Posts by Michael Wojcik 12252 publicly visible posts • joined 21 Dec 2007
why it's called carbon copy
Because of a folk etymology.
Carbon Copy (CC) and Blind Carbon Copy (BCC) fields
This is a myth. The abbreviation "cc" for "copies to" on memoranda was in use well before carbon paper was invented. Repeating an initial letter to indicate a plural in an abbreviation is long-established English usage, for example in "pp" for "pages".
The "Cc:" header field is an abbreviation for "copies", not for "carbon copy".
Interestingly, perhaps, in RFC 733 (et seq), Crocker does not use the term "carbon copy", merely describing the CC header as describing "secondary" recipients; but does incorrectly describe BCC as "Blind carbon". Neither CC nor BCC appeared in RFC 561; they're described in RFC 680, which (correctly) does not use the word "carbon". RFC 724, which updated 680, also avoids "carbon", so the error appears to have been introduced by Crocker in 733.
(And, yes, I've used carbon paper.)
Are they? Not saying they're not, but the sources I've found about this all come from 2021 and 2022. I'm curious to know if Uber and the other fake-taxi companies are still having trouble recruiting and retaining drivers.
(Personally, I have not and will not use Uber, as I dislike them as a company and dislike their business.)
When we lived in the Stately Manor, my wife got in the habit of having takeaway delivered. I didn't want to fight about it.
Here at Mountain Fastnesses 1.0 and 2.0, there is no delivery option, and I'm glad. I'm an introvert by nature, but if I want people to perform a service for me (even if I'm paying for it), I like to look them in the eye and acknowledge that they're people too, living their own lives. And I like to leave the house and be reminded that there's more to the world.
Frankly, I'm unable to conceive of a technical justification for mandatory cookies at all. During a session, requests can carry a session identifier by annotating href and XHR request-URIs with query parameters, and forms with hidden form fields, simply by injecting those when serving content. PHP has had support for doing this for many years, and that's as low a bar for an HTTP content-generation system as I can think of.
Across sessions, cookies support implicit authentication, but that's a user convenience, not necessary.
Cookies are a crutch, for any request where the server has control over the Request-URI or the body, and that's nearly everything. Non-browser UAs for custom applications could inject their own headers. Cookies ought to have been deprecated long ago.
By definition 50% has below average skills and IQ.
This canard again.
This is only true if by "average" you mean "median", or you have a distribution that happens to work this way. If by average you mean "arithmetic mean" or "mode", then your distribution may very well have more entities on one side of the average than the other.
And, of course, IQ is of debatable value, and there is no generally-accepted quantitative measure of "skills" (much less one that yields a single real number), so "average skills" is a meaningless phrase.
Perhaps if you understood the most basic matters of statistics you'd have a stronger argument.
Here the fiber drops from the pole to the house, for both Mountain Fastness 1.0 and 2.0 (about 100 feet / 30m) would have cost me around $100 each, except they waive the fee if you sign a 2-year contract for the ISP service. And I'm in a semi-rural neighborhood here. Both homes sit on around an acre each (~0.4 hectare) and there are sheep on the other side of the back fence; it's not like this is a dense urban area.
Anyone in the county who's on the grid can get fiber to their home under those terms.
The service itself is around $60/month, which is moderately expensive — but a quarter of what we used to pay for Internet plus cable TV at the Stately Manor. Since I work from home and we use our Internet service more or less constantly, it's a very reasonable expense.
Possibly to communicate with people and organizations who insist on using SMS? Most of my cellphone use is SMS. I have a couple of OTT messaging apps installed: Signal, which no one has ever used to contact me; and WhatsApp, which I reluctantly installed to communicate with my building contractor while he was on vacation outside the country, and haven't touched since. Occasionally I get or make voice calls, or get the occasional MMS,1 but SMS beats all of that by a couple of orders of magnitude.
That said, if OP had decent Internet at the hypothetical off-grid dwelling, a picocell might work for cellular service. We use a picocell supplied by AT&T here at the Mountain Fastness. I don't know if Starlink's latency is too high for one, though. (Some latency wouldn't matter for SMS, but can be annoying for voice calls.)
1MMS is a PITA, so I avoid it. Because of the many MMS-related security vulnerabilities (due to crap implementations of multimedia libraries and so forth), I have MMS set to explicit download only, and I only download them if they're from people I know and I'm expecting them. Even then I don't like it. MMS was a terrible idea, like most ideas in the mobile space.
That's why the electric co-op here spun out an ISP business, and ran fiber alongside its power lines to the whole county. For the most part, the Internet infrastructure is the electrical infrastructure here; damage to one is usually damage to the other, and can be fixed more or less in parallel. You still need to send two crews (electricians generally aren't fiber-optic repair techs too), but the site has already been identified and cleaned up by the first crew when the second arrives. If a house is on the grid, then there's already some arrangement for running the power line (easement, a drop and/or conduit, etc), which can generally be used for the fiber as well. Customers pay for installation, then they can either use the co-op's ISP service, or use a competitor's, and the co-op offers phone service as well, which is useful for people rural enough to not have good cellular service.
It's true there are still a lot of households in the county which are so rural they're off-grid, and can't make use of it, but the fiber rollout made a huge difference to high-speed Internet connectivity. And there are wireless-service ISPs (using microwave, maybe? I haven't investigated, but I think they mention line-of-sight in their materials) piggybacking off it for some of those off-grid households.
Add in lifestyle data and it would become exponentially more useful both on a personal basis and as a means of advising health policy. If it were opened up to wider sharing and analysis then the benefits could be, literally, life changing.
It certainly could be life-changing, since the combination of health and "lifestyle" data would likely make de-anonymization quite easy.
No thanks.
Yeah. There might be a few centuries' worth of history complicating the situation just a bit.
It's not even all ideological, of course. There were, and are, any number of personal motivations (revenge, for example), and some years prior to the Accords I read a lengthy analysis about how much of the conflict had devolved into criminal organizations running straight-up extortion rackets under the guise of adherence to one cause or the other.
War is pretty much never simple.
I wonder how many achieve these goals.
None. No organization does "everything within [its] power" to protect data, because if it did, it would have no resources remaining for any other purpose.
Spokespeople have to spout fine-sounding words with vague and overstated commitments. When it comes to actual security policies and practices, however, reality must intrude, or the effort will be unproductive.
Similarly, no one who understands security should ever give "absolute trust" in anything. That's fundamentally true at an epistemological level; you shouldn't assign probability 1 to any hypothesis,1 because, by Descartes' "evil genius" argument, your sensory and/or cognitive capabilities might be compromised.
1Except the hypothesis that something is considering hypotheses and assigning probabilities to them (cogito ergo sum). But even that might only be occurring for an infinitesimal period of time, as you might be a Boltzmann brain.
I find it slightly odd that HR 6611 was sponsored by Turner [OH], who's historically been one of the more moderate Republicans in the House, and who has a JD (and so shouldn't be unaware of the consequences of the bill, should it become law). But of course everyone has some inconsistent positions, and there are plenty of people in both parties who are far too enthusiastic about surveillance.
"Fake" sports reporting also has a long history.
As a young man, Ronald Reagan had a job for a time as a radio baseball reporter. He did the equivalent of play-by-play and color reporting, both reciting the events of the game and filling the audience in on peripheral details: the players' appearance, the crowd's reaction, and so on.
Of course Reagan was not physically present at these games. He was in a small room, reading a ticker tape with the bare facts of what was happening — who was at bat, what happened with each pitch and play, and so on. He invented the rest of it from whole cloth, in real time. Essentially he was telling a story about a baseball game, where the material facts of the story that affected the progress and outcome matched a real game.
For a human, this is a fairly impressive skill. For an LLM the "making stuff up" part is easy; it's just stringing together clichés. The challenge for automating it is eliminating material hallucinations so the audience doesn't catch you out.
Of course machine-generated sports reporting has been around for several years, just as with financial reporting and other sorts of specialized writing. The use of LLMs to do this sort of thing is merely evolutionary.
I've never read Sports Illustrated (or bothered with the legendary Swimsuit Issue, which has to be the most overrated, feeble excuse for titillation in modern publishing), so I have no idea of the quality of their work. (And I don't know enough about the subject to judge it on those grounds anyway.) But — and I freely admit this could be unfair prejudice — my impression is their audience is not the most discerning.
Is there a way to install Windows so that you don't have a GUI?
Yes. Windows Server Minimal.
It is also possible to remote-shell into Windows.
running a command shell just to use a command-line editor seems to be more effort than just using the GUI tools
Does it? Using anything GUI for most tasks seems like more effort to me. Maybe not everyone is you.
The IBM E family of editors is an interesting one. IBM PC-DOS included E from 6.0 onward, and so did OS/2.
I first used E in the late 1980s, when it was an internal IBM tool. It did a lot of things well. Column and rectangular select / copy / cut / paste was very handy. It had multiple file support, and could flip between the two most recent views very quickly with a single keypress, which was often handy as a visual-grep mechanism to find changes (just hold down the key combination and look for what's flashing on the screen). Full-fat E had a nice programming language which let you quickly build fancy macros and extensions.
I confess to borrowing a copy of E (including all the available add-ons) from IBM and installing it on the DOS and OS/2 machines at the startup where I was also working, and it became the preferred editor for DOS, Windows, and OS/2 for most of the developers there for several years.
Gmail was the first very-widely-used AJAX (well, XHR) web single-page application. While Microsoft invented XHR and the idea of AJAXy1, single-page, server-driven web applications, they weren't very successful at getting users to adopt them. Google were, and that made a huge change to the web as used by most people.
Whether that was a good change is debatable. And whether it represents innovation on Google's part, or just a better ability to market an approach, is also debatable.
1The acronym "AJAX" was coined by Jesse James Garrett in 2005, a year after Gmail launched and eight years (!) after Active Directory first appeared, and as it happened Garrett turned out to be largely wrong about the XML part (but then, so was Microsoft when they named XMLHttpRequest). But he recognized that the basic idea was going to become more common, and gave it a name that stuck.
That's the state of, eh, "the states"
No, it really isn't. 2.5 centuries of US jurisprudence firmly establish that the government can restrict expression in certain limited circumstances, and commercial fraud is definitely one of those.
it sort of sounds like he could get away with it
Not on First Amendment grounds.
Whinging about the First Amendment is Musk's go-to move. It hasn't worked well for him so far.
Oh, I remember plenty of hate. There was one ninny I kept seeing in some Usenet groups in the early '90s who was always trying to get people to meet him IRL to duel, for example. There was no shortage of heated words, both in the (unmoderated groups of) the Big Seven and particularly in the alt hierarchy.
At least as early as the late '90s, people on Usenet were doxxing supposed extremists, such as neo-Nazis; and there were sites doxxing abortion providers as well.
Talmadge wrote "The Flamers [sic] Bible" in 1987; flame wars were already well-established in Usenet and BBS culture then.
BBSes, by definition, are not "the internet". Usenet was originally UUCP only, and thus not "the internet"; even when NNTP became common, there were still UUCP links.
IRC wasn't even invented until four years after Flag Day, so the Internet was never just "Usenet and IRC".
As for "customization", whatever that's supposed to mean, there were no standards for BBSes, and so BBS organization and look & feel varied widely. They may have all been pure text,1 but there were many approaches to command-line and menu interfaces. Similarly, there were text-menu-based Internet applications over Telnet, Gopher, and other protocols. (Gopher plus WAIS had many of the capabilities of the early web.)
1Though that's debatable, since some might consider Sierra On-Line, for example, a BBS of sorts (it was dial-up; it wasn't really a "bulletin board"), and it had graphics as far back as 1980's Mystery House, though they were client-side resources selected by server-side instructions.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
