Correct. This is why the Product Manager role exists - to define "what problem are we solving, why, and for whom?" by engaging with the market. But if you already know what problem you want to solve (for yourself, or for fun), don't bother. But also don't expect others to pay for a solution to a problem they don't have.
> agile poker and code commenting and stuff
These are tools for team collaboration and business planning, i.e. when there is more than one person involved in a project. You don't _need_ them for solo projects (although I do think code commenting is still a good practice even for solo projects).
Side note/feel good moment: the author Raymond Chen is an OG beyond-super-engineer at Microsoft, and is an incredible person to work with. He assisted me with my intern project many years ago, and was amazingly eager to share his wealth of experience with a green-eared intern from outside his org.
Thanks Raymond! I decided to go the startup route instead of returning to MSFT, but you were an inspiration all the same.
I'll add that real world sensors (including cameras) are noisy in ways that are difficult to model 100% accurately in simulation. Same thing with aerodynamic effects, especially when the drone gets close to the ground or other objects, or y'know...it's windy outside. Dealing with these types of noise and unpredictably and building in resilience to them is a big part of testing and validating systems like these, and can only happen in the real world.
I'm pretty confident the drone linked in this article doesn't have high reliability. I'd be shocked if they managed to get >70% critical without significant real-world iteration.
I think it basically helps that quadcopters are already computer stabilized and highly manoeuvrable so the problem is more identifying spaces to move into and motion planning through them. What’s impressive since the last time this kind of thing was posted is the increase in speed and at least in the provided video the ability to avoid really quite small branches.
It looks from the video someone linked to above that these things don't use cameras, just depth sensors -- maybe they're easier to simulate? It looks like they do introduce noise into their simulated depth-sensor data, but maybe that's more straightforward than creating realistic visual noise.
Not sure if it's a fair comparison when the links you posted do not include any real evidence of flight agility or video of the actual drone. It's probably because of classified nature of this project, but still we don't know how similar these capabilities are.
> developers control the cryptocurrency issuance because despite the issuance is (usually) determined algorithmically, developers have the power to change these algorithms...If we compare this system with the US dollar system governed by the Federal Reserve, the Supreme Court of the US, the Congress, and the US government (in different ways), I’d say that the blockchain developers are at least as likely to become corrupt, coerced, or influenced by the system beneficiaries than the Fed and the Government officers and judges because the latter are 1) older on average, and more senior people tend to act more independently; 2) much better protected, physically and financially.
Don't miners and nodes then have to adopt those algorithmic changes though? It's essentially voting by adoption - if a majority of miners/nodes don't adopt a set of changes (e.g. increasing the supply cap or rate), then they aren't propagated to the blockchain.
This seems a lot more decentralized than the US/Fed monetary system that the article compares it to, where citizens have effectively zero influence on policy.
Unlike democratic policy, the votes for bitcoin policy are not 1 citizens 1 vote. You can claim that citizens have effectively no control (although they clearly do have some control). If you're not a miner you have absolutely no control. Even if you are a miner, your control is some function of your hash power.
> If you're not a miner you have absolutely no control.
The economic majority of Bitcoin users ultimately decides what "Bitcoin" even means and can enforce that against miners with e.g. a User Activated Soft Fork: https://uasf.co
I disagree. As a user, I have no real say. What ever the miners and exchanges decide if what I must use. And for the vast majority of users just don't care about which fork should be considered the true path.
As a user, I have about as much control over US currency. I can be very vocal, protest, and raise awareness for what I think is correct, but ultimately almost all decisions happen with governing bodies. And in the case of Bitcoin, the governing bodies are the miners and exchanges, which have very different incentives than the users.
You may have diminished say in the governance process, but you can vote with your money.
If enough people leave a contentious network, the price falls. If you decide to use another chain, exchanges/wallets/miners become profit incentivized to support those chains.
And this is the problem. Blockchain represents a revolution against existing financial power structures. I am personally far from convinced that this will go well, since we know from history that revolutions intent on removing power structures tend replace them. Even before we factor in geopolitics, I am not extremely keen on replacing my country's already too unchecked banking sector with some entirely unregulated, supranational line-up of blue-eyed ancaps and other speculators whose main qualification is owning enough GPUs in places where power is cheap enough to mine. I really don't see that working out well for me at all, regardless of the theory of democratizing access to financial services. Money is infrastructure, and those who own the roads set the tolls.
This is an advantage of proof-of-stake: in order to 'mine', you only have to own some of the coin. It's still not one vote per citizen, but it's much closer to it--and it's more stable, since people holding large amounts of the coin are motivated to ensure it's stability and security.
Developers control the means of software distribution, which does give them some staying power even if they make bad decisions. The tipping point to removing software developers from control is when it becomes preferable to set up new distribution channels instead of accept their code as-is.
> a lot more decentralized than the US/Fed monetary system [..]
The distribution of USD cash across the world would appear be (almost uniquely?) decentralized... If the Fed were to decide that as of midnight tonight, negative interest rates apply on all USD cash balances, how exactly could that affect the suitcase full of USD you buried at the bottom of your garden? [Spoiler: it wouldn't...]
> Physical cash is a tiny portion of the USD mass in circulation
We appear to be using the term "in circulation" somewhat differently.
"Currency in circulation refers to the amount of cash–in the form of paper notes or coins–within a country that is physically used to conduct transactions between consumers and businesses"[0]
"As of February 10, 2021, there is approximately US$2.10 trillion in circulation, $2.05 trillion of which is in the Federal Reserve Notes (the remaining $50 billion is in the form of U.S. notes and coins)"[1]
...of course you're correct to note that this ~$2 trillion is only a proportion of total money supply (assuming we can agree on which measure to use for 'money supply') but it's still a substantial sum. It's also decentralized.
How does ~$2 trillion of decentralized money compare with the total supply of one's favourite (allegedly decentralized) cryptocurrency?
"Noob gains" doesn't have anything to do with setting goals or quantitatively measuring progress. It refers to the speed at which unadapted athletes are able to progress (e.g. add weight to the bar, run farther) each workout when they first start training, compared to athletes who have been training longer, are more adapted, and progress slower.
Most professional and highly adapted athletes are extremely quantitative about their training programs, much more so at the advanced stages than in the early stages. They just progress much slower because it takes more training for their bodies to continue adapting.
Correct. This is why the Product Manager role exists - to define "what problem are we solving, why, and for whom?" by engaging with the market. But if you already know what problem you want to solve (for yourself, or for fun), don't bother. But also don't expect others to pay for a solution to a problem they don't have.
> agile poker and code commenting and stuff
These are tools for team collaboration and business planning, i.e. when there is more than one person involved in a project. You don't _need_ them for solo projects (although I do think code commenting is still a good practice even for solo projects).