> I should have mentioned Guam, Puerto Rico, or the US
> Virgin Islands instead.
Residents of those places don't pay money to the IRS. (Well, the
Virgin Islands at least; I imagine the others are handled in a
similar fashion.) They have to meet certain qualifications to be
considered residents, of course, such as actually living there, and
they do pay taxes, but not to the US Federal Government. See for
example here.
> The only thing which would make me marginally happier would have > been if these guys had known what an ochlophobe was.
From the Greek words "ochlos" (crowd) and phobeomai (I fear; you are familiar with "phobia" no doubt, from more common conditions such as arachnophobia). I'm very bothered by unordered or chaotic groups of people, especially large groups, especially of people I don't know very well. Mingling in a social setting with more than half a dozen people or so makes me physically uncomfortable.
My higher cognitive functions know that at the end of the event I will be just fine, that no horrific things are going to happen to me, but my higher cognitive functions only have control of small parts of my brain, since most of the rest of it is busy planning escape routes.
> You would be amazed at what kinds of things can disrupt a dialup > connection.
What amazes me is how _robust_ my dialup connection is. Sometimes it can go for several hours at a time before it drops. That may sound like a short time, but I have ears, and I know what the phone lines are like. Your wetware filters it out if you don't pay the right kind of attention, just like the stereo in the background when you are listening to somebody talking, but doing that filtering is AI-complete, or so I would think. But if you *listen* for line noise, you can hear it in any ten-minute phone call -- at least, around here you can. I would expect that kind of thing to disrupt a data link every single time it happens, but apparently most of the time it doesn't. Whoever designed ppp knew what he was doing, that's my take on it. I don't have any other explanation.
> Shouldn't it really be the RTFM policy, not the NAH policy?
The NAH policy is more holistic; besides RTFM, it also covers RTFS, DIY, and a general prohibition against writing HOWTO documents that cover a portion of code to which you have contributed, as well as some applications of LART. Of these, RTFS is widely considered to be the most important aspect of the policy.
> The problem is that in general many "volunteered" hours to > companies aren't voluntary.
This is true in almost all fields. Despite the Department of Labor's anality (and it is that in some cases), employers find ways to pressure employees into doing things off the clock. Fast food workers work through their breaks, or parts thereof -- unoffically, of course -- at least a third of the time in many establishments.
I work at a public library, and every year we have a Volunteer Recognition Banquet (to recognise the work of volunteers); attendance is mandatory for all staff, but it is considered volunteer time. I would no way EVER choose of my own volition to attend such an event; I'm a computer geek and an ochlophobe, and my tastes in food tend toward things cooked by human beings in real kitchens; high-class catered food is, IMO, just barely edible, on a good day; eating it together with a group of people I barely know is torture. On top of that, almost all of these volunteers are from a social class about three tax brackets higher than mine (or so it seems), so casual conversation is basically impossible between me and one of them even on an individual basis; dealing with a mingling group of them is sheer terror, and I am required to pretend I'm enjoying myself; such a banquet is very close to my personal idea of hell. (To complete my idea of hell, heat the room to about 100F, play country music at odd moments, and make me wear ill-fitting burlap.) Each year I go, but I vow to find another job before the next one rolls around. In most other respects, I like my job very much. But I have to do this mandatory thing voluntarily, as part of my job. (Please don't think I'm whining; the rest of my job is fairly cushy, and this is a once-a-year thing. But it is a good example of my point.)
The DOL may have a hard time pinning down specific infractions, but they are well aware of this general problem, I'm quite sure. So yes, employers have to be creatively careful about it.
And yes, as others have pointed out, there's the stupid liability thing. Which is dumb, but that's what we get for trying to entrust justice to juries made of random people off the street. (But that's another thread for another day.)
The techs in question should have used their brains and done their off-the-clock volunteering under an alias. The employer doesn't mind that people are being helped, and it doesn't mind that they are getting their help from someone other than on-the-clock employees; what they don't like is that it is visibly coming from off-the-clock employees. If they'd just been discreet, nobody would have ever cared. Certainly the employer would not have audited the employees' personal time to see whether they might also have secret lives as some of the helpful users in the fora.
As others have pointed out, there are an infinite number of such systems; however, none of them are natural numbers. IMO a more interesting question is, can you represent 1 (decimal) exactly in base pi (or any of the other bases in which pi can be expressed exactly)? Sure, 1 can be expressed _algebraically_ in base pi as pi^0, but my intuition says it is probably possible to prove by induction that you cannot express it exactly as a string of digits raised to a power of pi (i.e., conventional notation for floating point numbers -- I wanted to use the term "decimal point", but obviously it wouldn't be that; we could call it a pial point I suppose...) in any of the same bases in which pi can be so expressed -- i.e., my intuition says these two sets of bases are disjoint sets and can be considered as equivalence classes, in terms of which numbers can be expressed in them. (That is, I suspect that *none* of the numbers representable in the pi class of base notations can be represented in any of the natural class of base notations, and vice versa.) I also suspect that there are provably an infinite number of such classes and that an interesting study could be had from determining the cardinality of the set of all of them. (My first guess is aleph-sub-one, but that's a shot in the dark, as I haven't studied the question.)
Anybody know anyone who would know whether any work has been done on this concept?
Be aware, if you do this, that the links frequently have a unique ID embedded in them that uniquely identifies your address. So if you want to do this, use fake addresses that don't get real mail that you actually want to read. (You can ensure that spammers get your fake addresses by putting them in the From: fields of posts to alt.test or somesuch.) With a few dozen fake addresses, you could collect enough spammer website URLs to keep a large army of DDOS zombies busy, even if you expire all URLs every 48 hours or so.
Also, I'm not confident of the legality of doing this, so consult a lawyer before trying it, if you don't want to wind up in court.
> it certainly can't be very effective. if it is, how come?
The ROI can be significant as a percentage, only because the per-unit cost of sending is so very low. Consequently, if somebody is making millions spamming, you can bet they've sent trillions of messages to make those millions. Makes you want to... oh, wait, somebody did.
> I wonder what'd happen if he recieved a few > hundred of those.
He'd sell 'em on eBay, what do you think? Just because he's ethically impaired doesn't mean he can't spot a business opportunity.
Re:RAID can mean different things...
on
IDE RAID Examined
·
· Score: 2
> Where would random come into it?
Well, like I said, >> (The term "Random" means the same as in RAM -- i.e., that >> you can access any part... at any time.)
In other words, you can read or write the data in any order (Just like you can with a SLED, BTW.) These days, a non-random disk array is neigh unto inconceivable, of course.
> Alluding to the fact that if one drive fails you still have two > others that have a complete set of the data in the array.
No, that isn't how it works. The original concept was "redundant", but not _that_ redundant. Actually, what you describe is close to the kind of redundancy I want -- Federation redundancy, i.e., everything in triplicate. But current RAID designs are mostly not that way, and RAID 0 implementations have no redundancy at all; if any of the drives go bad, you'd better have backups. RAID 1 (and higher) are correctly described as "redundant array of inexpensive disks", but RAID 0 is non-redundant. This is what I meant when I said some RAID are done just for performance reasons.
If you want to understand partial redundancy better, read the article. In brief, RAID 1, 0+1, and 10 give you two copies of each piece of data; RAID 3 and 5 give you parity, which uses less disk space and can be just about as good.
Like I said, though, what I really want is everything in triplicate. I guess that's where offsite backups come in... which is better anyway, because if the building burns down, your whole RAID is, like, gone, man.
> Maybe they're also not smart enough to realize that they're > getting a computer without Windows?
Sure, but when they take it home and use it, will they notice even then? Not likely. Eventually someone will point it out to them: "Hey, I don't recognise your Windows, it's different from mine. I bet it's not Microsoft at all." Will they care? Or will they be like "Huh, well, I don't know about that, but when I click the little envelope thingy I get my mail."
If Wal*Mart gets a small enough number of returns on these things, and enough sales, maybe they'll decide to carry them in the actual _stores_...
> I think this is scary because people like us who actually need/use > higher end hardware will end up paying more. If only the cheap > hardware market moves units in large numbers, then higher end, > quality products will be manufactured in smaller numbers and be > harder to come by.
This is not news. It's the attack of the killer micros all over again. That's what the PC was 1980: a low-end machine that can just about almost manage the very basic things people need, so they don't have to fork over the bucks for a Real Computer. As a result, ecconomy of scale drives the cost of microcomputers down and the quality up, while minicomputers languish. "No one will survive the attack of the killer micros." That's what happened to DEC. Does that mean the people who needed the power of a minicomputer ten years ago are paying more now?
Don't worry about it. As the low-end micros get smaller and cheaper and better, you'll be able to buy twenty of them and make yourself a cluster, with a redundant distributed filesystem, for everyday use -- and whenever one goes bad, you'll just hot swap a replacement. And they'll fit in the space a large server case takes up today. (No, I'm not going to put a date on this prediction.)
I hope you are being sarcastic. The poorest of the poor here throw food away on a regular basis and waste unconscionable amounts of money on things like cable television and fast food. There is nothing in the US that even resembles poverty. once told a lady in my church that there's no real poverty in the United States, and she laughed at me. About a year later, as it happens, she travelled to the Cameroon and spent a couple of months there. When she came back, she was telling people that we don't have real poverty in the US. The third world redefined the word "poverty" for her.
> Mhmm... well, if New England was a british colony, then wasnt the > tea tax a Federal tax then? Or Hawaiians and Alaskans are not > subject to pay US Federal taxes?
Hawaii and Alaska each sport two US senators and elect a member of the US House of Representatives. The whole stupid war was mostly over whether the colonies could be taxed without representation in the House of Commons. (There were some other sticking points too... but they aren't germaine to this discussion.)
The article talks about OS support, and lists supported OSes. It seems people who use less common OSes are fresh out (though FreeBSD gets half-decent support). My question is, why does it make any difference what OS you use? Shouldn't the array be presented to the OS (indeed, to the BIOS) as a single disk? Isn't that the whole point of hardware RAID (as opposed to software RAID)? Why does the OS even need to know there _is_ a RAID? Doesn't make sense to me. I want to be able to set up a RAID (1, 3, or 5 are the schemes I'd be interested in), partition the resulting "drive" N ways, and install whatever OSes I want (provided they can run on x86 and support IDE, of course), just like I can on a single physical disk. Why won't that work?
Re:RAID can mean different things...
on
IDE RAID Examined
·
· Score: 5, Informative
> I don't quite understand where this Inexpensive crap came from.
RAID (Random Array of Inexpensive Disks) was as opposed to SLED (Single Large Expensive Disk). (The term "Random" means the same as in RAM -- i.e., that you can access any part (any drive, in this case) at any time.)
> RAID was around long before IDE RAID controllers started showing > up and of course SCSI RAID arrays almost always use very expesive > disks.
"expensive" is relative. (Instead of thinking of SCSI as the only other option besides RAID, try to remember that there were larger and more expensive disks at one time.)
> It's Redunant Array of Independent Disks, always has always > will be.
It's not necessarily "redundant" at all; some RAIDs are done just for performance reasons, with no redundancy. (Personally, I am more interested in the redundancy, however.)
Windows can be run _on_ thin clients. Whether it can act as the host for the thin-client-server I don't know, but I'm not sure that really matters.
> Frankly, I don't know of anyone doing this with Windows at all.
The school I attended had a network composed on Netware on the server end (this was before there was such a thing as Windows NT) with thin clients in the computer lab running DOS6 and Windows 3. They had floppy drives, but no hard drives. (If I were setting up a network of thin clients today, they wouldn't have any local drives at all, _especially_ not floppy, but at the time floppies were needed.)
The nature of the way thin clients work is that you can send them any OS you want to send them; they just boot whatever you give them via the network. (The network card and BIOS have to support this.) It doesn't have to be the same OS that runs on the server.
These days, of course, Microsoft has a thin client server offering. I haven't used it, but I'm sure it's _way_ easier to administer than numerious independent Windows installations on each client. If nothing else, most of the big weakness of needing to physically go to each computer to fix it would be greatly releaved. Of course, you could also use Windows for the client OS but have the server running something else.
The only downside to thin clients is Single Point Of Failure. For that reason, I would be inclined to want to run something reliable on the server (not Windows), have it on UPS, use fairly high-end hardware, and so on, to prevent it from going down all the time.
> the user does not perceive it to be an advertisement, but a > compulsory upgrade.
It's worse than that. Go to a public library and watch people who don't have a computer at home. By far and away the most common way they follow these deceptive dialog-banner links is by clicking the upper-right-hand corner, the part that resembles a close box.
They think it's modal. (They don't know the word "modal", or the term "dialog box" for that matter, but that's irrelevant.) They want it to go away, and the advertiser is deliberately harnessing the user's desire to make it go away. That's why it's presented as an error, rather than a positive message. The thing is designed so that if the user tries to make it go away, they will have the target content rammed down their throat -- obviously against their will, since they tried to close the thing.
> I'm still waiting for someone to sue CapitalOne over the dead tree > mailouts they send that try to look like a bill or some other > important information.
What about the mailings you get that are printed in two or three colours, with the main text in black and then script notes in the margins in another colour, things circled, underlined, comments added, made to appear as if a human being has gone over it for you and highlighted the good parts. Sure, you'd have to be nearsighted in the extreme to mistake those printings for actual handwritten comments, but it's still a transparent attempt at deception.
> I don't find anything deceptive in the ads though.
Then you are one of the 98% whom the ad targets: people who don't understand why it's bogus.
> You're computer is broadcasting an IP
No, it isn't. Your computer is narrowcasting an IP address only to specific computers with which it is communicating. However, even if your computer did not give out the address at all, other computers would still know that such an address exists and that there might be a computer at it. All the IP addresses on the internet (_all_ of them) are routinely checked by port scanners, even if there's _not_ a computer connected using that address. So even if you _were_ broadcasting your IP, crackers would not gain any advantage from that.
> an IP that can be used to attack you
Short answer: No, it can't.
> I would assume the InternetBoost really does increase download > speeds.
How, exactly, would it accomplish this? (Hint: it wouldn't.)
> They look like a windows pop-up box, but does that mean anything > that uses an actual pop-up box besides Windows in deceptive?
A program running on your computer is another matter. These are not programs running on your computer; they are advertisements trying to _appear_ to be such.
> When Kazaa pops up a box saying that there is an upgrade (or > whatever) is that deceptive?
To be fully analagous, the box in question would have to go out of its way to appear to have nothing to do with KaZaA, warn the user of something that sound like a problem (not just offer that "an upgrade is available", but more like "WARNING: Your computer is not functioning properly! You need to get an upgrade!"), and the "upgrade" in question would have to be something that the user is currently not using, nor has expressed any particular interest in using, something that doesn't do anything the dialog box predicted but instead surreptitously performs some other function that the user never intended.
That said, such an action wouldn't greatly surprise me coming from KaZaA.
> Only banner two is legit, because it says your current connection > MAY be capable of faster speeds.
These banners aren't fraudulent just because of what they say -- although what they say is certainly deceptive as well. My most serious objection to them is that they are deliberately designed to look like something much more important than advertising. If a company started putting up roadside advertisements made to look like road construction signs, little carts with blinking arrows made of individual lights (such as the DOT uses), and so on, in order to convince drivers that their driveway was the next exit, or that it was imperative to get off at their exit as part of a detour due to road construction, would we allow that? (Okay, Microsoft is not a government agency, but the importance of operating systems error messages on a computer is very similar to the importance of highway department messages on a highway.)
That the messages in these fake dialogs are deceptive is just the icing on the cake.
> Their 10-person IT staff supports 800 users running 400 devices > (as Dave calls the thin clients). There is no way they could > adequately support that many users and devices with such a small > staff if they ran Windows on individual desktops
I've got to call you on this. As much as I think Linux has a lower TCO than Windows, it's _completely_ unfair to compare Linux thin clients to Windows desktops. Thin clients are a huge win in the TCO department, regardless of which OS you use.
It's like saying, "Airplanes have a lower TCO than cars because t's cheaper to go to Florida once a year by plane than to drive there every week." Most of your savings there come from going once a year versus once a week; it would be unreasonable to attribute all of that to flying instead of driving.
If you want to compare, either compare thin clients with Linux to Windows thin clients, or else compare full Linux desktops to full Windows desktops. My take on the matter is that the Linux systems will require more initial configuration (but not hugely more), need less action later in the form of reboots, reinstalls, and so forth, be somewhat more secure (but not totally secure out of the box) and be easier to administer remotely, especially in bulk (i.e., it's trivial to write a script that does the same thing to a number of systems). However, more training is required since fewer people have previously used X11 than have used Win32. (Training is least relevant for servers, which explains why Linux has done better there than in other areas thus far.)
Where Windows really loses, IMO, is that a larger percentage of stuff that needs to be done requires physically going to the computer in question; with Linux, you do that when there is a hardware issue. (Once you get it set up initially.) Where Windows really wins is with training; it's what people use at home. (Which is mildly ironic, given that Microsoft achieved their home user share largely because "it's what I use at work", back in the days of DOS and Windows 3.x.)
Depending on what your users need to do, OS training _may_ be a total non-issue. This is especially true in cases where one big application is the computer's whole world, such as library catalog/circulation/automation systems.
Slashdot Mirror
Residents of those places don't pay money to the IRS. (Well, the Virgin Islands at least; I imagine the others are handled in a similar fashion.) They have to meet certain qualifications to be considered residents, of course, such as actually living there, and they do pay taxes, but not to the US Federal Government. See for example here.
> The only thing which would make me marginally happier would have
> been if these guys had known what an ochlophobe was.
From the Greek words "ochlos" (crowd) and phobeomai (I fear; you
are familiar with "phobia" no doubt, from more common conditions
such as arachnophobia). I'm very bothered by unordered or chaotic
groups of people, especially large groups, especially of people I
don't know very well. Mingling in a social setting with more than
half a dozen people or so makes me physically uncomfortable.
My higher cognitive functions know that at the end of the event I
will be just fine, that no horrific things are going to happen to
me, but my higher cognitive functions only have control of small
parts of my brain, since most of the rest of it is busy planning
escape routes.
> You would be amazed at what kinds of things can disrupt a dialup
> connection.
What amazes me is how _robust_ my dialup connection is. Sometimes
it can go for several hours at a time before it drops. That may
sound like a short time, but I have ears, and I know what the phone
lines are like. Your wetware filters it out if you don't pay the
right kind of attention, just like the stereo in the background when
you are listening to somebody talking, but doing that filtering is
AI-complete, or so I would think. But if you *listen* for line
noise, you can hear it in any ten-minute phone call -- at least,
around here you can. I would expect that kind of thing to disrupt
a data link every single time it happens, but apparently most of
the time it doesn't. Whoever designed ppp knew what he was doing,
that's my take on it. I don't have any other explanation.
> Shouldn't it really be the RTFM policy, not the NAH policy?
The NAH policy is more holistic; besides RTFM, it also covers RTFS,
DIY, and a general prohibition against writing HOWTO documents that
cover a portion of code to which you have contributed, as well as
some applications of LART. Of these, RTFS is widely considered to
be the most important aspect of the policy.
> This isn't microsoft sacrificing babies in the parking lot
> every morning.
Jeez, I knew Microsoft was bad, but _every morning_? I had no idea
they sacrificed babies in the parking lot so _often_!
> The problem is that in general many "volunteered" hours to
> companies aren't voluntary.
This is true in almost all fields. Despite the Department of Labor's
anality (and it is that in some cases), employers find ways to
pressure employees into doing things off the clock. Fast food
workers work through their breaks, or parts thereof -- unoffically,
of course -- at least a third of the time in many establishments.
I work at a public library, and every year we have a Volunteer
Recognition Banquet (to recognise the work of volunteers); attendance
is mandatory for all staff, but it is considered volunteer time. I
would no way EVER choose of my own volition to attend such an event;
I'm a computer geek and an ochlophobe, and my tastes in food tend
toward things cooked by human beings in real kitchens; high-class
catered food is, IMO, just barely edible, on a good day; eating it
together with a group of people I barely know is torture. On top
of that, almost all of these volunteers are from a social class
about three tax brackets higher than mine (or so it seems), so
casual conversation is basically impossible between me and one of
them even on an individual basis; dealing with a mingling group of
them is sheer terror, and I am required to pretend I'm enjoying
myself; such a banquet is very close to my personal idea of hell.
(To complete my idea of hell, heat the room to about 100F, play
country music at odd moments, and make me wear ill-fitting burlap.)
Each year I go, but I vow to find another job before the next one
rolls around. In most other respects, I like my job very much.
But I have to do this mandatory thing voluntarily, as part of my
job. (Please don't think I'm whining; the rest of my job is
fairly cushy, and this is a once-a-year thing. But it is a good
example of my point.)
The DOL may have a hard time pinning down specific infractions,
but they are well aware of this general problem, I'm quite sure.
So yes, employers have to be creatively careful about it.
And yes, as others have pointed out, there's the stupid liability
thing. Which is dumb, but that's what we get for trying to
entrust justice to juries made of random people off the street.
(But that's another thread for another day.)
The techs in question should have used their brains and done their
off-the-clock volunteering under an alias. The employer doesn't
mind that people are being helped, and it doesn't mind that they
are getting their help from someone other than on-the-clock
employees; what they don't like is that it is visibly coming from
off-the-clock employees. If they'd just been discreet, nobody
would have ever cared. Certainly the employer would not have
audited the employees' personal time to see whether they might
also have secret lives as some of the helpful users in the fora.
> You could probably do the same thing with rand() since it
> always outputs the same sequence of digits if you don't use
> srand()
On the same computer system it does...
As others have pointed out, there are an infinite number of such
systems; however, none of them are natural numbers. IMO a more
interesting question is, can you represent 1 (decimal) exactly
in base pi (or any of the other bases in which pi can be expressed
exactly)? Sure, 1 can be expressed _algebraically_ in base pi as
pi^0, but my intuition says it is probably possible to prove by
induction that you cannot express it exactly as a string of digits
raised to a power of pi (i.e., conventional notation for floating
point numbers -- I wanted to use the term "decimal point", but
obviously it wouldn't be that; we could call it a pial point I
suppose...) in any of the same bases in which pi can be so
expressed -- i.e., my intuition says these two sets of bases are
disjoint sets and can be considered as equivalence classes, in
terms of which numbers can be expressed in them. (That is, I
suspect that *none* of the numbers representable in the pi class
of base notations can be represented in any of the natural class
of base notations, and vice versa.) I also suspect that there
are provably an infinite number of such classes and that an
interesting study could be had from determining the cardinality
of the set of all of them. (My first guess is aleph-sub-one, but
that's a shot in the dark, as I haven't studied the question.)
Anybody know anyone who would know whether any work has been
done on this concept?
Be aware, if you do this, that the links frequently have a unique ID
embedded in them that uniquely identifies your address. So if you
want to do this, use fake addresses that don't get real mail that you
actually want to read. (You can ensure that spammers get your fake
addresses by putting them in the From: fields of posts to alt.test or
somesuch.) With a few dozen fake addresses, you could collect enough
spammer website URLs to keep a large army of DDOS zombies busy, even
if you expire all URLs every 48 hours or so.
Also, I'm not confident of the legality of doing this, so consult a
lawyer before trying it, if you don't want to wind up in court.
> it certainly can't be very effective. if it is, how come?
The ROI can be significant as a percentage, only because the per-unit
cost of sending is so very low. Consequently, if somebody is making
millions spamming, you can bet they've sent trillions of messages to
make those millions. Makes you want to... oh, wait, somebody did.
> I wonder what'd happen if he recieved a few
> hundred of those.
He'd sell 'em on eBay, what do you think? Just
because he's ethically impaired doesn't mean he
can't spot a business opportunity.
> Where would random come into it?
... at any time.)
Well, like I said,
>> (The term "Random" means the same as in RAM -- i.e., that
>> you can access any part
In other words, you can read or write the data in any order (Just
like you can with a SLED, BTW.) These days, a non-random disk
array is neigh unto inconceivable, of course.
> Alluding to the fact that if one drive fails you still have two
> others that have a complete set of the data in the array.
No, that isn't how it works. The original concept was "redundant",
but not _that_ redundant. Actually, what you describe is close
to the kind of redundancy I want -- Federation redundancy, i.e.,
everything in triplicate. But current RAID designs are mostly
not that way, and RAID 0 implementations have no redundancy at
all; if any of the drives go bad, you'd better have backups.
RAID 1 (and higher) are correctly described as "redundant array
of inexpensive disks", but RAID 0 is non-redundant. This is
what I meant when I said some RAID are done just for performance
reasons.
If you want to understand partial redundancy better, read the
article. In brief, RAID 1, 0+1, and 10 give you two copies of
each piece of data; RAID 3 and 5 give you parity, which uses
less disk space and can be just about as good.
Like I said, though, what I really want is everything in triplicate.
I guess that's where offsite backups come in... which is better
anyway, because if the building burns down, your whole RAID is,
like, gone, man.
> Maybe they're also not smart enough to realize that they're
> getting a computer without Windows?
Sure, but when they take it home and use it, will they notice even
then? Not likely. Eventually someone will point it out to them:
"Hey, I don't recognise your Windows, it's different from mine. I
bet it's not Microsoft at all." Will they care? Or will they be
like "Huh, well, I don't know about that, but when I click the
little envelope thingy I get my mail."
If Wal*Mart gets a small enough number of returns on these things,
and enough sales, maybe they'll decide to carry them in the actual
_stores_...
> I think this is scary because people like us who actually need/use
> higher end hardware will end up paying more. If only the cheap
> hardware market moves units in large numbers, then higher end,
> quality products will be manufactured in smaller numbers and be
> harder to come by.
This is not news. It's the attack of the killer micros all over
again. That's what the PC was 1980: a low-end machine that can
just about almost manage the very basic things people need, so
they don't have to fork over the bucks for a Real Computer. As
a result, ecconomy of scale drives the cost of microcomputers
down and the quality up, while minicomputers languish. "No one
will survive the attack of the killer micros." That's what
happened to DEC. Does that mean the people who needed the power
of a minicomputer ten years ago are paying more now?
Don't worry about it. As the low-end micros get smaller and
cheaper and better, you'll be able to buy twenty of them and
make yourself a cluster, with a redundant distributed filesystem,
for everyday use -- and whenever one goes bad, you'll just hot
swap a replacement. And they'll fit in the space a large server
case takes up today. (No, I'm not going to put a date on this
prediction.)
> In the US there is real poverty
I hope you are being sarcastic. The poorest of the poor here throw
food away on a regular basis and waste unconscionable amounts of
money on things like cable television and fast food. There is
nothing in the US that even resembles poverty. once told a lady
in my church that there's no real poverty in the United States, and
she laughed at me. About a year later, as it happens, she travelled
to the Cameroon and spent a couple of months there. When she came
back, she was telling people that we don't have real poverty in the
US. The third world redefined the word "poverty" for her.
> Mhmm... well, if New England was a british colony, then wasnt the
> tea tax a Federal tax then? Or Hawaiians and Alaskans are not
> subject to pay US Federal taxes?
Hawaii and Alaska each sport two US senators and elect a member of
the US House of Representatives. The whole stupid war was mostly
over whether the colonies could be taxed without representation in
the House of Commons. (There were some other sticking points too...
but they aren't germaine to this discussion.)
The article talks about OS support, and lists supported OSes. It
seems people who use less common OSes are fresh out (though FreeBSD
gets half-decent support). My question is, why does it make any
difference what OS you use? Shouldn't the array be presented to
the OS (indeed, to the BIOS) as a single disk? Isn't that the whole
point of hardware RAID (as opposed to software RAID)? Why does the
OS even need to know there _is_ a RAID? Doesn't make sense to me.
I want to be able to set up a RAID (1, 3, or 5 are the schemes I'd be
interested in), partition the resulting "drive" N ways, and install
whatever OSes I want (provided they can run on x86 and support IDE,
of course), just like I can on a single physical disk. Why won't
that work?
> I don't quite understand where this Inexpensive crap came from.
RAID (Random Array of Inexpensive Disks) was as opposed to SLED
(Single Large Expensive Disk). (The term "Random" means the same
as in RAM -- i.e., that you can access any part (any drive, in this
case) at any time.)
> RAID was around long before IDE RAID controllers started showing
> up and of course SCSI RAID arrays almost always use very expesive
> disks.
"expensive" is relative. (Instead of thinking of SCSI as the only
other option besides RAID, try to remember that there were larger
and more expensive disks at one time.)
> It's Redunant Array of Independent Disks, always has always
> will be.
It's not necessarily "redundant" at all; some RAIDs are done just
for performance reasons, with no redundancy. (Personally, I am
more interested in the redundancy, however.)
> But can Windows even do thin clients?
Windows can be run _on_ thin clients. Whether it can act as the host
for the thin-client-server I don't know, but I'm not sure that really
matters.
> Frankly, I don't know of anyone doing this with Windows at all.
The school I attended had a network composed on Netware on the server
end (this was before there was such a thing as Windows NT) with thin
clients in the computer lab running DOS6 and Windows 3. They had
floppy drives, but no hard drives. (If I were setting up a network
of thin clients today, they wouldn't have any local drives at all,
_especially_ not floppy, but at the time floppies were needed.)
The nature of the way thin clients work is that you can send them
any OS you want to send them; they just boot whatever you give them
via the network. (The network card and BIOS have to support this.)
It doesn't have to be the same OS that runs on the server.
These days, of course, Microsoft has a thin client server offering.
I haven't used it, but I'm sure it's _way_ easier to administer
than numerious independent Windows installations on each client.
If nothing else, most of the big weakness of needing to physically
go to each computer to fix it would be greatly releaved. Of course,
you could also use Windows for the client OS but have the server
running something else.
The only downside to thin clients is Single Point Of Failure. For
that reason, I would be inclined to want to run something reliable
on the server (not Windows), have it on UPS, use fairly high-end
hardware, and so on, to prevent it from going down all the time.
> The amount of harm done isn't really that great.
You have apparently never had to repair a computer afflicted with
the $#@! stuff Bonzi tricks people into installing.
> the user does not perceive it to be an advertisement, but a
> compulsory upgrade.
It's worse than that. Go to a public library and watch people who
don't have a computer at home. By far and away the most common way
they follow these deceptive dialog-banner links is by clicking the
upper-right-hand corner, the part that resembles a close box.
They think it's modal. (They don't know the word "modal", or the
term "dialog box" for that matter, but that's irrelevant.) They
want it to go away, and the advertiser is deliberately harnessing
the user's desire to make it go away. That's why it's presented
as an error, rather than a positive message. The thing is designed
so that if the user tries to make it go away, they will have the
target content rammed down their throat -- obviously against their
will, since they tried to close the thing.
> I'm still waiting for someone to sue CapitalOne over the dead tree
> mailouts they send that try to look like a bill or some other
> important information.
What about the mailings you get that are printed in two or three
colours, with the main text in black and then script notes in the
margins in another colour, things circled, underlined, comments
added, made to appear as if a human being has gone over it for you
and highlighted the good parts. Sure, you'd have to be nearsighted
in the extreme to mistake those printings for actual handwritten
comments, but it's still a transparent attempt at deception.
> I don't find anything deceptive in the ads though.
Then you are one of the 98% whom the ad targets: people who don't
understand why it's bogus.
> You're computer is broadcasting an IP
No, it isn't. Your computer is narrowcasting an IP address only
to specific computers with which it is communicating. However,
even if your computer did not give out the address at all, other
computers would still know that such an address exists and that
there might be a computer at it. All the IP addresses on the
internet (_all_ of them) are routinely checked by port scanners,
even if there's _not_ a computer connected using that address.
So even if you _were_ broadcasting your IP, crackers would not
gain any advantage from that.
> an IP that can be used to attack you
Short answer: No, it can't.
> I would assume the InternetBoost really does increase download
> speeds.
How, exactly, would it accomplish this? (Hint: it wouldn't.)
> They look like a windows pop-up box, but does that mean anything
> that uses an actual pop-up box besides Windows in deceptive?
A program running on your computer is another matter. These are
not programs running on your computer; they are advertisements
trying to _appear_ to be such.
> When Kazaa pops up a box saying that there is an upgrade (or
> whatever) is that deceptive?
To be fully analagous, the box in question would have to go out
of its way to appear to have nothing to do with KaZaA, warn the
user of something that sound like a problem (not just offer that
"an upgrade is available", but more like "WARNING: Your computer
is not functioning properly! You need to get an upgrade!"), and
the "upgrade" in question would have to be something that the
user is currently not using, nor has expressed any particular
interest in using, something that doesn't do anything the dialog
box predicted but instead surreptitously performs some other
function that the user never intended.
That said, such an action wouldn't greatly surprise me coming
from KaZaA.
> Only banner two is legit, because it says your current connection
> MAY be capable of faster speeds.
These banners aren't fraudulent just because of what they say --
although what they say is certainly deceptive as well. My most
serious objection to them is that they are deliberately designed
to look like something much more important than advertising. If
a company started putting up roadside advertisements made to look
like road construction signs, little carts with blinking arrows
made of individual lights (such as the DOT uses), and so on, in
order to convince drivers that their driveway was the next exit,
or that it was imperative to get off at their exit as part of a
detour due to road construction, would we allow that? (Okay,
Microsoft is not a government agency, but the importance of
operating systems error messages on a computer is very similar
to the importance of highway department messages on a highway.)
That the messages in these fake dialogs are deceptive is just
the icing on the cake.
> Their 10-person IT staff supports 800 users running 400 devices
> (as Dave calls the thin clients). There is no way they could
> adequately support that many users and devices with such a small
> staff if they ran Windows on individual desktops
I've got to call you on this. As much as I think Linux has a lower
TCO than Windows, it's _completely_ unfair to compare Linux thin
clients to Windows desktops. Thin clients are a huge win in the
TCO department, regardless of which OS you use.
It's like saying, "Airplanes have a lower TCO than cars because
t's cheaper to go to Florida once a year by plane than to drive
there every week." Most of your savings there come from going
once a year versus once a week; it would be unreasonable to
attribute all of that to flying instead of driving.
If you want to compare, either compare thin clients with Linux
to Windows thin clients, or else compare full Linux desktops to
full Windows desktops. My take on the matter is that the Linux
systems will require more initial configuration (but not hugely
more), need less action later in the form of reboots, reinstalls,
and so forth, be somewhat more secure (but not totally secure
out of the box) and be easier to administer remotely, especially
in bulk (i.e., it's trivial to write a script that does the
same thing to a number of systems). However, more training is
required since fewer people have previously used X11 than have
used Win32. (Training is least relevant for servers, which
explains why Linux has done better there than in other areas
thus far.)
Where Windows really loses, IMO, is that a larger percentage
of stuff that needs to be done requires physically going to
the computer in question; with Linux, you do that when there
is a hardware issue. (Once you get it set up initially.)
Where Windows really wins is with training; it's what people
use at home. (Which is mildly ironic, given that Microsoft
achieved their home user share largely because "it's what I
use at work", back in the days of DOS and Windows 3.x.)
Depending on what your users need to do, OS training _may_ be a
total non-issue. This is especially true in cases where one big
application is the computer's whole world, such as library
catalog/circulation/automation systems.