The boot.iso image is not part of the Fedora Core distribution images and does therfore not have it's MD5 sum listed. It is instead a part of disc1, since you can find it in the "images" subdirectory on the first disc.
The boot.iso in Fedora is the replacement for the many different boot diskette images that used to ship with distros. Using boot.iso you can perform a fully graphical installation of Fedora using many different sources. For example, you can do a network (FTP, HTTP, NFS..) installation or you can have it use the other Fedora discs (though this is pretty pointless, as disc1 is also a boot disk).
I prefer to simply burn the boot.iso image to a CD-RW disc with each release and doing a FTP installation with it, thus saving me 4 discs. It's also faster in my case than burning the discs, since my network is ridiculously fast:)
Finding good DVD media is a hit and miss on all drives in my experience. That's also why you see so many forums and web pages devoted simply to matching burners with known good media. Different brands work with varying success not only between drives and manufactures but also vary from one firmware version to the next. The trouble compounds when newer firwmare is not always better and may in fact render previously good disks unusable.
Anyway, for the NEC 1300A there exists RPC1 firmware that has been hacked for additional media compatability at http://www.herrie.org/. I've had great success with this, it improved my burners tollerance for different no-name media brands a lot, your mileage may vary. The first thing I do when I get a new drive is always upload a region-free RPC1 firmware in it, even though I have enough dvd-drives in the house to probably dedicate one to each region. Haven't had a single drive fail so far, and the hacked firmware is usually kept up to date based on the latest versions from the manufacturer and is likely to actually fix problems if you have a drive of older make.
Sun's "Java Desktop System" is a GNOME only Linux distro. Then there's some fringe ones, like Gnoppix and Progeny IIRC.
I don't think any distro of importance is GNOME or KDE only at this point, which is good. Hopefully after a few more years of Freedesktop.org cooperation the whole point will be moot and most will be using the best KDE (K3B, Kdevelop etc.) and GNOME (Evolution, Gimp etc.), software on a hardware compositioned X all jimmied together in some kind of nutty bouillabaisse with Mono thrown in for the hell of it.
I'm guessing Novell/SuSE will be one of the first, since they're now such a strange combination of KDE (SuSE) and GNOME (Ximian) people and neither faction is likely to back down.
The GNOME Desktop project has been putting serious effort into making the Linux desktop accessible for a few years now. It has a screen reader and a top notch maginifier. For instance, the magnifier can be made to watch an area of the desktop for change, so that it notifies you of changes in another part of the desktop while you're entering text somewhere else. It also has extensive tracking and cursor presentation options, much more so that its Windows counterpart. Check out some of the newer features in GNOME, it has improved a lot in just a year. There's also a long Accessibility Guide for GNOME, but it's not very good IMO. Still, all of the tools now have decent integrated help in 2.6.
KDE has a similar effort underway, but it's not as complete as GNOME's. In fact, the current roadmap seems to be to also use parts of the GNOME Acccessibility Toolkit (ATK) in KDE. Xzoom isn't the height of accessibility under Linux/X anymore:)
The problem is, there is no (decent) open Java VM, nor would I count on Sun opening theirs any time soon. On the other hand, Mono is evolving very fast..Net too is evolving, IMO it's not hard to see that a few major revisions down the line the.net VM will be a general purpose VM supporting many more programming languages and ideas. In this context, I feel that compiling Java code to Mono bytecode makes a lot more sense than the reverse.
The problem with Mono right now is that part of the.net libraries are patented. If we could just move the Classpath OpenSource Java libraries lock stock and barrel to Mono, get Java support close to 100%, we could use them for the base of the Linux desktop platform without patent liability. And hey, we get full support for C# and the limited support for.net libraries and Linux specific.net extensions such as GTK# thrown in as a bonus. In time, as the VM evolves, more languages like Python and Perl can be supported.
Yes, I know it's possible to compile some other languages to Java bytecode, but the difference IMO is that.net is actively moving towards a general purpose VM while the JCP is not taking this route. In short, Mono has got the VM, Java has got the platform and Open Source can marry the two.
Re:Hopeful about the post-X era
on
XFree86 4.4 Released
·
· Score: 5, Informative
Can we get rid of the X11R6 subdir? (once again, stop thinking X is a world to itself)?
Yes. This is being done for the Freedesktop.org X server. See here for more information. And about time! Another thing is that the horrible imake system is being removed (I personally would like to see it taken out behind the barn and shot). Looks like these guys are serious about dragging X into the 21st century.
If you're not seeing two logical processors in cpuinfo with 2.6, hyperthreading is not enabled for your processor. Most likely this is because you forgot to enable some kernel option, for example, you must enable ACPI device enumeration for HT to work. The kernel must also be SMP capable, even though you have an uniprocessor machine.
This used to be a point in favor of KDE didn't it?:)
Perhaps, but KDE really hasn't radically changed UI or HIG wise since 2.0 came out. What was state of the art back then is not acceptable today, so just by working as it always has done before it has actually become relatively worse. I'm not knocking KDE, they're really on the right track with 3.2, which fixes a lot of the problems I have with it (like the current on-crack context menus). The same can't be said for the "new and improved" Debian installer;)
The menubar isn't OS 9 style. KDE can do an OS9 style menubar up the top, GTK can't.
It used to be possible with a set of patches. I'd guess that they can't be applied anymore, since nobody seems to be interested in them anymore. In my opinion having a global menu becomes a bad idea as the desktop size keeps growing and the monitor resolution improving and you have more and more windows open.
Many sites already require you to use their own download manager applications if you want to download anything large, so requiring BitTorrent isn't so dramatic these days. But what if, instead of every site developing Yet Another Download Manager, they instead brandend their own versions of the BitTorrent client. The clients would be interchangable and the sites wouldn't have to maintain their own programs (I'm sure they've got better things to do).
Using BitTorrent for large files you need to serve seems almost like a no-brainer. The choices are basically: a) provide all the bandwidth for all the downloaders b) provide all the bandwidth in A, and on top of that all of the downloaders would also provide their upstream capability.
Your bandwidth would still be maxed out, but much more content would be distributed in the same amount of time. So what basically changes mod_torrent from regular torrenting is that the seeding would be essentially guaranteed.
What sort of qualifications does Diebold have to be making voting systems? If I as a customer saw these messages, bug rapports and horror stories, I wouldn't trust them to design a cup holder for my car, let alone for something as critical as a voting system.
Here's how you build a real voting system. - You get the best brains to really think about the problem. Forget the Diebold cubicle workers, you get someone like Rivest and pals to design the system. They solve the problems of audit trails, accountability, how to trust the machine etc.
- You get a collaboration of the top research institutes and universities to implement the system. Implementation must be done completely in the open. Every party and faction will have a great interest in eyeballing the system, so that no other faction can exploit it. With enough eyes, every bug is shallow.
- You don't design 52 systems, you design one or two. A well designed system will be used and paid for by virtually all the states. Done right it might cost as much as 30 bad systems, but it'll be worth it.
- You maintain the system troughout the year, not just 2 months before each election. You reuse improved versions of the system with each election.
The new up2date already available in rawhide and to be included in the next beta already includes APT and Yum repository support. The yum tool (very apt-get like) will also be included with the base distribution in addition to up2date.
AFAIK Red Hat will not sell support for the Fedora distribution. If you want support go with the Enterprise products, of which I'm sure we'll see more of in the future.
Re:Good news for KDE users...
on
Aethera 1.0
·
· Score: 2, Interesting
Actually it's against the new GNOME Human Interface Guidelines to name your app gWhatever unless it's an library. In the upcoming 2.4 version a lot of renaming has been done. There really isn't a whole lot of g* apps anymore and IMO all of the really good ones have avoided the stupid naming convention from the start.
Also, an apache mod where you could simply upload the file to your web server and not have to worry about running a bittorrent "seed" would be great
I've been thinking about a project like this for a while. Everyone who wants to help out, please see http://mod-torrent.sourceforge.net/ and get in touch with me.
If the seeding of files can be fully transparent (that's the easy part) and the tracking be made less resource intensive (the hard part) why would a company not want to distribute their own legal content with BitTorrent? Sure, the client must be installed first, but more and more sites are already requiring special download managers. The BitTorrent client is small and simple. It, or something like it, could easily become a standard requirement or the funtionality integrated into existing download mangers.
I have a T3 connection. Some might think that's fast but when you distribute content on even a moderate scale it won't cut it. With BitTorrent I've suddenly got a T3+whatever upload bandwith is not otherwise used by the people downloading from me. If even a couple of college kids with 10Mbit connections in their dorms download from me my effective serving capacity is multiplied. The base service, the T3, remains the same, the added capacity is pure free bandwith. Mini-Akamai networks for everyone!
I think MySQL is more popular because it is easier to set up and start working with. Many people claim that this is not the case and Postgres is just as easy, but these people don't look at it from the point of a newbie.
MySQL installation: Grab the RPM's, rpm -Uvh *, or use InstallShield on Windows. That's it. There is nothing else to set up, there isn't even a default password anymore when connecting from localhost. It's literally a 2 minute process.
Contrast this with PostgreSQL where you got to bootstrap the damn database as the user running the daemon process. Then you've got to set up some users for the database etc. MySQL, being a much simpler system than Postgres, also allows you to do radical things to your databases, you can drop and alter everything, don't worry about details like what indexes you've got, you can slap those on later. Postgres on the other hand enforces some limits, often you have to create a new table and move the data over just to make some bigger changes. Postgres also includes the notion of database maintenance. People who know Postgres might be shocked to learn that DB maintenance is a completely unknown concept for most MySQL users, there's no vacuuming to be done.
Finally, one must not forget the MySQL website, the documentation and even the names of the projects. All these things matter. You might not like it, databases shouldn't be chosen by how their websites look, but the truth is far stranger.
khttpd, the kernel webserver, has been removed in 2.5.x/2.6. I'm guessing NTLM support is part of the new kernel crypto/security API.
Re:It just may make me switch back from IE
on
Netscape 7.1 Released
·
· Score: 3, Informative
View->Show/Hide->Navigation Toolbar, voila, no more address bar. Google also has tons of projects, tips&trick and howtos on tailoring Mozilla especially for a kiosk environment.
The new nVidia installer will first try to locate a binary for your system, and if that fails it will recompile the kernel interface part of the driver. I'm using the driver with the new 2.4.21 kernel, even if nVidia does not "officially" support this version. I've seen it working on a 2.5.x kernel too (though I think it was a bit trickier than just running the installer).
With BitTorrent you start uploading to other people the moment the download begins. The file is split into chunks and downloaded in a completely random order, to avoid for example the end of the file being shared less than the beginning. There is no way to disable the upload and the more you upload the faster you are allowed to download. This is almost invisible to the user. The client looks a bit like the IE default download dialog, not at all like a typical P2P softare.
Let's face it. Most of the suggestions above are useless. Since when is a company going to officially distribute stuff via Kazaa or BitTorrent? Sorry, but when Microsoft says 'To download our latest Service Pack, use Kazaa' then pigs will be flying. It's so unprofessional.
They will do it if saves them money or if the alternative is simply not to host large files. Many sites already allow or even require you to have special download managers (Fileplanet etc.). What if they provide special rebranded versions of the p2p services? Instead of running "BitTorrent" you're using "Symantec AutoUpdater 4". Or some tiny general client could be delivered automatically like Macromedia does with Flash.
Using BitTorrent as the back end just make sense, you have a server serving the files, just like a standard web server (in fact, it could be the web server). That provides you with a base level of service. Then on top of this guaranteed bandwidth you add all the unused bandwith of whoever happens to be downloading at the same time. Suddenly your paltry T1 turns into a T3 and even the bean counters start to take notice. The key is hiding all this P2P stuff from the user, the end user just clicks on links and the download starts.
BitTorrent exists me a lot. I hope to create an apache module for drop-in deployment for torrent hosting some day.
This is because the Java plugin and Mozilla must be compiled with similiar verions of GCC to work (for C++ ABI compatibility). For example, you can't run (Sun's) Java 1.4.1, which is compiled with GCC 2.X, with Red Hat 9's Mozilla version which is build around GCC 3.X and a new glibc. Blackdown makes an Java version for users of newer glibc's and GCC's that had to be used in the past.
Now in Mozilla 1.4 the Linux builds are by default compiled with GCC 3.X so Sun's Java version no longer works. You got to either use Blackdown's 1.4.1, which is stable, or the go with the 1.4.2beta.
You need the following RPMs installed for up2date to work: python gnome-python2-gtkhtml2 gnome-pytho n2-canvas gnome-python2 gnome-python2-bonobo py thon-optik rpm-python libxml2-python
I've you been playing around with --force or --nodeps you might have several conflicting python versions installed. Do a "rpm -qa|grep python", remove the python packages with for example "rpm -e python-2.2.2-26". The version number is given as to remove conflicting packages with the same name. Then install the RPMs mentioned above.
Or you could just get apt4rpm. Using this tool you can do
I agree with most of your other points, though there are sometimes legitimate reasons to use --force (downgrading a package to an older version
Use --oldpackage instead. It's designed for just that.
and upgrading a package with a version that confuses RPM into thinking it's older than an already-installed package
I've encountered this. It's usually the filename of the RPM not matching the actual version number specified in the SPEC file (caused by the creator renaming the package after the build, always a bad idea). Use the --replacepkgs switch, it will allow you to install it over the old without breaking any deps.
--nodeps also has valid uses; if you don't want to pull in a dependency (e.g. a shared library, or application that provides the shared library) to satisfy a small non-critical sub-component of a critical package (and you're prepared to accept that sub-component being broken as a result).
Good point, but in my experience it always never works in practice, unless the sub-component is something separate from the program in question and will not be loaded by default. At that point it usually becomes easier to just rebuild the SRPM with the dependency disabled in the SPEC.
All RPMs created and distributed by Red Hat are signed with their PGP key. Assuming you have added their public key to your keyring (up2date does this for you automatically) it is completely safe to download the updates from the mirror sites.
Python-base is a package for Mandrake only, in RH there exists only two python packages, python and python-devel. You should never install Mandrake specific RPMs on Red Hat. You do not need python-base, I have bittorrent installed (from RPM) and it does not require python-base unless you are installing the Mandrake specific version!
Another tip for keeping your RPM database in good shape: don't ever use --force or --nodeps. If you do, you might as well go ahead and reinstall from scratch, it will come back and bite you later on. The RPM DB does not usually simply corrupt itself, in 90% of the cases, in my experience, it's because someone did a --force at some point, the rest is because of HW problems (sig 11's, flaky memory).
but where is the easy-to-set "we want to print on A4 DAMMIT!!"?
Interestingly, the really new HP LaserJet's have this feature, only slightly renamed;)
From the printer panel you can configure it so that it'll always use A4 size even if the machine doing the printing explicitly requests letter. This makes a lot of sense, I've never seen anybody use letter size (in Europe) but all software will still be configured for it by default. It really shouldn't be the printer's job to force some specific size, but whatever it takes to get rid of "Load letter" is ok in my book.
Slashdot Mirror
The boot.iso image is not part of the Fedora Core distribution images and does therfore not have it's MD5 sum listed. It is instead a part of disc1, since you can find it in the "images" subdirectory on the first disc.
:)
The boot.iso in Fedora is the replacement for the many different boot diskette images that used to ship with distros. Using boot.iso you can perform a fully graphical installation of Fedora using many different sources. For example, you can do a network (FTP, HTTP, NFS..) installation or you can have it use the other Fedora discs (though this is pretty pointless, as disc1 is also a boot disk).
I prefer to simply burn the boot.iso image to a CD-RW disc with each release and doing a FTP installation with it, thus saving me 4 discs. It's also faster in my case than burning the discs, since my network is ridiculously fast
Finding good DVD media is a hit and miss on all drives in my experience. That's also why you see so many forums and web pages devoted simply to matching burners with known good media. Different brands work with varying success not only between drives and manufactures but also vary from one firmware version to the next. The trouble compounds when newer firwmare is not always better and may in fact render previously good disks unusable.
Anyway, for the NEC 1300A there exists RPC1 firmware that has been hacked for additional media compatability at
http://www.herrie.org/. I've had great success with this, it improved my burners tollerance for different no-name media brands a lot, your mileage may vary. The first thing I do when I get a new drive is always upload a region-free RPC1 firmware in it, even though I have enough dvd-drives in the house to probably dedicate one to each region. Haven't had a single drive fail so far, and the hacked firmware is usually kept up to date based on the latest versions from the manufacturer and is likely to actually fix problems if you have a drive of older make.
Sun's "Java Desktop System" is a GNOME only Linux distro. Then there's some fringe ones, like Gnoppix and Progeny IIRC.
I don't think any distro of importance is GNOME or KDE only at this point, which is good. Hopefully after a few more years of Freedesktop.org cooperation the whole point will be moot and most will be using the best KDE (K3B, Kdevelop etc.) and GNOME (Evolution, Gimp etc.), software on a hardware compositioned X all jimmied together in some kind of nutty bouillabaisse with Mono thrown in for the hell of it.
I'm guessing Novell/SuSE will be one of the first, since they're now such a strange combination of KDE (SuSE) and GNOME (Ximian) people and neither faction is likely to back down.
The GNOME Desktop project has been putting serious effort into making the Linux desktop accessible for a few years now. It has a screen reader and a top notch maginifier. For instance, the magnifier can be made to watch an area of the desktop for change, so that it notifies you of changes in another part of the desktop while you're entering text somewhere else. It also has extensive tracking and cursor presentation options, much more so that its Windows counterpart. Check out some of the newer features in GNOME, it has improved a lot in just a year. There's also a long Accessibility Guide for GNOME, but it's not very good IMO. Still, all of the tools now have decent integrated help in 2.6.
:)
KDE has a similar effort underway, but it's not as complete as GNOME's. In fact, the current roadmap seems to be to also use parts of the GNOME Acccessibility Toolkit (ATK) in KDE. Xzoom isn't the height of accessibility under Linux/X anymore
The problem is, there is no (decent) open Java VM, nor would I count on Sun opening theirs any time soon. On the other hand, Mono is evolving very fast. .Net too is evolving, IMO it's not hard to see that a few major revisions down the line the .net VM will be a general purpose VM supporting many more programming languages and ideas. In this context, I feel that compiling Java code to Mono bytecode makes a lot more sense than the reverse.
.net libraries are patented. If we could just move the Classpath OpenSource Java libraries lock stock and barrel to Mono, get Java support close to 100%, we could use them for the base of the Linux desktop platform without patent liability. And hey, we get full support for C# and the limited support for .net libraries and Linux specific .net extensions such as GTK# thrown in as a bonus. In time, as the VM evolves, more languages like Python and Perl can be supported.
.net is actively moving towards a general purpose VM while the JCP is not taking this route. In short, Mono has got the VM, Java has got the platform and Open Source can marry the two.
The problem with Mono right now is that part of the
Yes, I know it's possible to compile some other languages to Java bytecode, but the difference IMO is that
Can we get rid of the X11R6 subdir? (once again, stop thinking X is a world to itself)?
Yes. This is being done for the Freedesktop.org X server. See here for more information. And about time! Another thing is that the horrible imake system is being removed (I personally would like to see it taken out behind the barn and shot). Looks like these guys are serious about dragging X into the 21st century.
If you're not seeing two logical processors in cpuinfo with 2.6, hyperthreading is not enabled for your processor. Most likely this is because you forgot to enable some kernel option, for example, you must enable ACPI device enumeration for HT to work. The kernel must also be SMP capable, even though you have an uniprocessor machine.
This used to be a point in favor of KDE didn't it? :)
;)
Perhaps, but KDE really hasn't radically changed UI or HIG wise since 2.0 came out. What was state of the art back then is not acceptable today, so just by working as it always has done before it has actually become relatively worse. I'm not knocking KDE, they're really on the right track with 3.2, which fixes a lot of the problems I have with it (like the current on-crack context menus). The same can't be said for the "new and improved" Debian installer
The menubar isn't OS 9 style. KDE can do an OS9 style menubar up the top, GTK can't.
It used to be possible with a set of patches. I'd guess that they can't be applied anymore, since nobody seems to be interested in them anymore. In my opinion having a global menu becomes a bad idea as the desktop size keeps growing and the monitor resolution improving and you have more and more windows open.
Many sites already require you to use their own download manager applications if you want to download anything large, so requiring BitTorrent isn't so dramatic these days. But what if, instead of every site developing Yet Another Download Manager, they instead brandend their own versions of the BitTorrent client. The clients would be interchangable and the sites wouldn't have to maintain their own programs (I'm sure they've got better things to do).
Using BitTorrent for large files you need to serve seems almost like a no-brainer. The choices are basically:
a) provide all the bandwidth for all the downloaders
b) provide all the bandwidth in A, and on top of that all of the downloaders would also provide their upstream capability.
Your bandwidth would still be maxed out, but much more content would be distributed in the same amount of time. So what basically changes mod_torrent from regular torrenting is that the seeding would be essentially guaranteed.
What sort of qualifications does Diebold have to be making voting systems? If I as a customer saw these messages, bug rapports and horror stories, I wouldn't trust them to design a cup holder for my car, let alone for something as critical as a voting system.
Here's how you build a real voting system.
- You get the best brains to really think about the problem. Forget the Diebold cubicle workers, you get someone like Rivest and pals to design the system. They solve the problems of audit trails, accountability, how to trust the machine etc.
- You get a collaboration of the top research institutes and universities to implement the system. Implementation must be done completely in the open. Every party and faction will have a great interest in eyeballing the system, so that no other faction can exploit it. With enough eyes, every bug is shallow.
- You don't design 52 systems, you design one or two. A well designed system will be used and paid for by virtually all the states. Done right it might cost as much as 30 bad systems, but it'll be worth it.
- You maintain the system troughout the year, not just 2 months before each election. You reuse improved versions of the system with each election.
The new up2date already available in rawhide and to be included in the next beta already includes APT and Yum repository support. The yum tool (very apt-get like) will also be included with the base distribution in addition to up2date.
AFAIK Red Hat will not sell support for the Fedora distribution. If you want support go with the Enterprise products, of which I'm sure we'll see more of in the future.
Actually it's against the new GNOME Human Interface Guidelines to name your app gWhatever unless it's an library. In the upcoming 2.4 version a lot of renaming has been done. There really isn't a whole lot of g* apps anymore and IMO all of the really good ones have avoided the stupid naming convention from the start.
I've been thinking about a project like this for a while. Everyone who wants to help out, please see http://mod-torrent.sourceforge.net/ and get in touch with me.
If the seeding of files can be fully transparent (that's the easy part) and the tracking be made less resource intensive (the hard part) why would a company not want to distribute their own legal content with BitTorrent? Sure, the client must be installed first, but more and more sites are already requiring special download managers. The BitTorrent client is small and simple. It, or something like it, could easily become a standard requirement or the funtionality integrated into existing download mangers.
I have a T3 connection. Some might think that's fast but when you distribute content on even a moderate scale it won't cut it. With BitTorrent I've suddenly got a T3+whatever upload bandwith is not otherwise used by the people downloading from me. If even a couple of college kids with 10Mbit connections in their dorms download from me my effective serving capacity is multiplied. The base service, the T3, remains the same, the added capacity is pure free bandwith. Mini-Akamai networks for everyone!
I think MySQL is more popular because it is easier to set up and start working with. Many people claim that this is not the case and Postgres is just as easy, but these people don't look at it from the point of a newbie.
MySQL installation: Grab the RPM's, rpm -Uvh *, or use InstallShield on Windows. That's it. There is nothing else to set up, there isn't even a default password anymore when connecting from localhost. It's literally a 2 minute process.
Contrast this with PostgreSQL where you got to bootstrap the damn database as the user running the daemon process. Then you've got to set up some users for the database etc. MySQL, being a much simpler system than Postgres, also allows you to do radical things to your databases, you can drop and alter everything, don't worry about details like what indexes you've got, you can slap those on later. Postgres on the other hand enforces some limits, often you have to create a new table and move the data over just to make some bigger changes. Postgres also includes the notion of database maintenance. People who know Postgres might be shocked to learn that DB maintenance is a completely unknown concept for most MySQL users, there's no vacuuming to be done.
Finally, one must not forget the MySQL website, the documentation and even the names of the projects. All these things matter. You might not like it, databases shouldn't be chosen by how their websites look, but the truth is far stranger.
khttpd, the kernel webserver, has been removed in 2.5.x/2.6. I'm guessing NTLM support is part of the new kernel crypto/security API.
View->Show/Hide->Navigation Toolbar, voila, no more address bar. Google also has tons of projects, tips&trick and howtos on tailoring Mozilla especially for a kiosk environment.
The new nVidia installer will first try to locate a binary for your system, and if that fails it will recompile the kernel interface part of the driver. I'm using the driver with the new 2.4.21 kernel, even if nVidia does not "officially" support this version. I've seen it working on a 2.5.x kernel too (though I think it was a bit trickier than just running the installer).
With BitTorrent you start uploading to other people the moment the download begins. The file is split into chunks and downloaded in a completely random order, to avoid for example the end of the file being shared less than the beginning. There is no way to disable the upload and the more you upload the faster you are allowed to download. This is almost invisible to the user. The client looks a bit like the IE default download dialog, not at all like a typical P2P softare.
They will do it if saves them money or if the alternative is simply not to host large files. Many sites already allow or even require you to have special download managers (Fileplanet etc.). What if they provide special rebranded versions of the p2p services? Instead of running "BitTorrent" you're using "Symantec AutoUpdater 4". Or some tiny general client could be delivered automatically like Macromedia does with Flash.
Using BitTorrent as the back end just make sense, you have a server serving the files, just like a standard web server (in fact, it could be the web server). That provides you with a base level of service. Then on top of this guaranteed bandwidth you add all the unused bandwith of whoever happens to be downloading at the same time. Suddenly your paltry T1 turns into a T3 and even the bean counters start to take notice. The key is hiding all this P2P stuff from the user, the end user just clicks on links and the download starts.
BitTorrent exists me a lot. I hope to create an apache module for drop-in deployment for torrent hosting some day.
This is because the Java plugin and Mozilla must be compiled with similiar verions of GCC to work (for C++ ABI compatibility). For example, you can't run (Sun's) Java 1.4.1, which is compiled with GCC 2.X, with Red Hat 9's Mozilla version which is build around GCC 3.X and a new glibc. Blackdown makes an Java version for users of newer glibc's and GCC's that had to be used in the past.
Now in Mozilla 1.4 the Linux builds are by default compiled with GCC 3.X so Sun's Java version no longer works. You got to either use Blackdown's 1.4.1, which is stable, or the go with the 1.4.2beta.
You need the following RPMs installed for up2date to work:o n2-canvasy thon-optik
python
gnome-python2-gtkhtml2
gnome-pyth
gnome-python2
gnome-python2-bonobo
p
rpm-python
libxml2-python
I've you been playing around with --force or --nodeps you might have several conflicting python versions installed. Do a "rpm -qa|grep python", remove the python packages with for example "rpm -e python-2.2.2-26". The version number is given as to remove conflicting packages with the same name. Then install the RPMs mentioned above.
Or you could just get apt4rpm. Using this tool you can do
apt-get update
apt-get remove python
apt-get remove up2date
apt-get install up2date
and you system should be back to normal. Python is reinstalled along with up2date in that last step.
Use --oldpackage instead. It's designed for just that.
I've encountered this. It's usually the filename of the RPM not matching the actual version number specified in the SPEC file (caused by the creator renaming the package after the build, always a bad idea). Use the --replacepkgs switch, it will allow you to install it over the old without breaking any deps.
Good point, but in my experience it always never works in practice, unless the sub-component is something separate from the program in question and will not be loaded by default. At that point it usually becomes easier to just rebuild the SRPM with the dependency disabled in the SPEC.
All RPMs created and distributed by Red Hat are signed with their PGP key. Assuming you have added their public key to your keyring (up2date does this for you automatically) it is completely safe to download the updates from the mirror sites.
Python-base is a package for Mandrake only, in RH there exists only two python packages, python and python-devel. You should never install Mandrake specific RPMs on Red Hat. You do not need python-base, I have bittorrent installed (from RPM) and it does not require python-base unless you are installing the Mandrake specific version!
Another tip for keeping your RPM database in good shape: don't ever use --force or --nodeps. If you do, you might as well go ahead and reinstall from scratch, it will come back and bite you later on. The RPM DB does not usually simply corrupt itself, in 90% of the cases, in my experience, it's because someone did a --force at some point, the rest is because of HW problems (sig 11's, flaky memory).
Interestingly, the really new HP LaserJet's have this feature, only slightly renamed
From the printer panel you can configure it so that it'll always use A4 size even if the machine doing the printing explicitly requests letter. This makes a lot of sense, I've never seen anybody use letter size (in Europe) but all software will still be configured for it by default. It really shouldn't be the printer's job to force some specific size, but whatever it takes to get rid of "Load letter" is ok in my book.