I must say, I find myself more and more fond of the small [x] in the corner of modal dialogs. If I'm presented with [OK] [Cancel], I'm forced to make a decision, not when I want to make it but when - some component decided it's the right time to bother me. So I press [x] in the corner telling the component to go fuck itself because it's not nearly important enough to deserve my attention at given time. In most cases the [x] means "ask me later." In some cases it means "crash, die and burn, and serves you well for getting in my way."
Only when his claims are true. In that case his statement was false on both accounts (that change would fix a potential bug on x86, and not affect x86 performance adversely).
Still, it was fixed in the port, leaving the trunk vulnerable, because it depended on assumptions about properties that can be changed. Also, the patch didn't impact performance of x86.
So not only he was wrong on both accounts (it did not at cost to supported architectures, and it didn't only benefit "embedded crap"), he was abusive.
In the other bug - the faulty/etc/hosts handling -
- He broke the standard - He claimed it's the badly written apps fault when the bug occured, even when shown a scenario where two correctly written apps communicating with each other are affected. - He claimed he has reasons to keep it that way instead of fixing it, stronger than reasons the bug reporter gave - He refused to disclose his reasons (demanded money to disclose them) - Upon examination it appeared his reason was a nanosecond saving in a once-a-hour function.
So, it wasn't just being abusive, or just wrong, or just lazy. That was a "La la la I don't hear you" style of stupid refusal to either fix your mistake or even allow others to do it for you.
I'm a Linux user too. I have a Microsoft bluetooth mouse (IntelliMouse Explorer.)
And I want to choke the authors of Linux Bluez to death.
The idiots decided that it's a good idea to require me to enter the PIN they suggest on the keypad of the bluetooth device, and outright refused to implement options for the user to provide their own PIN.
The procedure required to get my mouse to work in Linux involves emailing kernel messages and debug data to them and waiting till their updates to the hardcoded database are included in the distribution updates. Yes, Linux is completely ready for Desktop.
Absolutely NOT the cheap wireless. I had some and the experience was outright horrible. The battery fails fast, the range is like 40cm from the receiver, and random interferences make the mouse to fail. I don't know about "top of the line" wireless, but I got a decent bluetooth mouse recently and I don't want to look back. One significant minus is start-up time. The bluetooth service loads quite late and takes a few seconds to connect, which means it takes some 10s of staring at the loaded desktop screen before it starts working. It also took some work to hook up. But when it's working, it's working. I never noticed any delays, any loss of precision or reaction speed, it works just fine several meters from the receiver, and the battery life beats the old ones.
I still have the receiver of the old 'proprietary wireless' mouse plugged in. It has a charger built in, and I have two sets of accumulators for the bluetooth mouse - one in the mouse, one in the charger.
Unless you use initially pressurized gas as "fuel".
The "charger" may contain a pump that fills a small container with pressurized air - enough to last one battery charge. Then as GUI elements are required, the pre-charged container is being depleted. You don't need much of the air.
I mean, currently, if I can't enter a secure PIN on my bluetooth device it doesn't have in its database, it won't allow me to install it. A bluetooth mouse - a standard HID device - won't work because it lacks the keyboard. Solution: send us the ID the mouse sends over bluetooth (some very obscure console commands), it will be added with the next release.
There are several such "flowers" currently - "for the good of the user, we won't let them do this, the fact they need it be damned."
argh, I miss the day when Linux motto was "It should be fun".
Still they can take the same approach as Linden Labs did (in case of "firstlife" parody site, which used their logo): send a "Permit and proceed letter" - a one-time non-transferable free license to use the "infringing" trademarks.
By "protecting" the law means only "don't infringements leave unattended". Not "don't let anyone else use it, ever".
Considering the PIN is usually 4-8 digit number (about 30 bits of data) brute-forcing a hash you intercept instead of the PIN doesn't sound like much of a challenge.
Truth is absolute defense. Untruthful defamation is a subject of punishment. Any defamation is a subject of lawsuit. It's the same as with rape - the rare exception to "innocent until proven guilty". In this case the burden of proof of innocence is on the defense.
"Yeah, curving lasers, very viable". "Lasers shoot straight, stupid!" "So this beam is some kind of plasma or what? Laser? Are you kidding?" Now they will all see! My sharks will be able to shoot lasers from behind a corner!
Slashdot Mirror
I must say, I find myself more and more fond of the small [x] in the corner of modal dialogs.
If I'm presented with [OK] [Cancel], I'm forced to make a decision, not when I want to make it but when - some component decided it's the right time to bother me. So I press [x] in the corner telling the component to go fuck itself because it's not nearly important enough to deserve my attention at given time.
In most cases the [x] means "ask me later." In some cases it means "crash, die and burn, and serves you well for getting in my way."
"And the developer has every right"
Only when his claims are true.
In that case his statement was false on both accounts (that change would fix a potential bug on x86, and not affect x86 performance adversely).
As long as GLIBC does things outright WRONG...
(say, your app, conforming to RFC, allows values at which GLIBC crashes?)
Still, it was fixed in the port, leaving the trunk vulnerable, because it depended on assumptions about properties that can be changed. Also, the patch didn't impact performance of x86.
So not only he was wrong on both accounts (it did not at cost to supported architectures, and it didn't only benefit "embedded crap"), he was abusive.
In the other bug - the faulty /etc/hosts handling -
- He broke the standard
- He claimed it's the badly written apps fault when the bug occured, even when shown a scenario where two correctly written apps communicating with each other are affected.
- He claimed he has reasons to keep it that way instead of fixing it, stronger than reasons the bug reporter gave
- He refused to disclose his reasons (demanded money to disclose them)
- Upon examination it appeared his reason was a nanosecond saving in a once-a-hour function.
So, it wasn't just being abusive, or just wrong, or just lazy. That was a "La la la I don't hear you" style of stupid refusal to either fix your mistake or even allow others to do it for you.
Yeah, right. I bought this dongle which was supported only in Vista, so I installed Vista and...
You know, fuck you.
I'm a Linux user too. I have a Microsoft bluetooth mouse (IntelliMouse Explorer.)
And I want to choke the authors of Linux Bluez to death.
The idiots decided that it's a good idea to require me to enter the PIN they suggest on the keypad of the bluetooth device, and outright refused to implement options for the user to provide their own PIN.
The procedure required to get my mouse to work in Linux involves emailing kernel messages and debug data to them and waiting till their updates to the hardcoded database are included in the distribution updates. Yes, Linux is completely ready for Desktop.
Absolutely NOT the cheap wireless.
I had some and the experience was outright horrible. The battery fails fast, the range is like 40cm from the receiver, and random interferences make the mouse to fail.
I don't know about "top of the line" wireless, but I got a decent bluetooth mouse recently and I don't want to look back.
One significant minus is start-up time. The bluetooth service loads quite late and takes a few seconds to connect, which means it takes some 10s of staring at the loaded desktop screen before it starts working. It also took some work to hook up. But when it's working, it's working. I never noticed any delays, any loss of precision or reaction speed, it works just fine several meters from the receiver, and the battery life beats the old ones.
I still have the receiver of the old 'proprietary wireless' mouse plugged in. It has a charger built in, and I have two sets of accumulators for the bluetooth mouse - one in the mouse, one in the charger.
Unless you use initially pressurized gas as "fuel".
The "charger" may contain a pump that fills a small container with pressurized air - enough to last one battery charge. Then as GUI elements are required, the pre-charged container is being depleted. You don't need much of the air.
Also, a phone with a built-in lighter.
You won't see them in ATMs.
This seems entirely non-vandal-proof.
You're missing one important point:
How much would Google be losing to competition if they didn't have Youtube?
It's a war out there, and Youtube is an outpost - costly to keep, but if you don't keep it, the enemy will gain not only it but a lot of field.
but to my credit, I can claim I brought it to military spec
(duty cycle of 3/4th time on, 1/4th off.)
I mean, currently, if I can't enter a secure PIN on my bluetooth device it doesn't have in its database, it won't allow me to install it. A bluetooth mouse - a standard HID device - won't work because it lacks the keyboard. Solution: send us the ID the mouse sends over bluetooth (some very obscure console commands), it will be added with the next release.
There are several such "flowers" currently - "for the good of the user, we won't let them do this, the fact they need it be damned."
argh, I miss the day when Linux motto was "It should be fun".
Trademark needs to be protected by preventing unauthorized use of it.
But there are two ways of
- disallow/litigate
- authorize/license
Wikipedia choose the nasty way.
Linden Labs used the nice way.
http://games.slashdot.org/article.pl?sid=07/01/31/0216258&from=rss
Still they can take the same approach as Linden Labs did (in case of "firstlife" parody site, which used their logo): send a "Permit and proceed letter" - a one-time non-transferable free license to use the "infringing" trademarks.
By "protecting" the law means only "don't infringements leave unattended". Not "don't let anyone else use it, ever".
actually -outgoing- spam from infected computers has slowed down a network I was connected to down to a crawl.
The only reading model change that would help here is switching back to paper books and snailmail.
I thought it could address 16MB or RAM, not of diskspace?
Or does rendering a T-Rex have to be an atomic operation?
What about "minimize to tray" software?
I have at least 3 normal applications minimized to tray using a 3rd party tool most of the time here.
In fact, behind the Curtain, we had more freedom than we do now.
Touching a taboo topic hurt a lot, but the list of taboo topics was quite short and left little room for doubt what was a fair game and what wasn't.
Call me paranoid but the /cse?cx= part of the URL makes me distrustful.
Considering the PIN is usually 4-8 digit number (about 30 bits of data) brute-forcing a hash you intercept instead of the PIN doesn't sound like much of a challenge.
http://xkcd.com/556/
Just not even half as awesome.
http://en.wikipedia.org/wiki/File:Rdft0004.png
>>So, this is a neat twist on an older idea
They bent the laws of physics!
Truth is absolute defense. Untruthful defamation is a subject of punishment. Any defamation is a subject of lawsuit.
It's the same as with rape - the rare exception to "innocent until proven guilty". In this case the burden of proof of innocence is on the defense.
"Yeah, curving lasers, very viable".
"Lasers shoot straight, stupid!"
"So this beam is some kind of plasma or what? Laser? Are you kidding?"
Now they will all see! My sharks will be able to shoot lasers from behind a corner!