1. Instead of running 10 services on one physical machine the way we used to, you run one service per VM (one web server, one middleware server, one database server, etc) - you add the overhead of multiple operating system runtimes, but thanks to awesome hypervisor optimizations identical memory pages are merged, so you don't use any more RAM 2. If one server gets over-subscribed, you just live migrate a running service to a less loaded server. No more building the infrastructure on another server, and doing a dump/restore process, updating DNS, etc 3. If servers are under-subscribed, you can consolidate services on a smaller number of servers and spin down the under-utilized ones - saving power. 4. High availability - have services monitored, so that if they host they're on goes down they're immediately brought up on another host - you can even have redundant copies of the VM running all the time so that you don't have any downtime if a host goes down
As to your question about whether it makes sense to size applications to servers... servers these days are way more that your typical application needs. It would be wasteful to use a 16 core, 500GB RAM server for a web server. Virtualization gives you a way to run this hardware at capacity. I suppose you could run multiple services on one OS, either in the same OS or in containers (as another commenter suggested). That's certainly a valid approach, especially for scale-out applications which you find in global web app vendors like Google or Facebook. Virtualization allows you to get all of the benefits of mutualization, with the ability to run multiple host OSes, provide granular access to people using the infrastructure, and providing a high level of security against jailbreaking applications.
Our goal is to have OpenWengo be as open as possible - I want SIP to become the dominant VoIP protocol - and we definitely want the Wengophone to talk with other platforms, as well as share presence information and user directories.
For my part, I'd love to see various platform providers collaborate on things like directory services and presence so that we could have any SIP user look for any other SIP user, regardless of platform.
And aside from VoIP, we want to be as inclusive as possible - IM in many protocols, support for H323, Jingle and Skype (I can dream), and a very open and collaborative community.
We don't ask for copyright assignment on code contributions, because we don't want Wengo to be the sole proprietor of OpenWengo - the company is investing in the project and paying developers, but it really is a community-owned project.
TrollTech, as the copyright holder, can distribute the code under whatever licence it likes. By default, it distributes under the GPL (thus, you can modify, redistribute, etc) and according to both the FSF and Trolltech, you may not link (even dynamically) with GPL code, unless you GPL the result.
Therefore, if you don't want to GPL the result, you need Trolltech to distribute QT to you under a different licence. Which they will, for a price.
This mail from the GNOME Foundation mailing list is the end of an exchange between RMS and Jeff Waugh after the announcement of "Warty Warthog".
Quote: "I looked at the page you mentioned, which states Ubuntu's philosophy. It takes a strong stand in support of computer users' freedom; it couldn't possibly be better. If the project followed this philosophy completely, I would say it was great. (Well, it ought to be called Ubuntu GNU/Linux.)"
> 1) Did they waste time writing it all themselves, or are they interworking with SodiPodi? SodiPodi is an excellent piece of software if you want to edit SVG.
We (or rather Sven) used rsvg to read and render the SVG as a bitmap.
> 2) Does it just import them and make paths, or is it a full-featured SVG editor? Someone else commented on it now being Photoshop+Illustrator, but that's a whole different thing. Photoshop also supports importing SVG and AI format, it just doesn't edit them. (see question three)
It just imports SVG to a rastermap, and exports paths to SVG. There is no support for the funky stuff like gradient fills, object groups, etc. This is not a vector graphics program.
> 3) Does it make this simple?
Yes. You load your SVG, specifying the size of the bounding box, and there you go.
If you want your CVS users to connect (and read & write) to CVS as any user other than the user that CVS runs under, you need to run the pserver as root. This is because CVS's connect procedure is basically as follows...
1) inetd detects a request on 4901, spawns cvspserver 2) cvs checks CVSROOT/passwd for the username and passwd of the connector 3) unless it's disabled in CVSROOT/config, if the username isn't in CVSROOT/passwd, it will check/etc/passwd. 4) If the password is valid, it forks and does a setuid to the userid of the connector (or to the userid of the local user specified in CVSROOT/passwd 5) As this user, cvs does its thang, and performs all your requested server operations.
If you're not running your pserver as root, step 4 isn't possible, therefore all server operations must be done as the user as which the pserver runs.
In brief, if you run your pserver as a user other than root, you cannot implement user-level access to the repository. If you want to have user-level access, or module-level r/w access, you must run your pserver as root to allow the setuid to happen.
Not being a constitutional lawyer, is it possible to turn this on its head and while people are there anyway arguing that DMCA is an unconstitutional infringement on freedom of expression?
And although "fair use" and the right to reverse engineer aren't constitutional, could there be an argument for striking down DMCA because of the limitations it puts on those principles?
I noted with amusement that some of the songs (example, American Pie by Don Maclean) have covers which aren't banned (Madonna), and for some others (Dancing in the streets) some covers are mentioned, but not all (Mick Jagger & David Bowie) - does this mean that we could have a playlist that includded Ella Fitzgerald's version of "Mack the knife", followed by Joe Cocker singing "Hey Joe", with an encore of Michael Jackson's "Smooth criminal" without violating the banned song list?
In fact, about a quarter (from what I can see) of the songs on the list have been covered somewhere, by someone, and the covers would be OK? Or am I misreading the situation? Also - why ban anti-war/anti-terrorist songs like, say, "Imagine", "Sunday bloody Sunday" or "Blowin' in the wind" (but only the Peter Paul & Mary cover, mind)?
Regardless of what someone just said about watching the QT files grand under Linux, the Sorenson video codec is still not supported. This is not because of anything to do with xanim, it's because Sorenson signed an exclusive agreement with Apple, so the only product that supports the codec is QT4.
I e-mailed the main Sorenson support line (support@s-vision.com) a few days ago about this (after I couldn't watch the LotR trailer), and got a reply from Scott Wheeler (swheeler@s-vision.com) which said, among other stuff, Sorenson Vision has chosen to team up with Apple and their QuickTime product. Because of this, Sorenson Video is only available on supported QuickTime OS platforms.
If you're interested in QuickTime for platforms other than Mac or Windows, please contact your platform vendor and let them know that you would like them to license QuickTime from Apple. Once QuickTime is available on a platform, Sorenson Video will be there as well. I sent another e-mail, asking for (a) a contact in Apple and (b) any indication that there might be a way they could support a free platform. So far, I've gotten no reply to that one.
I think we should let people know how much proprietary video codecs restrict people, and encourage people to distribute in video formats that don't require us to buy new hardware and/or software to view them. As a start, we could mail Mr. Wheeler, nicely voicing our opinions, and asking him to pressure Apple to make a concession, and let Maark Podlipec include module support for the codec with xanim.
While I agree that the treatment both this guy and the DeCSS group in general are getting is both unethical and illegal, it has to be said that the big movie guys have something of a point...although a weak one.
OK, the fact, as I see them, are that at the moment, to view a (legally bought) DVD rom, you need one of (a) A DVD player and a TV (b) An x86 computer running Windows 95/98/NT with a licenced copy of DVD player software or (c) A mac running MacOS with a licenced copy of DVD player software. Which means that people who wish to watch their DVDs (in Ireland, anyway) are forking out a couple of hundred quid for software, even if they already have a computer with a DVD drive. So Linux, FreeBSd, etc users are SOL.
So *our* argument, as I see it, is that under fair use, we are entitled to decode the DVDs so that we can watch our disks on our computers. Grand. We know that, the press doesn't. *Their* argument isn't really about writing copies of DVDs at all, and that's something people are ignoring completely.
The reason the film industry is chasing so hard to keep DVDs under their control is the same reason the music industry is afraid of mp3s. If people crack DVDs to get raw mp2s, then those files will end up as ubiquitous web-wide as mp3s are for music...you'll have warez sites cropping up all over the place to distribute the latest movies fresh off the DVDs. *That's* the cracking/copying they're complaining about.
And the fact is that the only shot they *had* at stalling that was by nipping DeCSS in the bud. Sure, these guys are entitled to do what they did, but if the industry lets the opportunity slip while things are still this centralised, they'll be in a world of shit in a couple of years.
Now, I'm not advocating their actions, but when people say this has *nothing* to do with copying and illegally distributing DVDs, it gets up my wick. The case has two facets. The first is fair use, the second is the indusry's right to defend against illegal distribution. If they're to do it by the book, they'll have to wait until sites start cropping up and sue them one by one. And we all know where that got the music industry.
Long and short of it is, hate these guys for what they're doing to people, but understand where they're coming from, and remember it if you ever have the chance to download illegal mp2s.
I gotta say, I agree whole-heartedly with Mr. Katz here. I haven't been a prolific poster on this, or many other, web fora, IRC channels or mailing lists.
But those that I *have* participated in have tended to be smaller communities, which had a greater sense of community to them. Three years ago, I was a regular on #linpeople. Now I rarely visit. The same can be said of slashdot, alt.os.linux, and any other discussion forum I've become interested in. The main reason this happened is because with numbers came a low SNR (signal to noise ratio). When I started participating in #linpeople, I could barely turn away, because something of use to me was mentioned every minute or so. Towards the end (about 18 months ago) I found myself responding aggressively to people who were dominating the channel with chit-chat because the were (a) flaming newbies (b) hijacking help-session with trivia and chitchat or (c) not giving people a chance to get help, which was the purpose of the channel. Most of the better people from that channel now go to #debian, because that still keeps the quality level high.
The same can be said of Slashdot over the past 18 months. Moderation has meant that most people can follow a thread now without reading 400 comments, but moderation shouldn't be necessary. As Abe Lincoln once said, "Better to be believed an idiot than to open your mouth and remove all doubt" (sic). If people have nothing, or very little, to contribute, restraint should be expected.
The problem is that with numbers comes a degree of anonymity. If someone flamed in the old days, the response was typically "How could you? We trusted you...". Now, it seems like in large fora, flaming is suffered as a necessary evil. There is no acceptance, or apportioning, of responsibility. Little electronic words don't mean anything. "And who cares if a newbie can't get help, or I'm hurting someone's feelings, or driving people who don't have time to read my crap away - they shouldn't take it so seriously. And goddammit, it's my right!"
Maybe it's your right, but if you excercised rights like that with me anywhere near you in a street, I'd be going to jail for assault. That's no way to treat people. Just because we're on-line doesn't make us inhuman.
Someone was asking why such a powerful machine can only manage to predict the weather up to two weeks in advance. The answer comes from chaos theory, and particularly the work of Lorenz in the 60s. The crux of the matter is that the weather is one of the many real-life systems where miniscule changes in initial conditions result in huge differences over a relatively short period of time. He nicknamed this the "Butterfly effect" (the name comes from the theory that a change in the initial conditions as small as a butterfly flapping its wings in Europe could result in a typhoon developing over India, or something like that).
Think of a simple system, like a round-bottomed bowl turned curved side up. Put a marble on the top and record its path. The get the marble and put it close to the original starting point. It could, if you're sufficiently close to the centre of the bowl, end up going in any direction. That's the butterfly effect.
Since measuring instruments can't measure every contributing factor to the weather (temperature, pressure, moisture, wind) to arbitrary levels at a sufficient number of points to form an accurate and complete initial condition from which to predict the weather, it'll go close for a while (the better the measurements, the closer), but within a couple of weeks the values just diverge.
If people are interested in reading a bit more about this stuff, there are a few good books of introduction, like "Chaos" by James Gleich or "Does God play dice?" by an author I can't remember. A good article as a lead-in is here.
Dave Neary.
Eyes, ears and nose...now all we need's nerves.
on
Digital Nose
·
· Score: 2
This got me thinking about the story that was going around yesterday with the camera input wired into his brain...I read an article in the local press which predicted 256-bit greyscale within 2 or three years. and a market-ready product by q4 this year.
Added to this, modern hearing aids attach electronics directly into the inner ear, with vibration sensors outside the ear, to make people who would previously have been deaf, or almost deaf, hear perfectly again.
This chip means (I guess) that we could do to smell what has been done to sight and hearing. All we need to fill the five senses now are electronic tongues (and I believe those exist already) and electronic nerves, with an interface to the brain.
The way I see it, injury-related (as opposed to genetic dystrophic disease related) paralysis could be a thing of the past in about 30 years... now if we could only persuade people to have less fear of science, we might be able to preemp things like muscular dystrophy by then too.
Why is it that the Mary Shelleys of this world always turn up on the cusp of centuries?
...until the people who buy y2k on the first day are begging for something else?
Seems to me kind of silly to try to undercut the Redmond PR machine (that's what makes release days, after all), whereas if we hold off until March, it's no longer about PR, and more about quality.
You could have an "Uninstall y2k for St Patrick" day the month after... if people aren't *too* busy drowning the shamrock:)
Give Windows enough rope to hang itself before we get a lynch mob together:)
Slashdot Mirror
Some benefits:
1. Instead of running 10 services on one physical machine the way we used to, you run one service per VM (one web server, one middleware server, one database server, etc) - you add the overhead of multiple operating system runtimes, but thanks to awesome hypervisor optimizations identical memory pages are merged, so you don't use any more RAM
2. If one server gets over-subscribed, you just live migrate a running service to a less loaded server. No more building the infrastructure on another server, and doing a dump/restore process, updating DNS, etc
3. If servers are under-subscribed, you can consolidate services on a smaller number of servers and spin down the under-utilized ones - saving power.
4. High availability - have services monitored, so that if they host they're on goes down they're immediately brought up on another host - you can even have redundant copies of the VM running all the time so that you don't have any downtime if a host goes down
As to your question about whether it makes sense to size applications to servers... servers these days are way more that your typical application needs. It would be wasteful to use a 16 core, 500GB RAM server for a web server. Virtualization gives you a way to run this hardware at capacity. I suppose you could run multiple services on one OS, either in the same OS or in containers (as another commenter suggested). That's certainly a valid approach, especially for scale-out applications which you find in global web app vendors like Google or Facebook. Virtualization allows you to get all of the benefits of mutualization, with the ability to run multiple host OSes, provide granular access to people using the infrastructure, and providing a high level of security against jailbreaking applications.
Cheers,
Dave.
Our goal is to have OpenWengo be as open as possible - I want SIP to become the dominant VoIP protocol - and we definitely want the Wengophone to talk with other platforms, as well as share presence information and user directories.
For my part, I'd love to see various platform providers collaborate on things like directory services and presence so that we could have any SIP user look for any other SIP user, regardless of platform.
And aside from VoIP, we want to be as inclusive as possible - IM in many protocols, support for H323, Jingle and Skype (I can dream), and a very open and collaborative community.
We don't ask for copyright assignment on code contributions, because we don't want Wengo to be the sole proprietor of OpenWengo - the company is investing in the project and paying developers, but it really is a community-owned project.
Cheers,
Dave.
TrollTech, as the copyright holder, can distribute the code under whatever licence it likes. By default, it distributes under the GPL (thus, you can modify, redistribute, etc) and according to both the FSF and Trolltech, you may not link (even dynamically) with GPL code, unless you GPL the result.
Therefore, if you don't want to GPL the result, you need Trolltech to distribute QT to you under a different licence. Which they will, for a price.
Cygwin and MySQL use the same strategy.
This mail from the GNOME Foundation mailing list is the end of an exchange between RMS and Jeff Waugh after the announcement of "Warty Warthog".
Quote: "I looked at the page you mentioned, which states Ubuntu's philosophy. It takes a strong stand in support of computer users' freedom; it
couldn't possibly be better. If the project followed this philosophy completely, I would say it was great. (Well, it ought to be called
Ubuntu GNU/Linux.)"
Dave.
> 1) Did they waste time writing it all themselves, or are they interworking with SodiPodi? SodiPodi is an excellent piece of software if you want to edit SVG.
We (or rather Sven) used rsvg to read and render the SVG as a bitmap.
> 2) Does it just import them and make paths, or is it a full-featured SVG editor? Someone else commented on it now being Photoshop+Illustrator, but that's a whole different thing. Photoshop also supports importing SVG and AI format, it just doesn't edit them. (see question three)
It just imports SVG to a rastermap, and exports paths to SVG. There is no support for the funky stuff like gradient fills, object groups, etc. This is not a vector graphics program.
> 3) Does it make this simple?
Yes. You load your SVG, specifying the size of the bounding box, and there you go.
Cheers,
Dave.
If you want your CVS users to connect (and read & write) to CVS as any user other than the user that CVS runs under, you need to run the pserver as root. This is because CVS's connect procedure is basically as follows...
/etc/passwd.
1) inetd detects a request on 4901, spawns cvspserver
2) cvs checks CVSROOT/passwd for the username and passwd of the connector
3) unless it's disabled in CVSROOT/config, if the username isn't in CVSROOT/passwd, it will check
4) If the password is valid, it forks and does a setuid to the userid of the connector (or to the userid of the local user specified in CVSROOT/passwd
5) As this user, cvs does its thang, and performs all your requested server operations.
If you're not running your pserver as root, step 4 isn't possible, therefore all server operations must be done as the user as which the pserver runs.
In brief, if you run your pserver as a user other than root, you cannot implement user-level access to the repository. If you want to have user-level access, or module-level r/w access, you must run your pserver as root to allow the setuid to happen.
Cheers,
Dave.
Not being a constitutional lawyer, is it possible to turn this on its head and while people are there anyway arguing that DMCA is an unconstitutional infringement on freedom of expression?
And although "fair use" and the right to reverse engineer aren't constitutional, could there be an argument for striking down DMCA because of the limitations it puts on those principles?
Dave.
In fact, about a quarter (from what I can see) of the songs on the list have been covered somewhere, by someone, and the covers would be OK? Or am I misreading the situation? Also - why ban anti-war/anti-terrorist songs like, say, "Imagine", "Sunday bloody Sunday" or "Blowin' in the wind" (but only the Peter Paul & Mary cover, mind)?
Dave.
I e-mailed the main Sorenson support line (support@s-vision.com) a few days ago about this (after I couldn't watch the LotR trailer), and got a reply from Scott Wheeler (swheeler@s-vision.com) which said, among other stuff,
Sorenson Vision has chosen to team up with Apple and their QuickTime product. Because of this, Sorenson Video is only available on supported QuickTime OS platforms.
If you're interested in QuickTime for platforms other than Mac or Windows, please contact your platform vendor and let them know that you would like them to license QuickTime from Apple. Once QuickTime is available on a platform, Sorenson Video will be there as well.
I sent another e-mail, asking for (a) a contact in Apple and (b) any indication that there might be a way they could support a free platform. So far, I've gotten no reply to that one.
I think we should let people know how much proprietary video codecs restrict people, and encourage people to distribute in video formats that don't require us to buy new hardware and/or software to view them. As a start, we could mail Mr. Wheeler, nicely voicing our opinions, and asking him to pressure Apple to make a concession, and let Maark Podlipec include module support for the codec with xanim.
Dave.
OK, the fact, as I see them, are that at the moment, to view a (legally bought) DVD rom, you need one of
(a) A DVD player and a TV
(b) An x86 computer running Windows 95/98/NT with a licenced copy of DVD player software or
(c) A mac running MacOS with a licenced copy of DVD player software.
Which means that people who wish to watch their DVDs (in Ireland, anyway) are forking out a couple of hundred quid for software, even if they already have a computer with a DVD drive. So Linux, FreeBSd, etc users are SOL.
So *our* argument, as I see it, is that under fair use, we are entitled to decode the DVDs so that we can watch our disks on our computers. Grand. We know that, the press doesn't. *Their* argument isn't really about writing copies of DVDs at all, and that's something people are ignoring completely.
The reason the film industry is chasing so hard to keep DVDs under their control is the same reason the music industry is afraid of mp3s. If people crack DVDs to get raw mp2s, then those files will end up as ubiquitous web-wide as mp3s are for music...you'll have warez sites cropping up all over the place to distribute the latest movies fresh off the DVDs. *That's* the cracking/copying they're complaining about.
And the fact is that the only shot they *had* at stalling that was by nipping DeCSS in the bud. Sure, these guys are entitled to do what they did, but if the industry lets the opportunity slip while things are still this centralised, they'll be in a world of shit in a couple of years.
Now, I'm not advocating their actions, but when people say this has *nothing* to do with copying and illegally distributing DVDs, it gets up my wick. The case has two facets. The first is fair use, the second is the indusry's right to defend against illegal distribution. If they're to do it by the book, they'll have to wait until sites start cropping up and sue them one by one. And we all know where that got the music industry.
Long and short of it is, hate these guys for what they're doing to people, but understand where they're coming from, and remember it if you ever have the chance to download illegal mp2s.
Dave Neary.
But those that I *have* participated in have tended to be smaller communities, which had a greater sense of community to them. Three years ago, I was a regular on #linpeople. Now I rarely visit. The same can be said of slashdot, alt.os.linux, and any other discussion forum I've become interested in. The main reason this happened is because with numbers came a low SNR (signal to noise ratio). When I started participating in #linpeople, I could barely turn away, because something of use to me was mentioned every minute or so. Towards the end (about 18 months ago) I found myself responding aggressively to people who were dominating the channel with chit-chat because the were (a) flaming newbies (b) hijacking help-session with trivia and chitchat or (c) not giving people a chance to get help, which was the purpose of the channel. Most of the better people from that channel now go to #debian, because that still keeps the quality level high.
The same can be said of Slashdot over the past 18 months. Moderation has meant that most people can follow a thread now without reading 400 comments, but moderation shouldn't be necessary. As Abe Lincoln once said, "Better to be believed an idiot than to open your mouth and remove all doubt" (sic). If people have nothing, or very little, to contribute, restraint should be expected.
The problem is that with numbers comes a degree of anonymity. If someone flamed in the old days, the response was typically "How could you? We trusted you...". Now, it seems like in large fora, flaming is suffered as a necessary evil. There is no acceptance, or apportioning, of responsibility. Little electronic words don't mean anything. "And who cares if a newbie can't get help, or I'm hurting someone's feelings, or driving people who don't have time to read my crap away - they shouldn't take it so seriously. And goddammit, it's my right!"
Maybe it's your right, but if you excercised rights like that with me anywhere near you in a street, I'd be going to jail for assault. That's no way to treat people. Just because we're on-line doesn't make us inhuman.
Dave Neary.
Think of a simple system, like a round-bottomed bowl turned curved side up. Put a marble on the top and record its path. The get the marble and put it close to the original starting point. It could, if you're sufficiently close to the centre of the bowl, end up going in any direction. That's the butterfly effect.
Since measuring instruments can't measure every contributing factor to the weather (temperature, pressure, moisture, wind) to arbitrary levels at a sufficient number of points to form an accurate and complete initial condition from which to predict the weather, it'll go close for a while (the better the measurements, the closer), but within a couple of weeks the values just diverge.
If people are interested in reading a bit more about this stuff, there are a few good books of introduction, like "Chaos" by James Gleich or "Does God play dice?" by an author I can't remember. A good article as a lead-in is here.
Dave Neary.
Added to this, modern hearing aids attach electronics directly into the inner ear, with vibration sensors outside the ear, to make people who would previously have been deaf, or almost deaf, hear perfectly again.
This chip means (I guess) that we could do to smell what has been done to sight and hearing. All we need to fill the five senses now are electronic tongues (and I believe those exist already) and electronic nerves, with an interface to the brain.
The way I see it, injury-related (as opposed to genetic dystrophic disease related) paralysis could be a thing of the past in about 30 years... now if we could only persuade people to have less fear of science, we might be able to preemp things like muscular dystrophy by then too.
Why is it that the Mary Shelleys of this world always turn up on the cusp of centuries?
Dave Neary.
Seems to me kind of silly to try to undercut the Redmond PR machine (that's what makes release days, after all), whereas if we hold off until March, it's no longer about PR, and more about quality.
You could have an "Uninstall y2k for St Patrick" day the month after... if people aren't *too* busy drowning the shamrock :)
Give Windows enough rope to hang itself before we get a lynch mob together :)
Dave Neary.