vlowther

... writing the code required to write the code needed to implement indexes is more fragile than letting the compiler do it for you, most likely. I maintain a reflect-based indexing implementation for in-memory data, and look forward to seeing what I can do with generics to get rid of reflect based overhead.

Shades of NCR vs. IBM, to call back to an even older school acronym pissing match.

I hated the Reddit app so much I bought BaconReader.

... and the fastest way to reset the processor was via a triple fault -- arrange for the general protection fault and the double fault handlers to both fail.

Same, except just yay.

Legalize it and let "hardcore" heroin addicts get a prescription. It is expensive due to it being illegal and the willingness of addicts to pay damn near anything, not due to any intrinsic difficulty in manufacturing.

I am curious about where things broke down with the 301 based solution y'all used earlier.

Well, my insurance for a family of four cost slightly more than 10,000 last year (both employer and employee sides). You could triple the amount I paid into Medicare in exchange for Medicare covering everyone and I and my company would come out ahead. There has never been a year where I have been paying for insurance that it has not grown faster than inflation.

`map[t]struct{}` saves you a few bytes per entry. Just use the `_,found := foo[key]` form of lookup

The overwhelming majority of religions are not monotheistic and believe no such thing about any given deity. /nitpick

It will look good on the bookshelf next to https://www.goodreads.com/en/book/show/3113099-mr-bunny-s-bi...

"On any platform, pages are synchronised to disk atomically: even though larger chunks might only be partially synchronised, pages are guaranteed to be synchronised atomically."

Someone is being hopelessly optimistic about how reliable disk writes are.

Or better yet, condition yourself to hit [ a second time and use Bash's built-in conditionals instead of test and [

The GPU stats at https://www.gamingonlinux.com/index.php?module=statistics&vi... disagree with your first point. A few years ago I did all of my gaming direct on Linux with an nVidia GPU, and did so for over a decade, on various laptops with and without mixed Intel/nVidia GPU setups.

Boot time pre-OS malware is not a theoretical risk. At all.

The issue is that the Pi does something that is almost-but-not-quite PXE. In particular, the RPI firmware does not send anything that identifies itself as an RPI so that the DHCP server can determine what packet to send in response -- the option 60 string it sends in the DISCOVER packet claims to the DHCP server that it is an X86 server running in legacy BIOS mode. The only way the RPI will boot is if you send it that extra magic option 43. For those of us that write DHCP servers that try to be intelligent about how to respond to PXE requests, the plethora of stupid not-exactly-PXE mechanisms the various ARM SBC hobbyist boards are annoying as fuck. At least the server class ARM stuff usually has a UEFI implementation that adheres to the UEFI extensions to the PXE spec and properly identifies itself with options 60, 93, and 94.

Or you could just have separate branches for major versions. That works too, and everyone seems to forget it.

Excellent -- nil handling work as well?

Eh, last time I messed with I lost interest after noting that from-json | to-json would silently convert all numbers into quoted strings and do terrible things when nil was involved. It made non-trivial JSON manipulation not worth it. They might have fixed it by now.

re: filesystem journaling: Unless you were one of those crazy people that used fat32 in spite of Microsoft's defaults, filesystem journaling has been the default in Windows mainstream consumer OSes since the release of Windows XP using NTFS, which was journalled since day 1. (I am not counting Windows 2000 and before since they were not the default for consumer boxes).

What a weird slippery slope argument. There is a world of difference between "so complex that nobody understands it" and "I can telnet to port 80 and fetch a webpage using keyboard and fingers"

Yeah, it doesn't work that way, and no, most US citizens don't have any appreciable native ancestry, no matter what family mythology says.

There is also https://github.com/segmentio/encoding, which aims to have a zero allocation fastpath while preserving compatibility with encoding/json.

GTK came from the Gimp, not Gnome.

Zoom running on Plasma in X has worked fine for me for years. I would suggest that the problem (like so many others) is a Wayland ecosystem maturity thing, not a zoom thing.

Open source software is (I would argue) even more driven to fragmentation by political, ideological, and personality conflict driven squabbles than proprietary software, as there is no profit motive to also satisfy.

Also, the broken window thing asserts that small amounts of criminal activity lead to larger amounts of criminal activity via signaling that being bad or neglectful is OK, which is both not proven and irrelevant to software writers being prone to reinvent the wheel for whatever reasons they have.

Amazingly, so is http://www.zombo.com/. It still delivers all it promises

well, good thing there is https://www.flatcar-linux.org/

The only time I have used recover in production code is in the rollback path of a highly concurrent transaction manager.

Almost the same, except mine was a Precision 7520 -- no Xeon processor, no ECC, but an AMD Polaris 11 based dGPU. My general advice for Dells is to spring extra for the Precision laptops -- sure, they are heavier, but the build quality and ease of upgrades make it worth it. You can even claw some of that cash back by purchasing them via the outlet.dell.com -- it will be a return, but it will have the exact same warranty as a new one.