Well, that would add to the number of components that could fail, and require a high speed bus between the two controllers, as well as a shared cache and all the headaches that would bring with it (think SMP caches being ping-ponged). Then you've got to sync your interface to the system bus as well as the new internal buses. On the other hand, you can just crank the knob up to 11 and go 20K RPMs on known, tried and true, technology.
If someone decides that you're a trouble maker, or whatever, try to get your patch in a different branch of the tree. The kernel has a long tail, and all patches go upstream to Andrew Mortons tree and then hop to Linus' tree, provided that the code is decent and it serves a genuine purpose (A guy from Google has two spelling corrections that were merged and he got credited for).
You'll get yelled at for formatting and such before anyone of higher authority sees your code. You probably won't get your code past a subsystem maintainer without having it look presentable. It's just about getting your foot in the door at any node along the tail. I mean, even Namesys got code in the kernel with Hans shouting at everyone. The protocol for doing things is a bit flexible (I hear Andrew Morton still submits changes to Linus as tar balls from time to time... I'm not sure if he does this any more, though) if you stick to the key concepts.
Everyone becomes conservative with upgrades after the first time that a box 3000 miles away fails to come back up. Seriously, waiting for a remote reboot after a kernel update is always the longest two minutes of my life.
Even the headless boxes at my apartment wait for me to set aside time to haul out a monitor and keyboard if anything goes wrong during an update. It's better to assume that something will go wrong and be pleasantly surprised and ahead of schedule than to sit staring at pings that have been timing out for the last five minutes (while you think, maybe it's just taking a long time to init... yeah, right!).
And, regardless of what anyone says, a virtual machine test environment doesn't have anywhere near the complications that you get with heavy metal. A successful virtual machine test just means that nothing is assured to go wrong, nothing more.
No, the problem is that outside of their company, no one has any idea of how the thing works. They can't bring in a consultant, and there isn't anyone to turn to for advice. Even if they hired a small team of talented professionals to help them, that team would still need plenty of time to get up to speed on the whole system.
Just wondering if, by your own logic, you would say that string theory is a science or not. For that matter, how far in to physics would you assert your position?
Right. Just remember that the guy with his job on the line is never the one who made the poor decision. His boss did after the poor guy argued against it for months only to be forced to implement said Bad Idea.
How are you using vgetty? Are you using a hardware DAC?
From what I understood, Asterisk can't use regular PCI modems because almost all of them are winmodems. Meaning, they don't have a full firmware stack, but rely on their drivers to supply certain functionality. Namely, proprietary, closed source, non-free (beer or speech) Windows only drivers.
I've got a box with 5 PCI slots and as many old V.92's lying around that I'd love to get working on a Slackware build, if you know something I don't (not being sarcastic, I'd be really be thrilled if I didn't research far enough to find the answer).
You know Likewises' primary developer is Gerry Carter of the Samba project, as well as the author of OReilly's LDAP Administration, right?.
It's just like buying Red Hat support; you get the backing of a company that employs the people who are developers for that project. With Red Hat you get a bunch of kernel developers and Andrew Barlette (another key Samba developer). You can't get better support for your money than support from key developers. Also, it enables the developers to work on open source projects as a day job, too.
Eclipse. Every summer they do a multi project release on the same day. This year the release was called Ganymede. The release announcement follows (emphasis added):
The Eclipse Foundation and the entire Eclipse community are pleased to announce the availability of the Ganymede Release, the annual release train developed by the Eclipse community. The Ganymede Release is a coordinated release of 23 different Eclipse project teams that represents over 18 million lines of code.
You could also use S3Backer with an rsync script (or rsnapshot) on the host. That lets you mount the S3 bucket as a drive on your server through FUSE and then copy to it as if it were local. *NIX/BSD only, though.
The issue is more general than that. The usability experts who can't code *can* improve usability, by telling the developers what to do.
It's more of an issue of *just* criticising, instead of offering constructive opinions. If all you can do is say "this sucks" but not say why or how it can be improved, then I agree, you have no business in software. If, on the other hand, you can find a way to improve it, I'm sure many people will welcome your advice and implement it.
RE: Bugfix, Usability is a matter of opinion
*WONTFIX*
Usage works for me.
*Ducks and runs*
Note to self: fork codebase. Send snarky message at 1.0 release.;)
No doubt... I don't care who they are, low level hardware fixes against an OS in a week is impressive. I think Linus' window on RC's for fixing this stuff is two weeks, and the kernel team has been moving at breakneck speeds lately (averaging 4 LOC/hour, every hour, every day).
Have you looked in to Ruckus. It's DRM'ed and only runs on windows media player, but it's free for students with an.edu email account if the school signs up for it. Our technet at Kutztown University, Pennsylvania, has a huge banner on their home page saying something to the effect of "stop infesting the campus with your MP3s from torrents, we've signed you up for free music from this service". It seems to be working somewhat. Basically, Ruckus puts a skin on WMP and shows little adds in the bottom of the player, and that ad revenue supports all the music you can download.
Their network is always fast, and I think that if you setup some kind of caching server, you can cut down on bandwidth as music is mostly 'trendy' in the way that youTube links are; someone finds something and tells everyone else to check it out, and then you've got your pipe saturated with fifty copies of the same media.
At any rate, while I don't like DRM, it is a solution that at least gives you the ability to cut down on traffic externally, and internally (from infected machines reinfecting each other - I've seen the network room, most of the rack is lit up like a Christmas tree, solid green, even with hardware shapers and balancers over multiple backbones. They've even got IR on the roof as a failover... it's nuts what morons will do to your network, even when they've got $100,000 of hardware protecting them at the edge).
Ruckus is worth looking in to, IMHO, even if to compare what's on the market or solidify your choice for something else.
This is kinda' like a patch panel, I take it. Have a junction point for distribution of media and have it in place so that you can light up paths as you need them. Good idea, so long as no one gets the bright idea of proprietary junction boxes or something.
I know I wouldn't mind buying a house with my own fiber uplink to a distribution point, as it's probably cheaper when it's part of the 'package' with a house purchase than having it run by a contractor. Since it costs just about the same to run a single strand as it does to run 20, it would make sense for new neighborhoods to have someone run fiber to all the houses while the trench is open, like they do with cable now.
Slashdot Mirror
Well, that would add to the number of components that could fail, and require a high speed bus between the two controllers, as well as a shared cache and all the headaches that would bring with it (think SMP caches being ping-ponged). Then you've got to sync your interface to the system bus as well as the new internal buses. On the other hand, you can just crank the knob up to 11 and go 20K RPMs on known, tried and true, technology.
If someone decides that you're a trouble maker, or whatever, try to get your patch in a different branch of the tree. The kernel has a long tail, and all patches go upstream to Andrew Mortons tree and then hop to Linus' tree, provided that the code is decent and it serves a genuine purpose (A guy from Google has two spelling corrections that were merged and he got credited for).
You'll get yelled at for formatting and such before anyone of higher authority sees your code. You probably won't get your code past a subsystem maintainer without having it look presentable. It's just about getting your foot in the door at any node along the tail. I mean, even Namesys got code in the kernel with Hans shouting at everyone. The protocol for doing things is a bit flexible (I hear Andrew Morton still submits changes to Linus as tar balls from time to time... I'm not sure if he does this any more, though) if you stick to the key concepts.
Just my $0.02, I could be wrong.
To be honest, I've never heard of them. I'm assuming it's a low level VNC or something?
Everyone becomes conservative with upgrades after the first time that a box 3000 miles away fails to come back up. Seriously, waiting for a remote reboot after a kernel update is always the longest two minutes of my life.
Even the headless boxes at my apartment wait for me to set aside time to haul out a monitor and keyboard if anything goes wrong during an update. It's better to assume that something will go wrong and be pleasantly surprised and ahead of schedule than to sit staring at pings that have been timing out for the last five minutes (while you think, maybe it's just taking a long time to init... yeah, right!).
And, regardless of what anyone says, a virtual machine test environment doesn't have anywhere near the complications that you get with heavy metal. A successful virtual machine test just means that nothing is assured to go wrong, nothing more.
I think they have upstream patches for that as of last night... but, who knows. It does coincide well.
No, the problem is that outside of their company, no one has any idea of how the thing works. They can't bring in a consultant, and there isn't anyone to turn to for advice. Even if they hired a small team of talented professionals to help them, that team would still need plenty of time to get up to speed on the whole system.
*pssst*
I hear the guy that rooted you is at 127.0.0.1. Go get 'em, tiger!
This should be entertaining...
Just wondering if, by your own logic, you would say that string theory is a science or not. For that matter, how far in to physics would you assert your position?
Right. Just remember that the guy with his job on the line is never the one who made the poor decision. His boss did after the poor guy argued against it for months only to be forced to implement said Bad Idea.
That's too funny! I just replied to someone a thread up about trying to run a modem under Slackware
How are you using vgetty? Are you using a hardware DAC?
From what I understood, Asterisk can't use regular PCI modems because almost all of them are winmodems. Meaning, they don't have a full firmware stack, but rely on their drivers to supply certain functionality. Namely, proprietary, closed source, non-free (beer or speech) Windows only drivers.
I've got a box with 5 PCI slots and as many old V.92's lying around that I'd love to get working on a Slackware build, if you know something I don't (not being sarcastic, I'd be really be thrilled if I didn't research far enough to find the answer).
It's just like buying Red Hat support; you get the backing of a company that employs the people who are developers for that project. With Red Hat you get a bunch of kernel developers and Andrew Barlette (another key Samba developer). You can't get better support for your money than support from key developers. Also, it enables the developers to work on open source projects as a day job, too.
The Eclipse Foundation and the entire Eclipse community are pleased to announce the availability of the Ganymede Release, the annual release train developed by the Eclipse community. The Ganymede Release is a coordinated release of 23 different Eclipse project teams that represents over 18 million lines of code.
IBM didn't write all that code ;)
[...] two letter variable and save a handful [...]
dd if=/dev/floppy of=/media/floppy.img && mount -t auto /media/floppy.img /mnt/floppy
Why not?
If you use the word floppy all the way through, you could make it a two letter variable and same a handful of bytes, too =)
So, how much manufacturing does this stuff need to be a viable source of rocket fuel to fire rockets back to earth?
I thought plan9 was Lucent?
You mean except for the fact that it predates the founding of Lucent by a decade? Secondly, Bell Labs and Lucent are the same company.
Yeah, yeah... Minor details. Other than the two details you mentioned, it was Lucent. I'm sticking to my guns on this one!
I thought plan9 was Lucent?
At any rate, I greatly appreciated the joke.
FOSS S3Backer uses FUSE to mount a bucket as a disk :)
You could also use S3Backer with an rsync script (or rsnapshot) on the host. That lets you mount the S3 bucket as a drive on your server through FUSE and then copy to it as if it were local. *NIX/BSD only, though.
The issue is more general than that. The usability experts who can't code *can* improve usability, by telling the developers what to do.
It's more of an issue of *just* criticising, instead of offering constructive opinions. If all you can do is say "this sucks" but not say why or how it can be improved, then I agree, you have no business in software. If, on the other hand, you can find a way to improve it, I'm sure many people will welcome your advice and implement it.
RE: Bugfix, Usability is a matter of opinion
*WONTFIX*
Usage works for me.
*Ducks and runs*
Note to self: fork codebase. Send snarky message at 1.0 release. ;)
No doubt... I don't care who they are, low level hardware fixes against an OS in a week is impressive. I think Linus' window on RC's for fixing this stuff is two weeks, and the kernel team has been moving at breakneck speeds lately (averaging 4 LOC/hour, every hour, every day).
Have you looked in to Ruckus. It's DRM'ed and only runs on windows media player, but it's free for students with an .edu email account if the school signs up for it. Our technet at Kutztown University, Pennsylvania, has a huge banner on their home page saying something to the effect of "stop infesting the campus with your MP3s from torrents, we've signed you up for free music from this service". It seems to be working somewhat. Basically, Ruckus puts a skin on WMP and shows little adds in the bottom of the player, and that ad revenue supports all the music you can download.
Their network is always fast, and I think that if you setup some kind of caching server, you can cut down on bandwidth as music is mostly 'trendy' in the way that youTube links are; someone finds something and tells everyone else to check it out, and then you've got your pipe saturated with fifty copies of the same media.
At any rate, while I don't like DRM, it is a solution that at least gives you the ability to cut down on traffic externally, and internally (from infected machines reinfecting each other - I've seen the network room, most of the rack is lit up like a Christmas tree, solid green, even with hardware shapers and balancers over multiple backbones. They've even got IR on the roof as a failover... it's nuts what morons will do to your network, even when they've got $100,000 of hardware protecting them at the edge).
Ruckus is worth looking in to, IMHO, even if to compare what's on the market or solidify your choice for something else.
I'm pretty sure if you do away with software completely you'll be pretty safe.
Didn't you hear about the 0day abacus exploit?
This is kinda' like a patch panel, I take it. Have a junction point for distribution of media and have it in place so that you can light up paths as you need them. Good idea, so long as no one gets the bright idea of proprietary junction boxes or something.
I know I wouldn't mind buying a house with my own fiber uplink to a distribution point, as it's probably cheaper when it's part of the 'package' with a house purchase than having it run by a contractor. Since it costs just about the same to run a single strand as it does to run 20, it would make sense for new neighborhoods to have someone run fiber to all the houses while the trench is open, like they do with cable now.