Thanks to modern society, I realize you're quite divorced from the skills of animal husbandry, but think back to Oregon Trail. They hauled wagons, and plowed fields, I think you can get one to walk in a circle. I mean, you can get a camel to do something similar
Its possible that they're interested because, as Tim Gettys said, making software work better on OLPC will make it work better everywhere. Nobody notices a 1W software power consumption problem when the back light consumes 10W and the rest of the hardware another 5W. But OLPC is ambitious enough that you have to make it work. Microsoft might want to use the OLPC as a cheap platform to "test their mettle", so to speak, and bring the benefits back home to the desktop. With the XO hardware already determined, they'll only need to focus on the software, and mitigate potential competitive advantages Linux might have gained.
The other possibility I see is saving face. Microsoft is familiar with exactly why OLPC said they wouldn't pick Windows: hardware vendors are terrible at fixing their software. Even the WHQL can only do so much to twist their arms, imagine what little sway a mere laptop manufacturer holds. Now that the hardware is finalized, Microsoft can apply pressure to specific vendors in ways that OLPC couldn't, such as fixing their source code etc. And when it runs on OLPC, they'll point to their results and say "See, Microsoft software works fine, use it today", neglecting to mention the massive and unusual driver coordination effort that was needed.
The thing is, you're in a very narrow market. You're failing to find the people you need, possibly because their talents are already put to work elsewhere. Or perhaps you're simply over-discounting the ability of a good engineer to learn new domains. Nobody wants to know ACPI unless they have to, but that doesn't mean they can't spend time reading the specs and do a good job making things.
On the other hand, I've seen seminars on how to advertise for a position and meet "insufficient local talent" laws, because what they're really after is a small army of moderately trained, but more importantly, cheap programmers. One technique could be to require an unusual skill set, say something like, "Experience with: Python, Ruby on Rails, Apache and uClinux kernel programming".
Oil, gold and just about every commodity has seen large jumps in price over the past year. Core inflation is basically up 2.8 percent over the measurement last year, which discards food and oil prices. Food is up 4.8 percent and oil 5.3.
More importantly:
Consumer prices increased at a seasonally adjusted annual rate (SAAR) of 1.0 percent in the third quarter of 2007, following increases in the first and second quarters at annual rates of 4.7 and 5.2 percent, Wages are sticky -- they take time to adjust to market forces, for a large number of reasons, including "IANAEconomist". This suggests to me that wages were partly up 5.5 percent because of inflation, and if the credit crunch hadn't had a large effect on the market, wages could have been down or stable compared to inflation. In other words, this report not news at all. If you're celebrating an increased wage of 5.5 percent, stop.
Yes. Some states legalized earlier than others, and they look at these natural on / off switches as indicators and apparently it works out as affirming the hypothesis.
Good news, Bryce Herrington and other Ubuntu developers are working to make this sort of feat feasible for anyone. 7.10 includes a bulletproof X that launches a failsafe session with a very simple xorg config intended to work on all platforms, and runs a program (displayconfig-gtk) that lets you reconfigure the system graphically. It's not perfect, but it's a working start. This typically hasn't been looked at on Linux systems because the command line is passable if you're technically inclined enough to be familiar, and traditional linux users happen to be so technically inclined.
Of course, Macs are just as good -- whenever something goes wrong, you can just buy a new one!
If they merely want to get "their cut of that 10x marked up ticket", then why not auction to the highest bidder or simply price the tickets closer to market value?
Tickets are by definition a scarce resource So why then, is the trade of them barred? By getting tickets from those who value them less to those who value them more, you're increasing wealth in society. Which participant is the victim here? The team that could have charged more for tickets, the person who wishes to sell their tickets, or the person who wishes to buy tickets?
To draw a a parallel with the drug laws in my country, small possession for personal use of illegal drugs is treated a lot less harshly than running a drug distribution business. A punishment, increasing for repeat offenders; but not a life-destroying punishment for a first, small personal offence. I'm not familiar with the specifics of the case in the article, but I'd like to propose a question about distribution: if you use bittorrent as intended, is this puff and pass, or selling drugs to minors? Which way does such an analogy fall? And if you inhibit the upload is that like smoking but not inhaling?
There is a team called Laptop-Testing. They have a mailing list, that you might subscribe to and ask about a team effort. This sort of work just doesn't pay off for distros; they can document common power usage problems, but there are so many model specific bugs to fix that getting a "working great" system doesn't translate well to someone else's hardware. Ultimately, the vendors are the ones with the most incentives to make things happen.
But AFAIK, Canonical's relationship with Ubuntu as it currently stands is to: * Develop Launchpad, the bug tracker that Canonical guided Ubuntu into * Hire just enough developers to keep Ubuntu afloat * Take on contract work to improve Ubuntu itself for people's needs
In the current release, it feels a bit like Canonical stole from 2 to make 3 work for Intel's lpia initiative. Hence the underwhelming release.
Do your CDs often come with a write back cache? Because most programs should be able to handle a read error (file not found) situation, and it's a shitty excuse to lock down media because some program has an open read handle on the FS. You requested to remove it, the software should deal with it.
On the other hand, if your removal request conflicts with a write operation you presumably also requested, one hopes the system makes a point of honoring the write before the ejection. Sadly, USB drives don't have much they can do to stop this, aside from a transactional journal or something to at least keep the FS coherent.
The reason seems to be that the 3d accelerator on the GPU emits huge amounts of heat when being used. I think you mean 3d acceleration. 3d accelerometers measure tilt and motion in three dimensions. To address the heat problem you attempted to convey: nvidia is very cool for me -- the chips are designed to be efficient, whereas Intel offloads a great majority of work to the CPU. Admittedly, nvidia isn't the most power friendly driver around; powertop reports it as one of the bigger offenders, but nothing to where I can't watch a movie with the fans getting loud.
Indeed, Xubuntu can help with older PCs. And if you want, you can go even more minimal. The default CD's install (x|k?)ubuntu-desktop by default, but you can ask to install something like ubuntu-minimal instead. It might require the alternative installer, I haven't tried it.
Gentoo's case for optimizations and the like is massively overstated. If you dig into debian packaging, you'll see that debug packages have optimization off, and debhelper packages build with -O2 by default, and the gains to be had from recompiling for a specific arch are not as fantastic as you might expect. But if Gentoo is the excuse you needed to run off and be adventuresome with linux and configuration, hats off. Just don't think what you've learned is somehow specific to Gentoo. And you'll save a lot of power by reusing someone else's built packages;)
Gutsy early on (june till late september) suffered from a couple of critical bugs that would likely cause laptops problems. Trackerd, a search indexer similar to beagle and friends, would occasionally get into massive CPU loads, causing me to have to kill it. It also ignored indexing while on battery. I believe both those are fixed, albeit the battery check involves polling something in/proc last I checked. Compiz is also likely to be a heat contributer, as it puts the usually idle 3d chipsets to work.
I keep two installs and a shared/home on my laptop so if something like this happens, I can still fall back on something I feel reliable.
The question was not how to make the code available, any idiot can put the code on SF or Google's code hosting. It was about making it OSS appealing. Dead upstreams aren't appealing, with good cause. Nobody knows the code better than the author, and for something user centric and internet enabled like a phpBB replacement (itself already a security nightmare), you want someone to be watching over the code and accepting security patches at the least.
And SF.net provides little means to determine whether code is worthwhile or not -- it mostly just tells you whether something exists or not. Ohloh at least tells you things like how many comments there are, how many contributers there are and how significant their contributions have been, how long it's been under development, etc. Releasing the code's the first step, but not the only step to making code appealing.
Why are 4 player free for alls and tournaments exclusive? One of the more interesting parts, I think, is the strategies that evolve in a 4 player game. It turns out to also be quite suitable for fun play.
That didn't stop Miguel from trying. Or anyone else. The fundamental problem is that Mono is incomplete, and Miguel's efforts to port it really highlighted the problems. Enough that he said he still uses the old unpublished port as a test case for Mono.
I realize that it's been promised since forever, but the good news is that 2.4 is coming out shortly, allowing development to focus on 2.6, which is slated to bring in exactly that longstanding missing feature. Granted, the release could slip, but GIMP seems to be picking up a bit of steam -- GEGL was published in late 2006, and they're also hitting up the UI redesign. Both of these may woo people away from Photoshop. Granted, I doubt GIMP will ever see the massive plugin library that Photoshop has (short of a terrible wrapper plugin), but that's a tall order for any app, open or closed.
I dare say, if your undergraduate course doesn't teach you how the Power rule is derived, you deserve your money back. These are things I was taught in a public high school; I would hope your community college could at least keep pace with that.
Security critical logs where correlations need to be established between program use and even computers shouldn't be using local time. UNIX or UTC timestamps are a better approach, if a bit unreadable. There's already a lot of potential confusion because DST is particular to the region the server is in, the laws the regional government has passed, and the state of the admin's patches.
Have you actually used Debian testing? Or Debian at all? I admit it's a complex beast, and gotten more complex since I left, as this whole article revolves around a misunderstanding between sysadmins and developers. New packages start off in unstable (or experimental if the uploader thinks it's too buggy to consider for regular use), and if, after ten days without a major bug report, the package is placed in testing. Unstable is a rocky ride, but so many people use it that most developers are highly concerned about regressions like tab completion core dumping. So much so that during one cycle the DPL announced an unstable freeze until the new stable was released.
This misunderstanding about timezones is based on where changes go. Security updates to packages in stable go to the "security" repo. Things like clamav definitions change on a regular basis and reside in volatile. This particular repo is news to me, but I don't admin Debian boxes. The developers believe the update should belong in volatile and not security. That is all. Stable remains stable.
Slashdot Mirror
Thanks to modern society, I realize you're quite divorced from the skills of animal husbandry, but think back to Oregon Trail. They hauled wagons, and plowed fields, I think you can get one to walk in a circle. I mean, you can get a camel to do something similar
Its possible that they're interested because, as Tim Gettys said, making software work better on OLPC will make it work better everywhere. Nobody notices a 1W software power consumption problem when the back light consumes 10W and the rest of the hardware another 5W. But OLPC is ambitious enough that you have to make it work. Microsoft might want to use the OLPC as a cheap platform to "test their mettle", so to speak, and bring the benefits back home to the desktop. With the XO hardware already determined, they'll only need to focus on the software, and mitigate potential competitive advantages Linux might have gained.
The other possibility I see is saving face. Microsoft is familiar with exactly why OLPC said they wouldn't pick Windows: hardware vendors are terrible at fixing their software. Even the WHQL can only do so much to twist their arms, imagine what little sway a mere laptop manufacturer holds. Now that the hardware is finalized, Microsoft can apply pressure to specific vendors in ways that OLPC couldn't, such as fixing their source code etc. And when it runs on OLPC, they'll point to their results and say "See, Microsoft software works fine, use it today", neglecting to mention the massive and unusual driver coordination effort that was needed.
The thing is, you're in a very narrow market. You're failing to find the people you need, possibly because their talents are already put to work elsewhere. Or perhaps you're simply over-discounting the ability of a good engineer to learn new domains. Nobody wants to know ACPI unless they have to, but that doesn't mean they can't spend time reading the specs and do a good job making things.
On the other hand, I've seen seminars on how to advertise for a position and meet "insufficient local talent" laws, because what they're really after is a small army of moderately trained, but more importantly, cheap programmers. One technique could be to require an unusual skill set, say something like, "Experience with: Python, Ruby on Rails, Apache and uClinux kernel programming".
More importantly: Consumer prices increased at a seasonally adjusted annual rate (SAAR) of 1.0 percent in the third quarter of 2007, following increases in the first and second quarters at annual rates of 4.7 and 5.2 percent, Wages are sticky -- they take time to adjust to market forces, for a large number of reasons, including "IANAEconomist". This suggests to me that wages were partly up 5.5 percent because of inflation, and if the credit crunch hadn't had a large effect on the market, wages could have been down or stable compared to inflation. In other words, this report not news at all. If you're celebrating an increased wage of 5.5 percent, stop.
Yes. Some states legalized earlier than others, and they look at these natural on / off switches as indicators and apparently it works out as affirming the hypothesis.
Good news, Bryce Herrington and other Ubuntu developers are working to make this sort of feat feasible for anyone. 7.10 includes a bulletproof X that launches a failsafe session with a very simple xorg config intended to work on all platforms, and runs a program (displayconfig-gtk) that lets you reconfigure the system graphically. It's not perfect, but it's a working start. This typically hasn't been looked at on Linux systems because the command line is passable if you're technically inclined enough to be familiar, and traditional linux users happen to be so technically inclined.
Of course, Macs are just as good -- whenever something goes wrong, you can just buy a new one!
If they merely want to get "their cut of that 10x marked up ticket", then why not auction to the highest bidder or simply price the tickets closer to market value?
In which case, the FS is not mounted "ro."
Wow. Guess I took too much work home with me. At any rate, my comment about Intel 3d stands.
There is a team called Laptop-Testing. They have a mailing list, that you might subscribe to and ask about a team effort. This sort of work just doesn't pay off for distros; they can document common power usage problems, but there are so many model specific bugs to fix that getting a "working great" system doesn't translate well to someone else's hardware. Ultimately, the vendors are the ones with the most incentives to make things happen.
But AFAIK, Canonical's relationship with Ubuntu as it currently stands is to:
* Develop Launchpad, the bug tracker that Canonical guided Ubuntu into
* Hire just enough developers to keep Ubuntu afloat
* Take on contract work to improve Ubuntu itself for people's needs
In the current release, it feels a bit like Canonical stole from 2 to make 3 work for Intel's lpia initiative. Hence the underwhelming release.
Do your CDs often come with a write back cache? Because most programs should be able to handle a read error (file not found) situation, and it's a shitty excuse to lock down media because some program has an open read handle on the FS. You requested to remove it, the software should deal with it.
On the other hand, if your removal request conflicts with a write operation you presumably also requested, one hopes the system makes a point of honoring the write before the ejection. Sadly, USB drives don't have much they can do to stop this, aside from a transactional journal or something to at least keep the FS coherent.
Indeed, Xubuntu can help with older PCs. And if you want, you can go even more minimal. The default CD's install (x|k?)ubuntu-desktop by default, but you can ask to install something like ubuntu-minimal instead. It might require the alternative installer, I haven't tried it.
;)
Gentoo's case for optimizations and the like is massively overstated. If you dig into debian packaging, you'll see that debug packages have optimization off, and debhelper packages build with -O2 by default, and the gains to be had from recompiling for a specific arch are not as fantastic as you might expect. But if Gentoo is the excuse you needed to run off and be adventuresome with linux and configuration, hats off. Just don't think what you've learned is somehow specific to Gentoo. And you'll save a lot of power by reusing someone else's built packages
Gutsy early on (june till late september) suffered from a couple of critical bugs that would likely cause laptops problems. Trackerd, a search indexer similar to beagle and friends, would occasionally get into massive CPU loads, causing me to have to kill it. It also ignored indexing while on battery. I believe both those are fixed, albeit the battery check involves polling something in /proc last I checked. Compiz is also likely to be a heat contributer, as it puts the usually idle 3d chipsets to work.
/home on my laptop so if something like this happens, I can still fall back on something I feel reliable.
I keep two installs and a shared
The question was not how to make the code available, any idiot can put the code on SF or Google's code hosting. It was about making it OSS appealing. Dead upstreams aren't appealing, with good cause. Nobody knows the code better than the author, and for something user centric and internet enabled like a phpBB replacement (itself already a security nightmare), you want someone to be watching over the code and accepting security patches at the least.
And SF.net provides little means to determine whether code is worthwhile or not -- it mostly just tells you whether something exists or not. Ohloh at least tells you things like how many comments there are, how many contributers there are and how significant their contributions have been, how long it's been under development, etc. Releasing the code's the first step, but not the only step to making code appealing.
Why are 4 player free for alls and tournaments exclusive? One of the more interesting parts, I think, is the strategies that evolve in a 4 player game. It turns out to also be quite suitable for fun play.
I've always considered tactics how and when tactics, and what and where strategy. Who and why is politics ;)
Therefore, diplomacy is more strategy.
That didn't stop Miguel from trying. Or anyone else. The fundamental problem is that Mono is incomplete, and Miguel's efforts to port it really highlighted the problems. Enough that he said he still uses the old unpublished port as a test case for Mono.
I realize that it's been promised since forever, but the good news is that 2.4 is coming out shortly, allowing development to focus on 2.6, which is slated to bring in exactly that longstanding missing feature. Granted, the release could slip, but GIMP seems to be picking up a bit of steam -- GEGL was published in late 2006, and they're also hitting up the UI redesign. Both of these may woo people away from Photoshop. Granted, I doubt GIMP will ever see the massive plugin library that Photoshop has (short of a terrible wrapper plugin), but that's a tall order for any app, open or closed.
I dare say, if your undergraduate course doesn't teach you how the Power rule is derived, you deserve your money back. These are things I was taught in a public high school; I would hope your community college could at least keep pace with that.
Security critical logs where correlations need to be established between program use and even computers shouldn't be using local time. UNIX or UTC timestamps are a better approach, if a bit unreadable. There's already a lot of potential confusion because DST is particular to the region the server is in, the laws the regional government has passed, and the state of the admin's patches.
Have you actually used Debian testing? Or Debian at all? I admit it's a complex beast, and gotten more complex since I left, as this whole article revolves around a misunderstanding between sysadmins and developers. New packages start off in unstable (or experimental if the uploader thinks it's too buggy to consider for regular use), and if, after ten days without a major bug report, the package is placed in testing. Unstable is a rocky ride, but so many people use it that most developers are highly concerned about regressions like tab completion core dumping. So much so that during one cycle the DPL announced an unstable freeze until the new stable was released.
This misunderstanding about timezones is based on where changes go. Security updates to packages in stable go to the "security" repo. Things like clamav definitions change on a regular basis and reside in volatile. This particular repo is news to me, but I don't admin Debian boxes. The developers believe the update should belong in volatile and not security. That is all. Stable remains stable.