Live data from Hacker News

Ask HN: What weird or hard problems are you trying to solve?

news.ycombinator.com

411–420 of 831 posts

Re: Ask HN: What weird or hard problems are you trying to solve?

#411

Trying to find a solution to the maximum-entropy probability distribution Q(x,y,z) constrained to reproduce the marginal distributions P(x,y), P(y,z), and P(z,x) from some other distribution P(x,y,z). It is known that Q takes the form Q=a(x,y) b(y,z) c(z,x) for some functions a,b,c to be determined by solving the system of equations: P(x,y) = sum_z Q(x,y,z) P(y,z) = sum_x Q(x,y,z) P(z,x) = sum_y Q(x,y,z) It's not cle…

oh this is neat, it's like adding an additional dimension to the Wasserstein distance / optimal transport problem. Well, in the sense that you are using marginals as constraints. Kantorovich won a nobel prize for this kind of stuff, so it's definitely hard.

Re: Ask HN: What weird or hard problems are you trying to solve?

#412

Earlier quoted context omitted.

So that story shows the potential for long term damage (his store hasn't closed yet), but I agree that if it does, it would be the best example I've yet seen. However even if we assume the worst outcome in that situation, the negative impacts of cancel culture are tame compared to a lot of other systems. If we're calling for an end to cancel culture due to the one case of permanent damage, why aren't we calling for a…

There are a few reasons I’m particularly concerned about cancel culture. One is that there is a mechanism for me to change laws I disagree with. There’s lots of things I hate about the justice system that I believe are being worked on. But fundamentally, I accept I was born into a particular society, and I’ve implicitly agreed that I need to agree to certain rules. Cancel culture is mob justice. There’s no mechanism…

> One is that there is a mechanism for me to change laws I disagree with.

There are also mechanisms to address culture you disagree with (and you're exercising them!). The question was not why do you find cancel culture distasteful, nor was it even why do you personally find cancel culture potentially worse than unjust policing (which for now let's just agree to disagree on), but why it is that you are prioritizing the push back against cancellation over the pushback against unjust policing.

At this current moment, it is, I think, clear which unjust system causes more harm. It is the criminal justice institution. That cancel culture could grow worse is feasible, but it has not yet. People aren't routinely killed at the hands of twitter complaints.

> Finally, I don’t think we can trivialize the impact of tossing the idea of free speech out the window.

Here we disagree on premise: cancel culture is the result of people who previously did not feel empowered to speak freely taking advantage of a system that raises their voices more prominently. That it is extrajudicial is a failure of the institutional justice system, which continues to systemically fail underserved communities (women, minorities, poor people). Sharing controversial opinions has never been without risk, that's why they're "controversial". There have been privileged sects of society for whom the risk of holding controversial, and even outright despicable, opinions was low. I don't think more equality in that regard is a negative thing.

> In the example, blaming one person for the actions of a family member goes totally against the philosophies I believe in.

Do you believe the situation would be different if the employer had been unrelated to the girl who posted bad things? While the exact numbers might be different, I don't see the overall picture being that impacted by him being her father. And "punishing" an employer for the actions of an employee, while still perhaps fraught in some cases, is much less worrisome to me than targeting family.

Re: Ask HN: What weird or hard problems are you trying to solve?

#413
I'm building cad software for designing atomically precise objects.

I don't expect atomically precise manufacturing (at least in the form of molecular nanotechnology) to show up for at least a decade or two.

But the software to design things that can be built using nanotech can be written before the technology to synthesize them exists.

Re: Ask HN: What weird or hard problems are you trying to solve?

#414

Writing a computer program is often the best way to convince yourself that you really understand a problem. And an already-written computer program is often a good way to document the behavior of a complex system because you can play with it to see what happens in all the edge cases. So I'm writing curricula that use computer programs as the primary teaching tool. One is for computer science, where the idea is that a…

oh that's a neat idea! It's the next natural step in pedagogy. In my applied math undergraduate program we had a substantial programming component to better understand the concepts.

I like it.

Re: Ask HN: What weird or hard problems are you trying to solve?

#415
post #37

Earlier quoted context omitted.

I am optimistic :) The entire curriculum is extremely easy when properly explained at correct pace.

You could use automatic homework as a reward. Basically replacing broken curriculum with your own until they catch up. (Or even beyond that: homework takes time even when you know everything.)

Intriguing...

Re: Ask HN: What weird or hard problems are you trying to solve?

#416
post #374

Earlier quoted context omitted.

How is this different than a heavy-weight ORM like entity framework?

What I understood is that this like what TypeScript is to JavaScript (or to be more accurate Haxe to the few languages it transpiles to). It's a new language that will transpile to different SQL syntaxes instead of the method chaining approach of your typical ORM (that'll likely also include model abstractions over your tables and maybe also migration managing).

yea, I read it the same way, but ORM's like EF let you leverage an existing language to do basically the same thing as this does. Abstraction, data types, etc... I am unsure as to the value of something like that unless it is included with a client application that handles the conversion natively.

Re: Ask HN: What weird or hard problems are you trying to solve?

#418
post #289

Earlier quoted context omitted.

Take your child back to first grade and systematically re-test his knowledge. Identify where he got lost, catch up, and then today's assignments will not be as difficult. The child is likely stuck on something that wasn't thoroughly understood in the past. I did this for a child recently. He was in the special needs program. I looked at his homework and re-tested his math understanding. It turns out he had no solid u…

I totally agree that missing foundation is a huge, compounding issue, and that it can be solved with “surgical education”. I used to do this as a tutor. You dig deeper down the stack of fundamentals until you find something that’s missing, and then you build the stack back up. That being said, surgery is hard and can be traumatizing if taken on haphazardly. One of the harder problems, in my experience, is if an adole…

Exactly. Literacy will be the second prong. I want to annotate Strunk & White with background information. Writing the grammar engine will be difficult.

While reading S&W, I had to look up many terms that were assumed knowledge.

Re: Ask HN: What weird or hard problems are you trying to solve?

#419
A piece of music playing software that can react to the lead of an instrument. I'm picturing the software will be able to play a concerto (or simply a duet) with a real instrumentalist and react to dynamic and tempo changes in real time, like an orchestra under a conductor.

The same principle can also be used to create a real-time software harmonizer [1] for live performances, but this problem already has a reliable solution through hardware.

[1] - https://www.youtube.com/watch?v=DnpVAyPjxDA

Re: Ask HN: What weird or hard problems are you trying to solve?

#420

(It's a germ of an idea but I hope to work on this!) Problem: Public schools in India don't do justice to students. Private schools in India charge a bomb but most of the money ends up in the hands of the "owners" and not enough to teachers (For reference, an average primary school teacher earns lesser than what an Uber driver earns) . The solution : A network of "not-for-profit" schools where the fee structure is re…

I was thinking about something very similar. like, the teachers will get a lions share of the fee. this will instantly motivate teachers to better themselves, and move the school away from for-profit model.

but there are some real problems to solve. like, 1. how to get such school started ? who will provide initial capital ? 2. even if such a school is started, how to stop it from becoming for-profit again ? 3. is there a healthy balance between 'for-owers' and 'for-teachers' model ?

Post reply on HN