wrhn

Bill_Dimm

Joined 14 years ago

Karma
616
Submissions
207
Member since
Jun 11
The founder and CEO of Hot Neuron LLC. He developed the algorithms for conceptual clustering, near-dupe detection, and predictive coding used in the company's Clustify software. He has experience in the development and application of mathematical models to solve real-world problems in theoretical physics, mathematical finance, information retrieval, and e-discovery.
View on Hacker News

Recent submissions

Comment · 7 years ago
Very interesting statement, considering that Sargon (the guy Patreon banned) was directing his offensive language AT the alt-right (i.e., he is against the alt-right, not part of it).
Comment · 9 years ago
I don't think "100% in-house techs" is right. I have Comcast Business and my line was recently cut (probably an accident by apartment maintenance). The tech who came out and put in a temporary line was from Comcast, but he said a contractor would come out at some point in the future to put in a permanent one (hence, he could not guarantee that I could get a notification when it would happen).
Comment · 10 years ago
Actually looking forward to MS's UX folk cleaning up LinkedIn.

Along those lines, maybe they'll fix the very broken email alerts for group postings. For quite a while, my experience has been that if I have it set to weekly digests I do get them weekly as expected, but if I set a group to daily digests I get nothing at all. So I get nothing at all for the groups that I have the most interest in. I opened a support ticket about it but got nowhere with it. I don't know if this impacts everyone or just some subset of users, but I've seen others complain about it.

Comment · 10 years ago
I've found Vischeck to be helpful: http://www.vischeck.com/vischeck/
Comment · 10 years ago
Here is the breakdown for the quants at the bank I worked for in 1995-1999 (I may have forgotten a few people):

Cornell physics PhDs: 3

Harvard physics PhDs: 2

Princeton physics PhDs: 2

Univ of Penn physics PhD: 1

Other physics PhDs: 2

Cornell applied math PhD: 1

Other applied math PhD: 1

Cornell mechanical engineering PhD: 1

Univ of Chicago finance PhD: 1

So, if by "dropout" you mean that they didn't complete their PhDs, then no, that's wrong. If by "dropout" you mean people leaving physics, then yes, it's very common (especially for theorists -- all physicists on the list above where theorists except maybe one or two).

Comment · 10 years ago
The fpr and fnr should be swapped (note: I haven't read the rest of the paper yet). Specifically, it should be:

ROC curves plot (1-fnr) as a function of fpr.

Comment · 11 years ago
For the few that I looked at, clicking the "Description" tab tells you almost nothing. Clicking the word "Help" in the upper right gives a much more detailed description.
Comment · 11 years ago
DSL is available without telephone service, at least with Verizon in Pennsylvania. It is called "dry loop DSL." 3 Mbps / 768 kbps is costing me more than the price they are advertising for 50 Mbps / 50 Mbps FiOS (which I can't get) without contract. So it exists but it sucks, and I'm not sure you actually save much (or anything) by getting DSL without telephone service.
Comment · 11 years ago
About 20 years ago, when I was a grad student, I attended a colloquium where the speaker (a physicist, for better or worse) had computed the number of people that the planet could support if everyone consumed natural resources at the rate that Americans do. His number was 2 billion. I don't remember who the speaker was and I don't know what assumptions he made -- I don't know how much error there might be in his result or how improvements in technology might change it. If his number is even close to being right, "global populations will peak in my lifetime" isn't even close to being adequate unless you expect the bulk of the population to be satisfied with remaining poor while a small proportion of the population devours the planet.
Comment · 11 years ago
Neat. It would be helpful if you could go in the opposite direction (enter RGB values and get the name of the closest color). I find that when I'm making a chart or graph and I cook up some color in GIMP, I don't know how to refer to that color in the text of the document that references the chart.
Comment · 11 years ago
In a similar vein, but with physical mail, I made a donation to the local PBS TV station years ago during one of their fundraisers. They sent me a postcard with my name misspelled in a way I had never seen before. I then received solicitation mails from several other charities with the same misspelling. Nothing says "thank you for your donation" like being sold out for a few extra dollars. I never donated to them again.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton