Yeah I managed to get the file into the bucket. However I couldn't get the permissions to let it be publicly visible despite working through several tutorials.
I ended up on Netlify, it's 1000x more my speed.
305 karma · joined April 21, 2019
I ended up on Netlify, it's 1000x more my speed.
I spent 3 hours trying to get a bucket to host a static single page of html and failed completely.
I use amazon polly. I wanted to know how many characters I was using each month. I spent 2 hours searching through hundreds of pages and literally couldn't find that information.
I thought of trying to start a little text to speech service for dyslexics to make it easy to use Polly but one of the main thing putting me off is having to get my arms mangled in the AWS machine.
The whole thing is so totally maddening. I would love to be able to sit in on their meetings where they talk about usability, what do they say? Do they think everything is fine? Do they know it's totally broken and don't care? Are they unable to hire a UX designer? What is the problem?
How do you know if you hired well? Is this an excellent result or a terrible one? How do you know how well each of those 5 did and how well the team would have done with other people in the roles?
Like imagine 20 robots all trying to do some task, each one learning information as it goes. How much time should they spend doing the task vs how much time should they spend communicating with the other robots? (Assuming they are restricted to one or the other)
It's not really a solvable question. It's possible some other robot has a piece of information that would make you much more efficient, however you can't know if they have that information without asking them.
It's possible that taking the piece of information you just learned and gathering all the robots and sharing it with them will help many of them become more efficient, however you don't know if that's true unless you try it.
How can one of the robots work out the value of a piece of information to another robot? Surely all you can do is estimate?
Clearly 0% communication and 100% communication are terrible ideas. However working out which pieces of information to communicate and when is not particularly optimisable, you just have to muddle through as best you can.
https://arxiv.org/pdf/1608.02644.pdf
I'd really love to see someone like Deepmind get a hold of this. They are perfectly placed to apply a lot of resources to it and it's right in their wheelhouse.
Don't know how to contact them about it though.
I imagine there will be hold outs who will never switch to formal proofs, though they may simply take on co-authors who are just "formalizers".
I also am probably being over confident with the time scales, I think a process with a tipping point, everyone will use it once everyone starts using it.
However when that tipping point will occur is probably hard to predict. I think it will be soon, but I'm broke and not that sure.
https://jiggerwit.wordpress.com/2018/04/14/the-architecture-...
The two prerequisites are 1) making a fluid and intuitive interface for inputting proofs which can also display human readable proofs for any theorem known and 2) creating a database of all known mathematical proofs into which new results can be inserted.
Both tasks are ~50% complete. Metamath, for example, has all of mathematics formalised up to a basic undergraduate level. It has some reasonable editing software, but not close to good enough for mass adoption. There are also plenty of other systems being developed.
The time will come though relatively soon.
If everyone who used spotify was paying 79p per track you would see massive economic growth in the music sector. However because it's an all inclusive streaming service not a huge amount more money is moving through it.
However as a consumer I am getting 1000x more music now than when I was 15yrs old. That doesn't seem to get recorded though.
So much stuff on the internet (like stack overflow or github) has gargantuan economic value but doesn't figure in GDP stats because it's mostly free.
Though that does not take account of the latency of the webcam itself.
The wider and more frequently the drug is used the faster this process occurs?
1. Advances in fundamental science are what ultimately dictate the growth rate of an economy. For example when you discover quantum mechanics you can make lasers, circuit boards and therefore computers which boosts everything enormously.
2. The more science you have done the harder it is to do more. You can only discover a truth once. For example discovering new elements is extremely hard now compared to when Hennig Brand discovered phospherous in his urine.
Consequences:
1. When a society gets the scientific method right it goes through an S shape curve (logistic curve), where science accelerates for a while and then plateuas out.
2. When you hit the plateau of the curve the maximum growth rate of the economy becomes very limited, it's just not possible to have new ideas which are worth developing very fast.
3. Human labour still provides a surplus so what you get is more and more resources piling up with nowhere to invest them. Look at Apple's $245bn cash pile, if they had new ideas they would put that money to use.
This will cause bond yields to fall to almost nothing and loads of money to be pumped into any startup with a vague hope of accomplishing something.
If we are in this situation it would also mean that the 08 crisis is not needed to explain the slowing growth in the world, things are slowing because we don't have many new ideas.
Common objections:
1. But what about X invention that happened in the last few years? Invention still happens on the plateau, just slower, NN's, crispr, exoplanets, smartphones are cool, they're not relativity, quantum mechanics, electrification, aeroplanes etc.
2. Scientific progress is exponential! : Accurately fitting a curve to a logistic curve gives you an exponential up until it shifts on you and starts slowing.
3. China and India are still growing strong. : Essentially they are still deploying the previous discoveries. China's growth is slowing over time precisely as it has less and less discoveries to deploy.
More to read:
DOE finds it can't get supercomputers like it used to.
https://www.nextplatform.com/2019/05/06/doe-on-collision-cou...
https://www.theatlantic.com/science/archive/2018/11/diminish...
https://slatestarcodex.com/2018/11/26/is-science-slowing-dow...
Personally I think programmers who are "too clever" are cancer inside a codebase. Sure you saved a bunch of keystrokes but your code golf has locked the code forever unless someone is willing to rewrite it. This is not a bad example but in general I think people who strive to do things in less characters are highly problematic.
Luckily there's bacteriophages waiting in the wings.
If I have 20% more APM than you do I beat you even if I totally pick the wrong units and build order? Could a super smart bot win against pro's even if it's limited to 20% less APM?
So yeah it might be interesting if the bot gets superhuman to trail several versions with progressively lower APM caps to see how much you have to limit it before humans can beat it again.
Here are a lot of examples of good contributions by older mathematicians.
https://math.stackexchange.com/questions/1059235/great-contr...
I think one factor is that young people tend to be hungry to achieve something and also aren't burdened with family or admin which means they can work harder.
I keep trying to tell people Woking is the Vienna of the South East.
For example if a 25 yr old coder is $X per year surely, even if you have terrible age discrimination, there must be some salary level, say $0.75*X where having the older worker is preferable?
Is it that older workers are much more expensive and are bidding themselves out the market? Is it that people value youth so much they will pay a 3x markup for it or something?
If there's a large pool of super skilled people begging for work I don't quite understand why someone wouldn't want to leverage that. I get why maybe you pay older workers less if you discriminate, but to not hire them at all seems strange.
They were saying there is a critical mass where procedural tools feed into procedural tools and a lot of the pipeline is automated, then it becomes extremely cheap to make more content.
So back to your point: I think this is where we will see AI more in game dev, it will be as smart tools to empower creators to make stuff more easily. I don't think the AI will make the game, I think it will assist the humans with the creation of it.
Here is a talk about Spiderman and how much proc gen they used there https://www.youtube.com/watch?v=4aw9uyj9MAE
In 10 years I think every mathematician will use formal methods, it will be like typing your proofs with a type writer not to. Moreover the problems are very amenable to smart tools assisting in the work, there is a smooth path from where we are now to a fully ai mathematician.
There are a lot of positive things that can be done, for example educating women.
I'm not sure if anyone has seen these calls to action but it sounds like there are problems ahead with such a large population.
https://www.worldpopulationbalance.org/call-action-scientist...
Working at a reasonable, sustainable, pace requires big discipline. It's hard for a computer game studio with a release coming in 12 weeks to stick to 40 hours because that will get them 12 great weeks. They'll tend to slide into 80 hour weeks in panic but end up doing less.
I think two real issues are that you can't compare against a control version of the company and you can't tell you are slowing down because you are tired.
Another issue is management. Most managers think hours in means more productivity out, it's hard for them to not fall into that seeing as it is so widely held as true.
IMO people who work a large number of hours each week get cognitively slowed. They can't tell they've slowed down (partly because they don't have a control to compare against) and so think that if you do 80 hours per week you'll do twice as much as if you do 40 hours per week. You won't, you'll do less.
Executives who think the best way to measure business performance is by how much your employees are sat at their desks are idiots.
https://en.wikipedia.org/wiki/The_High_Frontier:_Human_Colon...
Intel and Nvidia have a hardware advantage. Ford, GM, Tesla etc have the advantage of making their own vechicles. Google is Google.
What do Uber and Lyft have that makes it likely they can figure it out before the market is conquered by others?
Maybe their assumption is that if driverless tech is widely available then their brand names will get them more business?