blast from the past. [1] https://www.macintoshr
178.
by davepeck 4y ago
These four paintings alone are sufficient reason to visit the National Gallery if you find yourself in DC.
179.
by davepeck 4y ago
character.ai was founded by one of the original authors of the "Attention Is All You Need" paper from 2017, which introduced the transformer model and really kicked the doors open to the past half-decade of language model advancem
180.
by davepeck 4y ago
> Natural language interfaces are [...] certainly not the next generation of computing. I'm game to take the other side of that wager. My instinct from the last 5 years of advances in ML language research is that we're right at

More ›