Convergence package PinePhone has more RAM -- 3Gig vs. separate phone which only come with 2Gig

I work at a place that writes new Fortran code today. Also c++, Python, Julia. Scientists will choose the least impedance mismatch to a library, some needed data set, or their brain.

To be sure. And I don't think that modern Fortran is overly obscure that you couldn't basically pick it up over a weekend from the language specification.

Every modern Linux distribution has a current gfortran bundled.

Modern Fortran is amazing. Force your company to move to Fortran 2008 and beyond. A good reference book is Modern Fortran Explained: Incorporating Fortran 2018.

Or just turn on paging and page in your kernel from the MBR, ignoring the BIOS: https://github.com/charlesap/bootpager

EFI eliminates the need for such trickery though, and who doesn't have EFI these days?

> who doesn't have EFI these days?

VM's hosted on VMware ESX with the default settings comes to mind.

About this much more complex -- https://github.com/charlesap/bootpager (example legacy boot block that jumps into protected mode, enables paging, and pages in and executes arbitrary code, no bios used.) As mentioned above the trick is what to do once you're there. And it is a UEFI world now, might as well just write 64-bit PE kernel and be done with it.

One or more kicks are over here: http://harvey-os.org/ is taking the open-source Plan 9 code and getting it to compile with GCC and clang, etc. They have a userspace with APE (posix) compatibility and they almost have Google Go ported and they debug it with gdb... very much a work in progress though.

Porting Plan 9 to compile with GCC and Clang, or giving it a POSIX compatibility layer seems like missing the point; Plan 9 was notable because of all the stuff it threw out despite being familiar and widely admired, from text-only output all the way down to ANSI C.

It's an environment that demands you adapt to it, rather than vice versa. POSIX doesn't need another quirky API bolted on the side, it's already got quite enough as it is.

You can go all the way to protected mode, enable paging, and then page in the rest of your kernel in just 512 bytes, and without touching the BIOS: https://github.com/charlesap/bootpager/blob/master/oiuboot.a...

That's pretty impressive. Unfortunately the BIOS is still required for getting a map of physical memory...

Well crap. Source code reputation is exactly the problem we are trying to solve at http://rputbl.com (sorry, just a splash page for now.)

We are nowhere near close to launch yet and there's a chicken-and-egg problem of having a sufficiently large database of hashes to make an arbitrary source code file reputation check worth your time, but we have some ideas about what to do about that too.

"your entire program has been garbage collected" I need this on a t-shirt.

I'm using it to do http traffic analysis on captured packets... about 2.5 million packets a second at peak. This is to provide realtime charts for our web service which didn't have realtime reporting hooks.

I am also using it for HTTP request reconstruction and retransmission in real-time to enable live-data and full load testing of pre-production code.

Is 30 close enough to 10?

I got a job with 30 lines of code appended to my resume, with a screenshot of the resultant output.

Perhaps the first two pages had something to do with it but I was told that the third page was what stood out.

This was a sysadmin position, the code includes bash, sed, awk, grep, tcpdump, netcat and gnuplot.

Hint to anyone who tries this sort of resume stunt: You WILL be asked what the code does.

Just the code: http://bbookkss.com/CharlesPerkinsCV3of3Pages.pdf

That's indeed an interesting example because, by any standard, this is terrible code.

Yet for a sysadmin position it still provides valuable insight into your a) pain threshold, b) familiarity with the tools, c) willingness and ability to bend reality for the boss.

I'd just add a big disclaimer that this kind of code only happens in emergencies.

(disclaimer: I'm guilty of these gnuplot hacks myself and they always stick around way longer than they should...)

> They always stick around way longer than they should...

But isn't that true for all 'terrible' code? Most of the elegant and well-formed code I write ends up getting butchered for the sake of 'features,' while the quick hacks that 'just work' stick around forever.

That's indeed an interesting example because, by any standard, this is terrible code.

It's not nice to criticize without being constructive. Care to explain why the code is bad, and offer better suggestions?

Useless 'cat' detected ;)

But seriously, nice work. Being able to put something useful together that easily (in terms of lines of code) is what most attracted me to UNIX/Linux coming from Windows. The amount of CLI tools one can take advantage of is simply astonishing.

How long would it have taken you to write that?