-
There is a glimmer of hope out there that normal people can scrap together enough gpu to really run the latest models themselves. The ui really appears to be having huge leaps forward such that doing things like rag is no longer such a research project that it was just a few years ago. So excited to see Prime go through this homelab [1] exercise.
References:
[1]: /homelab/
Publishing rhythm
Attrs does not like ‘/’ characters in its classes, so to use some tailwind classes with custom values we must make new classes in our tailwind input css.
.cinematic {
@apply aspect-[2.39/1];
}
Given the following markdown with attrs added to the image and to the paragraph block.
{.aspect-[2.39/1]}
{.cinematic}
{.cinematic}
We get the following output with only the middle one working correctly.
{.aspect-[2.39/1]}
Note
The inline version of `.cinematic` works, but `.aspect-[2.39/1]` does not,
it turns into text after the image. The block version with the class before the image applies to the paragraph, not the image.
-
Oh, this kills me to hear it. RSS is the OG [1] way to subscribe and share content out to others. It gives you control of what you subscribe to and reminds you when new content lands on your favorite sites. It is a huge component of web 1.0 and I feel is the most decentralized social media can ever hope to be.
References:
[1]: /og/
markata 0.8.0
I realize that I never did a post on markata 0.8.0, so here it is. 0.8.0 was
released on Jan 2, 2024, just over a year ago at this point. This was the
release that we got pydantic support, and multi feeds.
[1]
Pydantic Support Was kinda big # [2]
Markata is leaning on pydanttic for configuration and Post models. These
models are filled with validators such that you can give it an empty markdown
post and it will figure out some pretty sane default values for the
frontmatter. From there you can progressivly enhance your post with more
information like title, date, tags, slug, description.
validators are awesome!! for instance I don’t set the description on many
of my posts by hand, I let the auto_description grab the first bit of text
from the post most of the time.
Multi Feeds # [3]
Markata==0.8.1 brought multiple feeds into the mix, and started to take place
of rss and sitemap. Feeds are a list of posts that are configured with a slug,
filter, sort, reverse, description, and ...
markata 0.8.0
I realize that I never did a post on markata 0.8.0, so here it is. 0.8.0 was
released on Jan 2, 2024, just over a year ago at this point. This was the
release that we got pydantic support, and multi feeds.
[1]
Pydantic Support Was kinda big # [2]
Markata is leaning on pydantic for configuration and Post models. These
models are filled with validators such that you can give it an empty markdown
post and it will figure out some pretty sane default values for the
frontmatter. From there you can progressively enhance your post with more
information like title, date, tags, slug, description.
validators are awesome!! for instance I don’t set the description on many
of my posts by hand, I let the auto_description grab the first bit of text
from the post most of the time.
Multi Feeds # [3]
Markata==0.8.1 brought multiple feeds into the mix, and started to take place
of RSS and sitemap. Feeds are a list of posts that are configured with a slug,
filter, sort, reverse, description, and ...
asottile [1] has done a fantastic job with all-repos [2]. Highly recommend taking a look.
Clone all your repositories and apply sweeping changes.
References:
[1]: https://github.com/asottile
[2]: https://github.com/asottile/all-repos
If you’re into interesting projects, don’t miss out on tailpipe [1], created by turbot [2].
select * from logs; Tailpipe is an open source SIEM for instant log insights, powered by DuckDB. Analyze millions of events in seconds, right from your terminal.
References:
[1]: https://github.com/turbot/tailpipe
[2]: https://github.com/turbot
I recently discovered roomy [1] by muni-town [2], and it’s truly impressive.
ATproto-connected p2p group comms
References:
[1]: https://github.com/muni-town/roomy
[2]: https://github.com/muni-town
Behold, the Steam Brick
A modder has transformed the Steam Deck in a screen-less, controller-less Steam Brick.
Rock Paper Shotgun · rockpapershotgun.com [1]
I fully believe in our right to repair, ewaste reduction, and bringing a second life to still good hardware that is not up for it’s originally intended purpose. This is a sick console like experience you can strap to the back of a tv, throw in your back to take on a trip, or leave stuffed in your vehicle to game in the backseat. Sucks that it cant do 4k, but I’ve used mine on large screens, and it does quite well for a lot of games, maybe not AAA, but the cartoony multplayer games I play with my kids do quite well.
[2]
References:
[1]: https://www.rockpapershotgun.com/behold-the-steam-brick
[2]: https://dropper.waylonwalker.com/api/file/f3114f19-21cd-4ee6-84a8-06b83346d052.webp
-
Damn these deepseek memes go hard. Wild to see openai get played by their own game.
It’s crazy that the normie news that I have seen on deepseek shows that the Chinese made what the Americans did at a fraction of the price, without taking notice that they are building on the shoulders of openai.
markata search
A side effect of Markata DidYouMean [1] is that we are able to implement some
rudimentary search based on slug, title, tags, and description.
[2]
I made a video of the early work on using this side effect to search for content on markata.dev.
Replay markata-search-1.mp4 [3]
This was first implemented to solve 404 pages giving users a way to try to find
the page that they tried to access. But it turned out to be a decent way to
search through the docs.
References:
[1]: /markata-didyoumean/
[2]: https://dropper.waylonwalker.com/api/file/50cfa8dc-9d46-4f02-877b-688fa5510a83.png
[3]: https://dropper.waylonwalker.com/api/file/fd677374-5ef1-41c7-8845-6de0e10f224b.mp4
-
👏👏👏 This one is really good. I’m right there with him on most of this. I am very hesitant on subscription models, and all the ai tools feel like they are getting ready to be the next round of death by a thousand cuts, this time with pretty limited free tier and relatively high prices to run. I’m sure we will see companies get taken by huge bills soon by building off of someone else’s service.
On the flip side I’m definitely the guy that gets in a rut of just copy paste to the ai, wait for codeium to to inject. I feel like I have issues of momentum more than anything. When I’m on one side or the other I tend to stick it out for too long, but less so on going without because that llm drug is calling you when you hit a hard problem.
I’m excited to see him build out a homelab [1] for llm stuff that he mentioned at the top. I’m interested, but probably not building one out for myself until we start to see some cheaper maybe used hardware to do it.
References:
[1]: /homelab/
Markata DidYouMean
Coming in Markata 0.9.1 is far better documentation. i.e. Documentation that
actually exists for everything. As part of poking around I realized that I
often go to look up the docs for a plugin and forget that the path is
/markta/plugins/feeds, sometimes I might try /markata/feeds or
/plugins/feeds.py or /feeds or I might even forget the plugin name exactly
and try something like feed and get a 404. So I added a didyoumean plugin
to markata that takes care of this.
[1]
I made a quick recording of this early feature, pay close attention to the url
as it automatically updates to the correct page.
markata-didyoumean.mp4 [2]
Happy Path # [3]
direct forward
If you have one post called /markata/plugins/feeds, and it is the only post
called feeds, any combination of /markata/feeds or /plugins/feeds or
/feeds will all automatically redirect with an html [4] page (not a server 3xx)
to the /markata/plugins/feeds post.
Here is the snippet that does the redirect.
<div class="container ...
-
Kelsey says several times in this interview, you don’t need kubernetes. If you are running one node you don’t need kubernetes. My question though is, would you use kubernetes? Ya I get it if you are a web developer, data scientist, backend dev, but if you are looking to bee a whole ass engineer, or infrastructure engineer, you know kubernetes, Should you use kubernetes on single node?
Models
Pydantic Docs · docs.pydantic.dev [1]
I came accross from_attributes today it allows creation of pydantic models from objects such as a sqlalchemy Base Model or while nesting pydantic models. I believe in the past I have ran into some inconsistencies with nesting pydantic models and I’ll bet one had from_attributes set and another did not.
Arbitrary class instances¶
(Formerly known as “ORM Mode”/from_orm).
Pydantic models can also be created from arbitrary class instances by reading the instance > attributes corresponding to the model field names. One common application of this functionality is integration with object-relational mappings (ORMs).
To do this, set the from_attributes config value to True (see the documentation on Configuration for more details).
The example here uses SQLAlchemy, but the same approach should work for any ORM.
References:
[1]: https://docs.pydantic.dev/latest/concepts/models/#rebuilding-model-schema
-
Dang strong stance that tmux and zellij should not exist. I really do get his point though. Theres a good number of terminal features I often miss out on because I run tmux. Its an app that runs apps, and doesn’t let all of the signals back to the host. But its fantastic at what it does, and brings so much to the table that the little bit of downside it brings is well worth it to me. The other thing missing in this discussion is that I can take my hotkeys and session workflow to any machine just by running tmux. I do not need to run a certain terminal, or install it headlessly on a server to get special features just for it.
cold builds
Here are a list of some cold builds from my site. I’ve ran this site for a
long time and would like to have some references to go back to, and wish I had
kept a few profiles of cold builds laying around to compare with. The time is
now lets keep some cold build links around for reference.
[1]
01/27/2025 # [2]
- https://ec314b08.waylonwalker-com.pages.dev/_profile/ - 429s - markata==0.9.0.dev5 has md_video open cache on every post issue
- https://27f117fd.waylonwalker-com.pages.dev/_profile/ - 112s- markata==0.9.0.dev5 after fixing md_video issue
- https://e6b8f64a.waylonwalker-com.pages.dev/_profile/ - 16s - markata==0.9.0.dev5 1 post not skipped, this one
- https://10b778b4.waylonwalker-com.pages.dev/_profile/ - 5.65s - markata==0.9.0.dev5 fully hot cache build with no changes
References:
[1]: https://dropper.waylonwalker.com/api/file/cd8f9d96-948a-4e94-aa3d-2c352bb5657b.png
[2]: #01272025
pesos
Pesos is the act of Publish Elsewhere Syncicate to Own Site. It is an indieweb
concept that I recently started applying to my own site.
here does it skip again
Note
See <https://indieweb.org/PESOS> for more information, they have a ton of
information about the indieweb
In short it is the concept of pulling data from other sites that you use and
republishing it to your own site. This gives a single source of information
for you, and protection against sites and apis changing or rug pulling. Other
people might have a lot more use cases for this, but I already begin a lot of
my data right on my site.
GitHub stars # [1]
I am using the github api to get a list of my stars and then create posts in
the github repo for my blog. This allows me to keep track of things I star on
GitHub in my own way, and share them out with my rss feeds.
[2]
References:
[1]: #github-stars
[2]: https://dropper.waylonwalker.com/api/file/2380f814-a8cb-45d1-bb5b-538d171933e3.webp
Check out veekaybee [1] and their project gitfeed [2].
Feed of posts from Bluesky that have a GitHub link
References:
[1]: https://github.com/veekaybee
[2]: https://github.com/veekaybee/gitfeed
I came across llama.vim [1] from ggml-org [2], and it’s packed with great features and ideas.
Vim plugin for LLM-assisted code/text completion
References:
[1]: https://github.com/ggml-org/llama.vim
[2]: https://github.com/ggml-org