so she wants to be forgotten?

OpenCode Go/Zen have deepseek models that have the best cost-benefit out there. Then OpenRouter with selected providers if I want to try something else.

I don’t spend more than $20/month on personal projects and still manage to have a lot more functionality than I could have without AI.

Opencode Go and Zen are both much cheaper than anything from anthropic or openai. So much so that even paying per token can be cheaper than Anthropic’s subscription. Deepseek models specifically are 10x cheaper and almost as good as what you get with anthropic sonnet.

But that’s the thing with benchmarks, you run them because making assumptions about performance based on guesswork often fails. SQLite is very much architecturally unique for being a daemon-less database that doesn’t concern itself with concurrent writes.

Is UUID as pk slower than int or bigints? Probably - you’re storing 4x more data than a 32-bit integer. Does it matter? Probably not.

You filter it out to help with the bandwidth, because grain (or random noise in general) is expensive to transmit due to being incompressible. So adding it back gives you a result closer to the source material.

I got the second half

the irony of this comment 😅

It’s weird their main reason is performance, but then proceed to benchmark SQLite. Who’s inserting 10s of millions of records per minute on sqlite?

Even in production, client-server DBMSs, I’d wager that there are plenty of other things that dominate performance before you even get near your choice of a primary key, so it probably doesn’t matter until you get a large enough throughput in your database.

In most cases it’s more useful they’re not predictable, and I’m definitely not remembering private keys myself, so what’s the point. You can have preservation of creation order with UUID v6 and v7.

Brave lets people use Origin for free if they’re on Linux

lol

The vast majority of those we laid off last week were measurers. We cut middle managers across the organization because AI allows us to have more direct reports per manager while still measuring and mentoring our teams effectively. We consolidated our operations functions into a single group that can support teams across the business, using AI to gain specific expertise when needed. We significantly reduced our marketing team, which, like in most companies, was teeming with measurers. Across our finance team, we found opportunities to consolidate and automate.

But the layoff wasn’t about reducing headcount. In fact, we have a record number of open positions. In coming years I expect our number of employees will continue to grow. With fewer people needed for measuring, we can now invest more in people in the areas that drive growth.

that makes sense to me

You’d think so, but Windows runs better on snapdragon today than Linux. I was very interested in buying one, but in the current state it’s a toy for enthusiasts.

All we had to do was follow the damn train, CJ

it’s not that siml- gets taped

I did it that last month, not because I have any expectations of privacy (I wish we could move away from emails entirely), but because I don’t want to be so much at the mercy of what google decides, especially with their recent push on id verification left and right, and ties to this dystopian government. I’m gradually moving away from other of their products too.

If I was browsing options today, I’d also look into calendar and contact management / importing. Proton makes it easy to import existing calendars and they are kept in sync. They’re still improving the calendar features though, so maybe you’ll miss a thing or two there. Contacts are also easy to import, but there’s no feature to keep them in sync with what google has, if you need a transition period. There is a merging/deduplication feature though.

And if you’re using google workspaces, I couldn’t figure out how to send an email from proton using the work domain, so that’s something I still need to use the gmail web client or e.g. thunderbird.

me too, vending machine, me too

6.6.6 is mentioned

so group B doesn’t use frameworks but prefers making things according to a model? What’s a model here? I’m not following

IME some people (at least the ones who care) start more critical repos with good intentions, enabling most of these things.

But then comes a time where a hotfix must be merged asap to production and there’s no one to review it for 3 days; branch protection gets disabled.

or an update in the code quality tooling detects an additional dozen warnings without the codebase changing; check enforcement gets disabled.

I think in most cases this is completely fine. Whoever is already a contributor should know what’s the team policy merging things, and having ways to occasionally bypass these checks can be more beneficial than not.

dont tell me what to do