You don't. When you tell your network stream instance to interpret the next four bytes as integers (as in stream >> myInt32), the function responsible for that is perfectly capable of swapping the bytes for you. If you have to deal with endianness directly, you're definitely not using enough abstraction.
C++ is actually cross-platform, if you only use portable libraries - which brings us to the original poster's question. It sounds to me like he might want to use Qt - since 4.5 it's been LGPLed, and it has socket and threading abstractions, though I don't know if it has IPC functionality. If ACE has a sane API, then there's no need to look for something newer, but if you want to have a GUI for your software, you'll want to use Qt anyway.
Perhaps it was vindictive of me to call the previous hype a sham, but Netgear was trying to frame their router as the router to get for running third party firmware on, when in reality it is a poor choice in comparison to various others that are both cheaper and more well supported by third party firmware. Their own summary seemed deliberately misleading to me (it was certainly misleading, deliberate or not), so I wasn't feeling very charitable when I wrote this one.
My first question upon seeing this article was actually whether it uses Broadcom hardware. It does. Even the ethernet driver is closed source, let alone the wifi, according to the documentation from Netgear, except that instead of closed source or proprietary, they call it "precompiled". I'm disappointed, and given this, I think I might as well get the hardware from any vendor, because one can't count on the ability to run newer kernels on hardware with so many closed source drivers.
There are many comments posted before yours that recommend optimistic locking, which is better than the pessimistic locking you just suggested. I also suggest that to be complete, an application should also attempt to merge the new version of the data with the change that the user is trying to make. For examples of merge algorithm implementations, the developer should look at a distributed version control system, like Git, Mercurial, or Bazaar.
It was also not an easy "drop in" scheme, and required extensive modification of the source code for each web page / database table that we wanted to add it to.
I don't envy the crazy amounts of code duplication that you clearly had to deal with on a regular basis.
Agreed. I think that regardless of which interface is used for an app, the most important thing is making sure that the categories are unambiguous, so that the first category someone chooses to look at is the right one. Also, it's not like they would completely stop supporting the classic UI, so there is little to complain about. Assuming they go through with this, and not just in their Windows releases, I'd look forward to having the choice of trying the ribbon UI on an application I actually use. I resent Microsoft at least as much as the next guy here, but the ribbon UI is merely a new idea - it shouldn't matter where it came from, unless they've patented it.
Mod parent up. While I agree that there are plenty of open problems with education systems around the world, learning takes hard work no matter what. We should be looking for ways to make that work more efficient and palatable, e.g. flashcards vs. reading a dictionary front to back, or edutainment (when properly designed), like Mavis Beacon Teaches Typing.
In the case of the PS3 Slim though, it could just be that the PS3 had to be pushed out to meet its launch window, and that the Slim is what Sony was going for in the first place
One doesn't even have to have a PS3 to remember how long Sony delayed the PS3 at launch because it was waiting for enough supply of the BluRay drives. Also, lets not forget that the PS3 Slim is being released over two and a half years after the original PS3.
You wouldn't, I certainly don't. The point is that a developer is free to port something like that with little effort (keeping in mind that most apps like pidgin won't happen to be included in the base system), and then publish it in a form that can be installed by even novice users of the device, without any arcane tweaking as a prerequisite.
Sure, it runs just fine, but you have to go through hoops to run any such code on the Pre. Maemo allows any user to run native applications in whatever language, and installation is easy. See this for an example. If you have to wrestle with the platform in any way to run an app like pidgin, then said platform isn't a "normal Linux system" to any extent that would be considered exciting.
Go to the parent of your comment, and then go to the parent of that comment. At that point, hopefully you will realize that your comment was a mistake.
Also, being a "real distro" has everything to do with what language the apps are written in. In a real distribution (i.e. a real general purpose Linux distribution), you can choose whatever language you want, as long as the footprint isn't too big for the hardware. If a system doesn't offer this choice, then it isn't really a Linux distribution, not in the sense that people care about.
Firstly, if it only runs javascript applications as you say, then it's hardly any more of a "real distro" than Android is.
Why?
What's the difference between, Javascript as the language and HTML/CSS as the GUI, or using Python as the language and XML to do the GUI? And yet the first is "not a real distro", while the second is. Why?
Nobody is putting words in your mouth, you really do imply this above, by questioning why the Pre distribution shouldn't be considered a "real distro". The Pre distribution isn't a real Linux distribution because you are unable to run applications that are developed to run on the Linux platform. Instead, you are forced to use the web-based platform that Palm has chosen to be the only way to develop apps for the Pre. By suggesting that the Pre distribution should be considered real, you are implying that it's acceptable to require all apps to be written using Javascript exclusively for the logic.
I don't think you implied this intentionally. Rather, it looks like you simply have poor reading comprehension, at least in the context of this conversation. If you don't believe me, ask yourself why multiple Slashdotters are arguing with you about a couple of comments of yours.
This isn't really insightful so much as misinformed. I recently found the DownloadHelper extension for Firefox, which makes it easy to grab flash videos from websites in general. Using it reveals that when applicable, youtube actually provides medium quality 1280*720 video using the H.264 codec. I've only seen one video that is provided in this format - the bitrate for that one is around 240kbps, based on the file size minus the size of the audio data.
What notification do people expect? When you restart firefox, it will probably tell you that a new addon has been installed, at which point it's trivial to disable it. Also, this addon has so far only been installed on systems running Ubuntu 9.10, which hasn't been released yet, so it remains to be seen what notification they would provide if they decide to deploy these changes to a stable release.
There's nothing that suggests these problems are caused by release compatibility issues. This does suggest that you are groping for reasons to justify your choice of distribution, however.
I would have modded this troll or something, but it's at a 3 right now so I'm going to chime in. Defending someone takes time and effort, and perhaps NewYorkCountryLawyer can't be everywhere at once, seeing as how he is already defending other cases.
Also, the legal defense in this case tried to argue that exchanging copyrighted content was fair use and thus not an infringement, which is clearly a crazy defense given the courts' tolerance of RIAA antics in the past. He then got his client to admit his guilt on the stand, and is now giving up on the case. As previous replies have stated, this is incompetent.
And that's just CVS--I wonder what would happen if they encountered GIT or mercurial...
I think it's a misconception that somehow, because CVS (or SVN) is less flexible than git, that it's easier to learn. It is true that there is some added complexity, but it can be summed up as follows:
-a is needed for git commit, to tell it to automatically include all differences between tracked files in the working copy and the last commit (I rarely use this myself, it's better to instead make sure that nothing is accidentally included in the commit by learning how to use git add)
git push is needed in order to push commits back to a central repository
git rebase origin/master master is needed to deal with concurrency, since you might have multiple commits that you haven't pushed out to the remote repository
I think the added learning curve here is far less harmful than not having atomic commits, as with cvs, or not being able to edit and manage branches and commits without publishing them to the central server, or not being able to rebase branches on the latest code before publishing them. Merging in the usual way (including git merge) is harmful, because it becomes much harder to track history, but there's no alternative for this in cvs and svn.
Don't get me wrong, I respect Richard Stallman's ideals and achievements, but there comes a point when Linux should simply be called Linux and not GNU/Linux. Chances are that a system like Splashtop is past that point, since it would probably have few, if any GNU packages included, given that it's based on busybox.
Setting up a git repository requires typing git init, git add . (or git add on whatever files you want to track), and git commit.
Perhaps you mean publishing a git repository, which requires ssh and git on the server side for write access, and either http or git-daemon for read access. Is bzr easier than that?
cvs and svn are obsolete, especially CVS, which is utterly painful and missing many features that I consider basic and must-have, like the ability to check the status of your working tree easily, or only commit part of your changes. You should think about moving to git.
Slashdot Mirror
You don't. When you tell your network stream instance to interpret the next four bytes as integers (as in stream >> myInt32), the function responsible for that is perfectly capable of swapping the bytes for you. If you have to deal with endianness directly, you're definitely not using enough abstraction.
C++ is actually cross-platform, if you only use portable libraries - which brings us to the original poster's question. It sounds to me like he might want to use Qt - since 4.5 it's been LGPLed, and it has socket and threading abstractions, though I don't know if it has IPC functionality. If ACE has a sane API, then there's no need to look for something newer, but if you want to have a GUI for your software, you'll want to use Qt anyway.
Perhaps it was vindictive of me to call the previous hype a sham, but Netgear was trying to frame their router as the router to get for running third party firmware on, when in reality it is a poor choice in comparison to various others that are both cheaper and more well supported by third party firmware. Their own summary seemed deliberately misleading to me (it was certainly misleading, deliberate or not), so I wasn't feeling very charitable when I wrote this one.
Yay for slashvertisements that don't live up to their own hype.
My first question upon seeing this article was actually whether it uses Broadcom hardware. It does. Even the ethernet driver is closed source, let alone the wifi, according to the documentation from Netgear, except that instead of closed source or proprietary, they call it "precompiled". I'm disappointed, and given this, I think I might as well get the hardware from any vendor, because one can't count on the ability to run newer kernels on hardware with so many closed source drivers.
There are many comments posted before yours that recommend optimistic locking, which is better than the pessimistic locking you just suggested. I also suggest that to be complete, an application should also attempt to merge the new version of the data with the change that the user is trying to make. For examples of merge algorithm implementations, the developer should look at a distributed version control system, like Git, Mercurial, or Bazaar.
I don't envy the crazy amounts of code duplication that you clearly had to deal with on a regular basis.
Agreed. I think that regardless of which interface is used for an app, the most important thing is making sure that the categories are unambiguous, so that the first category someone chooses to look at is the right one. Also, it's not like they would completely stop supporting the classic UI, so there is little to complain about. Assuming they go through with this, and not just in their Windows releases, I'd look forward to having the choice of trying the ribbon UI on an application I actually use. I resent Microsoft at least as much as the next guy here, but the ribbon UI is merely a new idea - it shouldn't matter where it came from, unless they've patented it.
Mod parent up. While I agree that there are plenty of open problems with education systems around the world, learning takes hard work no matter what. We should be looking for ways to make that work more efficient and palatable, e.g. flashcards vs. reading a dictionary front to back, or edutainment (when properly designed), like Mavis Beacon Teaches Typing.
One doesn't even have to have a PS3 to remember how long Sony delayed the PS3 at launch because it was waiting for enough supply of the BluRay drives. Also, lets not forget that the PS3 Slim is being released over two and a half years after the original PS3.
You wouldn't, I certainly don't. The point is that a developer is free to port something like that with little effort (keeping in mind that most apps like pidgin won't happen to be included in the base system), and then publish it in a form that can be installed by even novice users of the device, without any arcane tweaking as a prerequisite.
Sure, it runs just fine, but you have to go through hoops to run any such code on the Pre. Maemo allows any user to run native applications in whatever language, and installation is easy. See this for an example. If you have to wrestle with the platform in any way to run an app like pidgin, then said platform isn't a "normal Linux system" to any extent that would be considered exciting.
Go to the parent of your comment, and then go to the parent of that comment. At that point, hopefully you will realize that your comment was a mistake.
Also, being a "real distro" has everything to do with what language the apps are written in. In a real distribution (i.e. a real general purpose Linux distribution), you can choose whatever language you want, as long as the footprint isn't too big for the hardware. If a system doesn't offer this choice, then it isn't really a Linux distribution, not in the sense that people care about.
Nobody is putting words in your mouth, you really do imply this above, by questioning why the Pre distribution shouldn't be considered a "real distro". The Pre distribution isn't a real Linux distribution because you are unable to run applications that are developed to run on the Linux platform. Instead, you are forced to use the web-based platform that Palm has chosen to be the only way to develop apps for the Pre. By suggesting that the Pre distribution should be considered real, you are implying that it's acceptable to require all apps to be written using Javascript exclusively for the logic.
I don't think you implied this intentionally. Rather, it looks like you simply have poor reading comprehension, at least in the context of this conversation. If you don't believe me, ask yourself why multiple Slashdotters are arguing with you about a couple of comments of yours.
Hmm, where have I seen that logo?
This isn't really insightful so much as misinformed. I recently found the DownloadHelper extension for Firefox, which makes it easy to grab flash videos from websites in general. Using it reveals that when applicable, youtube actually provides medium quality 1280*720 video using the H.264 codec. I've only seen one video that is provided in this format - the bitrate for that one is around 240kbps, based on the file size minus the size of the audio data.
What notification do people expect? When you restart firefox, it will probably tell you that a new addon has been installed, at which point it's trivial to disable it. Also, this addon has so far only been installed on systems running Ubuntu 9.10, which hasn't been released yet, so it remains to be seen what notification they would provide if they decide to deploy these changes to a stable release.
There's nothing that suggests these problems are caused by release compatibility issues. This does suggest that you are groping for reasons to justify your choice of distribution, however.
It really depends on what was done to lose the case.
I would have modded this troll or something, but it's at a 3 right now so I'm going to chime in. Defending someone takes time and effort, and perhaps NewYorkCountryLawyer can't be everywhere at once, seeing as how he is already defending other cases.
Also, the legal defense in this case tried to argue that exchanging copyrighted content was fair use and thus not an infringement, which is clearly a crazy defense given the courts' tolerance of RIAA antics in the past. He then got his client to admit his guilt on the stand, and is now giving up on the case. As previous replies have stated, this is incompetent.
I think it's a misconception that somehow, because CVS (or SVN) is less flexible than git, that it's easier to learn. It is true that there is some added complexity, but it can be summed up as follows:
I think the added learning curve here is far less harmful than not having atomic commits, as with cvs, or not being able to edit and manage branches and commits without publishing them to the central server, or not being able to rebase branches on the latest code before publishing them. Merging in the usual way (including git merge) is harmful, because it becomes much harder to track history, but there's no alternative for this in cvs and svn.
Why would you switch over it? Most development is unaffected by these petty squabbles.
Why would it include gcc? Splashtop clearly wasn't designed to be a free Unix.
Don't get me wrong, I respect Richard Stallman's ideals and achievements, but there comes a point when Linux should simply be called Linux and not GNU/Linux. Chances are that a system like Splashtop is past that point, since it would probably have few, if any GNU packages included, given that it's based on busybox.
Setting up a git repository requires typing git init, git add . (or git add on whatever files you want to track), and git commit. Perhaps you mean publishing a git repository, which requires ssh and git on the server side for write access, and either http or git-daemon for read access. Is bzr easier than that?
cvs and svn are obsolete, especially CVS, which is utterly painful and missing many features that I consider basic and must-have, like the ability to check the status of your working tree easily, or only commit part of your changes. You should think about moving to git.