If the author is right, OpenAI have room for improvement where they can further improve the fast models for correctness for certain tasks while Anthropic are left with scaling vertically. OFC, it is likely that over time both approaches will converge when the companies understand the problem space better and what tradeoofs are worth making.
My personal take is that they will need a big model to plan and break down tasks and schedule them to specialized smaller models while there is a good enough model for real time interactions with the user, but it is the naive take and many other things might be shaping the decisions.
From the beginning, B5 is like the UN with all the pettiness included. As a political storytelling, it was magnificent. The characters were also very high level.
DS9 has some wonderful episodes and fantastic characters, but the overall plot was weak. The world building was plot driven while in B5 it is vice versa and it made all the difference for me.
have you considered moving or having at least an alternative? asking as someone using caddy for personal hosting who likes to have their website secure. :)
We can of course host our code elsewhere, the problem is the community is kind of locked-in. It would be very "expensive" to move, and would have to be very worthwhile. So far the math doesn't support that kind of change.
Usually an outage is not a big deal, I can still work locally. Today I just happen to be in a very GH-centric workflow with the security reports and such.
I'm curious how other maintainers maintain productivity during GH outages.
For us the main shift was accepting that “being able to work locally” and “knowing whether users are affected” are two different problems.
Local dev usually survives outages just fine.
What hurts is losing external signals and assuming things are okay when they’re not.
After a few incidents like this, we stopped relying on a single monitoring setup.
One self-hosted probe plus at least one fully independent external check reduced blind spots a lot.
It doesn’t prevent outages, but it avoids flying blind during them.
As an alternative, I thought mainly as a secondary repo and ci in case that Github stops being reliable, not only as the current instability, but as an overall provider. I'm from the EU and recently catch myself evaluating every US company I interact with and I'm starting to realize that mine might not be the only risk vector to consider. Wondering how other people think about it.
> have you considered moving or having at least an alternative
Not who you're responding to, but my 2 cents: for a popular open-source project reliant on community contributions there is really no alternative. It's similar to social media - we all know it's trash and noxious, but if you're any kind of public figure you have to be there.
Scifi as a space based genre failed due to the end of the space race. The generation that grew in the post WWII era aged and died. The lack of big scientific breakthroughs for the last fifty years added to the lack of new dreams. The future is here, just it's not exciting.
For the last few decades the culturally significant fiction has been in anime and manga. Lots of it is trash, but lots of it captures the ya themes of friendship and adventure. Some of it captured better the ideas of the cyber age, I suppose. If one explores those genres with a bit of background, you can see how they have been inspired by the traditional sf works, but repackaged them for new audiences without the introspection and conventions of yesterday.
I'm not sure if it is all bad. Definitely, science is not an aspiration any more and those works lack the grounding hard scifi taught us. On the other hand, there is still romanticism in those stories and they teach kids to dream for the impossible. I choose to believe that this is still something to base some optimism on.
I suspect that the next generation of agenticly trained llm-s will have a mode where they first consider solving the problem by writing a program first before doing stuff by hand. At least, it would be interesting if in a few months the llm greets me with "Keep in mind that I run best on ubuntu with uv already installed!".
AI in space makes no sense because the data center will be distributed and half of the time your processors will be waiting for data (if you are lucky). When they are working, you need to cool them down. When they finally heed the call of gravity, your vibe coders can see the vapor of their work in the sky and you throw the entire unit as a loss (might be a net positive for tax purposes).
This is BS, everyone knows that this is BS, but because this is Elon, there are still people who don't call out the BS.
It might be distraction, he might be delusional, he might be asking his investors to stop asking for profit by giving them shares from SpaceX, but this is not him discovering new physics.
My personal take is that they will need a big model to plan and break down tasks and schedule them to specialized smaller models while there is a good enough model for real time interactions with the user, but it is the naive take and many other things might be shaping the decisions.
reply