My friend's dad works for the British Antarctic Survey. Apparently they once had some data analysis that needed doing and a really crappy computer to do it on, so they found a spare room on the outside of the building, stripped out the insulation, put the computer in it and something like doubled the clock speed.
Exponential growth is the same as a geometric increase, OK?
Doubling every [year, week, nanosecond] means x = A.2^{kt} for some suitable constants A and k; that's equivalent to x = A exp{kt log 2}, so it's also exponential growth.
(k depends on the unit you measure time in, and x=exp{kt} is exponential growth for any value of k: otherwise you'd get things that were exponential growth when you measure in years, but not when you measure in months or decades or Mars years or something:-)
In fact, that's how you define raising to any power, formally. The "definition" by repeated multiplication ("a^5 = a*a*a*a*a") starts to break when you want to raise to the power of a complex or irrational number, so you *define* a^b:= exp{b log a} - this works as expected when b is an integer or a rational, and also gives you sensible answers for b irrational, and answers that are no weirder than the rest of complex numbers for b complex.
(If you don't think this is necessary, please try multiplying a by itself pi+sqrt(-3) times and explain how you did it:-)
Someone wrote a CGI script for Apache that responded by using the exploit to fix it *and* make the admin aware of the problem. If I remember correctly, it did these:
- install the necessary MS patch to prevent the exploit
- set the computer up to display an explanatory message halfway through next time it booted
- shut down Windows
Not legal to use, of course, but if you're going to use the exploit, you might as well do it constructively:-)
In Europe, we have data protection law (google for Data Protection Act to see Britain's implementation, but it's an EU thing, so the other European countries have very similar laws with the same purpose).
Companies that deal with "personally identifiable information" have to be registered with a central authority (the Data Protection Registrar, a government office), and have to give a reason why they want to keep the information. If a company passes on the information, or even just uses it for something other than its original purpose, they're already breaking the law, so no new law is necessary.
(The registration thing might be different outside Britain, but the principle is probably the same in France)
When the country TLDs were allocated, the domains were issued individually (rather than "everyone gets their ISO code", it was a case of "the USA is hereby allocated.us, Germany is hereby allocated.de" and so on).
Whoever allocated the United Kingdom's TLD assumed the ISO code was UK, since that was the obvious code for it. (We even call ourselves "the UK", so we've already got a de facto 2-letter name like the US does, right?)
Unfortunately, it turns out that when ISO codes were handed out, nobody was thinking about Northern Ireland, so the United Kingdom of Great Britain and Northern Ireland was actually allocated GB. This matches the GB stickers seen on British cars which are temporarily abroad, the other commonly used name for the country is "Britain", and a lot of European languages call us things like Grosbrittanien or Grand Bretagne, so it does make sense to do that.
By the time the TLD authority realised there was a discrepancy,.uk domains were already in active use, and apparently they decided against deprecating.uk and allocating.gb instead. (Don't ask me why... maybe it took long enough to notice that the change would have seriously confused people.)
This is something MS just won't approve - a signed Linux binary would run arbitrary x86 code on an X-Box without mod chips, something MS have been very careful to avoid.
On the other hand, as the open letter makes pretty clear, they're just aiming for interoperability, they're not pestering Microsoft for technical information, and just like MS, they'd even prefer to avoid mod chips - so where's the problem?;-)
In the absence of flying pigs, this just makes Microsoft look rather silly, not to mention anticompetitive (again). Yes, true, their business model for the X-Box does rely on being anti-competitive. However, I fail to see how that's the X-Box Linux project's problem.
(he who lives by the legalese, dies by the legalese:-)
I've been playing with User-Mode Linux a bit recently; it's a port of Linux to run on Linux:-) (instead of running on real hardware, it does hardware-ish things via Linux syscalls). It runs as an unpriviledged user, but has its own internal users, permissions, even a root user.
I have a domain at a dedicated hosting company (i.e. not an ISP), which I want to use as my primary address. At the moment the host has a mail relay (using POP-before-SMTP to prevent abuse), so while at home, I ignore my ISP's mail services completely and use my web host's POP3 and relaying; when my domain was with a different host a while ago, I couldn't even do that, since they provided POP but not SMTP, so I had to use a "forged" header as you describe.
At university this is even more necessary, since my university blocks port 25 at most of their routers - the only exception is that anyone in the university can connect to a "server" (actually a load-balancing cluster) which acts as a central relay. This means it's impossible to send mail unless it's either tunnelled in some way (not an option for me, my web host charges extra for ssh), or through this relay server. The relay accepts mail with any faked From address, on the basis that some people (including some departments) need this functionality, and if someone spams through it, they have it logged and know who to blame.
(Before you ask whether my uni gives me an e-mail address: yes it does, but I do game modifications, and I don't want to use my uni address for that. Also, my domain is more permanent than an address that disappears when I graduate)
IANAPP*, but if I was doing browser detection for MSN, it'd work something like this:
- If it's Netscape or IE, version 4 or earlier (i.e. a browser I *know* fails to degrade gracefully), send a dumbed-down stylesheet
- If it's Opera 5 (wasn't that the one with the indenting bug they were working around?), send a hacked stylesheet which fixes it (just Opera 5 though, not <=4 or >=6)
- If The Boss says so, and the browser is IE6, send extra proprietary junk so it looks marginally nicer (hey, I'm trying to think like MSN here:-)
- Otherwise (default action), send a standards-compliant HTML4+CSS or XHTML+CSS page that does exactly what I want; anyone using a browser that's old *and* obscure probably knows what they're doing, if a new browser doesn't show it properly it's the browser author's fault, and "limited" browsers (text-only browsers, PDAs, screen readers,...) should just ignore the stuff they don't understand and show the content.
The design of HTML is meant to be such that browsers that don't understand a particular feature (like CSS) can ignore it and get a "gracefully degraded" version. The only ways to get a browser to actually show things *badly* are:
- Bugs in a particular feature (someone mentioned text-indent and margin-left not playing nicely together in Opera 5; MSIE doesn't add up border, margin and width quite right) - Half-implementations of standards (Netscape 4, basically.)
IMO the first priority when implementing a website's stylesheets, once you've decided what they're going to look like should be to make an "ideal" version which is valid and standards-compliant, and does what you wanted; hacks to make it still look nice in specific browsers should be optional and only done for those specific browsers (so browsers the designer doesn't specifically support, like Konqueror, Safari and "Oprah", get sent the "ideal" version).
My own site deliberately links in the stylesheet using a method NS4 doesn't understand (the @import feature in CSS), so NS4 gets no stylesheet at all. This makes it look boring, but at least it's legible, which makes it better than last time I tried to feed NS4 stylesheets - I gave up when inline images started overlapping my text, in a valid page which IE 4, an early Mozilla, and (IIRC) Opera rendered perfectly.
If you visit my site with a browser that knows about @import, you get the full-CSS version, which has no different content, but looks prettier. (I write the stylesheets using the CSS2 spec, with Mozilla as my "previewing" browser, then test in Konqueror and the W3C CSS and HTML validators occasionally, plus MSIE when I can be bothered).
Isn't the issue that the developers of MSIE/Opera7 and NS/Opera6 have chosen to implement their list indentation differently?
If I remember correctly, one strategy is to give the list a large (38px sounds right) left margin and no padding; the other is to give it a normal-size left margin and 38px of padding.
(The difference: padding is applied inside any borders, margins are applied outside the border. If you've used the presentational attributes on HTML tables, think of padding as like cellpadding and margins as like cellspacing.)
The consequence of these different choices is that if you want a list to stop being indented, you have to explicitly set the padding and the margin; but to be honest, if you're messing with padding and margins you should probably be setting them both anyway and not relying on the browser's defaults.
(My site does this, as well as some css-fu to collapse nested ul elements in the navigation into a single horizontal line; view a "deep" page in my site once in a modern browser and once in NS4.7, or use a "Disable CSS" bookmarklet, to see what I mean:-)
Unless the author of your webserver is particularly clueless, CGI scripts that run for more than a certain time (30 seconds?) get killed unceremoniously. 30-second uptimes are not suitable for running a botnet:-)
The Nullsoft Scriptable Install System is the open-source installer developed for Winamp. (Yes, I know Winamp is closed, but the installer it uses is under the zlib license).
I live in Britain, and most of that stuff sounds familiar (I can't remember the exact rules, but 6 weeks sounds right, and we certainly have subsidised party-political broadcasts and spending limits for individual politicians' ads).
The down side is that you don't get to vote on US laws, but you occasionally get them applied to you anyway.
mathematicians are pretty sure, but still cannot prove conclusively
In the best case, statistical analysis could come up with something like "there is a 99% probability that the numbers occur randomly". That's not a proof, that's just quantifying "pretty sure".
Pi is represented usually by a fraction or relatively simple equation, it's just the division that makes the number go on for ever.
Nope. If pi was rational (a fraction), it wouldn't go on for ever without repeating. (reference)
In fact pi is irrational, i.e. there are no integers p, q such that pi = p / q. (proof)
You can approximate pi as a fraction, which is what projects like this do. (pi is approximately equal to 31/10, or 314/100, or 31416/1000, or... but these are just approximations; 22/7 is a good enough approximation a lot of the time, but that's just an approximation too)
Britain (i.e. including England) is rapidly moving away from Imperial/"English" units. It's now illegal to price most fruit/vegetables/meat/etc. per pound unless the price per kilogram/gram/whatever is at least as prominent (although there's nothing to stop supermarkets charging per 454 grams). The running joke is that soon we'll get people in court for selling an ounce of marijuana, not because of what it is but because they sold it by the ounce:-)
We still use pints for milk, and pints and half-pints for beer, but most other capacity measures are in metric (actually, I'm not sure what the standard measures for spirits are, they might still be Imperial too).
The only other really common uses of Imperial units I can think of are measuring long distances (we still measure roads and speed limits in miles and miles per hour), and people's heights and weights (somehow it's still easier to visualise what someone looks like if you know they're 6'1" tall than if you know they're 1.85m tall).
Length - already covered by the AC Mass - already covered by parent Time - AC Angle - there are very good mathematical reasons for using radians Solid angle, luminous intensity - does anyone ever use these ones?:-) Amount of substance (mole in SI) - I'd hesitate to call this a dimensional unit, since it's just a very large number. Temperature - the AC quoted a unit of energy. Temperature is not energy, so you'll need a real unit of temperature.
The other missing one is either current or charge (intuitively, you'd think the charge on the electron would be the ideal base unit, but in fact SI defines that in terms of current (charge/time), because current is easier to measure, using the force between current-carrying wires).
An 'Ogg' is a tactical maneuver from the network game 'Netrek' that has entered common usage in a wider sense. From the definition:
3. To do anything forcefully, possibly without consideration of the drain on future resources. "I guess I'd better go ogg the problem set that's due tomorrow." "Whoops! I looked down at the map for a sec and almost ogged that oncoming car."
(see the rest of the definition for the original Netrek usage.)
At the time Ogg was starting out, most personal computers were i386s and the i486 was new. I remember thinking about the algorithms I was considering, "Woah, that's heavyweight. People are going to need a 486 to run that..." While the software ogged the music, there wasn't much processor left for anything else.
These days, Ogg is a larger multimedia project that does not only concern compression; Squish became the name of one of the Ogg codecs. For that reason, we usually just refer to it as Ogg when there's no Netrek context nearby. The Ogg project has nothing to do with the common surname 'Ogg'. Nor is it named after 'Nanny Ogg' from the Terry Pratchett book _Wyrd Sisters_.
[...]
Vorbis, on the other hand is named after the Terry Pratchett character from the book _Small Gods_. The name holds some significance, but it's an indirect, uninteresting story.
This is Britain's implementation of (probably) the previous reply, but the.gov.uk website for it seems to have more English and less legalese, so I'll post it anyway:
I had to do a bit of stuff on this during A-Level Computing (A-Level = the academic qualification most British 16-to-19-year-olds are studying for) and it seems very sane and well thought out. America could probably benefit from something like this.
Yes, you can arbitrarily block people if you really want to.
However, unless your website is so utterly mercenary that you probably don't have any of your own content anyway, it's a stupid thing to do - why alienate people for no good reason?
If the purpose of your site is to sell your products, you'll sell them better if your site is accessible. If the purpose of your site is to provide information, you'll provide that info to more people if you let them in. If the purpose of your site is solely to attract ad revenue, I don't want to be visiting it anyway, so I suppose you blocking me is actually sort of useful.
(For blocking read unnecessary Flash, excessive scripting, gratuitously incompatible HTML, or anything else that detracts from your site's purpose)
Our P2-233 (we ordered a P2-200, but apparently the manufacturer ran out of 200MHz chips and gave us a free upgrade) had USB ports, came with an OS that supported them at least slightly (Win95 OSR2), and was happy as a cable modem router/web browser terminal (with Debian 'woody' by now) for the last year and a bit.
I say "was" because it died recently; almost certainly flaky hardware (motherboard I think), since before installing Debian I had to swap the various expansion cards and the two sticks of RAM into exactly the right slots before it'd recognise them on boot*. For a month or so before it failed, the video card didn't work and I wasn't at home to try fixing it, so I told my parents how to shut it down "blind" (my Linux boxes always have Ctrl+Alt+Del set to shutdown rather than reboot) and it worked fine as a video-less router:-)
(*: I think it was 64MB in slot 1 and 32MB in slot 3, or something like that - every other combination didn't recognise all the RAM when it POSTed. As for the PCI slots, which needed to accomodate two net cards and a sound card, just don't go there:-)
Slashdot Mirror
My friend's dad works for the British Antarctic Survey. Apparently they once had some data analysis that needed doing and a really crappy computer to do it on, so they found a spare room on the outside of the building, stripped out the insulation, put the computer in it and something like doubled the clock speed.
Exponential growth is the same as a geometric increase, OK?
:-)
:= exp{b log a} - this works as expected when b is an integer or a rational, and also gives you sensible answers for b irrational, and answers that are no weirder than the rest of complex numbers for b complex.
:-)
Doubling every [year, week, nanosecond] means x = A.2^{kt} for some suitable constants A and k; that's equivalent to x = A exp{kt log 2}, so it's also exponential growth.
(k depends on the unit you measure time in, and x=exp{kt} is exponential growth for any value of k: otherwise you'd get things that were exponential growth when you measure in years, but not when you measure in months or decades or Mars years or something
In fact, that's how you define raising to any power, formally. The "definition" by repeated multiplication ("a^5 = a*a*a*a*a") starts to break when you want to raise to the power of a complex or irrational number, so you *define* a^b
(If you don't think this is necessary, please try multiplying a by itself pi+sqrt(-3) times and explain how you did it
Someone wrote a CGI script for Apache that responded by using the exploit to fix it *and* make the admin aware of the problem. If I remember correctly, it did these:
:-)
- install the necessary MS patch to prevent the exploit
- set the computer up to display an explanatory message halfway through next time it booted
- shut down Windows
Not legal to use, of course, but if you're going to use the exploit, you might as well do it constructively
No useful text here, read the AC's parent post instead :-)
In Europe, we have data protection law (google for Data Protection Act to see Britain's implementation, but it's an EU thing, so the other European countries have very similar laws with the same purpose).
Companies that deal with "personally identifiable information" have to be registered with a central authority (the Data Protection Registrar, a government office), and have to give a reason why they want to keep the information. If a company passes on the information, or even just uses it for something other than its original purpose, they're already breaking the law, so no new law is necessary.
(The registration thing might be different outside Britain, but the principle is probably the same in France)
When the country TLDs were allocated, the domains were issued individually (rather than "everyone gets their ISO code", it was a case of "the USA is hereby allocated .us, Germany is hereby allocated .de" and so on).
.uk domains were already in active use, and apparently they decided against deprecating .uk and allocating .gb instead. (Don't ask me why... maybe it took long enough to notice that the change would have seriously confused people.)
Whoever allocated the United Kingdom's TLD assumed the ISO code was UK, since that was the obvious code for it. (We even call ourselves "the UK", so we've already got a de facto 2-letter name like the US does, right?)
Unfortunately, it turns out that when ISO codes were handed out, nobody was thinking about Northern Ireland, so the United Kingdom of Great Britain and Northern Ireland was actually allocated GB. This matches the GB stickers seen on British cars which are temporarily abroad, the other commonly used name for the country is "Britain", and a lot of European languages call us things like Grosbrittanien or Grand Bretagne, so it does make sense to do that.
By the time the TLD authority realised there was a discrepancy,
This is something MS just won't approve - a signed Linux binary would run arbitrary x86 code on an X-Box without mod chips, something MS have been very careful to avoid.
;-)
:-)
On the other hand, as the open letter makes pretty clear, they're just aiming for interoperability, they're not pestering Microsoft for technical information, and just like MS, they'd even prefer to avoid mod chips - so where's the problem?
In the absence of flying pigs, this just makes Microsoft look rather silly, not to mention anticompetitive (again). Yes, true, their business model for the X-Box does rely on being anti-competitive. However, I fail to see how that's the X-Box Linux project's problem.
(he who lives by the legalese, dies by the legalese
I've been playing with User-Mode Linux a bit recently; it's a port of Linux to run on Linux :-) (instead of running on real hardware, it does hardware-ish things via Linux syscalls). It runs as an unpriviledged user, but has its own internal users, permissions, even a root user.
I have a domain at a dedicated hosting company (i.e. not an ISP), which I want to use as my primary address. At the moment the host has a mail relay (using POP-before-SMTP to prevent abuse), so while at home, I ignore my ISP's mail services completely and use my web host's POP3 and relaying; when my domain was with a different host a while ago, I couldn't even do that, since they provided POP but not SMTP, so I had to use a "forged" header as you describe.
At university this is even more necessary, since my university blocks port 25 at most of their routers - the only exception is that anyone in the university can connect to a "server" (actually a load-balancing cluster) which acts as a central relay. This means it's impossible to send mail unless it's either tunnelled in some way (not an option for me, my web host charges extra for ssh), or through this relay server. The relay accepts mail with any faked From address, on the basis that some people (including some departments) need this functionality, and if someone spams through it, they have it logged and know who to blame.
(Before you ask whether my uni gives me an e-mail address: yes it does, but I do game modifications, and I don't want to use my uni address for that. Also, my domain is more permanent than an address that disappears when I graduate)
IANAPP*, but if I was doing browser detection for MSN, it'd work something like this:
:-)
...) should just ignore the stuff they don't understand and show the content.
- If it's Netscape or IE, version 4 or earlier (i.e. a browser I *know* fails to degrade gracefully), send a dumbed-down stylesheet
- If it's Opera 5 (wasn't that the one with the indenting bug they were working around?), send a hacked stylesheet which fixes it (just Opera 5 though, not <=4 or >=6)
- If The Boss says so, and the browser is IE6, send extra proprietary junk so it looks marginally nicer (hey, I'm trying to think like MSN here
- Otherwise (default action), send a standards-compliant HTML4+CSS or XHTML+CSS page that does exactly what I want; anyone using a browser that's old *and* obscure probably knows what they're doing, if a new browser doesn't show it properly it's the browser author's fault, and "limited" browsers (text-only browsers, PDAs, screen readers,
The design of HTML is meant to be such that browsers that don't understand a particular feature (like CSS) can ignore it and get a "gracefully degraded" version. The only ways to get a browser to actually show things *badly* are:
- Bugs in a particular feature (someone mentioned text-indent and margin-left not playing nicely together in Opera 5; MSIE doesn't add up border, margin and width quite right)
- Half-implementations of standards (Netscape 4, basically.)
IMO the first priority when implementing a website's stylesheets, once you've decided what they're going to look like should be to make an "ideal" version which is valid and standards-compliant, and does what you wanted; hacks to make it still look nice in specific browsers should be optional and only done for those specific browsers (so browsers the designer doesn't specifically support, like Konqueror, Safari and "Oprah", get sent the "ideal" version).
My own site deliberately links in the stylesheet using a method NS4 doesn't understand (the @import feature in CSS), so NS4 gets no stylesheet at all. This makes it look boring, but at least it's legible, which makes it better than last time I tried to feed NS4 stylesheets - I gave up when inline images started overlapping my text, in a valid page which IE 4, an early Mozilla, and (IIRC) Opera rendered perfectly.
If you visit my site with a browser that knows about @import, you get the full-CSS version, which has no different content, but looks prettier. (I write the stylesheets using the CSS2 spec, with Mozilla as my "previewing" browser, then test in Konqueror and the W3C CSS and HTML validators occasionally, plus MSIE when I can be bothered).
* I am not a professional programmer
PS. I really liked learning about matrices and vectors but I have no idea wtf section of maths that this is a part of?
:-)
Linear algebra, I think.
(the courses I've done that specifically dealt with this sort of thing were called Algebra & Geometry and Linear Mathematics, if that's any help)
If you need to wait until a college professor instructs you about computing then its already late to get into computing IMHO. =)
As a second year maths student who writes programs in his spare time, I can agree with that one
Isn't the issue that the developers of MSIE/Opera7 and NS/Opera6 have chosen to implement their list indentation differently? If I remember correctly, one strategy is to give the list a large (38px sounds right) left margin and no padding; the other is to give it a normal-size left margin and 38px of padding. (The difference: padding is applied inside any borders, margins are applied outside the border. If you've used the presentational attributes on HTML tables, think of padding as like cellpadding and margins as like cellspacing.) The consequence of these different choices is that if you want a list to stop being indented, you have to explicitly set the padding and the margin; but to be honest, if you're messing with padding and margins you should probably be setting them both anyway and not relying on the browser's defaults. (My site does this, as well as some css-fu to collapse nested ul elements in the navigation into a single horizontal line; view a "deep" page in my site once in a modern browser and once in NS4.7, or use a "Disable CSS" bookmarklet, to see what I mean :-)
Unless the author of your webserver is particularly clueless, CGI scripts that run for more than a certain time (30 seconds?) get killed unceremoniously. 30-second uptimes are not suitable for running a botnet :-)
The Nullsoft Scriptable Install System is the open-source installer developed for Winamp. (Yes, I know Winamp is closed, but the installer it uses is under the zlib license).
I live in Britain, and most of that stuff sounds familiar (I can't remember the exact rules, but 6 weeks sounds right, and we certainly have subsidised party-political broadcasts and spending limits for individual politicians' ads).
The down side is that you don't get to vote on US laws, but you occasionally get them applied to you anyway.
mathematicians are pretty sure, but still cannot prove conclusively
In the best case, statistical analysis could come up with something like "there is a 99% probability that the numbers occur randomly". That's not a proof, that's just quantifying "pretty sure".
http://www.math.clemson.edu/~rsimms/neat/math/pipr oof.html
That's the great thing about maths, you can prove things like pi being infinitely long without actually calculating any digits.
Pi is represented usually by a fraction or relatively simple equation, it's just the division that makes the number go on for ever.
... but these are just approximations; 22/7 is a good enough approximation a lot of the time, but that's just an approximation too)
Nope. If pi was rational (a fraction), it wouldn't go on for ever without repeating. (reference)
In fact pi is irrational, i.e. there are no integers p, q such that pi = p / q. (proof)
You can approximate pi as a fraction, which is what projects like this do. (pi is approximately equal to 31/10, or 314/100, or 31416/1000, or
Britain (i.e. including England) is rapidly moving away from Imperial/"English" units. It's now illegal to price most fruit/vegetables/meat/etc. per pound unless the price per kilogram/gram/whatever is at least as prominent (although there's nothing to stop supermarkets charging per 454 grams). The running joke is that soon we'll get people in court for selling an ounce of marijuana, not because of what it is but because they sold it by the ounce :-)
We still use pints for milk, and pints and half-pints for beer, but most other capacity measures are in metric (actually, I'm not sure what the standard measures for spirits are, they might still be Imperial too).
The only other really common uses of Imperial units I can think of are measuring long distances (we still measure roads and speed limits in miles and miles per hour), and people's heights and weights (somehow it's still easier to visualise what someone looks like if you know they're 6'1" tall than if you know they're 1.85m tall).
SI base units:
:-)
Length - already covered by the AC
Mass - already covered by parent
Time - AC
Angle - there are very good mathematical reasons for using radians
Solid angle, luminous intensity - does anyone ever use these ones?
Amount of substance (mole in SI) - I'd hesitate to call this a dimensional unit, since it's just a very large number.
Temperature - the AC quoted a unit of energy. Temperature is not energy, so you'll need a real unit of temperature.
The other missing one is either current or charge (intuitively, you'd think the charge on the electron would be the ideal base unit, but in fact SI defines that in terms of current (charge/time), because current is easier to measure, using the force between current-carrying wires).
thanks to cofc.edu via Google
At least, I really hope so...
This is Britain's implementation of (probably) the previous reply, but the .gov.uk website for it seems to have more English and less legalese, so I'll post it anyway:
dataprotection.gov.uk
I had to do a bit of stuff on this during A-Level Computing (A-Level = the academic qualification most British 16-to-19-year-olds are studying for) and it seems very sane and well thought out. America could probably benefit from something like this.
</karma-whore>
Yes, you can arbitrarily block people if you really want to.
However, unless your website is so utterly mercenary that you probably don't have any of your own content anyway, it's a stupid thing to do - why alienate people for no good reason?
If the purpose of your site is to sell your products, you'll sell them better if your site is accessible. If the purpose of your site is to provide information, you'll provide that info to more people if you let them in. If the purpose of your site is solely to attract ad revenue, I don't want to be visiting it anyway, so I suppose you blocking me is actually sort of useful.
(For blocking read unnecessary Flash, excessive scripting, gratuitously incompatible HTML, or anything else that detracts from your site's purpose)
-- smcv, owner of an ad-free website
Our P2-233 (we ordered a P2-200, but apparently the manufacturer ran out of 200MHz chips and gave us a free upgrade) had USB ports, came with an OS that supported them at least slightly (Win95 OSR2), and was happy as a cable modem router/web browser terminal (with Debian 'woody' by now) for the last year and a bit.
:-)
:-)
I say "was" because it died recently; almost certainly flaky hardware (motherboard I think), since before installing Debian I had to swap the various expansion cards and the two sticks of RAM into exactly the right slots before it'd recognise them on boot*. For a month or so before it failed, the video card didn't work and I wasn't at home to try fixing it, so I told my parents how to shut it down "blind" (my Linux boxes always have Ctrl+Alt+Del set to shutdown rather than reboot) and it worked fine as a video-less router
(*: I think it was 64MB in slot 1 and 32MB in slot 3, or something like that - every other combination didn't recognise all the RAM when it POSTed. As for the PCI slots, which needed to accomodate two net cards and a sound card, just don't go there