PSXen don't fail as soon as they get some kind of error reading the CD, they just try again ad infinitum. As such, you can usually even remove the CD and put it back in while the game is going on and it will just pause while it is out. Besides, the drives are only 2x (are a PSOne's faster?, so it would take no time at all to spin back up.
Upgrade a system library in what way? If a library is upgraded to a newly-incompatible API/ABI, the version number will be bumped so everything which used to use it will remain doing so until it is recompiled (and possibly adapted if it used an API in a deprecated way). APIs don't usually get deprecated, but ABIs changing isn't unheard of.
Many in the FreeBSD community, including myself and the people that helped me way back when, are more than willing to help newbies with installation with the expectation after that that the new user will have enthusiasm to learn, ask good questions, and know the system well with the documentation that's also already there.
Or commit it themselves. There is no need to get approval from Core to commit new drivers to the system. If you can convince a developer of the quality of your driver there's no reason they can't decide to commit it themselves.
Falling off? Not gaining in number? Look at the CVS logs of, for example, our CVSROOT/access. You'll see what things are really like.
Wanting to code "a driver" doesn't help anybody. Having a piece of hardware you'd specifically like being supported, finding all the info you can on it and possible leads for support of something like it already in the system, and then investigating leads in the form of what everyone else thinks would be the best course of action to take would be.
It's not easy for someone without experience to code a driver from scratch. Then again, you may get that impression from looking at the sheer number of drivers existant in Linux, and not at their quality, and wondering why it's so much easier. FreeBSD, in specific, has a higher expectation of quality in drivers; if the device has a DMA mode, a driver with only PIO support is laughable. If it only "kinda-works" it's probably not going in.
The release isn't waiting for new drivers much if at all. The release is waiting for large API changes that affect every part of the kernel and have widespread ramifications through the entire system -- case in point, transition from processes to threads, processes, and KSEs.
If you want more specific examples, something quite common is someone wanting to support a new $10 ethernet card. We already have a developer that given the documents can crank them out faster and better than anyone else has proven possible.
When you luck out is when someone appears out of the woodwork, has been doing his/her homework, and pops up with a genuinely useful working driver because of having done the most important thing possible, going off examples and getting experience.
Asking vaguely "How do I write a driver?" isn't going to help anyone, but providing everything you know, what you want to do, and where you'd like to go next is both expected and welcome.
You raise some good points. Thanks for pointing out that this isn't snobbery, just expecting a reasonable amount of effort up front from someone expecting to get lots of other people's time (the most valuable asset one has) for free.
Also, FreeBSD has APIs that are definitely sometimes harder to learn in the first place. That's quite often due to them being designed properly and being able to do so much more. Would you like to compare the VFS of Linux to that of FreeBSD? One has definitely an "easier" bar of entry, whereas another definitely has a hell of a lot more functionality.
I quite _did_ mean that I run, and develop on, -CURRENT, and it is in a definite constant state of flux. There are many huge architectural design changes. Go read the -CURRENT mailing list for examples.
Do you know how to write modern FreeBSD device drivers? Have you tried before? You must not know the many, many hours of work it takes for someone to learn the APIs, even if it's someone else spending HIS own hours teaching him.
It's, quite frankly, more than a little bit easier for existing FreeBSD developers to write a given driver than it is for them to teach someone what to do. Writing drivers involves intimate familiarity with the system, especially with a system where the kernel API has been in a constant state of flux in the long-running development branch.
That said, intelligent questions about an arbitrary topic with non-obvious answers are USUALLY responded to politely. You can't just say, though, "I'd like to write some drivers. Can you tell me how?" or anything even moderately like that.
Actually, though the web server seems to be getting hit pretty hard, and the JavaScript screwed up in Mozilla and Konqueror (I had to use Opera to use the site), I just watched the movie with nary a hitch once I found the actual address
for the RealPlayer stream.
Actually, our favorite branch of our favorite government agency (DARPA, DoD) is funding a lot of work for both. For instance, CBOSS is a contract that NAI Labs recently won to start funding such things as SELinux and LOMAC.
For what it's worth, LOMAC is an example of a project currently underway andbeing developed for Linux and FreeBSD both, so it is not only Linux that is getting security projecs funded for it (^_^)
Disclaimer: I am an employee of NAI Labs, not that it makes this information less relevant.
I remember that game! It was fun, playing a futuristic 3D shooter back then with Sound Blaster (8-bit) compatible cards on 486 DXs, in full 256-color glory. Come to think of it, I remember all those games. I'm going to miss Dynamix, an incredible source of creativity in the gaming world for many years.
If you're going to be a grammar pedant, be pantically correct. "I" would be proper at the end of that sentence because it preceded an understood verb, and therefore was a subject pronoun and not an object pronoun.
--
Re:Just One Little Problem - I Can't Find It
on
FreeBSD 4.3 Released
·
· Score: 1
Actually, the policy nowadays is that the CD image for releases is concurrently made available in the normal places. It wasn't even until recently that the disc images were made available at all, and even more recently that made available by all mirrors, of course... but the way things work now most definitely is that the disc image is treated as just as important as the ability to do a net-install and download parts separately.
The previous/usual primary site, ftp.freesoftware.com, is what ftp.freebsd.org usually points to. However, for several days now, lightning.net has let connectivity to that point stay down. It's pretty bad not to have the usual primary mirror at release time, but pointing ftp.freebsd.org to another powerful server seems to be working:)
Here's to hoping that lightning.net restores connectivity to ftp.freesoftware.come soon.
The major reason is because supporting PAM is just not easy! It's just peachy if you are designing your apps around PAM's architecture, but to convert all current authentication-related programs to PAM's architecture will take time. Who does the work? The people who have both the time and necessary motivation! Were someone to care enough, they'd modify things to their ends and then probably submit the changes.
As far as OpenSSH itself supporting PAM, I have just MFCed a new version (2.3.0 + FreeBSD changes) to 4.2-STABLE after sitting in 5.0-CURRENT for a few weeks. This version includes PAM support for password authentication from OpenSSH "for Unix", and additional PAM support for TIS as well as integration were done by Eivind Eklund.
I imagine Linux and Solaris support PAM better mostly as a result of there being actual versions of PAM "for Linux" and "for Solaris", likely with PAM support having been done for GNU utilities by the PAM authors (you need something to support PAM before you can actually use it!), and the motivation of PAM for Solaris would probably be that someone was paid to do it.
I don't see where the misleading is, though. FreeBSD doesn't advertise that the entire system uses PAM, because it doesn't. If you are a sysadmin, you need to know your daemons (by reading the documentation, etc.) and of course test things. There is no "pretending" to support PAM; it is in the base system and used where it has been made to work. Patches for more complete PAM "support" are welcome from anyone who has the motivation to do the work!
What about various problems with DGA wrt keyboard input, as well? Last I recall, there were large problems that made the DGA2 unusable with emulators (snes9x, xmame). I don't want to give those up, and giving up full-screen VMware would be horrible, too.
Any idea if problems related to DGA and proper use of modifier keys have been fixed?
If you're using an unmodified program in its full, executable form, and there are no modifications to it, it is in full compliance with the GPL. It seems that many people like you to try to fault companies if they use free software in some large application.
Disclaimer: I work in a very small company -- only a few people. I'm not expressing a capitalistic viewpoint, but a realistic one: don't judge a user of software by whether they make money from using it.
Slashdot Mirror
PSXen don't fail as soon as they get some kind of error reading the CD, they just try again ad infinitum. As such, you can usually even remove the CD and put it back in while the game is going on and it will just pause while it is out. Besides, the drives are only 2x (are a PSOne's faster?, so it would take no time at all to spin back up.
Upgrade a system library in what way? If a library is upgraded to a newly-incompatible API/ABI, the version number will be bumped so everything which used to use it will remain doing so until it is recompiled (and possibly adapted if it used an API in a deprecated way). APIs don't usually get deprecated, but ABIs changing isn't unheard of.
And we should be comfortable with you equating Islam and support for terrorism?
Please stop trolling.
You know my dad can beat up your dad, and you just won't admit it.
Many in the FreeBSD community, including myself and the people that helped me way back when, are more than willing to help newbies with installation with the expectation after that that the new user will have enthusiasm to learn, ask good questions, and know the system well with the documentation that's also already there.
Or commit it themselves. There is no need to get approval from Core to commit new drivers to the system. If you can convince a developer of the quality of your driver there's no reason they can't decide to commit it themselves.
Falling off? Not gaining in number? Look at the CVS logs of, for example, our CVSROOT/access. You'll see what things are really like.
Wanting to code "a driver" doesn't help anybody. Having a piece of hardware you'd specifically like being supported, finding all the info you can on it and possible leads for support of something like it already in the system, and then investigating leads in the form of what everyone else thinks would be the best course of action to take would be.
It's not easy for someone without experience to code a driver from scratch. Then again, you may get that impression from looking at the sheer number of drivers existant in Linux, and not at their quality, and wondering why it's so much easier. FreeBSD, in specific, has a higher expectation of quality in drivers; if the device has a DMA mode, a driver with only PIO support is laughable. If it only "kinda-works" it's probably not going in.
... which is being done by a separate, none-FreeBSD project such as this one.
The release isn't waiting for new drivers much if at all. The release is waiting for large API changes that affect every part of the kernel and have widespread ramifications through the entire system -- case in point, transition from processes to threads, processes, and KSEs.
If you want more specific examples, something quite common is someone wanting to support a new $10 ethernet card. We already have a developer that given the documents can crank them out faster and better than anyone else has proven possible.
When you luck out is when someone appears out of the woodwork, has been doing his/her homework, and pops up with a genuinely useful working driver because of having done the most important thing possible, going off examples and getting experience.
Asking vaguely "How do I write a driver?" isn't going to help anyone, but providing everything you know, what you want to do, and where you'd like to go next is both expected and welcome.
You raise some good points. Thanks for pointing out that this isn't snobbery, just expecting a reasonable amount of effort up front from someone expecting to get lots of other people's time (the most valuable asset one has) for free.
Also, FreeBSD has APIs that are definitely sometimes harder to learn in the first place. That's quite often due to them being designed properly and being able to do so much more. Would you like to compare the VFS of Linux to that of FreeBSD? One has definitely an "easier" bar of entry, whereas another definitely has a hell of a lot more functionality.
I quite _did_ mean that I run, and develop on, -CURRENT, and it is in a definite constant state of flux. There are many huge architectural design changes. Go read the -CURRENT mailing list for examples.
Pre-10,000? I fart in your general direction!
You mean it's not because there can never be another office suite called "Office" since M$ would sue the hell out of anyone that tried?
Do you know how to write modern FreeBSD device drivers? Have you tried before? You must not know the many, many hours of work it takes for someone to learn the APIs, even if it's someone else spending HIS own hours teaching him.
It's, quite frankly, more than a little bit easier for existing FreeBSD developers to write a given driver than it is for them to teach someone what to do. Writing drivers involves intimate familiarity with the system, especially with a system where the kernel API has been in a constant state of flux in the long-running development branch.
That said, intelligent questions about an arbitrary topic with non-obvious answers are USUALLY responded to politely. You can't just say, though, "I'd like to write some drivers. Can you tell me how?" or anything even moderately like that.
Actually, though the web server seems to be getting hit pretty hard, and the JavaScript screwed up in Mozilla and Konqueror (I had to use Opera to use the site), I just watched the movie with nary a hitch once I found the actual address for the RealPlayer stream.
For what it's worth, LOMAC is an example of a project currently underway andbeing developed for Linux and FreeBSD both, so it is not only Linux that is getting security projecs funded for it (^_^)
Disclaimer: I am an employee of NAI Labs, not that it makes this information less relevant.
What's so special about paying money to buy off-the-shelf hardward and then installing software other people made, anyway?
I remember that game! It was fun, playing a futuristic 3D shooter back then with Sound Blaster (8-bit) compatible cards on 486 DXs, in full 256-color glory. Come to think of it, I remember all those games. I'm going to miss Dynamix, an incredible source of creativity in the gaming world for many years.
--
--
Here's to hoping that lightning.net restores connectivity to ftp.freesoftware.come soon.
--
--
As far as OpenSSH itself supporting PAM, I have just MFCed a new version (2.3.0 + FreeBSD changes) to 4.2-STABLE after sitting in 5.0-CURRENT for a few weeks. This version includes PAM support for password authentication from OpenSSH "for Unix", and additional PAM support for TIS as well as integration were done by Eivind Eklund.
I imagine Linux and Solaris support PAM better mostly as a result of there being actual versions of PAM "for Linux" and "for Solaris", likely with PAM support having been done for GNU utilities by the PAM authors (you need something to support PAM before you can actually use it!), and the motivation of PAM for Solaris would probably be that someone was paid to do it.
I don't see where the misleading is, though. FreeBSD doesn't advertise that the entire system uses PAM, because it doesn't. If you are a sysadmin, you need to know your daemons (by reading the documentation, etc.) and of course test things. There is no "pretending" to support PAM; it is in the base system and used where it has been made to work. Patches for more complete PAM "support" are welcome from anyone who has the motivation to do the work!
--
Any idea if problems related to DGA and proper use of modifier keys have been fixed?
--
Disclaimer: I work in a very small company -- only a few people. I'm not expressing a capitalistic viewpoint, but a realistic one: don't judge a user of software by whether they make money from using it.
--