Transliterated? Very strange. I tried my signature in the Mounce font, and a lot of the punctuation comes out as diacritical marks on the previous character. For example, where you see "map{my", the curley brace comes out as a rough breathing and accute accent, which looks pretty odd over the pi -- not to mention, the y comes out as a psi, so you end up with something that _if_ it could be pronounced would be roughly "maHPmps". On the other hand, what follows ("($a,$b)") comes out very nicely if you imagine someone using alpha and beta as variable names.
Heh. +1 Funny, but actually even transliteration isn't quite that straightforward, much less translation. If you take English and write it in a Greek-transliterative font, you get gibberish, which often isn't even pronounceable. Symbol is even worse, because some of the symbols aren't Greek letters or aren't even letters at all.
You will note, however, that the Symbol font only uses the Greek letters as symbols; it does not have provisions for diacritical marks, for example, or Greek punctuation (e.g., the ? glyph does not look like a semicolon) as any real transliterative font (Graeca, Mounce) would do.
And I *can* read Greek (albeit not quickly, and not modern Greek).
> In that case, shouldn't SGI be suing SCO for... ???
Not for profit. (Hello? Turnip? Blood?) However, they could sue them for assorted things (vague intellectual property violations come to mind) in an attempt to boost their stock price.
The linked usenet article seems to indicate (though not being a C
programmer it's hard to be certain) that the one bit of code (the one with the comment about allocating size units) was in BSD in 1984 and may have been changed at that time to fix a bug.
The other one (with the mutex and the assert) I don't know about,
but you can speculate about whether it may have been in BSD also.
> I don't see how the fact that Windows doesn't ship with > Apache is relevant.
I was trying to be nice. Allow me to rephrase so as to be more clear: "Doesn't ship with any web server software that isn't so infamous no sane sysadmin would permit it on his network".
> I'm also curious as to what constitutes a "decent" command > shell or text editor for you.
A decent command shell is one that's good enough you can easily do all your file management from it, so that you don't need a GUI file manager. One critical feature of a decent command shell is the ability to put the output of one command on the command line of another command, but there are other key features.
A decent text editor? Of course I mean Emacs, but I can make a *lengthy* list of important features that are present in twenty or thirty major text editors but absent in the ones that ship with Windows. The ones that ship with Windows don't even have basic macro capabilities, for crying out loud.
> There is also a secure shell included- terminal services.
I thought terminal services was a thinclient server? Maybe I was confused about that.
> why is there no grounds for pronouncing it as linnux?
Because it's neither etymologically nor phonetically correct.
> Lin is sometime prononced with a short i in English. Lint, > linear, lingo, liberty...
Lint is pronounced with a short i because it's phonetically correct that way. Lingo also is pronounced phonetically (rhymes with Ringo). Liberty is pronounced the way it is because it comes from French. (Compare with the pronunciation of Xouvert.) (Liberty's English pronunciation has drifted a bit from the French, but it's much closer to the French than it is to what would be phonetic in English.)
Linear is an interesting study. Etymologically, it comes from the Latin (wherein linea means line). It's pronounced that way, then, because the phonetic rendering in English is hard or impossible for most native speakers of the language to pronounce. The i and the e would both be long, and the a silent, thus: "line ear" -- easy to pronounce as two separate words, but when you run them together the dipthong gets mangled and comes out as two syllables. Also, some people have difficulty pronouncing the English long I in that combination. Anyway, the long and short of it is that people did not like saying it that way, so they used the Latin pronunciation, or something closer to it.
I prefer the other approach (speaking quickly): "Okay, I'm resetting your password. Your new password is capital Q period lowercase d forward slash seven tilde lowercase b uppercase I right curly brace exclamation mark zero uppercase N lowercase v backquote. Try to remember it this time, okay?"
There are two ways it makes sense to pronounce it. Either you say it the original way (the way Linus does), or you anglicise it (with an English long I). Either of these makes sense; the former, because it's the correct pronunciation etymologically, and the latter because it's correct phonetically in English.
The one that doesn't make sense is the one with an English short i, as if it were "linnux". There's no grounds for that one.
> Finnish as a written language is pretty young, only a couple > of hundred years. Therefore it has the convenient property > of words being written exactly as spoken (or vice versa, depends > on how you look at it), because the guy who invented it was > smart enough to do it like that.
Neither. It's pronounced router, just like route with er on the end. It would be rooter if it rooted, or rowter if it rowted (sounding like row, presumably), but since it routes, it's pronounced router, with the ou pronounced as in ouch and trounce.
That's actually fairly straightforward. The tricky ones are the oo words. Root, rooster, and roof are particularly troublesome: does the oo pronounce as in book and look, or as in spook and Hoover? (I prefer the latter, but it is arguable, and there are inconsistencies: for example, root beer is always pronounced with the oo as in book and look, even by people who pronounce tree roots with the oo as in spook and Hoover.)
I'm sure I've cleared up all confusion in this matter. HTH.HAND.
I pronounce the g in gnu, but I don't put a vowel between the g and the n like the other poster indicated; I just pronounce it the way it's spelled. You learn to pull such stunts after a couple of semesters of greek. I can pronounce psychiatry too. I still have trouble trilling my Rs. I was never able to do that at _all_, after two semesters of Spanish and much frustration, until my college roommate showed me the pronunciation section of his Klingon-English/English-Klingon dictionary. The instructions on how to pronounce the Klingon gh were very detailed (and very different from the pathetic instruction Spanish teachers give you on trilling), and with practice I found that not only can I pronounce the gh (a highly funky phoneme), but by a similar technique I can trill an rr. (Then I discovered that it's possible to trill other letters too... l, w, o,... actually, these are a bit easier than r, which still gives me trouble.)
This is naive. You can make it work if you're writing card games and such like, but the sorts of games that the other poster meant sometimes have to be a little more intimate with the system and rely on lower-level OS functionality than the Win32 API. Not most of the code of the game, mind you, but enough that it does matter.
Of course, once new systems stopped shipping with Win98SE and Me, all the game developers instantly saw the light, and they had several months to get their act together before the number of new WinXP systems grew enough to be deeply significant to their pocketbooks.
However, prior to the big release of XP, NT had nowhere near a 20% market share on the desktop.
20% desktop share for Linux? Sounds optimistic to me, but 2008 is still several years away, and an aweful lot can happen in that amount of time, so who knows. Idle speculation. Time will tell.
> when a Windows server can be had that can do it out of the > box with very little administration
That would represent a very radical change in Microsoft policy.
Don't get me wrong, NT has some things going for it, but "doing it all out of the box" isn't one of them. All that stuff is *available*, of course, and once you install it you have a pretty decent system, but it's not included OOTB. The reason for this goes directly back to Microsoft policy: the OOTB system is a base platform with basic functionality, suitable for the majority of users who have simple expectations. The minority who need features can obtain them separately. (Time was when they obtained them separately from third-party software vendors. These days with a few exceptions it's mostly either direct from MS or ports of OSS stuff free from the net. But the principle is the same.)
Out of the box, Windows systems are junk. You have to download and install a couple of gigabytes of software to make a Windows system useful. They don't ship with Apache, or a decent Java vm, no python, no decent command shell, no decent text editor, no secure shell server (critical for most servers, especially headless servers),... They don't even ship with Perl, for crying out loud. *Every* OS ships with Perl -- well, pretty much every non-handheld OS that matters, except Windows and VMS.
After you download and install a couple of gigs of software, then your Windows system starts to become useful.
Most Linux distros have the reverse problem -- three or four competing implementations of almost everything, with notable singleton exceptions like (oooh, back to topic) Samba, and ten or twelve competing implementations of some things, even more of certain key things (shells, window managers,...).
Samba IMO could use a competitor (that runs on something besides Windows). Just one competitor, though, not four or five or six. Preferably one written in a VHLL, and written in a more modular and flexible fashion so it can do things like support for multiple network/transport layers for compatibility with systems that are configured not to route NetBIOS over TCP/IP.
Depends what you put into it, and of course on what kind of learning we're talking about. Obviously, college won't teach you as much about the business world as the equivalent amount of time spent _in_ the business world, but the purpose of a classical education is not job training, at least not directly. (It does help, but that's not the primary immediate goal.)
College is (or was, at one time) designed to teach you how to think, teach you how to study, and expose you to a broad variety of topics. Will all of that help you in your first month on the job? A little. (You also learn other things that will help you more...) But the key thing you take away from college is the ability to learn more efficiently, so that the time you spend on the job (or anywhere else) subsequently will teach you more than it otherwise would have. Have I learned more technology in three years working in IT than I did in four years minoring in computer science? Yes, of course. Would I have learned as much, even assuming I had landed my current job, if I hadn't had the college background? I rather doubt it.
Of course, just as with anything, you to a large extent get out of college what you put into it, so this works better for some folks than for others.
Microsoft supports FAT12, FAT16, FAT32, and NTFS. NTFS is in many ways superior to FAT, but Linux support for NTFS is still limited to read-only unless you want to use highly experimental write support that is officially labelled DANGEROUS (i.e., more bleeding edge than code that is merely EXPERIMENTAL) and guaranteed to corrupt the filesystem. Then there's the issue of NTFS support in OS X; I'm not aware of any options there. Also, if you need to access it from Win9x/Me, those versions don't support NTFS either.
The ideal of course would be a journaling filesystem like Reiser. But Windows doesn't support those, last I knew.
Now, if you're willing to forego native support by some OSes, there are third-party apps to bring support to Windows and/or Mac for various filesystems that aren't supported natively. Explore2fs and the like. These are a kludge and a pain to use, and they don't integrate with the OS much less with apps, but they do (mostly) work. But when you go down that path you lose most of the nicer advantages of those filesystems. The real advantage of these apps is for multiboot systems, so that if you happen to be booted into the other OS and somebody on usenet asks how you got foo to work, you can pull your config file off the other filesystem and look. It's not really practical to use this as a regular filesystem.
FAT32 may suck, but it's supported by everything. Go with it. If you need to store things FAT32 can't store directly (symlinks, permissions, other attributes...), wrap them up in a ZIP archive.
> As for broadband...I dunno where you're from From the internet. Oh, you mean physically? Galion, Ohio.
> around here residential broadband is generally 1.5mbit+ down, > and 512kbit+ up The 1.5 down is as good as T1. 512 up isn't quite, but it's still way better than what I could get here.
> You should move Well, I'm going to colocate a server very soon; that'll help. For client stuff, in a pinch I can go in to work during my off time and use the T1. At home, I've adapted pretty well to my shared 33.6 dialup. wget rocks. Also, Mozilla's tabbed browsing lets me queue pages and continue reading the first page while they are retrieved. (Very nice for slashdot; scroll down the main page middle-clicking the interesting links, and by the time I get to the bottom the first story has loaded.) The new prefetching is nice too. The thing that really hurts is all the spam I get. Filtering helps with the time I have to spend _reading_ my mail, but it still takes thirty minutes to _download_ it all. I do something else while I wait for it, but it's annoying.
> I think it's safe to say that the majority of users have winamp, > realplayer, or quicktime.
I don't think that's a safe assumption at all. 84.3% of users don't have *anything* that wasn't either preinstalled when they bought the computer or automatically installed when they inserted the disk the ISP sent them in the Connection Kit. QT comes preinstalled on Mac, which gives it 4.7% share or so. Real and Winamp are AFAIK only used by people who install things. So, 4.7% for Mac users, plus the 15.7% who install things, but there's some overlap there, so only about 19.2% of users have any of these things installed.
OTOH, among slashdot readers the percentages are very different, since 98.4% of slashdot readers install software (and indeed almost 70% are not afraid to install entire operating systems). So among slashdot readers it is very likely that the majority are using one or another of the above-listed applications. (That said, I don't have or want any of them... I have timidity and xmms and what else do I need?)
Additionally, it is worth considering that 84.3% of users in the US are still on dialup (though, again, the number is lower on slashdot), which makes downloading audio news stories a somewhat less exciting proposition.
It used to be called schizophrenia. Then it was called multiple personality disorder. Then it was dissociative disorder. I'm not sure what they're calling it these days.
Much more likely, it could mean the same father impregnated two eggs at the same time, but it should be noted that one of the resulting sets of DNA might have a Y chromosome, and the other might not. The implications of this possibility are left as an exercise to the reader.
> After that, your cheap ide disk might have a hard time
You do realise, don't you, that a cheap IDE disk can handle more throughput than a T3 line or a 10/100 switch? Refer to the speed chart below.
Although, if you don't have adequate cooling, making it do so continually for a few hours could have a detrimental impact on its lifespan. But a university department server is probably in an air-conditioned building, which makes adequate cooling fairly easy, because ambient temperature stays low no matter how much heat you blow out the back of the box. So you slap a cooler on the HD and add one case fan and you're set, yes?
Speed Chart:
CPU, RAM, HD, Ethernet/T3, T1, Dialup, Keyboard Each step is _at least_ an order of magnitude slower than the previous step on the chart. Thus, if your bottleneck is the T3 line, even the cheapest IDE hard drive can easily keep up. Removable drives are well slower than the HD, but I don't know how they compare to the networking technologies. Residential broadband is generally not better than T1 and often is closer to dialup. (Dialup at "56k" gets real speeds up to 45kbps; T1 is about 1500kbps, so 128 or even 256kbps is closer to dialup than it is to T1. 512kbps is (geometrically) closer to the T1, however, and so can be classified as true broadband.)
> I find it hard to believe that a cheap dual channel switch > for 2 pc's could handle the slashdot effect.
You're on crack. A simple 100BaseT fast ethernet switch (which will run you $50 or so) running in full duplex mode can handle as much traffic as 66 T1 lines -- that's more traffic than one T3 line can carry. The slashdot effect is a lot of bandwidth, but it's not _that_ much bandwidth. (LAN technology carries much more bandwidth for the buck than WAN technology; the reason it's cheaper is mostly because it doesn't have to handle the long distances.)
The real danger of the slashdot effect is to people on narrow pipes (Cable/DSL -- or of course dialup) or with bandwidth caps, or to sites that carry ginormous bandwidth-intensive downloads with heavy geek appeal (ISO images anyone? What about a 500MB computer- generated animation of Darl McBride getting spanked?), or to sites with server-side active content that does something that consumes a lot of resources (such as database queries). Static HTML (and a few GIFs and sanely-sized PNGs) served over a university connection (probably T3 or better) isn't in any really serious danger.
Netscape 4, right? Yeah, it sucks. Netscape 7 is much better.
I also found that all internet-related apps crash many times as often as usual if you are running instant messaging software. I discovered this when I first experimented with ICQ, and Pegasus Mail started crashing. (Pegasus Mail, normally, does not crash.) I uninstalled ICQ, and Pegasus magically was healed.
Slashdot Mirror
Transliterated? Very strange. I tried my signature in the Mounce
font, and a lot of the punctuation comes out as diacritical marks
on the previous character. For example, where you see "map{my",
the curley brace comes out as a rough breathing and accute accent,
which looks pretty odd over the pi -- not to mention, the y comes
out as a psi, so you end up with something that _if_ it could be
pronounced would be roughly "maHPmps". On the other hand, what
follows ("($a,$b)") comes out very nicely if you imagine someone
using alpha and beta as variable names.
Heh. +1 Funny, but actually even transliteration isn't quite that
straightforward, much less translation. If you take English and
write it in a Greek-transliterative font, you get gibberish, which
often isn't even pronounceable. Symbol is even worse, because some
of the symbols aren't Greek letters or aren't even letters at all.
You will note, however, that the Symbol font only uses the Greek
letters as symbols; it does not have provisions for diacritical
marks, for example, or Greek punctuation (e.g., the ? glyph does
not look like a semicolon) as any real transliterative font (Graeca,
Mounce) would do.
And I *can* read Greek (albeit not quickly, and not modern Greek).
> In that case, shouldn't SGI be suing SCO for ... ???
Not for profit. (Hello? Turnip? Blood?) However, they could sue
them for assorted things (vague intellectual property violations
come to mind) in an attempt to boost their stock price.
The linked usenet article seems to indicate (though not being a C programmer it's hard to be certain) that the one bit of code (the one with the comment about allocating size units) was in BSD in 1984 and may have been changed at that time to fix a bug.
The other one (with the mutex and the assert) I don't know about, but you can speculate about whether it may have been in BSD also.
> I don't see how the fact that Windows doesn't ship with
> Apache is relevant.
I was trying to be nice. Allow me to rephrase so as to be more
clear: "Doesn't ship with any web server software that isn't
so infamous no sane sysadmin would permit it on his network".
> I'm also curious as to what constitutes a "decent" command
> shell or text editor for you.
A decent command shell is one that's good enough you can easily
do all your file management from it, so that you don't need a
GUI file manager. One critical feature of a decent command shell
is the ability to put the output of one command on the command
line of another command, but there are other key features.
A decent text editor? Of course I mean Emacs, but I can make a
*lengthy* list of important features that are present in twenty
or thirty major text editors but absent in the ones that ship
with Windows. The ones that ship with Windows don't even have
basic macro capabilities, for crying out loud.
> There is also a secure shell included- terminal services.
I thought terminal services was a thinclient server? Maybe I
was confused about that.
> why is there no grounds for pronouncing it as linnux?
Because it's neither etymologically nor phonetically correct.
> Lin is sometime prononced with a short i in English. Lint,
> linear, lingo, liberty...
Lint is pronounced with a short i because it's phonetically correct
that way. Lingo also is pronounced phonetically (rhymes with Ringo).
Liberty is pronounced the way it is because it comes from French.
(Compare with the pronunciation of Xouvert.) (Liberty's English
pronunciation has drifted a bit from the French, but it's much closer
to the French than it is to what would be phonetic in English.)
Linear is an interesting study. Etymologically, it comes from the
Latin (wherein linea means line). It's pronounced that way, then,
because the phonetic rendering in English is hard or impossible for
most native speakers of the language to pronounce. The i and the e
would both be long, and the a silent, thus: "line ear" -- easy to
pronounce as two separate words, but when you run them together the
dipthong gets mangled and comes out as two syllables. Also, some
people have difficulty pronouncing the English long I in that
combination. Anyway, the long and short of it is that people did
not like saying it that way, so they used the Latin pronunciation,
or something closer to it.
Just spell it out. "ess ell a ess aych dee oh tee dot oh are gee."
I prefer the other approach (speaking quickly): "Okay, I'm resetting
your password. Your new password is capital Q period lowercase d
forward slash seven tilde lowercase b uppercase I right curly brace
exclamation mark zero uppercase N lowercase v backquote. Try to
remember it this time, okay?"
There are two ways it makes sense to pronounce it. Either you say
it the original way (the way Linus does), or you anglicise it (with
an English long I). Either of these makes sense; the former, because
it's the correct pronunciation etymologically, and the latter because
it's correct phonetically in English.
The one that doesn't make sense is the one with an English short i,
as if it were "linnux". There's no grounds for that one.
> Finnish as a written language is pretty young, only a couple
> of hundred years. Therefore it has the convenient property
> of words being written exactly as spoken (or vice versa, depends
> on how you look at it), because the guy who invented it was
> smart enough to do it like that.
Songo also has this property.
> Is the pronunciation a rooter or a rowter.
Neither. It's pronounced router, just like route with er on the
end. It would be rooter if it rooted, or rowter if it rowted
(sounding like row, presumably), but since it routes, it's
pronounced router, with the ou pronounced as in ouch and trounce.
That's actually fairly straightforward. The tricky ones are the
oo words. Root, rooster, and roof are particularly troublesome:
does the oo pronounce as in book and look, or as in spook and
Hoover? (I prefer the latter, but it is arguable, and there are inconsistencies: for example, root beer is always pronounced with
the oo as in book and look, even by people who pronounce tree
roots with the oo as in spook and Hoover.)
I'm sure I've cleared up all confusion in this matter. HTH.HAND.
I pronounce the g in gnu, but I don't put a vowel between the g ... actually,
and the n like the other poster indicated; I just pronounce it
the way it's spelled. You learn to pull such stunts after a
couple of semesters of greek. I can pronounce psychiatry too.
I still have trouble trilling my Rs. I was never able to do that
at _all_, after two semesters of Spanish and much frustration,
until my college roommate showed me the pronunciation section of
his Klingon-English/English-Klingon dictionary. The instructions
on how to pronounce the Klingon gh were very detailed (and very
different from the pathetic instruction Spanish teachers give you
on trilling), and with practice I found that not only can I
pronounce the gh (a highly funky phoneme), but by a similar
technique I can trill an rr. (Then I discovered that it's
possible to trill other letters too... l, w, o,
these are a bit easier than r, which still gives me trouble.)
> win32 is win32 is win32
This is naive. You can make it work if you're writing card games
and such like, but the sorts of games that the other poster meant
sometimes have to be a little more intimate with the system and
rely on lower-level OS functionality than the Win32 API. Not most
of the code of the game, mind you, but enough that it does matter.
Of course, once new systems stopped shipping with Win98SE and Me,
all the game developers instantly saw the light, and they had
several months to get their act together before the number of
new WinXP systems grew enough to be deeply significant to their
pocketbooks.
However, prior to the big release of XP, NT had nowhere near a
20% market share on the desktop.
20% desktop share for Linux? Sounds optimistic to me, but 2008
is still several years away, and an aweful lot can happen in that
amount of time, so who knows. Idle speculation. Time will tell.
> when a Windows server can be had that can do it out of the
... They don't even ship with Perl, for crying
...).
> box with very little administration
That would represent a very radical change in Microsoft policy.
Don't get me wrong, NT has some things going for it, but "doing
it all out of the box" isn't one of them. All that stuff is
*available*, of course, and once you install it you have a
pretty decent system, but it's not included OOTB. The reason
for this goes directly back to Microsoft policy: the OOTB system
is a base platform with basic functionality, suitable for the
majority of users who have simple expectations. The minority
who need features can obtain them separately. (Time was when
they obtained them separately from third-party software vendors.
These days with a few exceptions it's mostly either direct from
MS or ports of OSS stuff free from the net. But the principle
is the same.)
Out of the box, Windows systems are junk. You have to download
and install a couple of gigabytes of software to make a Windows
system useful. They don't ship with Apache, or a decent Java
vm, no python, no decent command shell, no decent text editor,
no secure shell server (critical for most servers, especially
headless servers),
out loud. *Every* OS ships with Perl -- well, pretty much every
non-handheld OS that matters, except Windows and VMS.
After you download and install a couple of gigs of software,
then your Windows system starts to become useful.
Most Linux distros have the reverse problem -- three or four
competing implementations of almost everything, with notable
singleton exceptions like (oooh, back to topic) Samba, and
ten or twelve competing implementations of some things, even
more of certain key things (shells, window managers,
Samba IMO could use a competitor (that runs on something besides
Windows). Just one competitor, though, not four or five or six.
Preferably one written in a VHLL, and written in a more modular
and flexible fashion so it can do things like support for multiple
network/transport layers for compatibility with systems that are
configured not to route NetBIOS over TCP/IP.
> BTW I like the right tool for the problem beleive it or
;-)
> not clusters of Linux boxes dont work for any problem yet.
Of course not. For some problems you need BSD
> You don't learn in the classroom
Depends what you put into it, and of course on what kind of learning
we're talking about. Obviously, college won't teach you as much
about the business world as the equivalent amount of time spent
_in_ the business world, but the purpose of a classical education
is not job training, at least not directly. (It does help, but
that's not the primary immediate goal.)
College is (or was, at one time) designed to teach you how to think,
teach you how to study, and expose you to a broad variety of topics.
Will all of that help you in your first month on the job? A little.
(You also learn other things that will help you more...) But the
key thing you take away from college is the ability to learn more
efficiently, so that the time you spend on the job (or anywhere
else) subsequently will teach you more than it otherwise would
have. Have I learned more technology in three years working in
IT than I did in four years minoring in computer science? Yes,
of course. Would I have learned as much, even assuming I had
landed my current job, if I hadn't had the college background?
I rather doubt it.
Of course, just as with anything, you to a large extent get out
of college what you put into it, so this works better for some
folks than for others.
Microsoft supports FAT12, FAT16, FAT32, and NTFS. NTFS is in many
ways superior to FAT, but Linux support for NTFS is still limited
to read-only unless you want to use highly experimental write support
that is officially labelled DANGEROUS (i.e., more bleeding edge than
code that is merely EXPERIMENTAL) and guaranteed to corrupt the
filesystem. Then there's the issue of NTFS support in OS X; I'm
not aware of any options there. Also, if you need to access it
from Win9x/Me, those versions don't support NTFS either.
The ideal of course would be a journaling filesystem like Reiser.
But Windows doesn't support those, last I knew.
Now, if you're willing to forego native support by some OSes, there
are third-party apps to bring support to Windows and/or Mac for
various filesystems that aren't supported natively. Explore2fs and
the like. These are a kludge and a pain to use, and they don't
integrate with the OS much less with apps, but they do (mostly)
work. But when you go down that path you lose most of the nicer
advantages of those filesystems. The real advantage of these apps
is for multiboot systems, so that if you happen to be booted into
the other OS and somebody on usenet asks how you got foo to work,
you can pull your config file off the other filesystem and look.
It's not really practical to use this as a regular filesystem.
FAT32 may suck, but it's supported by everything. Go with it.
If you need to store things FAT32 can't store directly (symlinks,
permissions, other attributes...), wrap them up in a ZIP archive.
> As for broadband...I dunno where you're from
From the internet. Oh, you mean physically? Galion, Ohio.
> around here residential broadband is generally 1.5mbit+ down,
> and 512kbit+ up
The 1.5 down is as good as T1. 512 up isn't quite, but it's
still way better than what I could get here.
> You should move
Well, I'm going to colocate a server very soon; that'll help.
For client stuff, in a pinch I can go in to work during my off
time and use the T1. At home, I've adapted pretty well to my
shared 33.6 dialup. wget rocks. Also, Mozilla's tabbed browsing
lets me queue pages and continue reading the first page while
they are retrieved. (Very nice for slashdot; scroll down the
main page middle-clicking the interesting links, and by the time
I get to the bottom the first story has loaded.) The new
prefetching is nice too. The thing that really hurts is all the
spam I get. Filtering helps with the time I have to spend
_reading_ my mail, but it still takes thirty minutes to _download_
it all. I do something else while I wait for it, but it's annoying.
> I think it's safe to say that the majority of users have winamp,
> realplayer, or quicktime.
I don't think that's a safe assumption at all. 84.3% of users don't
have *anything* that wasn't either preinstalled when they bought the
computer or automatically installed when they inserted the disk the
ISP sent them in the Connection Kit. QT comes preinstalled on Mac,
which gives it 4.7% share or so. Real and Winamp are AFAIK only
used by people who install things. So, 4.7% for Mac users, plus
the 15.7% who install things, but there's some overlap there, so
only about 19.2% of users have any of these things installed.
OTOH, among slashdot readers the percentages are very different,
since 98.4% of slashdot readers install software (and indeed
almost 70% are not afraid to install entire operating systems).
So among slashdot readers it is very likely that the majority
are using one or another of the above-listed applications.
(That said, I don't have or want any of them... I have timidity
and xmms and what else do I need?)
Additionally, it is worth considering that 84.3% of users in
the US are still on dialup (though, again, the number is lower
on slashdot), which makes downloading audio news stories a
somewhat less exciting proposition.
Of course, 98.675% of all statistics are made up.
It used to be called schizophrenia. Then it was called multiple
personality disorder. Then it was dissociative disorder. I'm not
sure what they're calling it these days.
Much more likely, it could mean the same father impregnated two eggs
at the same time, but it should be noted that one of the resulting
sets of DNA might have a Y chromosome, and the other might not. The
implications of this possibility are left as an exercise to the reader.
> After that, your cheap ide disk might have a hard time
You do realise, don't you, that a cheap IDE disk can handle
more throughput than a T3 line or a 10/100 switch? Refer to
the speed chart below.
Although, if you don't have adequate cooling, making it do so
continually for a few hours could have a detrimental impact on
its lifespan. But a university department server is probably
in an air-conditioned building, which makes adequate cooling
fairly easy, because ambient temperature stays low no matter
how much heat you blow out the back of the box. So you slap
a cooler on the HD and add one case fan and you're set, yes?
Speed Chart:
CPU, RAM, HD, Ethernet/T3, T1, Dialup, Keyboard
Each step is _at least_ an order of magnitude slower than the
previous step on the chart. Thus, if your bottleneck is the
T3 line, even the cheapest IDE hard drive can easily keep up.
Removable drives are well slower than the HD, but I don't know
how they compare to the networking technologies. Residential
broadband is generally not better than T1 and often is closer
to dialup. (Dialup at "56k" gets real speeds up to 45kbps; T1
is about 1500kbps, so 128 or even 256kbps is closer to dialup
than it is to T1. 512kbps is (geometrically) closer to the T1,
however, and so can be classified as true broadband.)
> I find it hard to believe that a cheap dual channel switch
> for 2 pc's could handle the slashdot effect.
You're on crack. A simple 100BaseT fast ethernet switch (which
will run you $50 or so) running in full duplex mode can handle as
much traffic as 66 T1 lines -- that's more traffic than one T3
line can carry. The slashdot effect is a lot of bandwidth, but
it's not _that_ much bandwidth. (LAN technology carries much more
bandwidth for the buck than WAN technology; the reason it's cheaper
is mostly because it doesn't have to handle the long distances.)
The real danger of the slashdot effect is to people on narrow pipes
(Cable/DSL -- or of course dialup) or with bandwidth caps, or to
sites that carry ginormous bandwidth-intensive downloads with heavy
geek appeal (ISO images anyone? What about a 500MB computer-
generated animation of Darl McBride getting spanked?), or to sites
with server-side active content that does something that consumes
a lot of resources (such as database queries). Static HTML (and a
few GIFs and sanely-sized PNGs) served over a university connection
(probably T3 or better) isn't in any really serious danger.
Netscape 4, right? Yeah, it sucks. Netscape 7 is much better.
I also found that all internet-related apps crash many times as
often as usual if you are running instant messaging software. I
discovered this when I first experimented with ICQ, and Pegasus
Mail started crashing. (Pegasus Mail, normally, does not crash.)
I uninstalled ICQ, and Pegasus magically was healed.