Previously, I posted critically not because they were running businesses without humans, but because their post just described going through the motions without actually discussing if it really was effective or not. Sure the AI got through the day, checked off tasks on the list, but did it actually do that effectively or efficiently in any important way? Who knows... wasn't discussed.

I think where I come down now is that repeats of this same gimmick feel like just that: they're just playing a gimmick for attention. I can't tell that they're really demonstrating any special or significant capability... but man, just the story of trying to run a business without humans will get you that sweet, sweet attention.

Unfortunately, looking at least the first post, I stopped reading their "we let AI run X" posts. I think the only thing I really came away with is how thoughtless and mundane are most aspects of running a small business actually is; something I knew, but it really drove the point home. I didn't learn anything unexpected about AI tools or their products that seemed compelling or unexpected.

Comments URL: https://news.ycombinator.com/item?id=48180509

Points: 2

# Comments: 0

I couldn't more highly recommend it.

No, very technically possible... though, with good randomness, very, very unlikely.

But nothing technically prevents a UUIDv4 from generating a duplicate value.

Comments URL: https://news.ycombinator.com/item?id=47929542

Points: 1

# Comments: 0

1) That school is simply about landing a job.

2) That there is a value in students knowing how to have the AI do problems for them.

3) That follow-on effects of manually solving difficult problems is discountable compared to the direct output of the work.

I would say you're absolutely correct in that people pay for the result and they don't really care how you got there. But that's a pretty shallow rationale which overvalues the ability to be the conduit from the source of requirements to the final output and undervalues the individual ability to think for one's self when faced with the challenges of technological, geopolitical, or simply uncontrolled personal circumstances.

"The conduit", who you seem to be believe is the one with marketplace advantage, is exactly the person I would say is the most vulnerable. Not because getting the AI to produce demands is without value, but that its quickly becoming a task that doesn't need the intermediary at all. Those magicians that can prompt/agent/mcp/etc their way through to positive successes are actively being challenged by the very AI producers which our conduits people now depend on. Removing the need for intermediaries would be a great competitive advantage for any AI vendor able to achieve it. But insofar as intermediaries create output from LLMs, they'll not be very well differentiated: the common wisdom tends to be the output, lest the AI be accused of hallucination or being overly supportive. But when everyone is using AI for everything the opportunities will be in arbitraging that which is missed by common wisdom... filling in the cracks that any responsible AI would simply never venture to consider. Our conduit-person will be at a decided disadvantage because it takes real thought to know when it's best to color within the lines, and when it's best to not do so.

And that's really it. A good education is teaching you about the process of thought and becoming practiced at thinking. I would expect a better educated, thinking person to more easily adapt and make use of technology such as generative AI to solve problems more so than a person that just knows how to deal with today's prompting needs. The thinking person will be able to understand the bigger picture to better get a consistent and high quality series of results than the person just getting results as needed.

And that's really it. The output of a good education is you as a thoughtful & knowledgeable person: the output on the page is merely a means to that end. But if you focus solely on the answer on the page and the only important thing... you're really evaluating the AI, not the person that acted as intermediary.

In otherwords, if the person following your advice comes for a job, simply ask them which AIs they used in the interview and then just sign contracts with those vendors instead... you'll get better bang for your buck cutting out the middleman.

https://www.youtube.com/watch?v=2Vrhk5OjBP8

Good discussion in the comments there as well.

The only thing that I saw demonstrated, and again, I skimmed, is what many thousands of software developers using AI tools to write their boilerplate already know: these tools, as of now, are great at going through the motions. A successful retail business, and I spent many years in the retail industry, isn't about putting together a nice store front, hiring clerks, and selecting just any-old-products: it's about being profitable. In traditional retail one of most important things is getting the right real estate for your target market... seems like that choice was made already in this case. Yes, a nice store front and good clerks are important, but I've worked in chains which were immaculately designed and built stores with great clerks that failed... and some that opened little more than fluorescent lighted hellscapes with clerks that barely cared that succeeded. In both cases the overall quality of the decisions and strategies relative to the target markets mattered to the success of the business. Just going through the motions didn't.

So if all is this is to say AI can do the things people generally do in these circumstances then sure, you didn't need this much human effort to prove that.... developer types do that at scale everyday now. If there was something different that this company is trying to learn, I'd be much more interested in that.

Comments URL: https://news.ycombinator.com/item?id=47780998

Points: 1

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47698000

Points: 2

# Comments: 1

Or put another way, the context matters.

So yes, you are literally correct in that they haven't put one in orbit, but it's more out of caution than capability. What they've only demonstrated in the most recent tests is that they have good reason to believe to believe that they can deorbit in a controlled fashion. But... now they've upgraded everything: raptor 3, booster v3, starship v3. Those need to prove out those capabilities again.

So I wouldn't be surprised if they continue the suborbital program for the next 3 or 4 tests. Given all the redesign, they aren't exactly at the beginning, but they have to show that they haven't broken what they previously fixed.

I find Sal Mercogliano's "What's Going on With Shipping?" to be a better source to understanding what's happening in the Strait. Here's a link to yesterday's episode "Strait of Hormuz 3-Week Recap | What is the Status of the Ships, Transits and Escort Mission?": https://www.youtube.com/watch?v=q64cOs7GN_4

Someone also once said that the Azores are the remains of Atlantis. I simply didn't put any credence in it.

While behavioral changes around a self-driving car are plausible; they're common enough now that, at least where I live in San Francisco, regular human drivers should be pretty well acclimated to them.

The curious thing for me is that with jj I find that I don't ever really reach for a GUI/TUI anymore. At first I did try a GUI, but then I realized that the vast majority of what I want to get done conceptually didn't require it. So most of what I do now is just using the jj command line and very rarely do I reach for any sort of other tool.

The exceptions to the GUI/TUI use are resolving conflicts. For me being able to see the conflicts side by side and much more interactively choosing which I want is still more comfortable than simply hand-editing the file. And I also find I'm searching or chatting with an LLM anything I want to do something a little more advanced and less common, like rebasing all my feature branches on the most current mainline branch in one command... jj has a rich set of functions and pattern matching which I haven't (and may never) take the time to learn. But the majority of day to day interactions... just me and the command line. I would never say such a think using git by itself.

https://github.com/openai/symphony

I'm not very familiar with the space, I follow Elixir goings on more than some of the AI stuff.

It is curious... and refreshing... to see Elixir & the BEAM popping up for these sorts of orchestration type workloads.

https://en.wikipedia.org/wiki/Pierrot

From the Wikipedia entry...

*His character in contemporary popular culture—in poetry, fiction, and the visual arts, as well as works for the stage, screen, and concert hall—is that of the sad clown [...]*

A modern "Pierrot" style clown is:

https://en.wikipedia.org/wiki/Puddles_Pity_Party

I have to admit... I wouldn't know about the Pierrot style of clown if it wasn't for Puddles...

https://www.youtube.com/watch?v=Cf1w5GUturU

https://www.trialmonitors.org/meet-the-committee

Plenty of accomplished people there, but as a group "unbiased observers" isn't the first phrase that comes to mind.