rfergie

5 ms·

1.
by rfergie 3mo ago
> there are students whose performance on such assessments is not a good signal of their intellectual ability Is there a form of assessment that is a good signal of the intellectual ability of all students?
2.
by rfergie 4mo ago
> Probably 80% of the recent PhD grads I know are looking to leave academia, despite the fact that they went into it to pursue a career in academia Has this changed recently?
3.
by rfergie 2y ago
For most taxes you expect higher earners to pay more but this is not the case with student loans because high earners pay of their loans quickly whereas lower earners end up paying far more in interest. An actual graduate tax would be far l
4.
by rfergie 2y ago
> so either none of the current competitors are doping or all of them are The drafting effect in cycling means that a clean cyclist can finish very closely behind a doped cyclist
5.
by rfergie 2y ago
https://pypi.org/project/formulas/ might help with this without needing the OpenAI part
6.
by rfergie 3y ago
> It seems like ClosedAI wants their system to follow the desires of some people and not others, but without describing which ones or why If the problem is "unaligned AI will destroy humanity" then I'd take a system aligne
7.
by rfergie 3y ago
> What might be seen as exponential progress could actually be the accumulation of a few tremendous leaps This is what exponential progress is as long as the leaps continue happening
8.
by rfergie 4y ago
Thank you for the link to the paper; very interesting!
9.
by rfergie 4y ago
One construction of the natural numbers that I've seen is 1. There exists an element 0 that is a natural number 2. For every natural number there is a "sucessor" that is also a natural number. (i.e. if n is a natural number t
10.
by rfergie 4y ago
I think that at a networking event 1-2 minute conversations are normal so this isn't a sign you are doing something wrong
11.
by rfergie 4y ago
Don't you just end up on a "meaning treadmill" instead?
12.
by rfergie 4y ago
> There’s a huge difference between being there from day one vs day 2,555. Between being a founder and an employee who joined later on This is true but there is also a huge difference between 97% and 3%
13.
by rfergie 4y ago
Just saying "36000 die" by itself doesn't tell me very much about how dangerous it is - expressing that as a percentage gives much more information I think. As another person points out 0.03% is 3 times the risk of dying in
14.
by rfergie 4y ago
36000 Mexicans is 0.03% For more context https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r... has the risk of death by traffic in Mexico as 0.01%
15.
by rfergie 4y ago
> some of them talk about their gorgeous cars for the sake of talking about their cars > My coffee machine is all mine (Gaggia Babila)
16.
by rfergie 5y ago
Do a search and the websites that rank at the top are the optimised ones
17.
by rfergie 6y ago
so I took out a few Google Ads on companies in the same space This is considered an aggressive move. My guess is their CEO thinks he is responding in kind
18.
by rfergie 6y ago
The long slow SAAS ramp of death
19.
by rfergie 6y ago
I am very surprised by this. I suppose somewhere like Ben Nevis maybe 90% of the people you meet won't have a map and it is possible that places like this will represent a large proportion of the total people you meet when doing a roun
20.
by rfergie 6y ago
Ridiculous! It would take longer to get to the moon and back if you walked there. There is no place on Earth that you couldn't get to quicker if your budget was the same
21.
by rfergie 6y ago
It has been a long time since I've studied this; can you link me to something more about the reals being consistent and complete? Or write a quick explanation in a comment? I thought there were undecidable propositions in the reals (e.
22.
by rfergie 6y ago
I took his point to be that things like "feels readable" and "feels fast" and to do with feelings rather than actual quantifiable things that could reasonably be compared between languages. So it isn't that those
23.
by rfergie 6y ago
It is a signal the the person isn't top talent. Because if they were they wouldn't be desperate
24.
by rfergie 6y ago
What is crippleware? Does it mean it doesn't work?
25.
by rfergie 7y ago
You need a whole industry of people telling your customers that these things do work. I think it is quite difficult to accomplish entirely by yourself
26.
by rfergie 7y ago
Newish Polar bands (H10, maybe H7) send data on each beat. There are apps that use this to track heart rate variability (the variance in the gap between heart beats) which can be used as a measure of "readiness" in athletes
27.
by rfergie 7y ago
https://ico.org.uk/for-organisations/guide-to-data-protectio... says you only need a data protection officer in some circumstances. You don't say what your startup does, but unless you are processing a lot of data
28.
by rfergie 7y ago
I don't understand? Are you talking about days in winter where the sun doesn't rise at all?
29.
by rfergie 8y ago
Not qualified to speak on this myself, but I asked a similar question to a friend and he said something like: Part of good software development is knowing where and when to create abstractions. Category theory has a collection of interestin
30.
by rfergie 10y ago
I do some data science type work for a charity. This is only part time but I'd imagine larger charities could employ someone full time at this. The work is not cutting edge data science at all but I do find it meaningful. Happy to answ

More ›