Around here they sell quite a lot of gas powered hot water systems. Hot water on demand, no waste, etc.. Saves space, and you're not heating water just to have it disperse the heat uselessly.
Anyway, those hot water systems need a power plug to light the starter flame to ignite the gas. That adds $100 to the cost as well as one of the main causes of failure. But a relatively new invention is to use the cold water pressure to start the starter flame. More specifically, cold water pressure to turn a turbine to generate a spark, but the end effect is fire from water and all that (oh, and no external electricity either).
Not quite, the chipset is the product. Microsoft is the customer; Hollywood, Adobe, web sites, and a few others are microsoft's customer; You are Hollywood et. al's customer, not the product.
I.e., you are their customer's customer's cusotmer. That's why they pay at least lip service to you.
Y'know, I remember zmodem (sz -rb wasn't it?). One of the properties I remember about zmodem was that it took over you connection "in order to get you every last KB/sec so you can watch your content ASAP." Specifically, you could download data in other ways (kermit? uucp? ftp?) and continue using your connection, but people tended to use zmodem because it was faster.
Of course, this is over ten years ago now so my memory is more than a little hazy; correct me if I'm wrong.
Actually, yes I did. Logo, then ASM (6502), then Basic, then pascal, then Splus, then C, then ASM (386), then ASM (601), then C++, then ADA, then lisp, then postscript, then java, then prolog, then perl, then ML, then haskell, then python.
Apart from the odd detour playing with languages for a day or two (forth, robotalk, cobol, fortran, snobol, etc.), those I don't call languages (csh, bash, make, etc.) and implementing my own, that's pretty much it.
I would have thought that forth was an extension of ps -- ps is pretty primitive. The dictonaries in ps for example, you can save and load dictionaries (basically, collections of functions) but you can't do proper operations on them. Oh, and file IO is best described as awkward... I don't even know if ps has a syscall function.
Perhaps cat is not the easiest to use, but it easily the most powerful and easier to control from another program! It can also be trivially scripted to produce eps and pdf, or later updated with $EDITOR.
apple doesn't (or, at least, shouldn't) care about geeks running osx on their boxes. If apple codes for the boxes they produces then it is up to the geeks to emulate that hardware.
I mean c'mon, light DRM has been associated with apple's products for a long time now. This will make it harder to run apple on non-apple hardware, and harder to pirate movies (so apple can say to hollywood: sign with us and we'll respect your rights.) There is little incentive on apple getting the DRM watertight.
In the unlikely event that they do manage it, I'll just avoid buying their hardware. I imagine the x86 version of pearpc will run at almost native speeds if there are any apple apps I want to run.
Actually, I'm rather fond of my phone having a complete addressbook and using a different ringtone depending on who is ringing. That's available now.
Since I've had to manually enter that all into the computer in order to sync the cellphone anyway, I'd be pretty keen on the phone also syncing the information off the computer. That's the direction I'd like to see phones develop.
There's plenty of ways to get them cheaper if the cost is a real issue for you. A number of clinics will give them away, and (depending on your doctor) you can also get a perscription for them, which brings them down to something like 10c each.
I installed ubuntu on my brother's brand new a64 dream machine a few months ago and had quite a few problems -- nothing I wasn't able to handle after 10 years using linux, but I imagine many of them would have been a showstopper for a newcomer. One thing I noticed is that using the older version of ubuntu I had posted to my brother had resulted in significantly more problems than the one I downloaded when I was there.
I imagine it would be similar with your fancy new laptop; if you get the latest ubuntu, even the 'testing' version, you're likely to have fewer problems on new hardware than the stable version.
Yes, exactly. And that is why unstable is not showing any of the security problems that stable is despite there being no team to help with security patches in unstable.
It was always my impression that you didn't get security updates with 'testing' and 'unstable
This is (technically) correct. However, whenever a security bug is discovered in an unstable package, the uploaded version fixing it (usually just upgrading to the very latest package) is installed within a day -- some of the nomal double checking is bypassed for speed. Since fixing security bugs in unstable is so much easier than in stable, it happens quickly.
Similarly for testing, any bugfix that corrects a security update gets fast-tracked to testing and so is available within a couple days. There is even a process for making this faster if the vulnerability is spreading rapidly, but in general people don't seem to mind waiting 2-3 days for a security update.
Consider a situation where a server has been set up and is running well in a company. That server has been working for several years, and while it may not have whiz-bang features, it keeps working every day just as well as it did the day before -- nothing ever breaks.
Now, if a security issue is discovered in a package running on that machine, they do not want to upgrade to the latest release because they would worry about what it changes -- they want that one issue fixed and everything else to continue the same as before. Debian Stable is designed for people like this, the joke at the end of your post was actually close to the truth -- people really do want debian stable to be stable feature wise.
Consider another situation, where somebody wants a fairly reliable and a fairly up-to-date server. When a bug is discovered, and especially security-related bugs, they'd like an updated package. On the other hand, they don't want to be sent the latest buggy software, they'd like it restricted to software that appears pretty stable. Debian Testing is designed for people like this.
It sounds from your post that you cannot imagine people preferring a quirky, somewhat old, consistant distro over one kept up to date with bug fixes. I assure you that there is a large market for the stable distro, but if you are not in that market, there are plenty of others available.
I'm tempted to mod you down for this, but I'll reply instead. The article you linked to shows that Brazil disregards patents on AIDS.
Good on them! There is no way that drug companies factor profits from Brazil into their feasibility studies for the simple reason that Brazil cannot afford AIDS drugs. Since drug companies are making AIDS drugs, we know they are making a profit without support from Brazil. I'm glad that Brazil has enough guts to stand up to the US, Germany and France over drug patents.
As for respecting the GPL in kiddie porn, would it really make it that much better if they did respect the GPL? All they would have to do is provide the source code along with their kiddie porn program. The GPL permits anybody to do anything with the software, including run spam sites, distribute kiddie porn or program the guantanamo bay gas chambers.
If Brazil was taking my GPLed software, turning it into a commercial product, and then selling it around the world then yes I would be pissed at them ignoring the GPL. But guess what? Even in that hypothetical scenario I don't have to worry about Brazil not respecting copyright because I can just go to a local court to have their actions banned and a fine imposed.
Or, for the shorter story, your analogy is useless.
Even if you're wrong and no pirated copy can exist, the compatibility of pearpc combined with the flexibility of a hypervisor means you don't really need to crack OSX in order to install it on your clone.
I've installed ubuntu for a couple people, but haven't got around to trying it myself yet. I assume by "do anything in the gui that requires sudo..." you meant to say "do anything in the gui that requires root...". If so, this is exactly what I mean and I'm glad ubuntu is heading down this route. It will be nice if they get it working well, especially if kubuntu and linspire pick it up.
Oh, and IIRC, OSX gives permission denied rather than the dialogue box from the command line, though naturally you can type sudo in the same way you can in linux.
Yeah, OSX went the sudo route, and they did it a lot better than any of the linux distros.
Specifically, you never log in as root/admin on osx (actually, you can't unless you change a few settings). Every time you do something which gives permission denied, you're not told that permission was denied, instead a friendly little dialogue box pops up asking for the name and password of someone in the sudoers file.
Since requring root in osx is about as rare as it is in linux, this dialogue box popping up when you weren't expecting it causes mental alarm bells to start ringing. Of course, the standard dismissive argument here is that mac users will simply type their password whenever requested. *shrug*. I have found it is much easier to use a linux box without root than to use a mac without root, so there is some support for that argument.
Contrast OSX's integrated sudo with the situation in linux, where an attempt to do inpromptu admin instead gets a 'permission denied' message 90% of the time, and 10% of the time you have to explicitly click on the 'do this as root' button. Looking at it that way, I can see why Linspire runs as root by default, KDE etc. just don't (yet?) have sudo integrated tightly enough.
Er... no. Debian announced it will be moving to xorg as soon as xorg makes a proper release instead of a legacy release. I think debian was the first distro to announce a switch to xorg, though I may be wrong.
In order to get off the ground quickly, xorg has been releasing versions based on xmkmf that have only really been tested on x86 and ppc. That's great, and means 90% of the people reading this can run xorg now instead of waiting six months for a non-legacy version.
Debian has been about doing things right, and waiting until they can do things right. They don't want to change to the transitional version of xorg and then change to the non-legacy version of xorg in six months. When xorg gets around to a proper build script based around configure, and starts supporting all the architectures of xfree86, then debian will switch to them.
Slashdot Mirror
You must work in a strange field. I hear estimates of growth all the time. The connection between thise estimates and reality has always eluded me.
Around here they sell quite a lot of gas powered hot water systems. Hot water on demand, no waste, etc.. Saves space, and you're not heating water just to have it disperse the heat uselessly.
Anyway, those hot water systems need a power plug to light the starter flame to ignite the gas. That adds $100 to the cost as well as one of the main causes of failure. But a relatively new invention is to use the cold water pressure to start the starter flame. More specifically, cold water pressure to turn a turbine to generate a spark, but the end effect is fire from water and all that (oh, and no external electricity either).
Not quite, the chipset is the product. Microsoft is the customer; Hollywood, Adobe, web sites, and a few others are microsoft's customer; You are Hollywood et. al's customer, not the product.
I.e., you are their customer's customer's cusotmer. That's why they pay at least lip service to you.
Y'know, I remember zmodem (sz -rb wasn't it?). One of the properties I remember about zmodem was that it took over you connection "in order to get you every last KB/sec so you can watch your content ASAP." Specifically, you could download data in other ways (kermit? uucp? ftp?) and continue using your connection, but people tended to use zmodem because it was faster.
Of course, this is over ten years ago now so my memory is more than a little hazy; correct me if I'm wrong.
Actually, yes I did. Logo, then ASM (6502), then Basic, then pascal, then Splus, then C, then ASM (386), then ASM (601), then C++, then ADA, then lisp, then postscript, then java, then prolog, then perl, then ML, then haskell, then python.
Apart from the odd detour playing with languages for a day or two (forth, robotalk, cobol, fortran, snobol, etc.), those I don't call languages (csh, bash, make, etc.) and implementing my own, that's pretty much it.
I would have thought that forth was an extension of ps -- ps is pretty primitive. The dictonaries in ps for example, you can save and load dictionaries (basically, collections of functions) but you can't do proper operations on them. Oh, and file IO is best described as awkward... I don't even know if ps has a syscall function.
cat > file.ps
10 10 moveto
50 50 lineto
stroke
showpage
Perhaps cat is not the easiest to use, but it easily the most powerful and easier to control from another program! It can also be trivially scripted to produce eps and pdf, or later updated with $EDITOR.
apple doesn't (or, at least, shouldn't) care about geeks running osx on their boxes. If apple codes for the boxes they produces then it is up to the geeks to emulate that hardware.
I mean c'mon, light DRM has been associated with apple's products for a long time now. This will make it harder to run apple on non-apple hardware, and harder to pirate movies (so apple can say to hollywood: sign with us and we'll respect your rights.) There is little incentive on apple getting the DRM watertight.
In the unlikely event that they do manage it, I'll just avoid buying their hardware. I imagine the x86 version of pearpc will run at almost native speeds if there are any apple apps I want to run.
Absolutely. Shooting problems always makes them go away, just ask ...
Actually, I'm rather fond of my phone having a complete addressbook and using a different ringtone depending on who is ringing. That's available now.
Since I've had to manually enter that all into the computer in order to sync the cellphone anyway, I'd be pretty keen on the phone also syncing the information off the computer. That's the direction I'd like to see phones develop.
I guess finding the ultimate question is a fairly major achievement. I bet when it is found, they will feel it justifies a 0.43 release.
There's plenty of ways to get them cheaper if the cost is a real issue for you. A number of clinics will give them away, and (depending on your doctor) you can also get a perscription for them, which brings them down to something like 10c each.
Alternatively, the pill works out cheaper.
try spending less on your hardware and you'll start seeing kernel panics again.
I installed ubuntu on my brother's brand new a64 dream machine a few months ago and had quite a few problems -- nothing I wasn't able to handle after 10 years using linux, but I imagine many of them would have been a showstopper for a newcomer. One thing I noticed is that using the older version of ubuntu I had posted to my brother had resulted in significantly more problems than the one I downloaded when I was there.
I imagine it would be similar with your fancy new laptop; if you get the latest ubuntu, even the 'testing' version, you're likely to have fewer problems on new hardware than the stable version.
Yes, exactly. And that is why unstable is not showing any of the security problems that stable is despite there being no team to help with security patches in unstable.
It was always my impression that you didn't get security updates with 'testing' and 'unstable
This is (technically) correct. However, whenever a security bug is discovered in an unstable package, the uploaded version fixing it (usually just upgrading to the very latest package) is installed within a day -- some of the nomal double checking is bypassed for speed. Since fixing security bugs in unstable is so much easier than in stable, it happens quickly.
Similarly for testing, any bugfix that corrects a security update gets fast-tracked to testing and so is available within a couple days. There is even a process for making this faster if the vulnerability is spreading rapidly, but in general people don't seem to mind waiting 2-3 days for a security update.
Consider a situation where a server has been set up and is running well in a company. That server has been working for several years, and while it may not have whiz-bang features, it keeps working every day just as well as it did the day before -- nothing ever breaks.
Now, if a security issue is discovered in a package running on that machine, they do not want to upgrade to the latest release because they would worry about what it changes -- they want that one issue fixed and everything else to continue the same as before. Debian Stable is designed for people like this, the joke at the end of your post was actually close to the truth -- people really do want debian stable to be stable feature wise.
Consider another situation, where somebody wants a fairly reliable and a fairly up-to-date server. When a bug is discovered, and especially security-related bugs, they'd like an updated package. On the other hand, they don't want to be sent the latest buggy software, they'd like it restricted to software that appears pretty stable. Debian Testing is designed for people like this.
It sounds from your post that you cannot imagine people preferring a quirky, somewhat old, consistant distro over one kept up to date with bug fixes. I assure you that there is a large market for the stable distro, but if you are not in that market, there are plenty of others available.
That's pretty funny. I know I'd get one if I saw it for sale.
I'm tempted to mod you down for this, but I'll reply instead. The article you linked to shows that Brazil disregards patents on AIDS.
Good on them! There is no way that drug companies factor profits from Brazil into their feasibility studies for the simple reason that Brazil cannot afford AIDS drugs. Since drug companies are making AIDS drugs, we know they are making a profit without support from Brazil. I'm glad that Brazil has enough guts to stand up to the US, Germany and France over drug patents.
As for respecting the GPL in kiddie porn, would it really make it that much better if they did respect the GPL? All they would have to do is provide the source code along with their kiddie porn program. The GPL permits anybody to do anything with the software, including run spam sites, distribute kiddie porn or program the guantanamo bay gas chambers.
If Brazil was taking my GPLed software, turning it into a commercial product, and then selling it around the world then yes I would be pissed at them ignoring the GPL. But guess what? Even in that hypothetical scenario I don't have to worry about Brazil not respecting copyright because I can just go to a local court to have their actions banned and a fine imposed.
Or, for the shorter story, your analogy is useless.
Even if you're wrong and no pirated copy can exist, the compatibility of pearpc combined with the flexibility of a hypervisor means you don't really need to crack OSX in order to install it on your clone.
Some of the newer TI calcs will show working too...
I've installed ubuntu for a couple people, but haven't got around to trying it myself yet. I assume by "do anything in the gui that requires sudo..." you meant to say "do anything in the gui that requires root...". If so, this is exactly what I mean and I'm glad ubuntu is heading down this route. It will be nice if they get it working well, especially if kubuntu and linspire pick it up.
Oh, and IIRC, OSX gives permission denied rather than the dialogue box from the command line, though naturally you can type sudo in the same way you can in linux.
Yeah, OSX went the sudo route, and they did it a lot better than any of the linux distros.
Specifically, you never log in as root/admin on osx (actually, you can't unless you change a few settings). Every time you do something which gives permission denied, you're not told that permission was denied, instead a friendly little dialogue box pops up asking for the name and password of someone in the sudoers file.
Since requring root in osx is about as rare as it is in linux, this dialogue box popping up when you weren't expecting it causes mental alarm bells to start ringing. Of course, the standard dismissive argument here is that mac users will simply type their password whenever requested. *shrug*. I have found it is much easier to use a linux box without root than to use a mac without root, so there is some support for that argument.
Contrast OSX's integrated sudo with the situation in linux, where an attempt to do inpromptu admin instead gets a 'permission denied' message 90% of the time, and 10% of the time you have to explicitly click on the 'do this as root' button. Looking at it that way, I can see why Linspire runs as root by default, KDE etc. just don't (yet?) have sudo integrated tightly enough.
Er... no. Debian announced it will be moving to xorg as soon as xorg makes a proper release instead of a legacy release. I think debian was the first distro to announce a switch to xorg, though I may be wrong.
In order to get off the ground quickly, xorg has been releasing versions based on xmkmf that have only really been tested on x86 and ppc. That's great, and means 90% of the people reading this can run xorg now instead of waiting six months for a non-legacy version.
Debian has been about doing things right, and waiting until they can do things right. They don't want to change to the transitional version of xorg and then change to the non-legacy version of xorg in six months. When xorg gets around to a proper build script based around configure, and starts supporting all the architectures of xfree86, then debian will switch to them.