tevon

Sooo we’re building “SQLite for local-first development”, it’s here! Uses CRDTs, can be a partially replicated db, peer to peer networking and discovery.

Bruinen.co

Shoot me a note if you want an early build! Or if interested in building with us :)

tevon [at] bruinen.co

Bruinen | bruinen.co | (Senior) Software Eng | in-person / remote San Francisco

Bruinen is working on infrastructure technology for edge environments - we're starting with the database. Our database, Delta, is local-first and distributed. Each device (AV, robot, drone, sensor) can continue working without connectivity, and merge it's changes (conflict free) whenever it comes back online.

We target use-cases in the harshest environments - where connectivity is expected to drop: battlefields, at sea, in the air, and in the wilderness.

We're written completely in rust (though its not a requirement for this position, come learn!).

If interested, email me at tevon at bruinen dot co

This is wild.

And somehow at the end of this he comes to the conclusion that software engineers need a union? We're on of the highest paid, most flexible professions ever.

It seems like the thing missing here is the simple fact that engineers DO have significant bargaining power already, and have CHOSEN to use standardized solutions because they are good, well supported, and have wonderful communities of helpful people.

What happened to this guy? Seems incredibly bitter.

How is your license plate on public land your personal data?

This is awesome! Love that it is able to "plot" across a cylinder

My guess is that the "piracy" direction got killed by higher ups because we've likely already infected many of the foreign systems that we want to control. Piracy would just make them look a bit too hard at their existing systems, we don't want them to look, we want them to think they are winning because we're not retaliating.

Stuxnet was us. And there have been zero days that we know the NSA has had for years. I'd say we are more rarely being caught, our goal isn't to knock their internet offline for weeks but to be inside their machines for years.

The rarity of "being caught" could certainly mean that we're very good at this.

Currently if you visit the xz repository it is disabled for violating github's TOS.

While it should clearly be disabled, I feel like github should leave the code and history up, while displaying a banner (and disabled any features that could be exploited), so that researchers and others can learn about the exploit.

In more minor situations when a library is hosting malicious code, if I found the repo to be down I might not think anything of it.

I agree with this as well!

OpenAPI spec seems intended to be consumed, not written. Its a great way to convey what your API does, but is pretty awful to write from scratch.

I do wish there was a simpler language to write in... JSON-based as well that would allow this approach of writing the spec first. But alas, there is not, and I have looked a loooot. If anyone has suggestions for other spec languages I'd love to learn!

It could likely also be injected via malicious websites, force-shared google docs etc.

If a unknowing user asks a simple question, and Gemini reaches out to a malicious website for an answer, the prompt could be injected.

Additionally it could be taken out of an email / doc that was previously sent to the innocent user if the user asked Gemini to search their email or docs or something.

Kind of crazy the number of delivery vectors there are for these connected LLMs

Strangely also the site linked in the gitlab is not a live site. One other question: why gitlab instead of github? I think this is the first OS library I've seen using gitlab instead of github or a self-hosted site

This whole post excitingly reads like star wars.

Can someone with an electrical or physics background help translate what this means "Our propulsion system uses a cathodic arc discharge powered by a capacitor bank"?

I take it it uses capacitors to release larges amounts of electricity on a given period, thus producing plasma which creates propulsion?

Completely agree, however when I customized a standard keyboard, I DID have issues switching back. Now I've customized an ortholinear split hand keyboard, and my brain seems to keep this as a completely separate mental map and I have no issues switching to my laptop keyboard anymore.

i ended up customizing an ortholinear keyboard, took a little while to learn, but seems like my brain keeps a separate "mental map" of it, and it doesn't interfere with my typing on a standard laptop.

One thing that I've found in addressing the worry about training muscle memory on 99% of "standard" keyboards: use a keyboard that is sufficiently different from standard that you brain inherently recognizes the difference.

I use a dygma defy ortholinear keyboard and haven't seen any reduction in accuracy when using my normal mac keyboard. however when I customized a standard keyboard layout I DID find myself making mistakes. I put this up to my brain having a different mental map of the ortholinear keyboard, in the same way that typing with our thumbs on our phones doesn't impede our typing ability on a laptop.

This also doesn't fit with their recent announcements significantly lowering prices. If they were that worried about losing $ they wouldn't have cut prices, they're the clear leader from a performance perspective and can command a premium.

And up to today they probably had one of the best fundraising prospects of any private company in the world.

How do you think this will affect companies building single-tenant apps?

I can imagine SaaS businesses offering their product as a "box": making on-prem significantly easier. Deliver your software along with the hardware, hook it into their network.

Done.

Anyone with more experience have thoughts on if this will potentially become a common option for SaaS?

We've absolutely loved using duckDB, have paired it with PRQL to enable writing really simple to read pipelines that can be composed easily.

Next up adding a caching layer in.

Anyone know if duckdb is considering adding some sort of caching mechanism? i.e. determining which queries or intermediate queries to store in-memory?

Would be interesting to write this as an extension

Do we know if the phone was in lockdown mode? Anyone know how effective lockdown mode is in preventing most of these zero-days?

I was hopeful starting this article. Because I do think we need something better than OpenAPI for self-documenting APIs...

The idea of an interactive tutorial is great, but as others have mentioned, likely very difficult to maintain over time.

Is anyone working on this? I'd love be able to work with a full spec.

I think colang from Nvidia seems like the "closest" so far, and that is pretty specifically focused on chatbots which is to bad.

Introducing non-deterministic inputs into programs is already wild.

I think once we get past "make the LLM generate #$%!#@ JSON" we will start seeing a lot more of this, since we will then be able to constrain the code-paths that are followed.

I could absolutely see LLM-powered operators being introduced at some point, that we use just like an if statement.

Imagine "if(input like LLM(greeting))" or while(LLM(input) like "interested")

essentially distilling unstructured inputs into a canonical form to be used in control flows:

Switch(LLM(input)): - case request - case question - case query

I'd love to see some more examples!

This seems like it could be particularly useful in restricting the domain of the LLM. I.e. ensuring it outputs specific values and few-shot training these steps.

I think there may a really powerful integration here with LMQL. Has there been any thinking about doing so? LMQL seems quite powerful in constraining values and limiting the number of tokens used. This could be a great option for some steps. I.e. a categorization "is this task type X or type Y" before continuing in the logic flow.

I want this but instead of my CPU usage goes over a certain percentage I get that early 2010s MacBook Pro sound: like it’s a jet trying to take off from my desk.

That lets me know I have a hung process somewhere

Same I use them constantly, but just set one of your thumb keys to shift to your functions row. You can map all the way up to F24 then! and have any other IDE specific commands in that layer as well.

I use F all over the place in VS-code and have them all mapped on my dygma raise.

Surprised Dygma hasn't been mentioned yet.

https://dygma.com/pages/defy

Equally as expensive, but really good, team is very responsive and involved with the community, and their customization software is getting better and better.

Also wireless option.

Dygma is coming out with the Defy this month, which comes with wireless capability and is also an ortholinear split board!

I have their other keyboard, the Raise, and its unbelievably good. I modded the heck out of it and now folks come up to me just the press the keys.

Yea I hear this, I moved my ` and | symbols on my dygma raise, an mostly got ride of \ (its on a different layer), since I was only really using it for \n.

The biggest change for me on the Dygma raise was the thumb cluster. I moved my backspace to my thumb, as well as different layer shifting keys. The muscle memory is quite easy to learn, since your brain is already "expecting" to use your thumb to shift to other functionality (since we use our thumb to press command), so it simply makes all my vscode shortcuts 2 simple strokes instead of 3 or more.

Also moved my arrow keys under my fingers on a different layer, makes it so you don't need to move your hands when navigating around code. On that same layer I added option + arrows as well for fast navigation.

The design differences between the Japanese trains and the American ones is so sad.

The Japanese trains are significantly more aesthetic, let alone their superior function.

No chance, next restaurant you're at, 95% chance they don't have a single beer under 4.