Related question: I think I’d really like to work part-time so I can properly multi task on my personal projects and for-money work, rather than have this all-or-nothing choice every 3+ months. Do you have any advice you could give me about that?
64 karma · joined November 17, 2020
European countries can and should take them up on that. Thanks to GDPR people can export their data to a new network pretty easily, but I expect most won’t even care/expect to keep the old data.
It also has tree-shakeable imports which v2 doesn’t have, and a very extensible looking middleware architecture which I’ve yet to need but good to know that it’s there if I do. The auto-generated docs leave something to be desired, but are passable, with a few extra clicks here and there.
I’d much prefer a comfortable salary and working 3 days per week than more salary than I need and working 5 days a week, and I definitely want to keep those extra days to work on my own interests.
There’s a lot of a-moralistic attitude towards FAANG on Hacker News, which honestly I find strange; Google and Facebook in particular are just giant douchebags with lots of cash.
Remote: possible Willing to relocate: possible
Senior, self-taught, BSc Computer Science, MSc Information Security. Experience in a range of areas from C/C++ QUIC networking to React to penetration testing to AWS Lambda with TypeScript.
A trusted problem solver who is often asked to do odd jobs by senior management, in addition to larger scale work.
TypeScript, Node, Java, AWS, Serverless, Docker, ~Go, ~Flutter, ~React, ~C
Only able to work up to 3 days each week.
Contact: alastair - grokkie - com
London or Remote
Senior, self-taught, BSc Computer Science, MSc Information Security. Experience in a range of areas from C/C++ QUIC networking to React to penetration testing to AWS Lambda with TypeScript.
A trusted problem solver who is often asked to do odd jobs by senior management, in addition to larger scale work.
TypeScript, Node, Java, AWS, Serverless, Docker, ~Go, ~Flutter, ~React, ~C
Only able to work up to 3 days each week.
It seems important for compressibility to prepare the data for maximum self-similarity, in addition to the LZ algorithms (as evidenced by the sort in this article). Could someone point towards a good modern summary of the approaches or heuristics?
Senior Software Engineer and mostly-backend problem solver.
TypeScript/Node, Java, Go with Serverless, AWS, React, Flutter.
Up to 3 days per week.
Past services include:
* Code health audit for non-technical founders
* Candidate code test reviews
* Serverless microservice delivery
* Scaling plan for cost or performance
alastair grokkie.com
That’s been going on forever too, but at least my hand-wavey perspective is that whereas I might have thought in the past “huh, I sense my Cadbury’s bar is possibly 2% lower quality than 2 years ago :shrug:”, now it’s more like “huh, I can’t be the only one who’s had to stop buying this crap which was previously an ace product”.
I suppose both these forms of inflation are down to consumer buying power being weakened, and posit that nowadays it’s due to the “stock price being the product” - to lean on a reference to Silicon Valley’s Action Jack :)
When I ran the "if-else" statement of my programming language* by a user-study with my non-technical brother-in-law he said it reminded him of the binary nature of legalise and inherently limiting the nature of what is possible. I agree.
My sister also formed the study-group and consistently they both prefer "otherwise" to anything else (pardon the pun).
* The language in case anyone wants to take a look: https://github.com/ducklang-community/ducklang
As others have commented, this isn't sufficient though, and will over-estimate the complexity of worded passphrases or l33t-speak. Password crackers are wise to those generation techniques and will brute-force combinations of those with more direct generation methods instead of generating from the underlying character set at random.
Complexity rules backfire if the minimum is done to meet them, like capitalise the first letter, append a number 0 and exclamation mark. A tool like this could (and probably should) check for those special cases and discount them from entropy, assuming conservatively that it's a weak password, ie. "P4ssw0rd0!" ~= "p4ssw0rd" ~= "password" ~= 13 bits (a single English word at random).
I'd also add that while that's a cool graph, it's going to age poorly as FLOPs/Watt is still on an upward trend. I wrote a password generator* which takes into account GCloud/AWS GPU prices and wholesale energy costs, along with Hashcat metrics, to recommend the number of bits of entropy for passwords. Looking 20 years into the future you'd probably want something closer to 82 bits entropy than 76, based on my calculations.