774 karma · joined September 22, 2020
Was quite surprised something like this is possible and I love seeing new crypto tech on HN.
1. anything humans decide to do should be accepted because of invisible positive outcomes (who could have predicted Internet from oil combustion),
2. the moment negative outcomes become visible, let's wait until it becomes a problem, any attempts at prediction when it will become a problem are futile and fearmongering,
3. humans can solve any problem they encounter.
All three can be used to create an argument for solar/wind. Who knows what kind of positive stuff will solar/wind create (let's see), whatever happens (due to inherent volatility and energy storage problem of solar/wind) let's wait until it happens and then see, we are entirely capable of solving the issue.
0: https://nepis.epa.gov/Exe/ZyNET.exe/901V0100.txt?ZyActionD=Z...
Permafrost is speculation when it comes to time, not the size of the effect. There's no natural process that will reverse the release of 100 years of 2020 greenhouse emissions in a short amount of time.
There's no scientific way to demonstrate that human activity increases the planet's temperature -> therefore no reason to worry, but there's also no scientific way to demonstrate that there are interventions that work to stop the increase -> so no reason to worry?
Of course, the question of risk is asked before additive intervention, so in a way, combustion of oil should have never been done in the way it was, similarly raising of livestock (global pandemics), but the way unscientific think -> no data that demonstrates risk -> no risk at all. Then after the risk surfaces, we ask the question "Why should we change the way we live just because we became aware of the risk?". Sounds worrying to me too.
For example, we can estimate correctly (within the same order of magnitude) the emissions of CO2 equivalent gases a year that we do as humanity.
Then we can estimate correctly (within the same order of magnitude) the CO2 equivalent gases stored in permafrost or ice. [0]
Now we can ask a question. What if the ice melts? How many years of 2020 human activity is stored there. It ends up being more than 100 years of 2020 emissions.
It does not really take that much modelling to realize that a temperature increase that melts the permafrost can result in completely unpredictable outcomes, that are very probably not positive.
That is of course, if one believes in human caused climate change. The amount of CO2 equivalent gases that is emitted by everything else (not humans) is magnitudes more, but obviously it was not a problem for the Earth to handle before. Now, the n-th order effects will increase the rate of warming and will cause new invisible effects.
There's so much trash everywhere that is earning millions (book reading apps, silly chart-db software, proprietary SQL databases etc.).
I was quite shocked by the levels of disfunctionality in the software engineering sections of the company I worked at yet still, the company has huge success. We basically got requests from the product team to make something work in a day or two and even though it was completely horrible code that absolutely paralyzes us at some moment in the future, the company still reigns.
Constant time feasibility check of inserts and linear time update after insert.
Of course, there are tricks that support multiple simultaneous inserts and tricks that minimize the update of the data structure after insert.
But, if you want to do 2-opt (k-opt) with time windows, then you have to work out the details because no one in academia did.
Helsgaun's reports are pretty descriptive when it comes to LK algorithm.
But yes, I do agree that the original writing and the way the algorithm is taught in universities is a little bit cryptic and sometimes a completely different algorithm.
Simple explanation of the algorithm is:
1. 2-opt swap (A .. B -> .. -> C .. D ===> A .. C <- .. <- B .. D) - constant time compute of everything to check feasibility and new cost,
2. recursive 2-opt - now between C .. B with bounds (you can figure out which swaps are inferior and not recurse).
That's all there is to the algorithm. Helsgaun improves it a bunch (with preprocessing, deeper and more efficient k-opt moves, more efficient datastructures) but keeps the same idea.
There's another very flexible method called "ejection chains" started by Fred Glover and is very effective if you couple it with efficient data structures but also, very cryptic. It easily supports time windows, breaks, flexible work hours, flexible number of vehicles etc. Of course, the latest papers in "ejection chains" show no such thing :D
Sometimes you can stumble upon PhDs publishing their code and the tricks do not exist at all.
Just recently there's been a nice categorization of "timing problems" https://onlinelibrary.wiley.com/doi/abs/10.1002/net.21587
and there they try to categorize the time complexity of solving these problems and the tables are just not as up-to date with the industry. Some O(n log n) are O(n) or even better (depending on the local moves you do).
But it's a good start. A lot of these subproblems can be implemented really fast on a CPU and I guess the moment they are explicitly solved in a software library that's when the heuristics research will improve.
But yes, other than a good soy milk, most of plant-based milks are just empty calories.
I'm just laughing at the inconsistency. If people were really consistent in their views of (un)justified waste, I just don't see the same uproar to those other magnitudes more wasteful things (like eating meat, spending a life playing video games etc.).
It's like being mad about one of the fastest functions in your code when there's many others to optimized.
I also do not want to talk about the philosophical side to taking a life of an ant (delve into what waste is justified or not, and which degrees of dopamine response would justify it).
I was making an argument that deciding what is a valuable human activity does not include wastefulness at all. (OP was implying that someone from the future will value these mining efforts in a regretful way)
People are having fun earning money by burning gas and mining Bitcoin. There are so many other equivalent activities that do not receive any uproar but are even more wasteful.
For example, eating meat is insanely wasteful. Yet you will find people defending it.
Wasting energy is the thing we do for various reasons. The underlying "regret" that someone from the future should have due to it not being productive or useful is completely irrelevant. We rarely do things because they are energy efficient.
The only reason why I'd come somewhere is if I like the engineers or if I'm in need of money. If I need the money, I won't ask a thing.
Humans around the world are consuming food, heating/cooling themselves, all in the service of some gamified activity. Some of these activities are as useless as mining Bitcoin. Yes, these activities aren't burning through energy, but people are burning through energy, with the sole focus of doing the activity.
Any justification for these activities can most definitely be used to justify people wasting energy on mining Bitcoin.
Energy is being wasted everywhere on much sillier things at much grander scales.
If we ever make it to a cashless utopia, I'm pretty sure someone's going to look at us and see how much untapped human potential was wasted on years of watching a screen and experiencing dopamine release.
Works like Lin-Kernighan heuristic are extremely rare and a bunch of knowledge exists in industry only. Even the mentioned heuristic was for decades being implemented incorrectly until one individual came and demonstrated its superiority (K. Helsgaun).
I mean, most of the code that gets released today, doesn't even handle time windows constraint in the best way possible (optimal updates, constant time feasibility evaluations etc.). I believe all open source vehicle routing libraries do not have any data-structure relevant optimizations for handling time windows, task precedence, dynamic workshift starts etc. All are fairly simple implementation tricks that just got lost under giant amounts of population based heuristics or mathematical linear programming approaches.
Best example is discrete optimization research (traveling salesman, vehicle routing and its variants, schedule rostering etc.). Stuff you find in the papers there achieves state-of-the-art results very slowly (using integer linear programming or some rarely optimized heuristics) making you believe these instances of a general NP-hard problem can't be solved quickly.
When you start tinkering, you either find that data structures can be added that reduce the complexity significantly or that there are regularities in instances that, when exploited, support massive speedups.
I would say that TSP research is an exception but most of stuff coming out that has a lot citations is way too slow and is never as brilliantly implemented as Lin Kernighan heuristic or other stuff from the age of insanely slow computers.
I was quite surprised to find out there are thousands of women suffering decade long issues with no help or just enormous, health devastating doses of antibiotics (multi-year long therapies).
There's been so many approaches over the years (I watched the d-mannose just explode in sales when previously there was none available) but they do not seem to work completely.
There are also small research groups trying to figure out what's happening but there's still no widely accepted therapies. For example, doctors from most EU contries would never prescribe the amount of antibiotics required for multi-year therapies.
It was also quite surprising to me, as a male, when I got my first UTI, they gave me quite a lot of antibiotics, but then when my wife complains about it, she gets silly advice (get pregnant, drink multivitamins, it's probably not UTI [until she starts having kidney infections or urinates blood]).
Information theory came after people were already sending signals over the wire. Similarly, there were architectural feats before geometry or Newton’s laws were discovered.
The book sent me back to the college days where I remember doing similar things.
Can't wait to start working on the bytecode virtual machine.
It's really an excellent book.