I'll look at it when this shows up on https://aider.chat/docs/leaderboards/ I feel like keeping up with all the models is a full time job so I just use this instead and hopefully get 90% of the benefit I would by manually testing out every model.

Are these just leetcode exercises? What I would like to see is an independent benchmark based on real tasks in codebases of varying size.

Aider uses a dataset of 500 GitHub issues, so not LeetCode-style work.

It says right on that linked page:

> Aider’s polyglot benchmark tests LLMs on 225 challenging Exercism coding exercises across C++, Go, Java, JavaScript, Python, and Rust.

I looked up Exercism and they appear to be story problems that you solve by coding on mostly/entirely blank slates, unless I'm missing something? That format would seem to explain why the models are reportedly performing so well, because they definitely aren't that reliable on mature codebases.

Aider is not just leetcode exercises I think? livecodebench is leetcode exercises though.

At the end of every work day I attempt to add apointments for the next work day to a personal calander for the things I want to work on. I don't figure out individual tasks but will figure out what projects I want to work on and for how long. I dont' view these as hart committments but it definitely makes it easier for me to start the day with focus instead of trying to remember where I left everything yesterday.

We'd all be a lot less stressed if there was a clean separation between politics and technology platforms but that isn't the case. Arguments about the house settlement for college athletics and the politics around that are not a good topic of conversation on HN. How a major tech figure's platforms are running and what effects that may have is worth discussing. Just because someone did good things in the passed doesn't mean everything they do is good. It's irresponsible to not continue to judge people based on there current actions and give them a complete pass on any current actions based on passed behavior.

I'm disappointed, based on the headline I thought this was going to be a story about how military research somehow lead to the Warhead candy I enjoyed as a child.

I won't have to switch careers since I'm a software engineer instead of a developer that does nothing but implement specifications with no creativity or design work. I will have to keep up with the changing technology landscape though like I have been for my entire career.

Hardware that can assume the states you’d creatively conjure is on the way.

Your special literacy isn’t all that. It was a stop gap until hardware caught up. SWE gigs was something politics saw as a “create jobs” opportunity.

Chip makers see the opportunity is there to claim more of the tech valuations for themselves reducing the number of software “engineers” and are coming for ya with global politics on their side. Not just the normies sick of IT.

if robotics hasn't been solved yet, I'll become a robotics researcher. if robotics is solved, I'll research autonomous research. if automated research and robotics are solved, there are no more jobs. in any industry. at least not for long.

I figure the AI is a long way from understanding why the dev environment is pointed to the test database, while the test environment is pointed to the dev database. lol

As someone who's blind I've made this argument in the passed. I don't need 100% success as long as the failure mode won't injure me. Seven years ago I would have loved a car that would have driven me to and from work 95% of the time, and refused to take the other 5% if the weather forecast was bad enough that the self driving wouldn't work correctly. I'd also be fine with the car pulling over to the side of the road if it got confused and waiting for someone remote to take control and drive until it was out of the situation where autonomous driving wouldn't work. Given the fact that I now work remote and am married to someone who drives if you told me I could by a car with autonomous driving for $50000 now I don't think I'd do it. I'm interested to see just how good autonomous driving gets and if it drives down the prices of taxi services. At this point I'd rather see an autonomous taxi service offering lower rates then Uber instead of buying my own car with autonomous driving.

Is that because of DEI, or because Costco offers better deals and people are stocking up because of economic uncertainty? I know people who ahve stopped shopping at Target including me, but anecdotes don't prove anything.

Target also lost more than Walmart and basically all other retailers. There's a general downturn but Target's is steeper than others so it's likely the boycott is having an additive effect.

I have stopped shopping there as well, but continue to shop at Amazon. Amazon is clear that they are a corporation who only cares about the bottom line. I can respect that vs Target who is pandering.

I got excided looking at this hoping there was a laptop with out a screen. I'm totally blind so the power draw of a screen is pointless. I currently use my ROG Alli with a Bluetooth keyboard to connect to my more powerful laptop which has a keyboard that's going bad. While this setup works well and the battery life is pretty good it would be much nicer if I didn't have to put a keyboard on my lap, and the Alli on a table. At least the Alli doesn't need to be somewhere where I can look at it.

I'm not sure if this would work for you, but there are inexpensive devices that plug into an HDMI port. They appear to the computer as a monitor. I use them for screen sharing to a remote display, but they should enable to think there is a monitor attached. It negotiates the display information as if it was an actual monitor.

Here's the pack of three I purchased on Amazon.

Woieyeks 3 Pack HDMI Dummy Plug https://www.amazon.com/dp/B0CKKLTWMN

Would one of those computer in a keyboard set ups work, like the rapsberry pi one?

Google "headless macbook", there is a community of people making macbooks without displays.

The idea started from recovering macs with a broken display and using them like a mac mini. It's possible to find "broken macs" for cheap in second hand market and if the problem is only the display you can go for the headless approach and have macOS with Apple Silicon for very cheap.

Apple Silicon has outstanding battery life, without a screen I would think even more.

You could take a normal laptop and remove the screen.

I've seen several laptops used like this. The only issue is that the WiFi antennas are (usuall?) in the screen.

I wonder how the Framework laptops work with out a screen? I'm interested in the new Framework 12 for it's small size.

You can get some other antennas, place them in the chassis and connect them to the network card. On (at least the first run of) MNT Reform laptops that's how it's done.

On my Macbook Air, if I bring the screen brightness all the way down the screen appears to be completely off.

He says he uses the Khadas Mind / Khadas Mind 2 which is a mini pc that has a battery so its pretty much a screenless laptop. Not clear the battery is very large but he uses an external one too as its usb c powered.

It says that the battery is for when you want to move the PC between desks, so I'd imagine it only lasts for five minutes or so.

It goes into standby mode when unplugged. And lasts for 25 hours in standby. It is great for unplugging from one workplace and plugging into another.

Its a 90Wh battery powering a mobile SoC. It should last significantly more than 5 mins.

Where did you see 90? The site says 5.5.

We're talking about the battery inside the PC, not the external one.

Since you mentioned the ROG Ally, if you are looking for a handheld without a screen (basically a controller with a built in computer) you may like the Tecno Pocket Go.

Also, great pun with being blind and "excited looking at this".

> Also, great pun with being blind and "excited looking at this".

I'm also blind and this is not a pun. No one blind I know would change their usage of language to avoid using vision verbs for the sake of underlining how blind they are.

Obligatory SMBC comic: https://smbc-wiki.com/index.php/Dragons

The Tecno Pocket Go kickstarter seems to already be 4 months late and lots of complaints in the comments

There is a handheld keyboard you can get called the mini keyboard. It has a trackpad for a mouse. Connects by Bluetooth.

Honestly surprised no one's really leaned into that as a product category yet. Seems like there could be a small but very appreciative market for it