There was no leftover configuration on that system. Trust me on that. Besides, deleting files by hand is exactly what I don't want to do when the installer could warn me and even do it for me. Again, the test is for seeing what problems "users" will run into, not if the browser can be forced to work.
I am completely aware of why it's got the 0.x version number, but that reason is conveniently ignored by those who recommend this version to endusers.
I didn't delete my.mozilla directory because on Win98 there isn't one... And there was no old Windows-style configuration on the system, either. Must be a platform thing. The point here is that people are not willing to "make their browsers work". Therefore, when I test a browser, I am quite unforgiving. I give up early on purpose - without throwing all bughunting methods at the problem. My intention is to find out if the browser can be recommended to users, not to see if *I* can make it work.
What are the options for the majority of users (that's Windows-users) when it comes to choosing an almost-standards-compliant browser? IE5.5, Netscape 6, Mozilla 0.8 and Opera. Last time I checked, Opera was not free. It had banners. With a free contender in the ring, that is a no-go. Mozilla and Netscape are incomplete, crash disturbingly often (at least on Windows) and the user interface is slow (with the exception of the rendering engine itself). Leaves us with IE5.5...
Recommend the switch now and prepare to watch Unix users beg Microsoft to port IE7, 'cause by then the W3 consortium will have lost it's influence on the web-"standards".
HTML was meant to describe content. What most sites use HTML for is something entirely different. The problem is not that we can't figure out how things should be in the next version of webauthor paradise. It's just that we still haven't found the way that both users and webauthors are willing to walk together to get there.
Ok, all these "Mozilla is quite stable" messages like yours have tricked me into once again giving the latest milestone build a chance. It took me about 1 minute to find a stupid user interface misfeature, 3 minutes to crash the browser without Java and another 2 minutes to find a completely reproducible bug with Java - and I wasn't even actively looking for crashbugs. That is *not* a stable browser. It deserves to be called version 0.x. Recommending to "normal" users that they should switch to that generation of browsers now is going to effectively kill every browser with the exception of MS Internet Explorer and maybe Opera. And when that has happened, how much attention is MS going to pay to W3 standards?
The way is to implement backward-compatibility into *browsers*, not webpages. Why don't the all-so-standards-compliant browsers include a filter-module which transforms old-style code into valid current-generation code? With all the flexibility that CSS offers, this shouldn't be too hard. As a user I still have only these two choices:
1. A browser of the previous generation which shows most of the web just fine but fails to accept standards compliant modern code or
2. A current generation browser which renders standards compliant sites perfectly but fails to show me the rest of the web.
Webauthors can't recommend upgrading until there are browsers which don't force users to make an exclusive choice between the majority of pages which rely on tricks and the minority of standards-dependant pages.
That is only going to hurt those with low self esteem. Everybody else realizes that their English is probably much better than your Russian. My native tongue is German and I know that I tend to speak the harsh English dialect that is common to most Germans. I *do* laugh about jokes involving that accent. I don't see a reason why this should be different. Calm down.
Re:Maybe MS will write the software...
on
Laser-equipped 747
·
· Score: 1
This is not too far out. Consider what happened to the other project they were involved in:
http://www.wist.uni-linz.ac.at/~rene/comics/msguid e.jpg
The link in the story has obviously been corrected (it pointed to http://www.rheinfire.de/pamster.htm), so you might find this an interesting read:
From an interview by german computer and technology magazine c't with Rob Malda (aka CmdrTaco) (translation of translation follows):
c't: Haben Sie ein Problem damit, etwas zu korrigieren?
Malda: Das hängt von der Geschichte ab. Wenn ich einen Absatz
geschrieben habe, der einen Fehler enthält und es Hunderte von Kommentaren
über diesen Fehler gibt, würde eine Fehlerbereinigung diese Kommentare
sinnlos machen. Die Nutzer-Kommentare sind wirklich wichtig. Wenn man sie nicht
liest, ist es so, als würde man sich nur den Anfang eines New-York-Times-Artikels
anschauen, aber nicht auf Seite 8 umblättern, um ihn weiterzulesen.
Translation:
c't: Do you think that correcting things is problematic?
Malda: That depends on the story. When I have written a paragraph with an error and there are hundreds of comments about this error, correcting it would make all these comments inane. User comments are really important. Not reading them is like glancing at the beginning of a New York Times article and then not turning the pages over to page 8 to read the rest of it.
For practical reasons you may want more than just 0 and 1 superpositioned when it comes to implementing a quantum computer, but the idea of quantum computing is that all these values are considered at once in a calculation. It's not about storing more per storage cell.
The idea behind the light bulb is not "to use 110V", but "to use electric current to heat up a thin wire (so that it heats up and emits light)".
That's kind of the idea behind quantum computers. Each quantum bit would have 32 possible values, which is equivalent to 5 binary bits.
No, it's not. The idea of quantum computing is to have bits which are 0 and 1 at the same time. A "byte" made of 8 quantum bits is a superposition of all 256 states a normal binary byte can represent. So if you want to check which one out of the numbers 0-255 is the key in an encryption scheme, you can either test 256 values one after the other conventionally or check them all at once with a quantum byte.
You can always just use the disk in unencrypting mode if you don't want your data taken away from you. The problem is that there will be software which won't be willing to write in unencrypted mode. Then there is no difference between choosing not to use such software on one hand and not being able to use such software because you don't have an encrypting harddisk on the other hand. Not buying a copy control disk might serve you as a reminder, but it certainly isn't a necessity from a technical point of view. It is a vote of some kind though, because if the main manufacturers switch to producing only copy control drives, small manufacturers are not going to make a difference: Your computer is expected to have a copy control drive. That is where the trouble is aiming at you from. Legislators and software companies alike will have a point in saying that it's ok to leave the few protesters in the cold for the "good" of all.
Onion routing doesn't require every router to participate in the protocol, but it is not based on central proxies either. Sniffing in and out packets on onion routers doesn't work (you can't match packets because one side is encrypted). The proxy server closest to you does not know any more than that you're communicating; neither who you're talking to nor about what. The same is true for the other side of the connection. A single trustworthy onion routing hop is sufficient for privacy, more trustworthy hops add redundancy and thus overall trustworhtyness. The key feature of onion routing is that from an "attackers" point of view observing traffic, even all of it, gains you nothing unless you manage to compromise every single routing hop (the onion router itself, not the communication lines!).
> Better yet: One number, period.
Yes, you could think of the spare 2 digits as a "port"
> The nature of the call is detected and handled accordingly.
Yes, I don't see why we couldn't have a packet sent first identifying the device type & connection.
Take a look at ISDN. With ISDN you get service identification and multiple numbers. Theoretically you can assign the same number to your phone and your fax machine, set your fax to answer "fax calls" and the phone to ring on "voice calls". Everyone who has ever tried that reverts to assigning one number to the fax and a different number to the phone, because in real life, service identifiers don't work. It's a legacy problem: Analogue fax devices don't identify themselves as fax machines before the call is answered. To the ISDN system, they look just like a human caller. Thus they end up trying to synchronize with your ear...
BTW: Germany uses 4 digit area codes (preceded by 0 to leave your local area) and 5 to 7 digit phone numbers (new numbers are all 7 digits long).
CRTs don't have a fixed 1 to 1 relation between pixels and phosphor pixels by design (and due to thermal changes). Thus you can't use Cleartype reliably with CRTs.
The optimum is a high resolution display and anti-aliasing. High resolution provides sharpness, anti-aliasing provides correct black-white distribution. The latter is very important and can not be achieved with just (reasonably) higher resolution.
The reason for that is that Windows does not support sub-pixel positioning and does not do full supersampling. Horizontal or vertical lines are not "smoothed". The intention is to keep the font look as sharp as possible. The problem with this approach is that the local black and white distribution of characters is modified, and since the relative error gets bigger with smaller fonts, they would look too ugly with adaptive antialiasing enabled.
On the other hand, if you do full sub-pixel positioning and full antialiasing, small fonts really benefit from antialiasing.
What you want is a screen resolution such that you can almost but not quite tell two pixels apart from normal viewing distance, floating point screen addressing and full antialiasing.
There are 66 "active" Iridium satellites (plus some spares). The element Iridium has atomic number 77. After the design change, they kept the name instead of changing it to Dysprosium...
Still, anyway you slice it, that's a fuckload of IPs. Which is good, because we certainly don't want to run into the "nobody could possibly need more than 512k of ram" problem.
Just wait until someone finds good use for one-time IPs, preferably for very short, frequent tasks...;)
I have filed a patent for the phone number of the white house. I discovered it in the phone directory and created a version which exists outside of it's normal context. I do have a working model and know what it can be used for.
Don't you dare dial that number, or you'll have to pay an ungodly license fee to me.
How far do companies think they can push it before it breaks? If you publish it, it's public. If you really want to keep it to yourself, hide it in a safe or obfuscate it. The patent system was invented for a reason, but companies are abusing the system and now it does more harm than good. If they can't behave themselves, the whole patent system will be dumped, so watch it!
Slashdot Mirror
My bad. You're right, I wasn't looking at M18, it was Mozilla 0.8.
There was no leftover configuration on that system. Trust me on that. Besides, deleting files by hand is exactly what I don't want to do when the installer could warn me and even do it for me. Again, the test is for seeing what problems "users" will run into, not if the browser can be forced to work.
I am completely aware of why it's got the 0.x version number, but that reason is conveniently ignored by those who recommend this version to endusers. .mozilla directory because on Win98 there isn't one... And there was no old Windows-style configuration on the system, either. Must be a platform thing. The point here is that people are not willing to "make their browsers work". Therefore, when I test a browser, I am quite unforgiving. I give up early on purpose - without throwing all bughunting methods at the problem. My intention is to find out if the browser can be recommended to users, not to see if *I* can make it work.
I didn't delete my
What are the options for the majority of users (that's Windows-users) when it comes to choosing an almost-standards-compliant browser? IE5.5, Netscape 6, Mozilla 0.8 and Opera. Last time I checked, Opera was not free. It had banners. With a free contender in the ring, that is a no-go. Mozilla and Netscape are incomplete, crash disturbingly often (at least on Windows) and the user interface is slow (with the exception of the rendering engine itself). Leaves us with IE5.5... Recommend the switch now and prepare to watch Unix users beg Microsoft to port IE7, 'cause by then the W3 consortium will have lost it's influence on the web-"standards".
HTML was meant to describe content. What most sites use HTML for is something entirely different. The problem is not that we can't figure out how things should be in the next version of webauthor paradise. It's just that we still haven't found the way that both users and webauthors are willing to walk together to get there.
Ok, all these "Mozilla is quite stable" messages like yours have tricked me into once again giving the latest milestone build a chance. It took me about 1 minute to find a stupid user interface misfeature, 3 minutes to crash the browser without Java and another 2 minutes to find a completely reproducible bug with Java - and I wasn't even actively looking for crashbugs. That is *not* a stable browser. It deserves to be called version 0.x. Recommending to "normal" users that they should switch to that generation of browsers now is going to effectively kill every browser with the exception of MS Internet Explorer and maybe Opera. And when that has happened, how much attention is MS going to pay to W3 standards?
The way is to implement backward-compatibility into *browsers*, not webpages. Why don't the all-so-standards-compliant browsers include a filter-module which transforms old-style code into valid current-generation code? With all the flexibility that CSS offers, this shouldn't be too hard. As a user I still have only these two choices:
1. A browser of the previous generation which shows most of the web just fine but fails to accept standards compliant modern code or
2. A current generation browser which renders standards compliant sites perfectly but fails to show me the rest of the web.
Webauthors can't recommend upgrading until there are browsers which don't force users to make an exclusive choice between the majority of pages which rely on tricks and the minority of standards-dependant pages.
Talking about root servers: You do realize where napster.COM is based DNS-wise?
how do all of the effected flights get redirected?
Abruptly.
2001-03-10 18:23:37
Checking databases for tracking information is futile. Think they manage to hit the target area? Think again. And take cover. ;)
That is only going to hurt those with low self esteem. Everybody else realizes that their English is probably much better than your Russian. My native tongue is German and I know that I tend to speak the harsh English dialect that is common to most Germans. I *do* laugh about jokes involving that accent. I don't see a reason why this should be different. Calm down.
This is not too far out. Consider what happened to the other project they were involved in:d e.jpg
http://www.wist.uni-linz.ac.at/~rene/comics/msgui
The link in the story has obviously been corrected (it pointed to http://www.rheinfire.de/pamster.htm), so you might find this an interesting read:
From an interview by german computer and technology magazine c't with Rob Malda (aka CmdrTaco) (translation of translation follows):
c't: Haben Sie ein Problem damit, etwas zu korrigieren?
Malda: Das hängt von der Geschichte ab. Wenn ich einen Absatz geschrieben habe, der einen Fehler enthält und es Hunderte von Kommentaren über diesen Fehler gibt, würde eine Fehlerbereinigung diese Kommentare sinnlos machen. Die Nutzer-Kommentare sind wirklich wichtig. Wenn man sie nicht liest, ist es so, als würde man sich nur den Anfang eines New-York-Times-Artikels anschauen, aber nicht auf Seite 8 umblättern, um ihn weiterzulesen.
Translation:
c't: Do you think that correcting things is problematic?
Malda: That depends on the story. When I have written a paragraph with an error and there are hundreds of comments about this error, correcting it would make all these comments inane. User comments are really important. Not reading them is like glancing at the beginning of a New York Times article and then not turning the pages over to page 8 to read the rest of it.
If I had some modpoints now, this would go down the way of all off-topic posts... ;)
For practical reasons you may want more than just 0 and 1 superpositioned when it comes to implementing a quantum computer, but the idea of quantum computing is that all these values are considered at once in a calculation. It's not about storing more per storage cell.
The idea behind the light bulb is not "to use 110V", but "to use electric current to heat up a thin wire (so that it heats up and emits light)".
That's kind of the idea behind quantum computers. Each quantum bit would have 32 possible values, which is equivalent to 5 binary bits.
No, it's not. The idea of quantum computing is to have bits which are 0 and 1 at the same time. A "byte" made of 8 quantum bits is a superposition of all 256 states a normal binary byte can represent. So if you want to check which one out of the numbers 0-255 is the key in an encryption scheme, you can either test 256 values one after the other conventionally or check them all at once with a quantum byte.
You can always just use the disk in unencrypting mode if you don't want your data taken away from you. The problem is that there will be software which won't be willing to write in unencrypted mode. Then there is no difference between choosing not to use such software on one hand and not being able to use such software because you don't have an encrypting harddisk on the other hand. Not buying a copy control disk might serve you as a reminder, but it certainly isn't a necessity from a technical point of view. It is a vote of some kind though, because if the main manufacturers switch to producing only copy control drives, small manufacturers are not going to make a difference: Your computer is expected to have a copy control drive. That is where the trouble is aiming at you from. Legislators and software companies alike will have a point in saying that it's ok to leave the few protesters in the cold for the "good" of all.
imagine 6 lines crossing the poles at 60 degree intervals
Make that (idealized) 30 degree intervals. There are 6 planes, cutting mother earth into 12 pieces of cake...
http://www.ee.surrey.ac.uk/ has a nice picture for the imagination impaired.
Onion routing doesn't require every router to participate in the protocol, but it is not based on central proxies either. Sniffing in and out packets on onion routers doesn't work (you can't match packets because one side is encrypted). The proxy server closest to you does not know any more than that you're communicating; neither who you're talking to nor about what. The same is true for the other side of the connection. A single trustworthy onion routing hop is sufficient for privacy, more trustworthy hops add redundancy and thus overall trustworhtyness. The key feature of onion routing is that from an "attackers" point of view observing traffic, even all of it, gains you nothing unless you manage to compromise every single routing hop (the onion router itself, not the communication lines!).
> Better yet: One number, period.
Yes, you could think of the spare 2 digits as a "port"
> The nature of the call is detected and handled accordingly.
Yes, I don't see why we couldn't have a packet sent first identifying the device type & connection.
Take a look at ISDN. With ISDN you get service identification and multiple numbers. Theoretically you can assign the same number to your phone and your fax machine, set your fax to answer "fax calls" and the phone to ring on "voice calls". Everyone who has ever tried that reverts to assigning one number to the fax and a different number to the phone, because in real life, service identifiers don't work. It's a legacy problem: Analogue fax devices don't identify themselves as fax machines before the call is answered. To the ISDN system, they look just like a human caller. Thus they end up trying to synchronize with your ear...
BTW: Germany uses 4 digit area codes (preceded by 0 to leave your local area) and 5 to 7 digit phone numbers (new numbers are all 7 digits long).
CRTs don't have a fixed 1 to 1 relation between pixels and phosphor pixels by design (and due to thermal changes). Thus you can't use Cleartype reliably with CRTs.
The optimum is a high resolution display and anti-aliasing. High resolution provides sharpness, anti-aliasing provides correct black-white distribution. The latter is very important and can not be achieved with just (reasonably) higher resolution.
On the other hand, if you do full sub-pixel positioning and full antialiasing, small fonts really benefit from antialiasing.
What you want is a screen resolution such that you can almost but not quite tell two pixels apart from normal viewing distance, floating point screen addressing and full antialiasing.
http://www.ee.surrey.ac.uk/
http://www.skypub.com/
Still, anyway you slice it, that's a fuckload of IPs. Which is good, because we certainly don't want to run into the "nobody could possibly need more than 512k of ram" problem.
Just wait until someone finds good use for one-time IPs, preferably for very short, frequent tasks... ;)
I have filed a patent for the phone number of the white house. I discovered it in the phone directory and created a version which exists outside of it's normal context. I do have a working model and know what it can be used for.
Don't you dare dial that number, or you'll have to pay an ungodly license fee to me.
How far do companies think they can push it before it breaks? If you publish it, it's public. If you really want to keep it to yourself, hide it in a safe or obfuscate it. The patent system was invented for a reason, but companies are abusing the system and now it does more harm than good. If they can't behave themselves, the whole patent system will be dumped, so watch it!