I don’t rent anymore, but I’m 100% for transparent and up front pricing on everything. Next up lets get those restsurant service fees.

You are spending millions to expand fucking Flock cameras

Wait what? Is Seattle contracting with Flock?

The claims are “citing Wilson’s public safety approach to areas such as Aurora Avenue and Little Saigon, her move to turn off CCTV cameras in the Stadium District after the World Cup, and asking the police chief to resign two weeks ago.”

IMO These claims justifying a recall after a year in the role are tenuous at best. As somebody who lives downtown, public safety is a concern (not a OMG Seattle is hell-hole level,) but a remote police chief is not somebody who is invested in Seattle, and cameras–I ultimately think public safety should not come at the cost of mass surveillance.

It’s a small model that runs on your machine using a TFLite model. AI/ML models come in many different sizes. Some more conducive for running locally than others. There’s more info on the detection pipeline here.

I didn’t write this, just found and self hosted it.

Nice. What kind of microphone setup do you have? I’m just starting with security cameras because I have those, but the audio quality is ehh.

You’re right. I copied the wrong one out of the commitee agenda doc.

Here’s the right one:

https://seattle.legistar.com/View.ashx?M=F&ID=15745977&GUID=30A54712-2A1D-4EFE-BFDE-8BCCE843730F

Now you have two package managers. The problem is that Nix needs to create a folder structure that matches what your language package manager wants to see.

Even if the language package manager goes away, languages have slight differences that can’t be totally abstracted away. C/C++ have .h and .c files. JS has its own thing, C# has a .dll. Each language has difference opinions on how symbols get resolved.

The email also mentions: Billing and shipping address information was leaked.

Like yes, it’s technically not Framework’s fault

I know it’s a third party vendor that got hacked, but I really think the blame still falls on the company that contracted with the vendor.

Framework decided to hire this company and companies need to recognize the risk in hiring SaaS companies. I know at my previous job we had to go through security and privacy compliance reviews to decide it was even worth sharing data with a third party vendor.

Too many companies have a ton of SaaS providers and shovel customer data to each one.

Thanks for posting. I wasn’t able to load the capitolhillseattle.com link because CloudFlare blocked me.

I know some grocery stores have been switching to app-based coupons as a way to save money on paper mailers, but also because they can tailor the coupons based on each individual customer’s likelihood of buying. I sent a message to the city council in support.

The actual proposed bill is here

Looks like it only applies to Home Assistant OS which would be a dedicated machine/VM, not container or other installation strategies.

EDIT: The internal listen port is still 8123: https://github.com/home-assistant/core/blob/938a78275e9a858a0364fe13390d2f583f70ebe4/homeassistant/const.py#L970

One of the most complicated systems I worked on was based around event sourcing with a bunch if AWS DynamoDB tables and Kinesis streams. The original designers thought the inherent auditing capabilities and ability to generate view tables of anything would be a great for data durability and performance. In reality, nobody really understood how to effectively use it and ended up with events that were giant entity blobs so you couldn’t understand changes, expensive AWS bills, numerous view tables that were impossible to know if they were in sync and correct, and difficult to evolve systems.

Event sourcing is a cool system, but friends don’t let friends build systems they don’t understand. There’s a lot to event sourcing and I agree with the gist of this article of not modifying events. Though you can end up with a lot of events in a complex data model.

50k satellites? Who is going to pay for that much light? At least with Starlink/Satellite Internet there’s wide spread demand from many different consumers, but for sunlight on demand? That just sounds like marketing to investors instead of a legitimate plan.

The problem is still concurrency. If the application, in this case the git CLI, is not thread-safe, then you could run into unsafe data mutations causing data corruption.

Imagine two pushes that come in around the same time. S3 doesn’t have the locking mechanics you need to ensure that only one of those writes win and the other fails. I’ve done a lot of work with distributed systems in AWS and my experience has always been to layer something like DynamoDB on-top of S3 to provide this.

Codeberg runs on Forgejo which is a fork of Gitea.

I’m running Forgejo myself and am running into difficulties with High Availability. Many data files can be stored in S3 compatible interfaces, but the Git repos are stored as Git files on disk. That’s hard to distribute across multiple hosts. Something like NFS seems to be prone corruption and Forgejo has to be able to lock.

An open issue on clustering : https://codeberg.org/forgejo/discussions/issues/259

I looked quickly, but didn’t see an answer. Does Sure.am let you download your transactions into something like Actual or Firefly-iii or is it all within the hosted platform? I saw an option to self host, but didn’t see the integrations.

I’m trying to find something to pull into my firefly-iii instance and am currently abusing the Monarch API but it has some issues. I also wrote my own scrapers but they are fragile and I’d rather just integrate directly with plaid.

I don’t see how trump could pardon a company in a civil suit. He’d have to get a sympathetic judge or pressure the judge presiding.

The funny thing about this is that an LLM can only do as much as the harness allows.

So another way to phrase this is OpenAI was reckless and did not constrain or monitor their model’s behavior.

Connecting the wires is a little finicky at times, but I used a pogo pin clamp which had the right pitch size to make it easy to connect and avoid soldering. Once you get it on ESPhome the first time, you can keep OTA updating new firmware

This walks through what I did

I installed an Emporia Vue2 flashed with ESPHome in my home breaker box and put CT clamps on all my circuits which makes installing this integration easy.

This integration is pretty well polished and was much more polished than my simple crossing trigger I had before.

Just started using it so it hasn’t learned any cycles yet.