> I wonder what other words are _not_ so frequently used then.
Use Google, and try to get the lowest number you can get for the number of pages. Yes, this is a variant on GoogleWhacking, but with only one word.
Some quick attempts: Google finds 76,500 pages using 'rotund' (round), 31,000 for 'pneumatology' (the study of the [sS]pirit), 13,900 for 'cromulent' (valid), 818 for 'pimola' (a stuffed olive), 242 for 'anatopism' (something that is out of place), and only 31 for 'propretonic' (preceding the syllable before the accent).
I chose "pimola" because I happen to know that it's not listed in the OED, so I figured it was fairly uncommon, but it turns out that a couple of the other words I tried are even less common. I was surprised that "propretonic" isn't used more often. FWIW, the sites that do use it probably use it numerous times each.
> I would have thought things like "the", "and" or "or"
"the" isn't nearly as common in English as the definite article in many other languages. We tend to use it once where other languages would use it several times. For example, what might be "the words the common the most" in another language comes out as "the most common words" in English -- so the article occurs only once instead of three times. This hurts its rank considerably. Also we don't use it on proper nouns, and sometimes we just leave it off and let the reader figure out from context that the word is definite. (Having indefinite articles helps here; if we wanted the word to be indefinite we could explicitely tag it as indefinite, so with no article at all the word's definiteness is ambiguous.)
"and" is fairly common, but I would think it would be beat out by words like "to", "for", "from", and "that". ("that" has an unfair advantage because it's actually about several different words spelled and pronounced the same way; most notably it's a relative pronoun ("the word that introduces a relative clause"), the most common of the demonstratives ("yes, that word"), and a very common subordinating adverb ("I provided examples so that you might know what I mean."); it would probably be _the_ most commonly used word in the language if we didn't elide it more than half the time.)
Also, some of the uses of "and" are picked up by other words. English has separate words for "even", "also", and half a dozen other concepts that tend to be the same word as "and" in some languages.
> One Palmyra Atoll dollar = 17 pieces of mithril, or approximately twenty > kilograms of fairy dust. There's no such thing, people. This is a joke.
Actually, mithril does exist. It's also known as titanium ore or titanium steel (depending on whether it's being mined or whether a worked object is being described).
It's Adamantium that hasn't been invented yet. Unlike mithril, adamantium is not vulnerable to the liquid nitrogen freeze-and-shatter attack. It is speculated to contain plastic polymer in addition to metal alloy, but we don't know how to actually make it.
> What's your source for the 9% transmission losses? I have always heard a > figure of about 2%, which is a lot more reasonable.
It varies, depending on stuff. Among other things, the further you are from the power plant, the more loss there is on the way to you. It can be as high as 20% in some cases, but the average is lower than that. 2% sounds like a minimum, best-case-scenerio to me.
> > the entire Microsoft organisation (which undoubtedly employs some of > > the world's finest software engineers and quality assurance experts) > What ever happened to judging people by their results...
He was judging by _quantity_ of results. Microsoft produces *lots* of software.
> if you're talking PC, why wouldn't you install XP?
I can think of some reasons. It's not safe to connect it to the internet without an external firewall. File and print sharing don't work properly. It takes half an hour or more to completely remove Outlook Express, even if you know what you're doing. (And if you're installing for a clueless person, removing Outlook Express is *top* priority; otherwise you'll be back to do a reinstall next week or next month.)
Windows 98 SE may crash more often, but you reboot it and everything is all better again, so for most home users that's really not a big deal. (It would drive me crazy, but that's because I'm the sort of geek who leaves web browser windows open for months at a time.) If you need System Restore, Windows Me has that, without Windows XP's bugs and nastiness.
It's not always true that the latest version is best -- and that's not unique to Microsoft. Heck, Gnome2 *still* hasn't reached the level of functionality of Gnome 1.4. Don't be a slave to version numbers. Use the version that meets your needs best, whether it's got the biggest number or not.
Here are some books I enjoyed, where the father figure is white-collar:
* Ordinary Jack. Not only is this one of the best books ever written, but
the father is a writer, who works at home in his study. Also, Uncle
Parker's job is something to do with the stock market, so I'd call that
white collar too. This one is my number-one top recommendation. * I think the father in A Wrinkle in Time is a scientist. * I don't know if Calvin & Hobbes is the sort of literature you had in mind,
but Calvin's dad is a patent attourney. * The adult male in The Chronicles of Narnia is a professor. He's an uncle or
something rather than the father, but the children are living in his house. * In the Lord of the Rings, Frodo's father-figure Bilbo (though not technically
his father) doesn't have a very well-defined occupation, but whatever it is
it's definitely white-collar. He teaches Frodo to read and write Elvish.
Also, nine years old isn't too young to start reading biographies and nonfiction.
> Anyone who thinks Microsoft is going to announce its GENUINE thoughts > about Linux to the public world is deceiving themselves.
You can read between the lines. If you analyze what he said, you can tell which questions the inverviewee found personally interesting and has been thinking about and let some of his actual thoughts slip out. Other questions you can tell he wasn't really interested in the question, or his interest in the question was professional, and in those cases he just gave the official line or whatever. (The most obvious case of this was the question about commissioning research. The answer about made me dizzy, it was spun so hard.)
Really, you should read the interview. There's spin there, sure, but there's also some quite interesting stuff in it. The interviewee understands some things that I would not have expected a Microsoft exec to understand; he is clearly the right guy at MS for his job.
> Admins don't (and shouldn't) rely on Microsoft's or anybody else's > regression and breakage testing anyway.
Indeed, they shouldn't rely on it, but that doesn't mean they don't want it done. The system vendor absolutely should be doing regression testing. If they don't, then the admins are going to be finding regressions nearly every time, which means not deploying, contacting the vendor, filing a trouble ticket, waiting...
It is absolutely true that the admins must do their own regression testing. But they should *usually* not find any regressions, if the vendor is doing a proper job. When they do find an occasional regression, it should be due to something special about their setup that caused an issue to manifest itself that was missed in the vendor's testing.
And yes, vendors like IBM and Novel do testing, as well they should.
> Seriously, look at the list of features being added to Perl 6 and you get > the idea that they're being heavily influenced by RubySeriously, look at > the list of features being added to Perl 6 and you get the idea that > they're being heavily influenced by Ruby
When the Ruby people look at Perl6, they see Ruby. When the Scheme people look at Perl6, they see Functional Programming. When the Smalltalk people look at Perl6, they see Smalltalk. Indeed, all of these have contributed heavily to the design of Perl6 (though, none as much as Perl5). Perl is fundamentally a multiparadigmatic language. The Perl community actively hunts down other languages and takes their nifty features. We're not just getting vastly improved OO -- we're also getting *much* better FP, and other things as well.
But it's true that Parrot is a really cool thing about Perl6 and, indeed, running on Parrot is going to be the most important feature of every language that runs on Parrot, because running on Parrot will get you the ability to easily use libraries written in any language that runs on Parrot. The best very feature of Perl5 is the CPAN, and in Perl6 it's going to be a whole lot better.
> If it were just the desktop, I like KDE better. But I prefer the Gnome apps.
I prefer some Gnome apps (most notably, gnome-terminal and Gimp), some KDE apps (e.g., the calculator -- the Gnome calculator sucks), some independent apps (Emacs/Gnus, Mozilla, OpenOffice, Inkscape).
As far as the actual desktops, the last time I tried KDE it *still* didn't have panel drawers, which makes it unusable for me. Actually, for the panels, Gnome 1.4 is still the best and most configurable.
For the window manager, I use sawfish. Metacity is even less configurable than the Microsoft window management stuff, if that's possible.
The worst thing about Gnome, though, is that it virtually insists on running Nautilus 24/7. You kill it, and it comes back. You killall -9 it, and it comes back. It took me nearly an hour to figure out how to remove Nautilus from my Gnome session. (By the end, my family could hear me muttering, "Die, already, die, Nautilus, die, die, die".) The worst part about this is, Nautilus serves no useful purpose, other than to set the desktop background color and/or wallpaper once at the beginning of the session (which is largely irrelevant for me, since I almost never *see* the desktop, burried as it always is behind umpteen windows that I never bother to minimize (much less close)). I suppose some people use Nautilus for copying and moving files, but I've done my file management from the commandline since DOS 3.3, because it was faster and more flexible (think wildcards) than drag-and-drop file management and the tab completion in *nix shells makes that even *more* efficient, to say nothing of what you can do with a little Perl one-liner. Stuff that would take *hours* with Nautilus or the equivalent can be done in seconds.
Add to this that Nautilus is a system-resource hog. I'm not normally very critical of resource consumption. I run OpenOffice and six Mozilla windows (each with 20-30 tabs), several Emacsen, half a dozen terminals... leave MySQL and Apache running in the background all the time on my desktop even though I go days without using them, because it's handy to have them there when I want to test something... I have a lot of RAM and don't mind using it, because that's what it's there for. But Nautilus uses *way* more of the system's resources than is anywhere near reasonable for an app that doesn't actually *do* much. Performance is MUCH better with it removed from my session.
> So, uh, you run IE on your Gnome1 system instead?
No, silly, I'm using SeaMonkey.
> The GP was comparing running Firefox vs IE/Avant/XP SP2/Ad-blocker software, > on Windows.
Yes, but I was pointing out that Windows isn't the only system you ever have to fight to update. I loathe Windows, don't get me wrong, but I prefer to criticize it for things it does wrong that the competition gets right.
> As I understand it, the Thunderbird spam filter is bayesian.
Yes, that's right. I'm not sure whether it's completely naive, or whether it attempts to compensate for basic bayesian-filter-evasion techniques.
> Given the right stimuli, this can work very well.
It gets *way* too many false positives. Bayesian filters in general have this problem. You end up having to go looking through the spam filter to find the real messages that got there by mistake -- at which point, the filter is doing you no practical good at *all*. A filter that only catches half of the spam but never gets a false positive is in practice much better.
I filter based on character set (it's a truism that anything I can't read isn't worth my time to look at to see if it's spam or not), which by itself filters out about a third of my spam. (Almost a quarter of the spam I get is GB2312 alone.) Then on top of that I run regular expression filters to remove certain other perpetual classes of spam (e.g., P@Xi1 and \/i@gra), plus filters to take out the easy-to-filter stuff (e.g. Columbus Streetmail, Hermess Newsletter, and anything else that comes several times a day and follows a reliable pattern). After all that is said and done, about 20% of my spam lands in my inbox, and I go through that 20% manually.
> Thunderbird spam filtering didn't meet my needs either
Heck, Thunderbird's filtering doesn't meet my needs even for non-spam mail. But then, I use Gnus, so I don't think I'm really part of the target market for Thunderbird. That'd be like trying to sell a kayak as a replacement for a submarine, because it's leaner and meaner and easier to learn to operate and has "all the features most people really need".
> How can MS compete with a name like Firefox? Or Thunderbird
If they wanted the "cool name" factor, their marketing department wouldn't have *any* trouble doing better than "Firefox" or "Thunderbird". However, MS doesn't really *want* that image. They're trying to be corporate and respectable and stuff, so "Explorer" is just about as wacky and out-there as they want to go with a product name. If anything, they'll probably rename it "Microsoft Internet" one of these releases.
> if someone wrote a Firefox extension to correct for Slashdot's new IT > colour-scheme
1. Install the thEmacs GTK theme. 2. In Firefox, open the prefs, click the "Fonts and Colors" button. 3. Check "Use system colors" and Select "always use my colors". Hit Ok. 4. ??? 5. No more unpleasant color schemes, on *any* site.
I haven't browsed with page colors enabled since I gave up Navigator 4.08.
I installed Firesomething at home, set all the modifiers to words like "Fire", "Fear", "Death", "Pain", and so on, and set all the noun words to weapons and stuff. So now I browse with things like FearBlayde, QuickKnife, PowerLance, BrazenAxe, DeathCannon, PainHalberd, LightningSword,...
Slashdot Mirror
> I wonder what other words are _not_ so frequently used then.
Use Google, and try to get the lowest number you can get for the number of
pages. Yes, this is a variant on GoogleWhacking, but with only one word.
Some quick attempts: Google finds 76,500 pages using 'rotund' (round),
31,000 for 'pneumatology' (the study of the [sS]pirit), 13,900 for 'cromulent'
(valid), 818 for 'pimola' (a stuffed olive), 242 for 'anatopism' (something
that is out of place), and only 31 for 'propretonic' (preceding the syllable
before the accent).
I chose "pimola" because I happen to know that it's not listed in the OED, so
I figured it was fairly uncommon, but it turns out that a couple of the other
words I tried are even less common. I was surprised that "propretonic" isn't
used more often. FWIW, the sites that do use it probably use it numerous
times each.
> I would have thought things like "the", "and" or "or"
"the" isn't nearly as common in English as the definite article in many other
languages. We tend to use it once where other languages would use it several
times. For example, what might be "the words the common the most" in another
language comes out as "the most common words" in English -- so the article
occurs only once instead of three times. This hurts its rank considerably.
Also we don't use it on proper nouns, and sometimes we just leave it off and
let the reader figure out from context that the word is definite. (Having
indefinite articles helps here; if we wanted the word to be indefinite we
could explicitely tag it as indefinite, so with no article at all the word's
definiteness is ambiguous.)
"and" is fairly common, but I would think it would be beat out by words like
"to", "for", "from", and "that". ("that" has an unfair advantage because it's
actually about several different words spelled and pronounced the same way;
most notably it's a relative pronoun ("the word that introduces a relative
clause"), the most common of the demonstratives ("yes, that word"), and a
very common subordinating adverb ("I provided examples so that you might know
what I mean."); it would probably be _the_ most commonly used word in the
language if we didn't elide it more than half the time.)
Also, some of the uses of "and" are picked up by other words. English has
separate words for "even", "also", and half a dozen other concepts that tend
to be the same word as "and" in some languages.
> One Palmyra Atoll dollar = 17 pieces of mithril, or approximately twenty
> kilograms of fairy dust. There's no such thing, people. This is a joke.
Actually, mithril does exist. It's also known as titanium ore or titanium
steel (depending on whether it's being mined or whether a worked object is
being described).
It's Adamantium that hasn't been invented yet. Unlike mithril, adamantium is
not vulnerable to the liquid nitrogen freeze-and-shatter attack. It is
speculated to contain plastic polymer in addition to metal alloy, but we don't
know how to actually make it.
HTH.HAND.
> would like a bunch of buzzword bozos to READ EVERY DAMN MAIL YOU GET
That wouldn't bother me. But I'd be very worried about their accuracy.
> The daughter is nine MONTHS old.
Oh. Heh. Store away all my suggestions for half a dozen years or so then.
> What's your source for the 9% transmission losses? I have always heard a
> figure of about 2%, which is a lot more reasonable.
It varies, depending on stuff. Among other things, the further you are from
the power plant, the more loss there is on the way to you. It can be as high
as 20% in some cases, but the average is lower than that. 2% sounds like a
minimum, best-case-scenerio to me.
> > the entire Microsoft organisation (which undoubtedly employs some of
> > the world's finest software engineers and quality assurance experts)
> What ever happened to judging people by their results...
He was judging by _quantity_ of results. Microsoft produces *lots* of software.
> if you're talking PC, why wouldn't you install XP?
I can think of some reasons. It's not safe to connect it to the internet
without an external firewall. File and print sharing don't work properly.
It takes half an hour or more to completely remove Outlook Express, even if
you know what you're doing. (And if you're installing for a clueless person,
removing Outlook Express is *top* priority; otherwise you'll be back to do a
reinstall next week or next month.)
Windows 98 SE may crash more often, but you reboot it and everything is all
better again, so for most home users that's really not a big deal. (It would
drive me crazy, but that's because I'm the sort of geek who leaves web browser
windows open for months at a time.) If you need System Restore, Windows Me
has that, without Windows XP's bugs and nastiness.
It's not always true that the latest version is best -- and that's not unique
to Microsoft. Heck, Gnome2 *still* hasn't reached the level of functionality
of Gnome 1.4. Don't be a slave to version numbers. Use the version that
meets your needs best, whether it's got the biggest number or not.
> my $jane = Girl->new(age => 7);
> my $spot = Doggie::JackRusselTerrier->new();
> $jane->see($s pot); $jane->see($spot->run);
The Inform language would be ideally suited for this:
object jane "Jane"
class Girl,
with age 7
react_before [;
Run: if (actor == spot) { <<See spot>>; }
];
object spot "Spot"
class JackRusselTerrier;
initialize [; ChangePlayer(jane); ];
Here are some books I enjoyed, where the father figure is white-collar:
* Ordinary Jack. Not only is this one of the best books ever written, but
the father is a writer, who works at home in his study. Also, Uncle
Parker's job is something to do with the stock market, so I'd call that
white collar too. This one is my number-one top recommendation.
* I think the father in A Wrinkle in Time is a scientist.
* I don't know if Calvin & Hobbes is the sort of literature you had in mind,
but Calvin's dad is a patent attourney.
* The adult male in The Chronicles of Narnia is a professor. He's an uncle or
something rather than the father, but the children are living in his house.
* In the Lord of the Rings, Frodo's father-figure Bilbo (though not technically
his father) doesn't have a very well-defined occupation, but whatever it is
it's definitely white-collar. He teaches Frodo to read and write Elvish.
Also, nine years old isn't too young to start reading biographies and
nonfiction.
> I think you should be educating your users on things not to do
Yeah, like 1: Don't use Outlook and 2: Don't use Outlook Express.
Any other education you do is for nothing if your users are using Outlook.
> Anyone who thinks Microsoft is going to announce its GENUINE thoughts
> about Linux to the public world is deceiving themselves.
You can read between the lines. If you analyze what he said, you can tell
which questions the inverviewee found personally interesting and has been
thinking about and let some of his actual thoughts slip out. Other questions
you can tell he wasn't really interested in the question, or his interest in
the question was professional, and in those cases he just gave the official
line or whatever. (The most obvious case of this was the question about
commissioning research. The answer about made me dizzy, it was spun so hard.)
Really, you should read the interview. There's spin there, sure, but there's
also some quite interesting stuff in it. The interviewee understands some
things that I would not have expected a Microsoft exec to understand; he is
clearly the right guy at MS for his job.
> Admins don't (and shouldn't) rely on Microsoft's or anybody else's
> regression and breakage testing anyway.
Indeed, they shouldn't rely on it, but that doesn't mean they don't want it
done. The system vendor absolutely should be doing regression testing. If
they don't, then the admins are going to be finding regressions nearly every
time, which means not deploying, contacting the vendor, filing a trouble
ticket, waiting...
It is absolutely true that the admins must do their own regression testing.
But they should *usually* not find any regressions, if the vendor is doing a
proper job. When they do find an occasional regression, it should be due to
something special about their setup that caused an issue to manifest itself
that was missed in the vendor's testing.
And yes, vendors like IBM and Novel do testing, as well they should.
> Seriously, look at the list of features being added to Perl 6 and you get
> the idea that they're being heavily influenced by RubySeriously, look at
> the list of features being added to Perl 6 and you get the idea that
> they're being heavily influenced by Ruby
When the Ruby people look at Perl6, they see Ruby. When the Scheme people
look at Perl6, they see Functional Programming. When the Smalltalk people
look at Perl6, they see Smalltalk. Indeed, all of these have contributed
heavily to the design of Perl6 (though, none as much as Perl5). Perl is
fundamentally a multiparadigmatic language. The Perl community actively
hunts down other languages and takes their nifty features. We're not just
getting vastly improved OO -- we're also getting *much* better FP, and
other things as well.
But it's true that Parrot is a really cool thing about Perl6 and, indeed,
running on Parrot is going to be the most important feature of every language
that runs on Parrot, because running on Parrot will get you the ability to
easily use libraries written in any language that runs on Parrot. The best
very feature of Perl5 is the CPAN, and in Perl6 it's going to be a whole lot
better.
> If it were just the desktop, I like KDE better. But I prefer the Gnome apps.
I prefer some Gnome apps (most notably, gnome-terminal and Gimp), some KDE
apps (e.g., the calculator -- the Gnome calculator sucks), some independent
apps (Emacs/Gnus, Mozilla, OpenOffice, Inkscape).
As far as the actual desktops, the last time I tried KDE it *still* didn't
have panel drawers, which makes it unusable for me. Actually, for the panels,
Gnome 1.4 is still the best and most configurable.
For the window manager, I use sawfish. Metacity is even less configurable
than the Microsoft window management stuff, if that's possible.
The worst thing about Gnome, though, is that it virtually insists on running
Nautilus 24/7. You kill it, and it comes back. You killall -9 it, and it
comes back. It took me nearly an hour to figure out how to remove Nautilus
from my Gnome session. (By the end, my family could hear me muttering, "Die,
already, die, Nautilus, die, die, die".) The worst part about this is,
Nautilus serves no useful purpose, other than to set the desktop background
color and/or wallpaper once at the beginning of the session (which is largely
irrelevant for me, since I almost never *see* the desktop, burried as it
always is behind umpteen windows that I never bother to minimize (much less
close)). I suppose some people use Nautilus for copying and moving files,
but I've done my file management from the commandline since DOS 3.3, because
it was faster and more flexible (think wildcards) than drag-and-drop file
management and the tab completion in *nix shells makes that even *more*
efficient, to say nothing of what you can do with a little Perl one-liner.
Stuff that would take *hours* with Nautilus or the equivalent can be done
in seconds.
Add to this that Nautilus is a system-resource hog. I'm not normally very
critical of resource consumption. I run OpenOffice and six Mozilla windows
(each with 20-30 tabs), several Emacsen, half a dozen terminals... leave
MySQL and Apache running in the background all the time on my desktop even
though I go days without using them, because it's handy to have them there
when I want to test something... I have a lot of RAM and don't mind using
it, because that's what it's there for. But Nautilus uses *way* more of
the system's resources than is anywhere near reasonable for an app that
doesn't actually *do* much. Performance is MUCH better with it removed from
my session.
> shouldn't the GNOME guys worry more about basic functionality
That would violate the fundamental "no features" principle that has governed
Gnome development since 2.0.
I want my always-on-top clock panel that apps don't avoid on maximize back.
> So, uh, you run IE on your Gnome1 system instead?
No, silly, I'm using SeaMonkey.
> The GP was comparing running Firefox vs IE/Avant/XP SP2/Ad-blocker software,
> on Windows.
Yes, but I was pointing out that Windows isn't the only system you ever have
to fight to update. I loathe Windows, don't get me wrong, but I prefer to
criticize it for things it does wrong that the competition gets right.
> ctrl + D on firefox focuses the address bar, but doesn't highlight the text.
What you want is Ctrl + L.
> Backspace doesn't go 'back'
And, more generally, there's no easy UI for remapping the keyboard shortcuts.
There should be.
> As I understand it, the Thunderbird spam filter is bayesian.
Yes, that's right. I'm not sure whether it's completely naive, or whether
it attempts to compensate for basic bayesian-filter-evasion techniques.
> Given the right stimuli, this can work very well.
It gets *way* too many false positives. Bayesian filters in general have
this problem. You end up having to go looking through the spam filter to
find the real messages that got there by mistake -- at which point, the
filter is doing you no practical good at *all*. A filter that only catches
half of the spam but never gets a false positive is in practice much better.
I filter based on character set (it's a truism that anything I can't read
isn't worth my time to look at to see if it's spam or not), which by itself
filters out about a third of my spam. (Almost a quarter of the spam I get
is GB2312 alone.) Then on top of that I run regular expression filters to
remove certain other perpetual classes of spam (e.g., P@Xi1 and \/i@gra),
plus filters to take out the easy-to-filter stuff (e.g. Columbus Streetmail,
Hermess Newsletter, and anything else that comes several times a day and
follows a reliable pattern). After all that is said and done, about 20%
of my spam lands in my inbox, and I go through that 20% manually.
> Thunderbird spam filtering didn't meet my needs either
Heck, Thunderbird's filtering doesn't meet my needs even for non-spam mail.
But then, I use Gnus, so I don't think I'm really part of the target market
for Thunderbird. That'd be like trying to sell a kayak as a replacement for
a submarine, because it's leaner and meaner and easier to learn to operate
and has "all the features most people really need".
> How can MS compete with a name like Firefox? Or Thunderbird
If they wanted the "cool name" factor, their marketing department wouldn't
have *any* trouble doing better than "Firefox" or "Thunderbird". However,
MS doesn't really *want* that image. They're trying to be corporate and
respectable and stuff, so "Explorer" is just about as wacky and out-there
as they want to go with a product name. If anything, they'll probably
rename it "Microsoft Internet" one of these releases.
> if someone wrote a Firefox extension to correct for Slashdot's new IT
> colour-scheme
1. Install the thEmacs GTK theme.
2. In Firefox, open the prefs, click the "Fonts and Colors" button.
3. Check "Use system colors" and Select "always use my colors". Hit Ok.
4. ???
5. No more unpleasant color schemes, on *any* site.
I haven't browsed with page colors enabled since I gave up Navigator 4.08.
> focus problems are partly mandated by the javascript standards
A page's Javascript should only even be active when that tab is selected. It
*certainly* shouldn't be able to steal focus away from another tab.
I installed Firesomething at home, set all the modifiers to words like ...
"Fire", "Fear", "Death", "Pain", and so on, and set all the noun words to
weapons and stuff. So now I browse with things like FearBlayde, QuickKnife,
PowerLance, BrazenAxe, DeathCannon, PainHalberd, LightningSword,