I remember going to Holmdel to install our network backup product (which later became something named ABARS in AT&T if I remember well), in the 80s. For this young French Unix geek, this was like walking into the Olympe, among the gods :) Edit: Souvenirs, souvenirs. I remember how impressive the atrium was, nothing like it in France at the time (or now probably). However I was working in a basement room, which was incidentally full of Cisco routers. If I had seized the hint and bought a bit of stock, I'd be a richer man... Our software had a memory usage issue on the big Bell Labs file systems, and the very nice AT&T manager insisted that I would not be allowed to go back to France until I fixed the issue. This was a trivial linear reallocation problem, but there was obviously no Stack Overflow at the time so I had to invent exponential reallocation on the spot to get home. I still have the UUCP email paths to our correspondents in my mail archive.
Around 1999, I was testing a still young MySQL as an INFORMIX replacement, and network queries needed a very suspect and quite exact 100 mS. A bug report and message to mysql@lists.mysql.com, and this is how MySQL got to set TCP_NODELAY on its network sockets...
Just as a note, and I checked that it's not the case with the GNU coreutils: on some systems, cp (and maybe cat) would mmap() the source file. When the output is the devnull driver, no read occurs because of course its write function does nothing... So, using a pipe (or dd) maybe a good idea in all cases (I did not check the current BSDs).
The fact that TP-Link products are vastly better and cheaper than all their numerous competitors is indeed a bit strange. You have to either think that all the people at Linksys, Netgear, D-link, etc. are incompetents or that something a bit out of the ordinary is going on at TP-Link...
Anti-hail netting is definitely a thing (protecting cars or fruit trees), for the reasons you state. Even big hailstones are rather slower than fast golf balls.
What rclgrep does is run the recoll text extraction and do a grep-like operation on the extracted texts. If you want to give it a try, don't hesitate to contact me if you have any trouble building or using it, or think it could be improved. It's more or less a (working) prototype at the moment, but I'm willing to expand it if it looks useful. The "Usage" string is much better in the latest git source than in the tar, and it sorely needs a man page.
By the way Recoll also has a utility named rclgrep which is an index-less search. It does everything that Recoll can do which can reasonably done without an index (e.g.: no proximity search, no stem expansion etc.). It will search all file types supported by Recoll, including embedded documents (email attachments, archive members, etc.). It is not built or distributed by default, because I think that building an index is a better approach, but it's in the source tar distribution and can be built with -Drclgrep=true. Disclosure: I am the Recoll developper.
This is true but it was certainly not considered good practise even at the time. I've been on the seller part of a few software companies acquisitions from the early nineties, and checking what kind of source control we were using and how was part of every audit. A long history of sccs -> cvs -> subversion ->mercurial or git...
> You still get solar power on cloudy days, it just takes more panels to generate some specific level of power.
That is wrong. On cloudy winter days the inverters often just stop. Source: have 20kW of panels on a house in the south of France.
Fun anecdote, in the mid-1980s, I was bored at work and wrote a similar simulation in Pascal running on a Vax (11-780 if I remember well), and displaying on a tektro (emulation probably). I discovered after a bit that every time I played with it, all other users wondered why the Vax was suddenly frozen (not sure exactly why, probably the serial I/O) ... Oops.
I used ed on video terminals connected to VME 68000 Unix V7 machines around 1986. For some reason, I don't remember using vi on them, and in any case, ed remained very useful when the terminal type was misconfigured or unknown to the system, which happened quite a lot (I guess that not everything emulated a VT100 at the time). We actually jumped directly from ed to some Emacs relative (Jove probably), which was the beginning of paradise, and then we received our Sun 3 and it was Emacs all the way to present :)
For making a good GUI, the tech guys need to be ordered around by non-tech guys who think like, or talk to, regular users and know what is needed. Tech guys don't like this, so someone needs to pay for their suffering.
Your last point is a bit off. Text pages are shared iff they come from the same file. So your point works if the 90k processes are the same program. Else: shared libs...
fork() also presents performance issues for programs with a large virtual space. Here vfork() helps, but it has even more pitfalls than fork(). I had written a small doc about converting the recollindex Recoll indexer from fork() to vfork() a while ago: https://www.lesbonscomptes.com/recoll/pages/idxthreads/forki...