> Look at Win95 -- would anyone have ever switched to it from DOS if > it didn't run DOS and Win3.11 apps perfectly?
Well, it didn't (not by a *long* shot), and yet many people did.
Still, it "mostly" (translation: almost) ran "many" (translation: certain approved) legacy apps, and that was enough for the marketroids to _claim_ compatibility, and without that _claim_ many fewer people would have accepted Win95 as an upgrade path.
Precious metals and gems fall into the category of contingency investments. You don't want to invest in them for the purposes of growing your money, because they don't increase in value that much. Sometimes they decrease in value (though not by a lot). But regardless of what happens to the ecconomy, they continue to hold value. So if you have a lot of money, you put some of it in precious metals as a safety net in case the ecconomy goes completely to pot and your stocks and bonds and whatnot become so much used paper. The only time precious metals lose their value is in the case of prolonged severe famine (which, with modern transportation being what it is, would have to be global to have that effect these days).
Blue chip stocks and mutual funds are a different category. They are very resilient to small turmoils that impact only segments of the ecconomy, but if the whole stock marget goes to pieces they will too. They rise and fall with the ups and downs of the whole market, but (so far at least) they increase significantly in value over the long term. They are considered to be good investments for the 10-years-plus timeframe. Real estate is generally also good for a long-term investment, assuming you can afford the property tax. IBM is a blue chip stock.
Then there are companies like SCO, relative nobodies (when compared to the blue chips), but not really new. You can invest in these for short term gains if you think they are about to increase in value. This is guesswork. They're riskier than the long-term stocks, because they fluctuate more, but they're (usually) not as risky as startups.
With startups, you invest only a small amount (unless you _really_ believe in them or something), because chances are you'll lose it, but there's a small chance they'll grow hugely and you'll make a killing.
Of course, I'm not an econ major, much less an investment analyst, but these are just basic categories and overall guidelines. There are always specific deviations. Such as, right now, SCO is a riskier choice than usual for a company its size and age. Unless you know something we don't.
> David Boies (ahem, didn't the justice department win the battle > but lose the war in their anti-trust suit against microsoft??)
I was going to point that out myself. SCO is represented by the same guy who got basically nothing of any significance out of Microsoft (when MS was _clearly_ in the wrong). IBM has is just as well-represented legally as MS, and SCO thinks they're going to get something out of this? Haha. They'd better have one heck of a good case. I for one find it difficult to believe that SCO's case against IBM is anything like as strong as the DOJ's case against Microsoft was. And if it's not, they'll get jack squat.
I have thought for a long time now that Id ought to do a new, spiffy, 3D first-person version of Commander Keen. Keen had several things going for it. The characters were cartooney, and everyone knows that cartooney violence is much more fun than stuff that looks real. In addition, the pogo stick was really innovative and added a great deal to the gameplay. I'd particularly like to see that item done in a first-person 3D version of the game.
> Depending on just how serious you are about being without power
Exactly. Is your friend going to England and France, Mexico and China, or Bangledesh and the Cameroun? Urban, or rural? Are we talking about being without power during the day and sometimes for 2-3 days because not _all_ of the hotels have power, or are we talking about being sixty miles (on nothing that we would consider to be roads in any North-American sense of that word) from the nearest place that has power certain days of the week, except when doesn't work when it's supposed to?
Regarding sending things back: are we talking about Europe, where you can always get to a phone line within two days (usually much more easily than that) and there's fairly reliable postal service in nearly every country? Or is he going to South America, where the mail service is abysmal, or Africa where (in large areas) there are no phone lines? "Around the world" covers a lot of territory. The difference between Paris and Abong Mbong is the difference between "make sure the recharger can handle 220/50 current" and "buy solar".
Also bear in mind that outside of North America, what power you get is unlikely to be 110-120 volts at 60 Hertz. Anything you get MUST be able to run on 50-Hertz current anywhere from 100-250 volts, and it may not exactly (ahem) be a smooth sine wave either. Either that or batteries.
If you're going to the third world, I suggest either solar, or stuff that runs entirely on flashlight batteries (a common size, not smaller than AA, not larger than D, and not obscure like B), since you can go weeks without seeing power in some areas. And carry a couple spare sets of batteries all the time, because sometimes you may not be able to buy them on short notice. Bonus points if it uses rechargeable batteries and you have a solar recharger. In that case, take three sets of batteries (one that is in the device, one that is recharged and ready, and one that is in the recharger).
If you're going to only first and second world nations and the occasional breif stop in a major metropolis in the third world, you can probably get away with equipment that charges from either 110 or 220 volts, provided it can handle 50-Hertz current -- which you must not take for granted; a lot of stuff you buy in the US will fry on 50-Hertz current, or at least not work properly. And you don't want to carry around a converter, because it will be heavy. (To convert the cycle, you actually have to convert to DC and back, so the equimpent that does it contains an inverter and therefore has significant mass. Not for backpacking.)
One other thing, should be obvious: get a camera that lets you preview your pictures and delete some you don't want (to make room on your current storage whatsit) so that you don't have to wait to get back to a place where you can use your laptop/whatever before consolidating. You will want to take at *least* two of every picture you want and keep the better one. At home you would keep them all until you get back to your PC at least, but on the road you may need the space for other pics then and there.
I like the political cartoon I saw recently. The Dem donkey is holding up a sign listing several things W has done (or is perceived as having done), and a couple of people labelled as voters are looking at the sign, smiling. The caption reads, "No, no, those are reasons _not_ to reelect him." Uh-huh.
It remains to be seen whether he'll retain that popularity long enough to make the re-election. Of course, he's going to be running against either Gore or Hillary. If Gore gets the party nomination, W is basically a shoe-in for re-election. (Yeah, it was really close before and all that, but it's been four years and life has gone on and stuff has happened and neither of them has the same public image as then.) If the Dems go with Hillary, I can't predict what will happen for sure, except that there will be a lot of people with strong feelings about the election.
I'm hoping Gore gets the nomination, because I _really_ don't want a Clinton to be our first female president. (I was hoping for Liz Dole, but that didn't happen. Yeah, I know people who read slashdot aren't supposed to be conservative, but too bad. Just be glad I didn't openly admit that I liked Quayle, since that would surely start a flamefest.)
> it's not like Microsoft have updated IE for the Mac in forever
I believe the exact timeframe is forever and a day. HTH.HAND.
Seriously, does anybody really believe Microsoft was planning, prior to the release of Safari, to do any further work on Mac IE? I had already concluded months ago that they had stopped development of it after OS X came out and that apart from recompiling it under Carbon libs they were done working on it.
I've seen the latest version of MSIE for the Mac, and it's a toss-up between that and Navigator 4.08 which is better. I just feel sorry for the people stuck on a MacOS version prior to 8.6, because there is AFAIK no even remotely decent browser available for it anywhere. Time to buy and OS upgrade, I guess... but jeez, needing to buy an OS upgrade just to get a half-decent web browser? Bummer. (Okay, so there are other reasons to upgrade from MacOS 8... but it's the principle of the thing.)
> The future is so bright I gotta wear shades!
And just think, in a few years Perl6 will come too!
They had better use 1.4 for the next major Netscape release), because with the dropping of XPFE for 1.5 according to the new roadmap, it's going to be at _least_ six months before the next reasonably usable browser codebase (Firebird, which is extremely beta still) hits release quality, and two years or more before we have a decent mail/news client (from Thunderbird, which at this point is only barely alpha and won't be able to replace Messenger (to say nothing of a real mailreader) for months and months).
I was (in the 1.2 days) predicting 1.5 or 1.6 for the next Netscape release, but at this point they need to work with 1.4, because 1.5 and 1.6 aren't going to be gamma quality. Then again, neither was the branch that they based 6.x on, really, so who knows what they will do. But what they *ought* to do is take 1.4 and work with it for the next branch. Or 1.3, which is stable in my experience. 1.5 will feature serious loss of functionality compared to 1.4, as a lot of things will still need to be ported over from XPFE to the new toolkit. Quite a few browser features (all minor, but lots of minor features add up), plus almost everything outside the browser component.
That means we're probably looking at Netscape 9.x in circa 2006 before Netscape users get the flexibile architecture features (extensions and whatnot) that were developed for Phoenix.
This is all entirely off-topic in a discussion about Mac IE, of course, but hey, this is slashdot.
Huh. I wasn't able to get it to work with IE6 on WinXP; in fact, the page at mongus.net doesn't work for me either. It does work in Mac IE 5.0, though.
Weird.
> I'd still prefer not dealing with IE at all.
It wouldn't be so bad if it were possible to have *one* PC and keep several versions of IE installed on it (say, 5.0, 5.5, and 6.0). Right now the only way to do that AFAIK is with VMWare or the equivalent. (Eventually, I'm going to have to break down and buy VMWare anyway, but I was hoping to wait until I get some better hardware so it won't be unbearably slow...)
See, I can have one PC and one OS installation and test my stuff in assorted versions of Mozilla, assorted versions of Opera, assorted minor browsers (Lynx, Links, Konqueror, W3, and even in Navigator 4 if I care), but if I want to test in multiple versions of MSIE I basically need a separate PC for each. This makes it highly impractical for a small-time webmaster to support more than one version of MSIE. You can petition friends for screenshots (and I do), but that only shows you a problem; it doesn't help you reduce the page to a testcase, isolate the problem, and work around it.
If I can justify working on something at work (like if I'm testing a technology, like PNG alpha transparency, which I might potentially use at work), then I have access to two versions of IE on Windows (two different PCs) and one on a Mac. But I can't justify testing my personal websites at work -- at least, not at all extensively. At home, I only can keep one version of MSIE installed at a time, so testing in multiple versions is totally impractical.
Someday I'll break down and buy VMWare. But it annoys me that I should have to do that.
> How much time do you think it's needed to take a snapshot of > the Web? Most certainly much longer than a day or even a week. > My bet would be several months at the very least.
I strongly suspect some pages get re-evaluated much more often than others. At least, it _ought_ to be that way. For one thing, pages with a higher PageRank are more important (to the evaluation of other pages) and thus should get redone more often. Additionally, pages that are known to change frequently should be done more often than pages that are not known to have changed previously. How to ballance those two considerations is an open question.
Backwards you have it. You start at leaf nodes (pages that don't link to anything) and work backwards to pages that link to the leaf nodes, pages that link to those, and so forth.
Initially you have lots of webs -- one for each leaf node. As you add pages that link to them, these will tend to get joined as you find that many of them belong to some of the same trees.
Oh, and you reserve known search engines for last.
> I wonder if this data would be hugely different from the > number of visits a page receives
Dunno. But there's no way for Google to know how many visits a page receives; whereas, they can calculate how well-linked it is.
It would be interesting to do a study on the relationship between pagerank and page loads and number of distinct visitors, but that would require having log data from all the servers involved; it would be pretty easy to get log data from a small number of servers, but getting those data for a decent subset of the internet would be significantly less easy.
I've managed to verify that this works in Mac IE 5.0, but I can't
seem to get it to work in either version of MSIE for Windows that
I have access to at work (5.0 on Win98 or 6.0 on XP Home). Am I
doing something wrong? Here's where I'm trying...
Like I said, IE6 on XP Home seems to do the same thing as IE5 on
Win98, namely, ignore the hack. Does it work only in 5.5 and on
the Mac, or am I applying it incorrectly?
There are also some layout issues with Mac IE 5, but I'll break
those down into testcases separately another time. Right now
I'm mainly interested in the PNG transparency.
Address blocking is worthless anyhow. The spammers who send 99.75% of the spam[1] use software that automatically generates a new random From address for each message. Something like this...
open WORDS, "<listofnames.txt"; @word = map {chomp;$_} <WORDS>; close WORDS; @tld = qw(com net org); foreach (@messages_to_send) {
my $from = $word[rand @word]
. "@"
. $word[rand @word]
. "."
. $tld[rand @tld];
sendspam($_, $from); }
Some of the less sophisticated ones don't even bother to use a namelist, just generate random letters, so the address comes out looking like oliejlamvr@lcjoiwleru.com
This is why you start a class action suit. If you can get a few hundred thousand people to go in on it, the $11000 per call starts to sound like real money, enough to pay some legal fees almost. Then your lawyers agree to settle out of court, take their cut, and leave you with $1.50 for each plaintiff, which still isn't much, but it's a positive number.
I prefer to just say something along the lines of "can you explain that in detail?" and then gently set the phone down on the counter and go do something else for a while.
Not looseleaf paper. Get a separate book of paper (the perforated notebooks are good) for each class, and make sure the cover is a different color for each class during any given semester. Take a big fat magic marker and put the name of the class (or a suitable abbreviation) on the cover in about 96-point lettering.
Do you want a computer also? Absolutely. Do you want to take notes on it? Probably not. You can sketch charts faster on paper, circle things, draw arrows,...
If you're the kind of person who has much trouble with grabbing the wrong thing in the morning, you can make sure each course has the same colour of notebook, folder, and book cover(s). Then you just have to remember that Tuesdays and Thursdays you grab Blue and Red, and Mondays, Wednesdays, and Fridays you grab Yellow and Black knd the morning and Green in the afternoon, or whatever.
I can't see that. For the desktop, you want... RH or mdk or whatnot, not TurboLinux. Yes, I've used TurboLinux, though it was a couple of versions ago. But that was on a cgi server. I can't see putting it on a desktop system intended for end users.
I remapped my keyboard so that there is no numlock; it's just always off. (Actually, what I did was remap it so that the cursor movement keys located physically on the keypad look to the computer like the other cursor movement keys that don't double as numbers.) I touch-type numbers on the top row, so the keypad has always been about cursor movement for me.
Yes, I don't think ratpoison is the best window manager for using Gimp. But then, Gimp is not exactly minimalist itself. (I'm not saying that's a bad thing. I use Gimp daily. But as I said, minimalism isn't my thing. I also use Mozilla, not Lynx or Links.)
Slashdot Mirror
All levels?
...
Sure, protocols, ports, source code, passwords,
Oh, wait...
Obscurity is a fine contribution to security, provided you have
something more to go with it. It just can't stand alone.
Denithore. Their father is Denithore. Arathorn is Strider's dad.
> Look at Win95 -- would anyone have ever switched to it from DOS if
> it didn't run DOS and Win3.11 apps perfectly?
Well, it didn't (not by a *long* shot), and yet many people did.
Still, it "mostly" (translation: almost) ran "many" (translation:
certain approved) legacy apps, and that was enough for the
marketroids to _claim_ compatibility, and without that _claim_
many fewer people would have accepted Win95 as an upgrade path.
Precious metals and gems fall into the category of contingency
investments. You don't want to invest in them for the purposes
of growing your money, because they don't increase in value that
much. Sometimes they decrease in value (though not by a lot).
But regardless of what happens to the ecconomy, they continue
to hold value. So if you have a lot of money, you put some of
it in precious metals as a safety net in case the ecconomy goes
completely to pot and your stocks and bonds and whatnot become
so much used paper. The only time precious metals lose their
value is in the case of prolonged severe famine (which, with
modern transportation being what it is, would have to be
global to have that effect these days).
Blue chip stocks and mutual funds are a different category. They
are very resilient to small turmoils that impact only segments of
the ecconomy, but if the whole stock marget goes to pieces they
will too. They rise and fall with the ups and downs of the whole
market, but (so far at least) they increase significantly in value
over the long term. They are considered to be good investments
for the 10-years-plus timeframe. Real estate is generally also
good for a long-term investment, assuming you can afford the
property tax. IBM is a blue chip stock.
Then there are companies like SCO, relative nobodies (when compared
to the blue chips), but not really new. You can invest in these
for short term gains if you think they are about to increase in
value. This is guesswork. They're riskier than the long-term
stocks, because they fluctuate more, but they're (usually) not
as risky as startups.
With startups, you invest only a small amount (unless you
_really_ believe in them or something), because chances are
you'll lose it, but there's a small chance they'll grow hugely
and you'll make a killing.
Of course, I'm not an econ major, much less an investment analyst,
but these are just basic categories and overall guidelines. There
are always specific deviations. Such as, right now, SCO is a
riskier choice than usual for a company its size and age. Unless
you know something we don't.
> David Boies (ahem, didn't the justice department win the battle
> but lose the war in their anti-trust suit against microsoft??)
I was going to point that out myself. SCO is represented by the
same guy who got basically nothing of any significance out of
Microsoft (when MS was _clearly_ in the wrong). IBM has is just
as well-represented legally as MS, and SCO thinks they're going
to get something out of this? Haha. They'd better have one heck
of a good case. I for one find it difficult to believe that SCO's
case against IBM is anything like as strong as the DOJ's case
against Microsoft was. And if it's not, they'll get jack squat.
I have thought for a long time now that Id ought to do a new, spiffy,
3D first-person version of Commander Keen. Keen had several things
going for it. The characters were cartooney, and everyone knows that
cartooney violence is much more fun than stuff that looks real. In
addition, the pogo stick was really innovative and added a great
deal to the gameplay. I'd particularly like to see that item done
in a first-person 3D version of the game.
> Depending on just how serious you are about being without power
Exactly. Is your friend going to England and France, Mexico and
China, or Bangledesh and the Cameroun? Urban, or rural? Are we
talking about being without power during the day and sometimes for
2-3 days because not _all_ of the hotels have power, or are we
talking about being sixty miles (on nothing that we would consider
to be roads in any North-American sense of that word) from the
nearest place that has power certain days of the week, except when
doesn't work when it's supposed to?
Regarding sending things back: are we talking about Europe,
where you can always get to a phone line within two days (usually
much more easily than that) and there's fairly reliable postal
service in nearly every country? Or is he going to South America,
where the mail service is abysmal, or Africa where (in large areas)
there are no phone lines? "Around the world" covers a lot of
territory. The difference between Paris and Abong Mbong is
the difference between "make sure the recharger can handle 220/50
current" and "buy solar".
Also bear in mind that outside of North America, what power you get
is unlikely to be 110-120 volts at 60 Hertz. Anything you get MUST
be able to run on 50-Hertz current anywhere from 100-250 volts, and
it may not exactly (ahem) be a smooth sine wave either. Either that
or batteries.
If you're going to the third world, I suggest either solar, or
stuff that runs entirely on flashlight batteries (a common size,
not smaller than AA, not larger than D, and not obscure like B),
since you can go weeks without seeing power in some areas. And
carry a couple spare sets of batteries all the time, because
sometimes you may not be able to buy them on short notice. Bonus
points if it uses rechargeable batteries and you have a solar
recharger. In that case, take three sets of batteries (one that
is in the device, one that is recharged and ready, and one that
is in the recharger).
If you're going to only first and second world nations and the
occasional breif stop in a major metropolis in the third world,
you can probably get away with equipment that charges from either
110 or 220 volts, provided it can handle 50-Hertz current -- which
you must not take for granted; a lot of stuff you buy in the US
will fry on 50-Hertz current, or at least not work properly. And
you don't want to carry around a converter, because it will be
heavy. (To convert the cycle, you actually have to convert to DC
and back, so the equimpent that does it contains an inverter and
therefore has significant mass. Not for backpacking.)
One other thing, should be obvious: get a camera that lets you
preview your pictures and delete some you don't want (to make room
on your current storage whatsit) so that you don't have to wait
to get back to a place where you can use your laptop/whatever
before consolidating. You will want to take at *least* two of
every picture you want and keep the better one. At home you
would keep them all until you get back to your PC at least, but
on the road you may need the space for other pics then and there.
I like the political cartoon I saw recently. The Dem donkey is
holding up a sign listing several things W has done (or is perceived
as having done), and a couple of people labelled as voters are
looking at the sign, smiling. The caption reads, "No, no, those
are reasons _not_ to reelect him." Uh-huh.
It remains to be seen whether he'll retain that popularity long
enough to make the re-election. Of course, he's going to be
running against either Gore or Hillary. If Gore gets the party
nomination, W is basically a shoe-in for re-election. (Yeah, it
was really close before and all that, but it's been four years
and life has gone on and stuff has happened and neither of them
has the same public image as then.) If the Dems go with Hillary,
I can't predict what will happen for sure, except that there will
be a lot of people with strong feelings about the election.
I'm hoping Gore gets the nomination, because I _really_ don't
want a Clinton to be our first female president. (I was hoping
for Liz Dole, but that didn't happen. Yeah, I know people who
read slashdot aren't supposed to be conservative, but too bad.
Just be glad I didn't openly admit that I liked Quayle, since
that would surely start a flamefest.)
> it's not like Microsoft have updated IE for the Mac in forever
I believe the exact timeframe is forever and a day. HTH.HAND.
Seriously, does anybody really believe Microsoft was planning,
prior to the release of Safari, to do any further work on Mac IE?
I had already concluded months ago that they had stopped development
of it after OS X came out and that apart from recompiling it under
Carbon libs they were done working on it.
I've seen the latest version of MSIE for the Mac, and it's a toss-up
between that and Navigator 4.08 which is better. I just feel sorry
for the people stuck on a MacOS version prior to 8.6, because there
is AFAIK no even remotely decent browser available for it anywhere.
Time to buy and OS upgrade, I guess... but jeez, needing to buy an
OS upgrade just to get a half-decent web browser? Bummer. (Okay,
so there are other reasons to upgrade from MacOS 8... but it's the
principle of the thing.)
> The future is so bright I gotta wear shades!
And just think, in a few years Perl6 will come too!
They had better use 1.4 for the next major Netscape release), because
with the dropping of XPFE for 1.5 according to the new roadmap, it's
going to be at _least_ six months before the next reasonably usable
browser codebase (Firebird, which is extremely beta still) hits
release quality, and two years or more before we have a decent
mail/news client (from Thunderbird, which at this point is only
barely alpha and won't be able to replace Messenger (to say nothing
of a real mailreader) for months and months).
I was (in the 1.2 days) predicting 1.5 or 1.6 for the next Netscape
release, but at this point they need to work with 1.4, because 1.5
and 1.6 aren't going to be gamma quality. Then again, neither was
the branch that they based 6.x on, really, so who knows what they
will do. But what they *ought* to do is take 1.4 and work with it
for the next branch. Or 1.3, which is stable in my experience.
1.5 will feature serious loss of functionality compared to 1.4, as
a lot of things will still need to be ported over from XPFE to the
new toolkit. Quite a few browser features (all minor, but lots
of minor features add up), plus almost everything outside the
browser component.
That means we're probably looking at Netscape 9.x in circa 2006
before Netscape users get the flexibile architecture features
(extensions and whatnot) that were developed for Phoenix.
This is all entirely off-topic in a discussion about Mac IE,
of course, but hey, this is slashdot.
Huh. I wasn't able to get it to work with IE6 on WinXP; in fact,
the page at mongus.net doesn't work for me either. It does work
in Mac IE 5.0, though.
Weird.
> I'd still prefer not dealing with IE at all.
It wouldn't be so bad if it were possible to have *one* PC and
keep several versions of IE installed on it (say, 5.0, 5.5, and
6.0). Right now the only way to do that AFAIK is with VMWare or
the equivalent. (Eventually, I'm going to have to break down and
buy VMWare anyway, but I was hoping to wait until I get some
better hardware so it won't be unbearably slow...)
See, I can have one PC and one OS installation and test my stuff
in assorted versions of Mozilla, assorted versions of Opera,
assorted minor browsers (Lynx, Links, Konqueror, W3, and even in
Navigator 4 if I care), but if I want to test in multiple versions
of MSIE I basically need a separate PC for each. This makes it
highly impractical for a small-time webmaster to support more than
one version of MSIE. You can petition friends for screenshots
(and I do), but that only shows you a problem; it doesn't help
you reduce the page to a testcase, isolate the problem, and work
around it.
If I can justify working on something at work (like if I'm
testing a technology, like PNG alpha transparency, which I might
potentially use at work), then I have access to two versions of
IE on Windows (two different PCs) and one on a Mac. But I can't
justify testing my personal websites at work -- at least, not at
all extensively. At home, I only can keep one version of MSIE
installed at a time, so testing in multiple versions is totally
impractical.
Someday I'll break down and buy VMWare. But it annoys me that
I should have to do that.
> How much time do you think it's needed to take a snapshot of
> the Web? Most certainly much longer than a day or even a week.
> My bet would be several months at the very least.
I strongly suspect some pages get re-evaluated much more often than
others. At least, it _ought_ to be that way. For one thing, pages
with a higher PageRank are more important (to the evaluation of
other pages) and thus should get redone more often. Additionally,
pages that are known to change frequently should be done more often
than pages that are not known to have changed previously. How to
ballance those two considerations is an open question.
Backwards you have it. You start at leaf nodes (pages that don't
link to anything) and work backwards to pages that link to the leaf
nodes, pages that link to those, and so forth.
Initially you have lots of webs -- one for each leaf node. As you
add pages that link to them, these will tend to get joined as you
find that many of them belong to some of the same trees.
Oh, and you reserve known search engines for last.
> I wonder if this data would be hugely different from the
> number of visits a page receives
Dunno. But there's no way for Google to know how many visits
a page receives; whereas, they can calculate how well-linked
it is.
It would be interesting to do a study on the relationship between
pagerank and page loads and number of distinct visitors, but that
would require having log data from all the servers involved; it
would be pretty easy to get log data from a small number of servers,
but getting those data for a decent subset of the internet would be
significantly less easy.
You forgot the shoes. We will feed them death, hell, and shoes.
I've managed to verify that this works in Mac IE 5.0, but I can't seem to get it to work in either version of MSIE for Windows that I have access to at work (5.0 on Win98 or 6.0 on XP Home). Am I doing something wrong? Here's where I'm trying...
my test page screenshot from IE5 on Windows screenshot from IE5 on Mac screenshot from geckoLike I said, IE6 on XP Home seems to do the same thing as IE5 on Win98, namely, ignore the hack. Does it work only in 5.5 and on the Mac, or am I applying it incorrectly?
There are also some layout issues with Mac IE 5, but I'll break those down into testcases separately another time. Right now I'm mainly interested in the PNG transparency.
Address blocking is worthless anyhow. The spammers who send 99.75%
of the spam[1] use software that automatically generates a new random
From address for each message. Something like this...
open WORDS, "<listofnames.txt";
@word = map {chomp;$_} <WORDS>; close WORDS;
@tld = qw(com net org);
foreach (@messages_to_send) {
my $from = $word[rand @word]
. "@"
. $word[rand @word]
. "."
. $tld[rand @tld];
sendspam($_, $from);
}
Some of the less sophisticated ones don't even bother to use
a namelist, just generate random letters, so the address comes
out looking like oliejlamvr@lcjoiwleru.com
[1] 96.785% of statistics are made up.
> This would be more interesting had they actually tested it on
> a standard 512kbs connection
A standard what?
Now, if they test it on a 33.6 ppp dialup connection and get
speeds 6000 times faster, or even 2 times faster, then I'll
be interested.
> In the end, very few 16 bit apps had problems on Win95.
Ehh? I must have tried to use every single one of them then.
This is why you start a class action suit. If you can get a few
hundred thousand people to go in on it, the $11000 per call starts
to sound like real money, enough to pay some legal fees almost.
Then your lawyers agree to settle out of court, take their cut,
and leave you with $1.50 for each plaintiff, which still isn't
much, but it's a positive number.
> i have better things to do...
I prefer to just say something along the lines of "can you explain
that in detail?" and then gently set the phone down on the counter
and go do something else for a while.
> The lost art of paper and pen?
...
Not looseleaf paper. Get a separate book of paper (the perforated
notebooks are good) for each class, and make sure the cover is a
different color for each class during any given semester. Take a
big fat magic marker and put the name of the class (or a suitable
abbreviation) on the cover in about 96-point lettering.
Do you want a computer also? Absolutely. Do you want to take
notes on it? Probably not. You can sketch charts faster on
paper, circle things, draw arrows,
If you're the kind of person who has much trouble with grabbing
the wrong thing in the morning, you can make sure each course has
the same colour of notebook, folder, and book cover(s). Then you
just have to remember that Tuesdays and Thursdays you grab Blue
and Red, and Mondays, Wednesdays, and Fridays you grab Yellow
and Black knd the morning and Green in the afternoon, or whatever.
> In the past, I've seen IBM pimping TurboLinux.
For the desktop?
I can't see that. For the desktop, you want... RH or mdk or
whatnot, not TurboLinux. Yes, I've used TurboLinux, though it
was a couple of versions ago. But that was on a cgi server.
I can't see putting it on a desktop system intended for end
users.
I remapped my keyboard so that there is no numlock; it's just
always off. (Actually, what I did was remap it so that the cursor
movement keys located physically on the keypad look to the computer
like the other cursor movement keys that don't double as numbers.)
I touch-type numbers on the top row, so the keypad has always been
about cursor movement for me.
Yes, I don't think ratpoison is the best window manager for using
Gimp. But then, Gimp is not exactly minimalist itself. (I'm not
saying that's a bad thing. I use Gimp daily. But as I said,
minimalism isn't my thing. I also use Mozilla, not Lynx or Links.)