dhruvrrp

Both clause and codex through bedrock seem to be working fine.

Uber isn't cheaper in India, but most people in urban areas prefer taking it over local alternatives or taxis. The cars are actually regular taxis, but the Uber tax is worth it over taxis for the convenience and potential safety/driver verification.

Tangentially, their steps listed to verify PGP signature is the prime example of why PGP has such dismal adoption.

I currently use hetzner and DO, but a while back when I was trying to find cheaper alternatives, and hetzner pretty good for its price to performance ratio. Smaller vps providers were crazy oversubscribed that would get pretty inconsistent performance.

None of the dedicated machines I found were cheaper, so I'm surprised someone is offering that for $35/month.

The problem is there are a class of problems that only appear after you go over the tip of what your system can handle, which are very difficult to predict or model.

Did we read the same article? This seemed pretty positive about UPI as a whole, while covering the expected drawbacks of adding fees.

Screenshots are much better for archival. So many times you go back to a url/link you shared, and poof it's gone or the content has changed from when you sent it.

Screenshots allows you to get a point in time reference to what you are sharing.

A self validating loop goes a long way with getting AI to optimize stuff. We had some builds that used to take a long time (15-20~ minutes), that i just asked Claude to reduce the build time. It brought it down to 2 minutes and under 10 seconds for incremental builds.

Turns out folks had been copy pasting some setup code which took 2 minutes to run, and only needed to be run once per environment. Claude refactored the code to get it to run once, and since we had tests which validated the build artifacts, it was able to roll back if its change had broke the build.

Dunno if this is the same issue, but someone found malware in their projector. I'm not sure about the accuracy since the report is blatantly AI generated: https://github.com/jrm360seclab/aodin-vo1d-malware

Use openWrt (https://openwrt.org), and use their hardware list to pick a consumer router with the feature set you need that can be flashed to use openWrt.

This will fly in EU. As long as the company states the time period for which it will keep data and clean it afterwards, gdpr has no issues with the data retention.

Their carve-outs for safety (public interest) and legal are also valid exceptions in gdpr as well.

How is this different from Jellyfish? I recall reading a paper around that back in the 2010s https://www.usenix.org/system/files/conference/nsdi12/nsdi12...

Edit: Answering my own question, Jellyfish proved theoretically that random networks can be better, and this is a working implementation based on that solves the problems with creating/operating random networks.

I work at AWS (disclaimer opinions are my own, do not reflect views of my employer) and i think the existence of a leaderboard has led to folks gamifying it. People see peers in a higher tier on the leaderboard and start burning tokens to catch up.

I think the company realizes this and is actively trying to avoid this, since for the new tools there isn't a leaderboard.

This blog is basically openAI admitting that Hey we parse and store your prompts!

I know it's well known in tech crowds that they use all prompt data for training, but most non tech people I talk to almost consider their chats with openAI to be private.

Why did they think this was a good blog post?

Some really big/rich markets on the list (Brazil, India, ME..).

I don't think LEGO is big in most of those countries (at least not in India), so they might be trying to slow down the secondary market in order to grow sales for new products.

Well, not that great for current workers either. Microsoft RSUs are usually 20% (compared to 40-80% in other big tech) or less for most employees, and on average msft has 30-50% less total compensation than other big tech.

All this stock upside might put some people's compensation to their true market value, but even that is based on tenure and historical stock vests.

My sports sedan gives 25 mpg, and my parent's Toyota hybrid SUV gives 58 mpg.

It's enshittification because most people don't need the 2 year upgrade cycle. For most individuals and small businesses, it was more like buy once and use forever.

Isn't it exactly the opposite? It used to be low salary + gambling, but now that most people know how easy it is to screw early employees, folks ask for high(er) salary + gambling

I went digging for this, and their SEC filing for 2024 mention they have a stay from the Irish High Court.

> The IDPC issued an administrative fine of EUR €1.2 billion as well as corrective orders requiring Meta Platforms Ireland to suspend the relevant transfers and to bring its processing operations into compliance with Chapter V GDPR by ceasing the unlawful processing, including storage, of such data in the United States. We are appealing this Final Decision and it is currently subject to an interim stay from the Irish High Court.

https://www.sec.gov/Archives/edgar/data/1326801/000132680125...

I would be really surprised if Amazon didn't have algorithms running that price matched other stores automatically.

I thought the CEO was trying to take the company private, for pennies to the dollar which caused the independent board to resign (see: https://investors.23andme.com/news-releases/news-release-det...)

> Financial literacy itself is quite simple: spend less than you make.

I'd disagree with this. Financial literacy includes understanding how /how much to save as well. Too many people spend less than what they earn, but would be screwed if an emergency hit, because they have basically 0 savings/emergency fund.

And the consequences of this are dire, I've seen people in their 70s working because they basically cannot afford to not work due to a lack of savings.

Interesting read, the fix seems to be straightforward, but I'd have a few more questions if I was trying to do something similar.

Is software deployed regularly on this cluster? Does that deployment happen faster than the rate at which they were losing CPUs? Why not just periodically force a deployment, given it's a repeated process that probably already happens frequently.

What happens to the clients trying to connect to the stuck instances? Did they just get stuck/timeout? Would it have been better to have more targeted terminations/full terminations instead?

>every single team is distributed. Unfortunately this isn't true. Amazon already forced all teams to co-locate, calling it RTT (return to team). So majority of teams are in the same location/building.

> Although the problem is with the hood latch, as with many Tesla safety recalls, the problem can be fixed with an over-the-air software patch. The new software is able to detect if the hood is open and, if so, will display a warning to the driver to alert them to stop their vehicle and secure the hood.

So the OTA update doesn't fix the problem, just gives passengers a warning to stop and close the hood.

But does this mean they are not actually fixing the hood latches themselves? I don't think a warning will be very useful if the hood opens up while someone is driving down a freeway at 80 mph.

Link to the Reuters report the article is based off: https://www.reuters.com/investigates/special-report/tesla-mu...

Seems to be specifically the Dutch entity. I wonder how this would affect their USA business, since i see VanMoof bikes pretty often in Seattle

In this context (the blog post) and the reason firecracker was created, was to isolate workloads.

And if youre running untrusted code, then using a virtualized environment is the easiest (id even say best) way to go about it.

I'm curious what the terms of usage would be, since eventually people will figure out how to block all the ads one way or another.

Also after having seen a few cheap Walmart tvs, some with very intrusive ads I think there might be a market for this