aleda145

Yeah deep down I know you're right

:(

I've done this for internal dev tools! It's amazing how many assumptions you have about everything.

I've great success with friction logs: https://mikebifulco.com/posts/how-stripe-uses-friction-logs

If you are a platform team, going through this with your internal customers is both driving adoption and making your tools better. Highly recommended!

Thank you for pointing this out! I always love a good competition and since I already have a workers paid plan it seemed fun. But I'm not eligible, so I guess I'll rebuild github without giving any IP to Cloudflare

I really wanted to like using a Macbook. I've given it a go a few times, amazing battery life and great build quality.

But I can't get over how it locks you down completely software wise. I wanted a sane WM and tried skhd and yabai. They are a collection of hacks on the accessibility API, and they will break whenever Apple decides to. To me that's unacceptable as a daily driver

I open sourced kavla, the SQL canvas I'm working on: https://github.com/aleda145/kavla

README is very bare bones, needs a video or a picture too. The website should tell you some more things: https://kavla.dev (and now the demo does not require a login (also need to remake the demo to fit the new self hosted version))

It's still lacking an analytics agent for the local version though. I've explored using codex CLI, but it feels like its extensive system prompt cripples the agent a lot vs Kimi 2.7 that I've been using in the cloud version.

I think it would also be nice to lean into the fully local LLM setup too, especially for data that shouldn't leave the device.

Excited about a stable C++ API for extensions!

I made a dry run extension a few months ago (https://github.com/aleda145/duckdb-dryrun), will be so nice to build it just once and know that it will always work.

Also urge anyone to make an extension, the template makes it quite smooth: https://github.com/duckdb/extension-template

Hard disagree. Sure their docs are bad, and it's sometimes buggy. But to me cloudflare is actually innovating and trying to offer "new" infrastructure. Durable objects and workers are super cool. R2 has free egress, isn't it insane that they could pull that off?

If what you want is a VM to run postgres then there are other offerings that would be much better.

Not saying they aren't evil though, they probably are, but the infra stuff is cool.

This stance is a breath of fresh air. In my experience change management is the first thing to slap on when a bad release happens.

I've worked at a large enterprise that have a "Change Advisory Board", that you need to convince when you want to bump the major version on your linter. It has the effect of velocity slowing down to a crawl. Changes have to large, since otherwise it wouldn't be approved by the CAB. A slow mess.

At my current place we have to loudly declare "I CONFIRM COMPLIANCE" in every PR description. I'm not sure that anyone knows why, but it keeps the bureaucrats happy. Shrug

I'm very impressed how well durable objects work for multiplayer experience!

Also looks like the table shapes broke, there should be data that you can browse and download. I've replaced those with screenshots just in case anyone else sees it

100% agree, it's on the todo list!

If you just want to see what it looks like here's a demo looking at hackernews data: https://kavla.dev/hn

There's also no email verification gatekeeping it so feel free to just input some bogus data when signing up

10 years ago I rode my bicycle from Connecticut to Central America. I stayed in Antigua for a few days to just relax and reflect. Thanks for bringing those memories back!

Working on open sourcing my data analytics canvas: https://kavla.dev/

I've worked on it for about a year, with 1700 commits. I've been going back and forth if I should just open up the canvas or also the multiplayer experience.

I've always had the dream to make some money from a side project, but I don't think this one is it, so I'll just make the entire repo public. Just want to clean up some stuff first.

Lots of fun tech used: duckDB WASM, Cloudflare Durable Objects, tldraw and pocketbase

Yeah I think it boils down to trust issues. I can't know how the text was created. Anything created before 2023 I know that at least a human had to invest a significant amount of time to produce the story.

Maybe I'm a bit too reactionary though. Thinking about it some more I would absolutely read something if it was recommended by a trusted source (which in my case is almost always a human I know in the physical world).

I concur with the author. Personally I think I just won't read fiction produced after 2023. There's so much good literature already produced, why bother with the new stuff?

I feel very sorry to any budding new authors. Maybe we are entering a literary dark age?

Pretty clever to keep the existing parquet files as is (assuming daily partitioned) to not break any analytical queries. Also creating this index for the first time would make me sweat. With exabytes in the lake, the backfill bill will be $$$ (even just scanning the key column).

I'm not sure about the AI agent angle though, why would it need that specific context on a user level? I guess they are exploring a "chat with my spotify history" or user generated playlists?

(also the article has several LLM smells. It's technically interesting but reads like slop at times)

As a data engineering person I can say that this is a great write up!

Some thoughts:

A "bubbling" topic right now is conversational analytics (i.e. talk to your data). There has been an explosion of tools in the last 6 months. YC is backing one too: https://getnao.io/

I feel like pandas is also somewhat frowned upon, the industry has moved on from that. Most SQL tools can now do everything that we could only do with pandas.

In my network everyone is talking about DuckDB. As long as you are under a 1TB it will have everything you need. I think most people should start with that vs locking themselves into something like Snowflake

I have never heard of Denodo before. "The only real semantic layer" is a bold claim. From the website it just looks like another data governance tool?

Got another daughter last week!

There were a lot of complications post delivery, and I want to make some sort of interactive story about it. We'll see how it goes

(Everyone is safe and sound)

I let agents write my code but not my commits. I think I gain some understanding by doing `git diff` and then writing the (brief) commit message myself.

Örebropartiet is like the weirdest party in Sweden. It's named after "Örebro", a Swedish city with 125k population. The party's founder, Markus Allard, used to be far left politician before turning far right.

Their party program is all over the place. They stand for free dental care, direct democracy and deporting immigrants.

Marcus is also known for profanity and foul language in council meetings.

An oddity in Swedish politics is that if a local party manages to get 12% of the votes in a constituency they are eligible for getting a seat in parliament, and can skip the regular 4% popular vote rule.

Örebropartiet actually has a chance to get into national government next election (Fall 2026) since their local support is quite strong. Times are weird

I think the animosity comes from most people just encountering the "slop". Generated ads, fake content or low quality music.

I just started doing this last week!

I'm not very good at C++, but coupled with the extension template and codex I got a basic version of my extension working within an hour. Go for it!

Are you doing a bespoke canvas engine or using tldraw/excalidraw?

Adding agents to my SQL canvas (https://kavla.dev)

Here's a live example of it figuring out when to post on HN: https://kavla.dev/hn (spoiler, its noon UTC on Sundays)

And here's it generating an interactive map of 20000 earthquakes: https://kavla.dev/quakes

I feel like the canvas is actually a great way to interact with an agent, everything it does is visible, so auditing what it did is (relatively) easy.

I still got some credits to burn so agent usage is free atm (you still have to sign up to use it though)

This sent me on a trip down memory lane! So nice that they are still going.

I solved a few of their problems almost 10 years ago. Looking at my repo I have absolutely regressed mathematically. I apparently solved project 15 without any code: https://github.com/aleda145/projecteuler/blob/master/problem...

(spoilers for: https://projecteuler.net/problem=15)

No way I would have reached for combinatorics if I got this problem at work.

This makes me a bit sad :(

That's a great point! "Speed to insight" feels more important than ever

Cool! But as a data engineer I don't know when I would ever use this. Getting data into a centralized place so it can be joined and queried easily is like prio 1 for any data team.

I'm sure SREs will really love me doing expensive adhoc queries against production postgres /s

I've yet to work in enterprises big enough to have multi cloud data warehouses though, maybe it's more useful in that setting?

I recently picked up writing short stories again. I briefly looked at different editors, but ended up just doing it in vscode (daily driver). I'll make sure to look at cheese paper for the next one, looks like it has some cool features!

A feature that I have been dreaming about is making an editor that treats each paragraph like a unit of work, and the full text is created by linking together different paragraphs. That way you can easily try different ways without deleting any text. Sort of like nodes in a graph.

And here's my a corporate themed short story: https://dahl.dev/capacity

Really nice! I tried out a similar thing but with tldraw as the canvas engine: https://github.com/aleda145/terminaldraw

I see Cate also uses node-pty. I didn't know what psuedo terminals were before, it's cool stuff

I don't know, it is a useful signal that the person did not think deeply about their code changes, and should be treated as such.