uf00lme

its due out tomorrow right?

Cursor is saying we need to get improve the harness and tooling, but the models are getting better. One would have to assume that will continue and that cheaper/smaller models will continue to become even more useful.

This tends to be the kind of thing you can get thrown in for free when negotiating. It's highly unlikely to be a deal breaker for anyone buying Perforce, if anything they may have done this to stop annoying cheap customers buying from them.

It's surprising usable and cheap enough to run in 'fast' mode when vibing something quick. For simple code I find I prefer the code it writes over GLM or Gemini family.

You can thank the us gov for a list of companies to invest in.

I wonder why they didn’t train off Kimi 2.6, I hope is it because they already had a good base and not that they messed up that relationship.

Woah, is this part of the future of models? Basically little models you can use as tools.

The speed is impressive, I wish it could be setup for similar to speculative decoding

A 2011 MBP is likely a better a general purpose PC, those early models had some great engineering. Wait for the reviews and benchmarks but the M1/M2 based MBPs are still great daily drivers.

Channels matter a lot, quad channel ddr4 is going to beat ddr5 in dual channel most of the time.

I would be more impressed if they found issues the Apple password service or 1Password, you always have to assume that no software is complete secure but personal I only trust those two especially after the Lastpass hack https://blog.lastpass.com/posts/security-incident-update-rec...

Do you know which MRI you used? Not all are equal. Most MRI are 1.5T powered, and you can’t get fine details until you hit 3T. And there are differences even in the 3T power range. There are higher powered MRI which are mostly only used in research, whilst it is a bit scary thinking about the sheer power of them but a 7T machine doing a full scan of you, would be guaranteed nearly to find anything wrong with you.

When I last looked the full body scans for sale seemed to used 1.5T setup, which seems like a waste. The 3T advanced scans looks much more detailed, but it just depends on where you live - I couldn’t find any around.

I think having kids asking if they can scan the face of their parents/older sibling will probably give rise to some good conversations.

Mirrors are not effective enough. Shielding drones from energy weapons seems like a similar problem to entering Earth’s atmosphere, you want to shield it in a way that will blast away safely and ideally diffuse the laser, so the energy is spread over a larger space. I suspect larger lasers will likely aways win, since there is only so much shielding can do. At which point we could end up with transformers like drones that are built to be broken apart mid flight and yet still deliver damage. I feel like defending drones could become possible with energy weapons but only under ideal weather conditions.

It might feels natural to try and use terraform to deploy kubernetes resources since you’ve likely configured the cluster using it, but the helm/kubeneters/kubectl providers are limited by terraform’s way of working. So whilst the providers try to marry the two when deploying anything complex it generally ends up feels like a hack and you lose a lot of the benefits of using terraform in the first place.

In my experience, it’s best to bootstrap ArgoCD/flux, rbac and cloud permissions those services need in Terraform and then move on to do everything else can via Kustomize via gitop. This keeps everything sane and relatively easy to debug on the fly, using the right tool for the job.

This has been my argument for years now.

I think it is because most engineers learn to use Kubernetes by spinning up a cluster and then deploying a couple of helm charts. It makes it feel like that’s the natural way without understanding the pain and complexity of having to create and maintain those charts.

Then there are centralised ‘platform’ teams which use helm to try and enforce their own templating onto everything even small simple micro services. Functionally it works and can scale, so the centralised team can justify their existence but as a pattern it costs everyone a little bit of sanity.

The quality of code and reputation of the authors is excellent in this new release.

I’ve never looked at it before but this seems pretty solid, definitely worth keeping an eye on or testing.

I would be surprised if it is worse than the audio latency in any of the standard video conference software. I expect it would have a ton less processing and less hops.

Cloudflare and similarly placed CDNs will likely make it enabling it for sites a checkbox like option, e.g., https://blog.cloudflare.com/this-is-brotli-from-origin/ That's where it will have the most savings globally.

A lot of safety features are mandatory in Eu/Aus/US/NA etc, making imported Chinese cars much more expensive than first thought. Bidirectional charging does not add much cost if planned ahead, since most of the effort is in implementing ISO 15118-20 rather than hardware.

I feel like the main issue with bidirectional adoption currently is the number of groups trying to maximise their profit, everyone wants a cut of the pie: the car companies, energy companies, ev charging companies, ev charging networks, solar/inverter manufactures and government standard groups. The needle won't move until California or other countries with influence force it. In Japan, from a technology perspective CHAdeMO has allowed for EVs to do bidirectional charging for years now.

Side note - I'm all for any extra safety features but I do find that a lot of the software driven features are poorly designed and implemented. I've had bad experiences with automatic braking and lane keeping when freeways driving where I was lucky to avoid having an accidents, so not being able to disable them permanently is a major annoyance to me. It seems like these features have very little real world testing.

The first Dune had an epic soundscape, it was the first movie I've watched where I thought the audio was doing a better job than the visuals in pulling viewers into the world. I'm going to see if I can pay a bit extra to see it in a cinema this time, and hopefully find one that has a good audio setup.

600% more, 0.60$ for extended support vs 0.10$

Still that might be cheaper if a product is going eol then fixing issues blocking an eks upgrade path. Inversely it’s hard to see any justification for the price increase, I can’t see support for these versions costing 600% more but I guess this is why AWS make so much money…

Has a 1-3 team moat ever really existed with new technologies? My best guess would be to somehow get your tech embedded into the defence industry.

My goto moat test: is there an obvious ‘death by Amazon’ route? Amazon can strike fear into entire industries with scale and capital. Very few tech companies could withstand a full frontal assault by Amazon. I can’t see how this trend won’t get worse if AGI occurs during our lifetime.

The only reason there is not more AI or robots doing the majority of the work available to humans, is either it’s cheaper to hire humans or companies just haven’t gotten around to it yet.

At least in theory it should be possible to recondition these batteries to make them useful again, I'm not sure who/when/how much but I suspect they will never be completely worthless.

Not many people use compute 24/7, so if your use cases can be ephemeral than spot/savings plans can be a good saving and using the cloud allows for opex rather than capex controls and taxes

As per other comments, I will likely net profit today from maxing out usage during negative pricing via octopus agile.

I have octopus agile for electricity and octopus tracker for gas. Been on both since February when my last fixed rate ended.

It has worked out great so far, some days when there isn’t much wind can be painful but overall it works out better. Even if it didn’t, I’d rather pay a fair amount than have energy companies rip me off.

I’d be very interested to see if you send two samples in fully anonymously to see how much they differ. I don’t trust the accuracy of these low cost, high throughput methods.

It took scientists ~18 years to sequence the first genome, I know that was ~15 years ago but they did have huge budgets.

The normal approach is exceptionally high taxes on product that cause economic and social harm to a country but that only goes so far and in my opinion often under estimates the true cost.

Maybe try paper and pen? Some simple fun exercises. Documenting cooking or making simple food, like the logic behind making a sandwich and then make it. As a bonus, you get to eat some food, and the kid learns some life skills. I guess that is not really programming, but the process, organisation and problem-solving skill set isn't that different for a kid.