There have been a range from quite basic things like social media posting, that were a real pain to organise, but actually took no time at all - to external APIs around text summarisation or modification, right the way through to modifying images etc. - the external APIs had a range of different benefits, but always were priced annoyingly enough that it was faster to just build what was needed.
Totally agree. And beyond that human life assumes it’s own sentience through an incredibly narrow view. Isn’t it fair/logical to assume that if giant mushrooms did over the earth that given the passing of considerable time they’ve likely evolved to a far more considerable state of intelligence/harmony than humans have. Their survivability/adaptability being an illustration.
Likewise, one ends up sounding like an anti climate nut, but the existential fear for the planet seems arrogant at best and massively over emphasises our place in the universe. See, I sound like an anti climate nut.
Yes, there’s a really sensible approach here. A recent experiment saw a fairly sophisticated internal tool use a number of sheets as a backend. Aside from speed of implementation the other real benefit was the ease of which non-technical users could not only make easy edits to the backend, but also that they could onward integrate to other APIs/data sources through sheets.
There’s an irony in the fact that the more the slips that are covered by the press and the Twitter bubble chamber -the more desensitised everyones’s getting to them - no matter how outlandish, offensive or inappropriate. Leading to a more dangerous position where there’s a growing acceptance of the obvious fallibility of current LLMs.
Maybe Google is right to be cautious. In fact, they almost certainly are, I hope they hold strong for a while longer at least.
Not being a software engineer I’ve been relearning to code again, and after a deep dive dived into a new framework. Which was a headf*k to put it mildly including server config etc.
What I re-learned was the “arc of despair” that initial period of complete disaster where everything wasn’t just confusing but incredibly difficult. And many many times of asking myself of whether this was worth it and whether I shouldn’t just pay a “professional” to do it. Then within days I was swimming fluently. Was a strong reminder of the various stages of being auto-didactic. The need for faith :)
And yes, in the end it was worth it, because the lateral connections and first principles you learn (complete with new learnings and questions) allows you to go much deeper than simply paying someone else to do it. Was a reasonably profound experience.
It was a few years ago when Google got absolutely panned for claiming they could call a restaurant up and book a table using an LLM + voice. And here we are where it now feels not just acceptable but fun.
So starting point is identify key trending global news event eg major terrorist attack, election, major domestic incident, maybe even pop cultural. And then rather than relying on disparate news sources to update, bring them all into one place.
Yep, that’s sensible, if not easy. What about rumours? Is there value in adding news that the major publishers can’t, given the diligence they must apply, but here could be brought in, albeit with a clear UI that makes it clear it’s unverified.