I would say 300 is more than slightly longer. Varies by institution and style, but 100k words, 4-5 collated papers, 100-150 pages maximum are quite common.
I would be concerned as an examiner if this came across my desk, more so if I read the colophon where the author comments that they intended to write their own typesetting system, before reading Knuth and wisely concluding that they were unlikely to do anything better than TeX. Top tier yak shaving there.
How do you ensure people don't think you're not good at your job, if you continually propose poor solutions? Since you're emailing other teams here, they probably(?) don't get to work with you day-to-day to realize you're doing this on purpose.
Also, what's the point of Y being a poor solution? Even if the other team isn't very responsive, can't you always propose your
> best attempt at solving it
such that even if they don't respond, you can implement a decent solution?
> How do you ensure people don't think you're not good at your job
Great point. In the beginning, it was fine because I was new and didn't have domain knowledge. As time went by, I had decided I hate the team/job and I need to find a new job anyway.
> Also, what's the point of Y being a poor solution?
In this scenario, I was never in a point of knowing if my good solution was good enough. My best attempt may be something that would get me 80-90% of the way but needed small modifications.
People speed reading my email might not see the nuance and think I already had the correct approach, and wouldn't respond. That risks going forth with a problematic solution. If, however, you say something that is clearly wrong, people are quick to respond.
More concisely, if I go with my best approach, and no one responds, I still don't know if my approach is correct or not. The point of the email is I need their feedback. So I have to set up the situation to virtually guarantee I get their feedback.
(Another approach I would do is list 3 options I was thinking of, and make the default one a poor one - at least this way they know I'm capable of thinking up good solutions but haven't developed judgement yet).
> such that even if they don't respond, you can implement a decent solution?
The goal is to elicit a response. If I say clearly "I need a response", then I'm not going to get it.
Again, do this only with problematic teams. This isn't a guide for normal work. For normally functioning teams, I would recommend going with the best approach and making that the clear default.
I think it's just to hype Anthropic. Check it out, we have products so dangerous the government banned them, we must be so advanced. (Their competitors cannot make such a claim.)
You think Dario called up Andy Jassy and told him "Hey, we're trying to get Fable banned, so can you please go talk to the government and tell them that they need to ban it"?
He's always talking about how dangerous AI is, how the models he's building could be used for cyber attacks, and how if his company is successful then at least 50% of the white-collar workforce will lose their jobs.
Doesn't seem that unlikely he might say something like that.. Unless he's super-villain evil it sounds like he believes the government needs to do something?
wrt. the end of the story, it will be interesting to see if people start noticing their Dunning-Kruger bias as a result of LLMs.
Specifically: LLMs make it really easy to misunderestimate the complexity of fields other than your own. (You can see this with a lot of vibecoded projects, for example – once they hit the wall of complexity, they stall out or start finding ugly patches for fundamental design issues, etc.)
I don't think this sort of cultural change will happen short-term, though.
> LLMs make it really easy to misunderestimate the complexity
In my experience this is a real problem. Just yesterday I asked my LLM to create a piece of software that could help me build an 'ambilight-like experience' through my home assistant. It did something that seems to work as I expected, but there is a lot of theory that I just brushed past. It would be pretty easy for me to assume that I would be able to replicate this feature from scratch 'now that I understand the problem'.
Agreed. LLMs are really terrific at sounding like they know exactly what they are talking about. Fable is the best yet. Beautiful, thorough explanations with absolute certainty, which under even light scrutiny turn out to be mostly bullshit.
I still love the tool, but remain as convinced as ever that AGI does not lie at the end of this particular path.
e.g. attribute, increase, conduct
reply