As far as I can tell, believe it or not, the August spike was Mozilla Foundation being reorganised. You can also see a clear spike when Firefox 1.0 was released. In fact there is a generally high correlation between Slashdot.org readership and Mozilla.org readership:
I would hazard a guess that the odds of randomly picking 100 numbers in a row without duplicates randomly from a population of 100 are low enough that the prisoners would die first...
Also, based on the question description, there doesn't seem to be any way for the prisoners to count days passing (windowless room, no guarentee that anything will happen to all prisoners each day).
IMHO a better solution is for the prisoners to keep a tally in the living room of how many people have been to that room at least once, e.g. by marking the walls in the way that prisoners in films always do... When the tally hits 100, the prisoner makes his assertion. Assuming no prisoner comits suicides by marking the wall twice, they should all be free once each prisoner has been there at least once. No need to use the lamp...
The problem isn't the design of X's copy-and-paste support. The problem is that the people who implement copy-and-paste all do it differently so nothing of what you describe actually works reliably and intuitively.
Um, the whole point of an open standard format is that it is an open standard -- you can't just add random new things to it, it wouldn't be compliant to the open standard any more.
What you're asking Microsoft to do here would in fact be exactly what they normall do; embrace and extend. And it would be bad for all the reasons it is normally bad: it would make files that claim to be OpenDocument files no longer work with OpenDocument-compliant UAs. Thus fragmenting the marketplace.
Give her a command line, a compiler, some basic instructions on how to program, and teach her how to find out more information herself... worked for me at that age (course back then I was on CP/M and using compas (a Pascal compiler)... I expect it's probably easier with a real OS).
Re:Some other (more useful) comments.
on
Hacking Hotels 101
·
· Score: 1
WEP isn't secure. There are several WEP-cracking tools available that will get you a key, either through an active attack or even through a passive attack.
Personally my home wireless network is non-encrypted and is serving DHCP to anyone who asks (although the router's configuration page is password-protected), mostly because I think sharing connectivity is fine, and I'm thankful for people who do that same when I'm stuck in a town I don't know.
But all my traffic is highly encrypted at the application layer (e.g. SSH for mail and editing my documents, SSL for IRC, TLS for any important Web sites I have to log into), so it doesn't matter. There's nothing to sniff except my/. password.
Yeah, I'm guessing they just had everyone on a hub. (Even if they used a switch, though, you could still get to see this stuff using ARP floods to redirect the traffic through your machine.)
I was in a hotel a few months ago, plugged into the free ethernet (for which I was very thankful), checking my e-mail, editing my documents on a remote server, chatting on IRC and browsing work sites (all over SSH, TLS, and SSL). My work consists amongst other things of testing Web browsers, and at one point I had to determine why one browser was not handling some HTTP headers correctly, so I fired up tcpdump to check exactly what headers were going over the wire.
What I saw scared the heck out of me. SQL queries from the hotel reservation system, including things like the results of "SELECT * FROM RESERVATIONS" and "INSERT INTO ROOMS..." and so on, with full credit card numbers, addresses, names, room numbers, lengths of stays, the works.
Not only was it all unencrypted, but they were broadcasting all that information to every ethernet port in every room. You can just imagine the potential for identity theft and burgalary networks ("he'll be gone til tuesday!"). And I wouldn't be surprised if you could actually just send out your own SQL queries if you wanted to ("I'll be staying for another week, honest!").
The Acid test is in Standards mode, Slashdot is in Quirks mode, so there should be no compatibility-breaking necessary. Do you know exactly what it is that would break compatibility if it wasn't for the quirks/standards thing?
I believe some versions of Konqueror do pass it. However, the point isn't so much that IE7 fails it; as you say, most browsers do (which was the intention, after all! Not much point in a test that every passes). The point is more that IE7 fails it so much more dramatically than every other modern browser.
Also, every other modern browser is getting closer with every release, whereas the IE team seems to have, as yet, made no effort whatsoever. Of course it's early days yet.
CSS2.1 has had as much ratification as CSS2. When CSS2 was originally published, "CR" was the last phase, except it was called "REC". Now there is another phase after "CR", and to get there you have to have two implementations. Nobody ever made a complete implementation of CSS2, so it would never have got there. That's why we made CSS2.1 in the first place, to create a version of CSS2 that had the opportunity to actually make it past that stage and reach the new final stage.
Incidentally, the software I use is on my site: http://software.hixie.ch/utilities/unix/trains/ Not exactly the prettiest code, but it works. Consider it covered by the GPL if you want to use it.
Buy a Märklin Digital starter set, then buy the 6051 Interface. It comes with a manual describing the serial API used.
Märklin actually recently came up with a new digital system and a new API to go with it (the entire layout becomes an actual network instead of just a one-way communications channel, it's pretty cool) but they haven't documented it yet.
I assumed everyone with a train set did this. My own train set (a Märklin Digital HO set of C track that I take out and build into various layouts when I get the inclination) is run by a TCP/IP server speaking a custom line-based protocol, and I've written a Web interface and an IRC bot interface to speak to it, as well as a couple of perl scripts that just run the trains around on specific schedules (using sensors in the track to detect when the train gets to a particular station).
So when I have my layout out people can just come to visit with their wireless laptops and immediately can control everything on the layout (trains, points, decouplers, etc).
I'm in Norway. I once had someone try to play a simplified Timesaver layout from Sweden, over IRC. That was not a pretty sight.
Admittedly I don't have a Web cam, which I assume is the attraction here.
(It's actually really hard to run any train set remotely, simply due to latency issues. Two seconds can easily be the difference between a neat arrival in a station and overshooting and hitting a freight train doing operations in a nearby yard.)
Slashdot Mirror
You know, meeting people you have never seen at an airport is a solved problem -- you hold up a card with their name on it. :-P
As far as I can tell, believe it or not, the August spike was Mozilla Foundation being reorganised. You can also see a clear spike when Firefox 1.0 was released. In fact there is a generally high correlation between Slashdot.org readership and Mozilla.org readership:
y =r&u=slashdot.org&u=mozilla.org&u=digg.com
http://traffic.alexa.com/graph?w=900&h=350&r=18m&
http://www.mozilla.org/news.html
Note that Alexa numbers only count users browsing with IE and with the Alexa toolbar installed.
Not to mention that the power to heat his water to oxidise this magnesium is generated completely environmentally cleanly too! It's amazing!
I would hazard a guess that the odds of randomly picking 100 numbers in a row without duplicates randomly from a population of 100 are low enough that the prisoners would die first...
Also, based on the question description, there doesn't seem to be any way for the prisoners to count days passing (windowless room, no guarentee that anything will happen to all prisoners each day).
IMHO a better solution is for the prisoners to keep a tally in the living room of how many people have been to that room at least once, e.g. by marking the walls in the way that prisoners in films always do... When the tally hits 100, the prisoner makes his assertion. Assuming no prisoner comits suicides by marking the wall twice, they should all be free once each prisoner has been there at least once. No need to use the lamp...
The problem isn't the design of X's copy-and-paste support. The problem is that the people who implement copy-and-paste all do it differently so nothing of what you describe actually works reliably and intuitively.
All apps on Vista are 3D. Even notepad. Vista uses 3D rendering for everything. It even renders Truetype fonts into models to render text.
Dude, that's a Red Dwarf reference, whereas this article is about Supernovaes. Quite inappropriate.
Um, the whole point of an open standard format is that it is an open standard -- you can't just add random new things to it, it wouldn't be compliant to the open standard any more.
What you're asking Microsoft to do here would in fact be exactly what they normall do; embrace and extend. And it would be bad for all the reasons it is normally bad: it would make files that claim to be OpenDocument files no longer work with OpenDocument-compliant UAs. Thus fragmenting the marketplace.
You know, like they did with HTML and the DOM.
Give her a command line, a compiler, some basic instructions on how to program, and teach her how to find out more information herself... worked for me at that age (course back then I was on CP/M and using compas (a Pascal compiler)... I expect it's probably easier with a real OS).
WEP isn't secure. There are several WEP-cracking tools available that will get you a key, either through an active attack or even through a passive attack.
/. password.
Personally my home wireless network is non-encrypted and is serving DHCP to anyone who asks (although the router's configuration page is password-protected), mostly because I think sharing connectivity is fine, and I'm thankful for people who do that same when I'm stuck in a town I don't know.
But all my traffic is highly encrypted at the application layer (e.g. SSH for mail and editing my documents, SSL for IRC, TLS for any important Web sites I have to log into), so it doesn't matter. There's nothing to sniff except my
Yeah, I'm guessing they just had everyone on a hub. (Even if they used a switch, though, you could still get to see this stuff using ARP floods to redirect the traffic through your machine.)
I was in a hotel a few months ago, plugged into the free ethernet (for which I was very thankful), checking my e-mail, editing my documents on a remote server, chatting on IRC and browsing work sites (all over SSH, TLS, and SSL). My work consists amongst other things of testing Web browsers, and at one point I had to determine why one browser was not handling some HTTP headers correctly, so I fired up tcpdump to check exactly what headers were going over the wire.
..." and so on, with full credit card numbers, addresses, names, room numbers, lengths of stays, the works.
What I saw scared the heck out of me. SQL queries from the hotel reservation system, including things like the results of "SELECT * FROM RESERVATIONS" and "INSERT INTO ROOMS
Not only was it all unencrypted, but they were broadcasting all that information to every ethernet port in every room. You can just imagine the potential for identity theft and burgalary networks ("he'll be gone til tuesday!"). And I wouldn't be surprised if you could actually just send out your own SQL queries if you wanted to ("I'll be staying for another week, honest!").
You can tunnel NTP over SSH if you first tunnel the UDP over TCP, e.g. using tcptunnel(1).
The Acid test is in Standards mode, Slashdot is in Quirks mode, so there should be no compatibility-breaking necessary. Do you know exactly what it is that would break compatibility if it wasn't for the quirks/standards thing?
I believe some versions of Konqueror do pass it. However, the point isn't so much that IE7 fails it; as you say, most browsers do (which was the intention, after all! Not much point in a test that every passes). The point is more that IE7 fails it so much more dramatically than every other modern browser.
Also, every other modern browser is getting closer with every release, whereas the IE team seems to have, as yet, made no effort whatsoever. Of course it's early days yet.
Well, it's a big company. My point is just that it isn't 100% IE only. There might be misguided parts that are, but not the entire place.
IBM is the company that published the article in question. I can assure you they are no longer an IE-only shop. Quite the contrary in fact.
CSS2.1 has had as much ratification as CSS2. When CSS2 was originally published, "CR" was the last phase, except it was called "REC". Now there is another phase after "CR", and to get there you have to have two implementations. Nobody ever made a complete implementation of CSS2, so it would never have got there. That's why we made CSS2.1 in the first place, to create a version of CSS2 that had the opportunity to actually make it past that stage and reach the new final stage.
CSS2.1 is CSS2.
(Disclosure: I'm one of CSS2.1 editors and I wrote that blog post.)
Well, everyone with a train set who would read or post to Slashdot, anyway.
Incidentally, the software I use is on my site:
http://software.hixie.ch/utilities/unix/trains/
Not exactly the prettiest code, but it works. Consider it covered by the GPL if you want to use it.
Buy a Märklin Digital starter set, then buy the 6051 Interface. It comes with a manual describing the serial API used.
Märklin actually recently came up with a new digital system and a new API to go with it (the entire layout becomes an actual network instead of just a one-way communications channel, it's pretty cool) but they haven't documented it yet.
I assumed everyone with a train set did this. My own train set (a Märklin Digital HO set of C track that I take out and build into various layouts when I get the inclination) is run by a TCP/IP server speaking a custom line-based protocol, and I've written a Web interface and an IRC bot interface to speak to it, as well as a couple of perl scripts that just run the trains around on specific schedules (using sensors in the track to detect when the train gets to a particular station).
So when I have my layout out people can just come to visit with their wireless laptops and immediately can control everything on the layout (trains, points, decouplers, etc).
I'm in Norway. I once had someone try to play a simplified Timesaver layout from Sweden, over IRC. That was not a pretty sight.
Admittedly I don't have a Web cam, which I assume is the attraction here.
(It's actually really hard to run any train set remotely, simply due to latency issues. Two seconds can easily be the difference between a neat arrival in a station and overshooting and hitting a freight train doing operations in a nearby yard.)
Indeed. And as has been pointed out many times in these comments, the idea that violence is somehow less bad than sex is quite ridiculous.
No, there is (implied, non-explicit) sex in the game even without the patch.