dahcryn

"All you have to do is follow the damn train, user"

I think you overestimate the power of a blogpost and the speed of bugfixing at Apple for something like this.

I almost guarantee there is no way they can read this blogpost, escalate it internally, get the appropriate approval to the work item, actually work on the fix, get it through QA and get it live in production in 3 days. That would only happen on really critical issues, and this is definitely not critical enough for that.

when I want it to google stuff, I just use the deep research mode. Not as instant, but it googles a lot of stuff then

4.1 is great for our stuff at work. It's quite stable (doesn't change personality every month, and one word difference doesn't change the behaviour). IT doesn't think, so it's still reasonably fast.

Is there anything as good in the 5 series? likely, but doing the full QA testing again for no added business value, just because the model disappears, is just a hard sell. But the ones we tested were just slower, or tried to have more personality, which is useless for automation projects.

They don't do full reverse osmosis to the purest extent. There are still quite some minerals left. That's actually better for the end product

what more do you actually need?

Unless you want to actually develop ON the device (and build binaries etc...), this completely allows you to use the device and connect it to whatever, so I don't know what more we should expect.

No one else is doing this, so yeay applause

but cost is critical. It's been proven customers are willing to pay +- 20/month, no matter how much underlying cost there is to the provider.

Google is almost an order of magnitude cheaper to serve GenAI compared to ChatGPT. Long term, this will be a big competitive advantage to them. Look at their very generous free tier compared to others. And the products are not subpar, they do compete on quality. OpenAI had the early mover advantage, but it's clear the crowd who is willing to pay for these services, is not very sticky and churn is really high when a new model is release, it's one of the more competitive markets.

Claude Code > Gemini CLI, fair enough

But I actually find Gemini Pro (not the free one) extremely capable, especially since you can throw any conversation into notebooklm and deep thinking mode to go in depth

Opus is great, especially for coding and writing, but for actual productivity outside of that (e.g. working with PDF, images, screenshots, design stuff like marketing, tshirts, ...,...) I prefer Gemini. It's also the fastest.

Nowhere do I feel like GPT 5.2 is as capable as these two, although admittedly I just stopped using it frequently around november.

many people try to go too much around puddles, bits of mud, rocks, ... Switching direction is what makes you slip/fall.

Sometimes, you should just focus on going straight, or at least keep your center of mass relatively going in a smooth line.

Yes, sometimes that means going through a puddle or do a small jump, but I find it much safer. Of course, on a potential slippery surface, try to make that 1 step lower impact, basically like an in-between step. This can also imply vary big changes to your cadence, which is not always optimal from aerobic perspective.

The good ones don't accept. Sadly there's just many more idiots out there trying to make a quick buck

literally 0 issues on my macbook

It has different issues, but wireless headsets nor hibernation are among them

we often forget how great Yahoo engineering was back in the day, sad it was destroyed by bad management and horrible business cases prioritization

at least they were specific

Today it's usually implicitly designed for iphone, designed for 1080p, or ipad, and you have to guess, strong correlation with whatever device the designer uses in his personal life.

why? It's the industry standard as far as my reach goes.

What other framework would you replace it with?

No, polars or spark is not a good answer, those are optimized for data engineering performance, not a holistic approach to data science.

also, don't forget to include QA and release management overhead, as well as projectmanagement etc...

the 60k buffer probably just covers the salaries of the multiple layers of management and facilities (building, cleaning...)

it's the management structure that's broken. Plenty of decent engineers around microsoft who could fix it, plenty of customer and enterprises willing to pay, but they are not allowed to work on it because of prioritization bullshit, allegedly they could get more money elsewhere

That's literally the issue, management by KPI frameworks

buy a pixel and you get it basically unlimited for free for a year ;)

Speculation online as to the why in this case, it's pure advertisement of their capabilities.

I find it weird they are allowed to do anything but destroy end-of-life HDDs

If our legal team finds out about this, we are never going to be allowed to use Backblaze lol.

but how do you know that before the swap?

As long as this information balance exists, they are worth equal value

as if an independent expert panel has no financial incentive to declare something AGI... yeah, this is gonna end well

at that point, it's about portability, not battery life

simple CRUD, is as common in many many business applications or backend portals, are a good fit for AI assistance imho. And fix some designs here and there, where you can't be bothered to keep track of the latest JS/CSS framework

that's usually in areas where they are very certain.

I'd be surprised if they'd do that for GenAI projects, maybe only for really good clients that pay them 50mln+ a year anyway

both make a lot of sense, but the biggest mistake they make is to see people as capacity, or as a counter.

Each human can be a bit more productive, I fully believe 10-15% is possible with today's tools if we do it right. But each human has it unique set of experience and knowledge. If I do my job a bit faster, and you do your job a bit faster. But if we are a team of 10, and we do all our job 10% faster, doesn't mean you can let one of us go. It just means, we all do our job 10% faster, which we probably waste by drinking more coffee or taking longer lunch breaks

your geographical whereabouts are part of your medical history yes, that's important. Far from the most important thing, but definitely important in case of diagnosing weird stuff

and they save versions every edit or every few minutes too. It's free version control that you cannot forget to turn on

Huge contrast with MS approach to this

this is common practice in general no? People ask for references, or try to contact former bosses, when hiring critical profiles. Obviously nobody will say anything bad, so HR is trained, and giving trainings to the hiring managers, how to "grade" the level of positivity.

There's a difference in saying "Yes I confirm person X worked here, he did a good job on all the tasks that we have asked him to do" vs "Yes, he was amazing at his job, he was proactive and really drove innovation, we are sad to see him leave"

The thing is, Salesforce understands sales people, and the product is designed to make their lives easier and more effective. And you know what, they are good at it, that's why they are so big.

But they are horrible at integrating with anything else, making engineers happy, make data and AI people happy. They wall everything in. Guess what, you are not their customer. The sales people are.

So yeah, I hate them, but even more reluctantly, I admit that despite the multi million dollar invoice they send each year, we haven't really found a worthwile replacement. And most of our staff is actually quite positive about them because the old system was MS Dynamics, which is even worse.

they tend to be smarter about this. Instead of a rug pull, they apply the boiling frog principle. Much more gradual and opaque in their increases. It all adds up of course