Yeah I’m more thinking about the four wheel drive Gen 4.

They’ll never be allowed to compete head to head. It’s like Formula E, that will always be enough different from Formula 1 so as to never show what a pure matchup would look like.

15t/s is workable IMHO. What’s your system specs? I have 96GB DDR5 but never thought about going to an ever higher MoE.

Of course it’s not necessary. I’m a way-beyond-senior dev who laughed at LLMs up until a few months ago when trusted friends, whose competence is not in question, told me they got good usage out of them.

I decided to challenge my convictions and sat down and took the time to learn how to use LLM assistants (I tried everyting from full vibe coding to manual gatekeeping of suggestions).

Now I use them for my own personal projects, and I’m much more productive (for various reasons - but one is that the initial friction of oh yet another thing I have to learn just to do X is much lower. I have no boss telling me what to do, and I select my projects myself. If they didn’t bring any benefits I wouldn’t use them.

That’s my personal view as well and so all apps I write are “licensed” as CC0. Other maintainers are taking a different view.

Having a policy against bad apps is fine - but that’s not what this policy says.

It’s trivial to write a good app using LLM aids - but that new app won’t have such a history.

That’s … incredibly stupid. It will also make Flathub completely irrelevant shortly since app developers won’t change what they’re doing just because Flathub throws a hissy fit.

/had an LLM-assisted PR merged into an open source project yesterday

It’s similar to being an assembler coder when higher level languages with compilers came. No need for management purging, you’ll simply be competing for a smaller segment of assigments.

I don’t know of a single developer that has actually used LLM aids say there’s no benefit to them. Those that refuse do so for some other convictions and don’t really know the difference between LLM aiding in tasks and full on yolo vibe coding.

When they figure it out and become capable of reading and writing. Tablets, phones and computers are not locked down. Parental guidening and open communication means they know what it is, that there’s good and there’s bad content and people etc.

Working great.

/Swedish

I guess it didn’t happen during the race?

Well it’s also that apparently all spectators have to be evacuated from the grandstands in the US when there’s lightning. No idea why that’s a thing there and not anywhere else.

That was a race.

I remember the good old days when it was enough to watch the start to know what the final results would be. I like these new regs.

Haven’t we decided AOC/Stewart is the strongest ticket?

/European

I run a quant of Qwen 35B A3B (Qwen3.6-35B-A3B-GGUF:UD_Q4_K_XL) at the moment, using Opencode and llama.cpp. I’m getting useful work out of it - but it’s of course not Claude. My hardware is a 5060Ti with 16GB VRAM and then ~20GB or so of system mem is getting used as well.

It’s important to put boundaries on less capable models though, so I have two plugins in Opencode as well that really makes a big difference to the results: @tarquinen/opencode-dcp@latest and superpowers@git+https://github.com/obra/superpowers.git.

I want to work in small steps with good control over what the models do so it’s not very similar to what you describe with just having them run away for half an hour and do everything.

Best I can do is just point to source code written by an old colleague and friend of mine from back in the day: https://sourceforge.net/p/qbubblemp/code/HEAD/tree/

I wrote system apps and UI frameworks and it’s 25 years ago so I don’t want to pretend to remember too much :D

ex-Symbian employee/developer here:

Symbian/Epoc C++ is vastly different from modern C++. Enjoy yourself, as long as you are ok with learning a “dead” eco system where a lot of the things you learn aren’t useful on other more modern platforms.

I have since many years back documented most tasks in a digital note taker app. Over the years I’ve transferred them between systems, right now they’re in Nextcloud.

I also have configured a local AI on the server and since Nextcloud has built in support for using local AI it has now indexed all my notes so I can pose queries to a local AI that has all that info I’ve ever written down backing it.

“When is it time to re-apply for the garage building permit” and “Which one of our kids was the tallest at birth” are examples of questions I can now do.

Centralized profit seeking service: We platform the trolls because “user number go up!”

Decentralized community-run services: We throw out the bad actors since users cost money to host.

I like one of them better than the other.

Depression: SSRIs, within a few days Anxiety: Promethazine, within a few days

Never - he was a Jewish prophet. Christianity was invented by Paulus.