You are assuming reasonableness on behalf of the regulators in question. In an ideal world, they hopefully said "you cannot give information to your own advertising/analytics division that you don't give to others". However, they could just as easily have said "you must provide either third-party cookies or a replacement for them", without offering the much more reasonable alternative of turning both off and not giving Google advertising/analytics any information either.
So, what did the regulators actually say, and does it in fact allow Google to turn off third-party cookies without any replacement? If it does, then this is Google's fault for adding this feature in Chrome. If it doesn't, then this is the fault of bad regulation.
The trade off has changed over time as memory access has become a bigger and bigger bottleneck. But caring about this is still "premature optimisation".
The claim of 20x program performance difference is overblown. Compilers can often remove virtual function calls, JITs can also do it at runtime. Virtual function calls in a tight loop are slow but most of your program isn't in a tight loop and few programs have compute as a bottleneck.
Measure your program, find the tight loops in your program and optimise that small part of your program.
Buildings do not have moving parts. Their core structures are almost all fundamentally the same. There are problems to solve, but again, not dynamic systems, and always variations on very well known themes.
Programming is usually like building a new type of interdimensional alien spacecraft engine that interfaces with some other alien artifacts. There are usually a lot of unknowns, new concepts, many moving parts, unsolved problems, to build a new invention.
Depends, really. Most software is doing basic CRUD and efforts can be estimated reasonably well. Some other software, like your example suggests, is literally rocket science. But even here estimates are typically not too far off.
> Programming is usually like building a new type of interdimensional alien spacecraft engine that interfaces with some other alien artifacts.
If programming indeed was "usually" like that, then there shouldn't be much software getting produced at all. Since building "a new type of interdimensional alien spacecraft engine" is something that's very very unusual, to say the least. :)
Except wind and solar cannot be used to sustain base load energy demands. Until battery technology improves drastically the issue with solar and wind is the variability of output. One of the greatest challenges of operating a power grid is matching the supply of electricity with demand. Thats why they build gas and coal plants to meet peak energy demand periods. Nuclear energy supplemented by win and solar is by far the cleanest way to meet our energy needs currently.
Alternately, if wind and solar can be made absurdly cheap, there are are less-energy-efficient but lower-cost per unit output forms of storage that can help. Pneumatic storage is promising for this for example. So like instead of one wind turbine and one battery, you build two wind turbines and one pneumatic energy storage vessel and get the same amount out.
Oh no. You only get one unit of wind, one unit of solar, one unit of methane by electrolysis, and one unit of combined cycle gas plant for the same price rather than the usual of double all that.
However can we figure out how to get the same result with twice the energy and the ability to store it indefinitely at 40% efficiency. Especially given that the costs of all these technologies are going down at two digit percent per year.
Coal plants in particular are dramatically unsuited for the rapid changes in output to respond to changing demands for power.
Gas plants on the other hand are. And gas plants are cheap!
Overbuilding renewables with combinations of solar, geothermal (where available) and wind gives pretty good availability on it's own and with gas plants as a backup it gives you plenty of grid stability.
Edit: A link you posted elsewhere (thanks!) points out how well this works:
> If other sources meet demand 5% of the time, electricity costs fall and the energy capacity cost target rises to $150/kWh.
Battery storage is already well below this $150/kWh price.
Again, this is only for very short term storage. We are comparing apples and oranges.
As an example, for most of Aus, 5 hrs storage will only work for <80% renewables [0]. It's only practical on smaller scales which doesn't mean it isn't useful now but just means that the ~$100/MWh systems will not be practical everywhere, especially when renewables become more ubiquitous.
Edit: I had a deeper look at the NREL paper and most of their costs seem to be in the $100s/kWh rather than MWh. E.g ~$300/ kWh for 6 hour systems on Fig. 6.
The caveat is mentioned in the link, right after the price comparison. "provided the 1,500- to-1,800 acres of land needed was free and the sun shone steadily" considering the price of land recently... that may need a better estimate than just the cost of the tech.
Also the land requirements for huge solar / wind installations is another downside. Nuclear at scale is a clear winner especially in seismically stable regions of NA.
It's not critical, but I think you are looking at the wrong page. $55 is on the /MW basis, the next page shows $131/MWh which is still good.
The source is "Lazard estimates", but I can't find anything close to this in reality. Pumped hydro is meant to be super cheap, but is currently ~ $100/kWh. I'm starting to agree with the other commenter that Lazard is bunk.
"Pumped hydro is already the cheapest energy storage technology in the world in terms of cost per installed kilowatt-hour of capacity. Total project costs range between $106 and $200 per kilowatt-hour, compared to between $393 and $581 for lithium-ion batteries, World Bank figures show."
https://www.greentechmedia.com/articles/read/pumped-hydro-mo...
Its not bunk. People are comparing two widely different things and not understanding the difference.
The price of $150 is capacity of a battery. If i want a battery that can store 4 kwh i pay $600.
The other number referencing $55/MWh is referencing levelized cost of storage.
I have a a MWh but instead of using it right now i store it for 4 hours.
Now this MWh costs its initial price + $55.
Apply this example to a KWh where i would pay $0.1 for it if i used it right now
if i want to store it for 4 hours and use it later i have to add $55/1000
to the price or about $0.055.
So a directly used KWh would cost me $0.1. One stored for 4 hours would cost me
$0.15.
Hope that makes sense and explains these different numbers.
> If you do that calculation at the global level, we evaluate the LCOE for recently financed projects is at US$150/MWh including charging costs. That’s our benchmark. We have a range around that benchmark which goes from US$115/MWh in China.
Thanks for the link. I am trying to get my head around the ~1000x improvement and I think it is because we are comparing apples and oranges. ~$100/kWh (like pumped hydro) can last for as long as the water in the reservoir lasts. ~$100/MWh (like the batteries you mentioned) are for <4 hrs reserve.
So the critical question is: how long will grids connected to renewables need reserve? 4 hours doesn't seem like much but I'm not sure the best way to find this info.
Edit: from a 30s google search, 4 hours is only good for <80% renewables in Aus.
[0] https://reneweconomy.com.au/much-storage-back-high-renewable...
80% renewables is huge! It's some years before we get to that level by which time we can expect the 20 hours required for all states to be economically viable.
And in the mean time gas peaker plants provide a viable, cheap and safe compromise without needing nuclear!
> We found in
some cases the battery requirement becomes
very large relative to the load, at greater than
20 hours. In these cases, it was concluded that
additional gas peaking capacity would be more
economic (and biogas was used when the
emission constraint did not allow for natural gas).
Yeah, we can use gas plants until batteries become economical, but why use gas when we could start the switch to a low-emission alternative now. The risks of Nuclear are tiny compared to the existential risks of delaying emissions reductions.
Biogas use isn't practical to replace existing gas supply.
I honestly really wish renewables + batteries could take over but it's too early. Aus has heaps of Uranium, is geologically stable, has strong regulating authorities and geopolitically secure. The perfect spot for low-emissions Nuclear which is practical and possible now in all respects except for politically.
Not to mention the other comments that refute those numbers using the same paper - projects would be closer to ~$100/KWh when storage is taken into account, as shown graphed in the paper itself.
* $200-400 million for equivalent onshore wind (they don't mention offshore wind which is more cost effective over it's lifetime)
* $1.5 billion for the SMR
Assuming solar and wind stay the same price for the next decade (which they definitely won't) that's a 4-5x construction price difference. The article didn't mention lifetime, decommissioning or running costs, which is likely a significant difference too.
If you have any unauthenticated routes that you don't want arbitrary websites calling.
> using JWT in a typical SPA <-> API scenario.
Is this typical? It's a pretty horrible setup.
Cookies have a lot of great features that 'store a JWT in LocalStorage' just doesn't have.
This doesn't actually prevent arbitrary websites from calling them, it just makes it a tiny bit hard. They could always just proxy your endpoint and add the CORS headers.
I'm still interested in the original question: if you use localstorage for auth tokens and you have proper CSRF protection, what does allowing all CORS actually make you vulnerable to?
Sure, but the needle has moved. The thing that took you two weeks to write in 1987 isn't even considered anything useful by today's standards. The parsing needs to happen in real time, it needs to provide machine learning assisted auto complete, it needs a web UI and UI needs to show you an animated graphic of the scene you're describing, etc.
