"Because this version of AI is worth 10 trillion dollars."

i wonder how much the real version of AI is worth. I've got a hinch we're going to find out pretty soon.

Probably still a lot. If not, then all of these people who praise them are just bad developers.

wondering : apart from the aesthetics, is there still some technology from beos that would be still considered an improvement compared to what today's OS provide ?

of course, nothing magically prevents the app from sending keys or decrypted content to a third party.

That's why if you're really serious about e2ee you have to install the app from source.

Yeah, makes sense.

I wonder if for closed-source apps if governments can not just force the key collection the same way they would force decryption with centralized keys.

I like to co-opt the expression: not your keys, not your privacy.

i think using AI is going to transform our definition of programming into something more abstract, but it's still going to require human.

We're going to start thinking in terms of program general properties ( termination conditions, invariants, symetries, etc), and code quality indicators, to drive our work into certain directions, depending on the project.

Then in 20 years we're going to consider the task of actually typing code something of really low value and uninteresting.

I can feel it's already happening to me after just a few months of vibe coding.

I think System Design and Operational aspect of Software Engineering is now defines programming job (IC position) today

"used to be" ?? What are engineering team doing nowadays when discussing architecturing their systems ?

In my experience the last several years, primarily we’re all on Zoom waving our hands and making false promises to update Confluence with what we talked about. I miss offices with walls and whiteboards.

Miro.com is one of the few SaaS products that our team's collaboration could not live without.

Perfect for a distributed team to replace the DO NOT ERASE white boards of yore.

yes miro is also what i'm using. It's really a digital whiteboard.

Working remotely.

i tried to find an API pricing for GLM 5.1 but couldn't find any on the homepage. How are you using it ?

per-token via DeepInfra, who hosts it as one of their models.

https://deepinfra.com/zai-org/GLM-5.1

you forgot the memory model. Which is an absolutely essential and hard to design part of the agent.

and occasionally, UI prompts with QA.

Memory model? I would not want agent to remember previous conversations.

Ever? You would not want to accumulate any useful context at all?

Maybe in the future but with the current models I found the constantly accessible memories to be an impediment. I don't want models to record and repeat mistakes or suboptimal strategies.

Gemini (just in the browser) has been really bad about conflating a bunch of similar projects. It remembers "oh, you have a home server that does XYZ", so my new home server that's doing ZYX instead must be the same system.

could it be that anthropic is experiencing a massive shortage of compute capacity, and is desperately trying to find means to overcome it ?

All the news i hear about this company for the past weeks made it sound like they're really desperate.

does it? have you been trying to use signal while disconnected from the internet?

I had a friend who complained about this too. I never understood it. She had a really cheap old android phone. Maybe that’s the issue?

I primarily use a nearly-bottom end android phone that's a few years old and just recently switched to an even older, even lower end android phone that is six years old. Neither has that issue.

Obviously, I'm not really claiming that it's not possible people are experiencing this issue, but it can't possibly be widespread.

I feel like most likely people are using android skins that aggressively kill apps in the background.

I have that exact issue on a couple of not exactly low end Samsung phones. Holding them side by side with signal open. Delivery times vary wildly. Whereas WhatsApp just works (though I hate it for other reasons)

I can see why that would be incredibly frustrating. Have you reported it to them, either via github or email?

nope, iphone here, and quite recent. But it's not just me, all the people i communicate with on this app have the same kind of problems. With a group of friends we even had a totally weird ordering of messages, making the conversation quite absurd.

There's something deeply wrong with the way signal delivers messages...

same here... The API costs are absolutely insane for any real usage. This is either high prices to make sure no profitable competitor to claude workspace or other agent system emerges, or heavily sponsoring of their own soluions.