Before anyone comments "fusion is always 10 years away" I'll just say that we've made significant progress in the last 40 years (energy generated by fusion is on par with moore's law) and it is simultaneously the single hardest and the most rewarding engineering problem ever attempted by humans. Any news on this front is good news.
The phrase "we've made significant progress in the last X years" has been used for a long time nuclear fusion research. Quote mining from Google Scholar:
1999 - "Over the past several decades, significant and steady progress has been made in the development of fusion energy and its associated technology and in the understanding of the physics of high-temperature plasmas." - https://e-reports-ext.llnl.gov/pdf/237013.pdf
As such, the phrase, while true, cannot be used to infer any sort of near-time (within the next few decades) prediction for an effective fusion power system.
Second, I do not like the appeal to Moore's Law because even if true (I don't know the baseline numbers), all you are saying is that there has been an exponential growth over the decades. However, there are many S-curve growth patterns which appear exponential near the start. What is the basis for assuming that nuclear fusion will be like Moore's law, with exponential growth over many decades, and not an S-curve?
For example, from the 1800s to the 1970s we saw huge increases in transport speed; from trains and wooden ships to moon travel and supersonic passenger airlines. Someone in 1965 (ahem, Heinlein) might point to the exponential growth in rocket ships over a single lifetime and easily assume it would result in Moon bases and tourism within another two generations.
It's not meant to, it's meant to dispel the common notion that no progress is being made in this area of research.
And the total power output doesn't need to remain exponential for fusion to be viable, we just need a way to sustain fusion reactions for a long enough period of time (china recently set a record with 101 seconds).
Seriously look into all the work being done on this currently by Government projects and private companies. I'm not saying it'll happen soon but significant progress has indeed been happening for a long time, and is only picking up pace as time goes on. As I said, this is likely the most difficult thing humans have ever attempted.
But the quip that "fusion is always 10 years away" is not an expression that no progress is being made. Rather, it's an expression that the most optimistic views for fusion are always for at least 10 years in future, and as we get more understanding of the difficulty, that horizon always stays 10 years away.
I've been following fusion from the sidelines since I first read about it in SciAm in the 1980s.
I think that understanding how the human body (or any complex biological system) works is a more complex project.
On a project I started last year, I used SHA512 (probably overkill but who cares) and then bcrypt. Allows users to have passwords of arbitrary size with the goodness of bcrypt.
Then AES256 the result and use that. I read somewhere that's what Facebook does.
While somewhat better than plaintext, SHA-256 is far too easy/fast to slow down dictionary attacks.
Practically speaking, the best practice is to use a previously vetted library that provides high-level password storage solutions based on the recommendations below. Developers should not be reinventing the wheel to solve these problems for each job.
Having said that, here are the main factors:
- Choose a function that's designed to be slow to compute (and preferably that can't be easily optimized by throwing memory at the problem). Bcrypt, scrypt, and pbkdf2 are the most common solutions. Argon2 was the winning candidate of the 2013 - 2015 password hashing competition and is gradually starting to appear in more places.
- In addition to choosing the right function, hashes should be salted with a random value (that's stored with the hash) to prevent pre-computation and attacking multiple hashes in parallel.
Marijuana taxes are obscene and immoral. I pay 30-40% on a plant that has medicinal properties in Colorado. The whole "sin" tax on marijuana makes no sense. God forbid this gets set at a Federal level.
>Of course, you might feel differently, but one of my biggest takeaways from all of this side-by-side code is that Rust is clear, expressive
As someone who doesn't use either of these languages, I couldn't disagree more. The Javascript was intuitive and immediately made sense, while the Rust code gave me a headache.
> My only hope is that this isn't so successful that they neglect the installer because nobody is using it anymore :-)
Wouldn't that be a good thing? I mean, you'd have a tough time finding a computer for sale in 2019 that even has a CD drive, why care about it so much?
My initial thought was that sometimes you want to install it to your hardware, not just a USB stick. If this is a finished install that lacks the installer, then you wouldn't be able to use it to install to a hard drive. I could be wrong tho.
