bobm_kite9

7 ms·

31.
by bobm_kite9 5y ago
The law of large number works well for some things (say, shoe sizes) and fails to work for others (e.g. income distributions, which follow a power-law). So it's not odd at all that it would fail to adequately account for people. It&#x
32.
by bobm_kite9 5y ago
My takeaway is that danluu is saying that, at the small scale, people understand how individuals contribute to teams. But at the large scale (like the HR department) we don't. At the small scale, we're able to apply systems thin
33.
by bobm_kite9 5y ago
I just need to write a GAN to generate plausible looking irises, then - Profit!
34.
by bobm_kite9 5y ago
Can't someone turn this into a feature of HN?
35.
by bobm_kite9 5y ago
I guess I'm interested to see if this performs qualitatively better than GPT-3, given how many more parameters it has. However, I think this is really a dead-end: throwing more hardware at this is just going to generate better-soundin
36.
by bobm_kite9 5y ago
If you’ve ruled out mental causes (like perhaps ADHD, as discussed in other comments) then perhaps it’s something about the work. I’ve often found in these periods that there’s some underlying part of the work that makes me not want to do i
37.
by bobm_kite9 5y ago
> Notably, GPT-3 is licensed by Microsoft, and is therefore closely guarded. You interact with it only via apps which act as oracles. It’s basically a genie stuffed in Microsoft’s basement you can Zoom with. This brings to mind Eliezer Y
38.
by bobm_kite9 5y ago
I think you're right. This is discussed in the article- GPT-3 does very well at the small-scale, like short poems. The architecture of GPT-3 is based on Word2Vec, and attention - it doesn't really have a good understanding of l
39.
by bobm_kite9 5y ago
I have no idea why this has been down-voted. Sure, it's a strong opinion but that's what HN is all about. Couple of questions: - Do we have any estimates yet on how long olivine would take to be effective? - Can we do HFC-capt
40.
by bobm_kite9 5y ago
Is it just accidentally?
41.
by bobm_kite9 5y ago
Is there any explanation of how this works, beyond ‘ai’? I mean, I understand (I think) how Bert/gpt works, but I can’t quite see how you would train this.
42.
by bobm_kite9 5y ago
Hell yes. After 20 or more years doing this, you build a reasonable bullshit filter. A lot of the time, if the ask doesn’t pass that filter I find it very, very hard to commit. It’s hard enough being useful without that. Life is too sh
43.
Story Points(riskfirst.org)
1 points by bobm_kite9 5y ago|0 comments
44.
by bobm_kite9 5y ago
This really implies that the whole idea of chrome extensions is broken: a mix of attempted malware and adware. It seems like they should be more carefully sandboxed in some way. Are there any proposals for fixing this?
45.
by bobm_kite9 5y ago
Lots of people here saying they don't see the point of NFTs. I totally agree for digital artefacts like this video. After all, the original can be copied perfectly as many times as necessary. Wouldn't they actually be more useful
46.
On Story Points(riskfirst.org)
4 points by bobm_kite9 5y ago|0 comments
47.
Emotion Typology(emotiontypology.com)
1 points by bobm_kite9 5y ago|0 comments
48.
On Story Points(riskfirst.org)
2 points by bobm_kite9 5y ago|1 comments
49.
by bobm_kite9 6y ago
Another way to think of this: imagine two equally-sized countries A and B. A implements Sam’s suggestion, taxing everything at 2.5% pa, B doesn’t. In A, people are happy, we’ll fed, Pursuing their interests, living meaningful lives. In B
50.
by bobm_kite9 6y ago
What about the 51% attack problem? Why doesn’t this matter? Can someone explain how this currency is a good bet when there is this seemly huge gaping hole in the system. Honest question!
51.
by bobm_kite9 6y ago
This is generally true. However, sometimes I like to ‘bury’ a hard-to-answer question in a follow up. The reason is, if you ask the question straight off, people have the option of pretending they haven’t seen it. They can’t do that if
52.
by bobm_kite9 6y ago
This looks really nice, but what’s to stop it getting blocked like all the other trackers once apple/uBlock/etc. add it to their database?
53.
Covid, Agile, ChatOps(riskfirst.org)
1 points by bobm_kite9 6y ago|0 comments
54.
Risk-First Software Development(riskfirst.org)
135 points by bobm_kite9 6y ago|40 comments
55.
by bobm_kite9 6y ago
I think COVID is finally killing off what is left of scrum: https://riskfirst.org/presentations/BotsChatOpsCovid/chatops...
56.
Agile, Bots, ChatOps and Covid(riskfirst.org)
1 points by bobm_kite9 6y ago|0 comments
57.
by bobm_kite9 6y ago
This is exactly right - and that initial website was written during the "peak of inflated expectations" part. I still think that chatbots provide an entirely new way of interacting with software. We're very used to dividing
58.
by bobm_kite9 6y ago
I’m not sure I understand the point. This is to somehow make developers want to fix bugs, rather than build features, right? It seems like an odd way to go about it, but kudos to them if they’re all enjoying it. The problem will come when
59.
by bobm_kite9 6y ago
Can someone explain what's going on here? I feel like this might be really interesting but I don't quite get the notation
60.
by bobm_kite9 6y ago
This article is very light on explaining why this might be necessary for the police. Does anyone have any suggestions as to why this advice was issued?

More ›