No, it seriously is. Not only is it dead easy to navigate, but you can just type php.net/function_name_here, and it'll redirect you to the documentation for that function.
Dead awesome. And yeah, the documentation is extremely clear, and there is an unusually high number of code examples (plus the area for user-contributed notes, of which 50% are complete noob, and 50% are pure genius).
Ubuntu does not, it distributes a script which allows the user to fetch the drivers directly from nVidia and install them.
Actually, it doesn't. The contents of nvidia-glx-* and nvidia-*-kernel-source (including the proprietary nv-kernel.o file) are packaged up inside the.deb files in the repository like other packages (albeit being in the restricted repository).
Fibre is just that -- fibre. What defines the speed is what you put on either end of the fibre cable -- i.e. the modems. If they choose to upgrade to a 1Gbps connection in the future, it won't involve digging up the cables.
Firefox doesn't display a UI to deal with the tags, but it will perform prefetching on a <link rel="next"/> page. This is especially handy when browsing Pipermail archives, as clicking on the next in thread link is always fast and snappy.
What interests me is the fact that in these discussions about Theora being an old and antiquated codec, nobody seems to know about Dirac, which is a modern video codec quite comparable to H.264 developed by the BBC.
Dirac is specifically designed to be free in the sense we love, and they have specifically checked to make sure it doesn't violate any patents, etc.
It is supported in recent versions of FFMPEG, and since VLC 0.9.2. Support for it is maturing quite fast, and I don't understand why Mozilla didn't include support for it in their HTML5 video implementation.
Since Opera implements <video> with GStreamer, it should already support Dirac if you have the support installed.
Seriously man, what exactly is the fascination with twitter? Everyone is on twitter and if you want to hang with the "cool kids", OMG, you must be on twitter!
Even so, it is cool that such inertia exists.
I would like to find out what generates such inertia, and apply it to something worthwhile, like free and open source software. Wouldn't it be cool every other week to hear about which new celebrity has switched to a GNOME desktop, or has chosen to release their new album exclusively on Ogg Vorbis?
Didn't think so.
Sounds pretty stupid, and that's probably why there's so little traction with regards to FOSS among mainstream. Firefox is probably the most mainstream open source software there is -- the rest of the community could learn a thing or two from Firefox and Twitter.
Firstly, it's per-user specific, so not only will other users on your system get a dynamic IP still, if you are not logged on, you won't have a network connection at all.
And yes, there is an "Available to all users" checkbox, which in theory should do what the G.P. wants, but in practice, it's broken. Methinks some PolicyKit rules are screwed, but it's never worked for me on the several systems I've tried it on -- I've been in the same boat.
My solution was to set the static IPs on the DHCP server instead, so the client wouldn't have to worry. But that's not feasible for everyone.
It can't be done with Windows (at least not to my knowledge), but multiseat on Linux these days is a cinch. Google has tons of resources on the topic -- basically it involves a bit of xorg.conf hacking, and then Bob's your uncle.
I myself have done it before on an amd64 dual-core 2.2GHz system with two video cards, a GeForce 7600GT on PCI-e, and a GeForce 6200 on plain PCI. Worked beautifully. I could multiplayer FlightGear by running one instance on each seat. Each user can log on and off independently with their own keyboard and mouse.
This is a (blurry and fuzzy) picture of my setup (1280x1024 JPG). You can see glxgears running on each screen -- handled by the same computer. Cool thing about using two video cards is that each user gets his own GPU -- running two FPSes simultaneously (I tested Nexuiz) had absolutely zero slowdown.
What? Letting some company dictate what you can and cannot do with a device you legitimately bought is a "good thing"? A "good thing" because you can't see "what possible advantage is there to using such a device as a server"?
As long as you are not breaking the law, Apple has no right to block you using such applications. (And even if you are breaking the law, it's law enforcement's problem, not Apple's.)
I myself do see an advantage in having a web server on an iPhone, if only over WiFi, and not 3G. It could be an interface for, say, transferring files, or maybe some diagnostics.
I do realise that Apple doesn't have any obligation to endorse apps in its app store. But because of the DRM, isn't it impossible to install apps from any other source? (Correct me if I'm wrong. My whole post depends on that flimsy piece of memory.)
Yeah, except it confuses integers and floating points on PowerPC. Try running Lavacap on x86 under DOSBox, and it'll run fine: on the first level, the score requirement is 360 points.
Try the same on the PowerPC version of DOSBox, and it'll tell you that your score needs to be 359.99999999999...
This has everything to do with DRM. If there was no DRM, then when the guy got banned from the Amazon store, he would still have been able to read the ebooks that he'd already bought (that he could have backed up himself). But he lost them too!
Running end-to-end encryption would certainly prevent proxies from stashing away frequently accessed objects.
Your web browser can cache the data in its own cache, but you are correct in guessing that a proxy cannot transparently cache the data.
However, recently, I was looking at Riverbed Steelhead, which claims to be able to cache SSL-encrypted data. I'm guessing it would work similar to a replay attack -- you don't know what the data means, but you still know what its encrypted form looks like, so you can still cache it. Might be worth a look.
Though I know it's probably redundant to say so by now, but you do need to have access to your online account to be able to actually use the backup (or be online in order to go into "offline" mode).
Otherwise, you get stuck at a login dialog. Extremely frustrating for LAN parties without Internet access.
Wrong. [...] Truly, NAT in IPv6 is, from a technical perspective, no more difficult than it has been with IPv4.
Oh, absolutely. Not saying it can't be done, or not even that it hasn't. AFAIK there's even an RFC for IPv6 NAT. However, I don't think anybody's implemented it yet, because nobody likes it.;)
There may still be uses for NAT (like routing a subnet when you don't have the ability to configure routes for your upstream routers, or, as you say, your home appliances), but 99% of the case, your typical NAT'ed network can logically be replaced with an IPv6/64 subnet.
For those that want IPv6 and NAT, you can do that too.
Not 100% sure what you mean by that.
You can do IPv4+NAT and IPv6 on the same network, but there is no such thing as NAT for IPv6. (Well, there might be a specification for it, but nobody's implemented it.)
In XP, it's an add-on protocol that you must specifically install though it's on the install disk.
Not that it really matters, but I should point out that you don't need the install disc to enable IPv6. It just needs enabling.
The fact that it's not enabled by default doesn't really matter. If a consumer ISP today wanted to offer a residential DSL plan with IPv6, they'd make enabling IPv6 part of the "setup process" (along with anti-virus trialware crap) on the install disc that ISPs tend to give out these days.
No citation provided, but I saw one running Windows 98 in the touristy district of the Spanish city of Santiago de Compostela (right near the cathedral).
I wasn't game enough to trust my debit card with it, but a passerby used it, and boy was it slow. You could see the individual images redrawing on the screen. It's been so long since it was last updated that the CRT monitor has the text burnt into its screen. (Although I thought modern CRTs were supposed to be immune to burn-in.)
I'm referring to patents that cover the codecs inside the MP4. I guess you'd be able to use Theora/Dirac + Vorbis inside an MP4, but saying "MP4" implies DivX or MPEG video. Hence the patents.
No, it won't. You're getting confused with the <object> tag.
By default (and most users use defaults) Flash videos (mime type "video/shockwave-flash" I think) will play in Flash.
Okay, I haven't tested this to back up my statement, so I am as bad as you, but AFAIK Flash won't play FLV videos directly. If it did, then nobody would be making all these.swf applets to do the job, such as the JW FLV Player.
To play a Flash video in Flash, you must have a.swf, and either hardcode the.flv URL in the.swf, or pass it as a parameter to the object.
I can pass a.flv to the <object> tag, tell it to use the "video/flv" mime-type to tell it to play the.flv directly. But guess what: it doesn't work! (Well, actually, the VLC nsplugin registers itself for the video/flv mime-type, so if you have the VLC plugin installed, it'll work -- but that's not using Flash!)
Go to about:plugins if you use an nsplugin-enabled browser. Look at the mime-types Flash registers itself for:
application/x-shockwave-flash
application/futuresplash
You see? It only handles Shockwave Flash files, a.k.a..swf files. Not.flv videos directly.
That is what the G.G.P. was getting at. With the <video> tag, you can upload a video to your web server, reference it in the HTML, and be done with it.
The whole concept of "renting" digital content is absurd.
When you go to Video Ezy and borrow a DVD or VHS (DVDs don't really count as "digital" as they physically leave the store), you're physically taking it from the store into your home, where other customers cannot access it (unless Video Ezy has a couple in stock). Thus it makes sense to return it.
Renting online content does not make sense, because of the simple fact that you are getting a copy of the information. If you keep your copy of the video for 50 years, it doesn't stop other customers from purchasing and watching it at the same time.
Renting is obsolete. These days, online distribution costs are so cheap, you can offer a movie that you can keep for life for <$5 a pop, and still be making a profit.
The <video> tag doesn't work like that. You reference a video file directly, such as: <video src="http://videos.example.com/the_video.ogg"></video>
Your browser displays the UI for the video -- Flash isn't involved at all. (Unless Adobe made Flash interpret the <video> tag while running in IE, which would be cool, but at the same time contradictory to their lock-in philosophy.)
Grab a copy of the latest Firefox 3.1 beta and start playing with it. Without Flash installed.
Slashdot Mirror
No, it seriously is. Not only is it dead easy to navigate, but you can just type php.net/function_name_here, and it'll redirect you to the documentation for that function.
Dead awesome. And yeah, the documentation is extremely clear, and there is an unusually high number of code examples (plus the area for user-contributed notes, of which 50% are complete noob, and 50% are pure genius).
Ubuntu does not, it distributes a script which allows the user to fetch the drivers directly from nVidia and install them.
Actually, it doesn't. The contents of nvidia-glx-* and nvidia-*-kernel-source (including the proprietary nv-kernel.o file) are packaged up inside the .deb files in the repository like other packages (albeit being in the restricted repository).
Proof here.
Fibre is just that -- fibre. What defines the speed is what you put on either end of the fibre cable -- i.e. the modems. If they choose to upgrade to a 1Gbps connection in the future, it won't involve digging up the cables.
I'd rather poke my eyes out with needles
Warning: do not poke remaining eye with needle.
Firefox doesn't display a UI to deal with the tags, but it will perform prefetching on a <link rel="next"/> page. This is especially handy when browsing Pipermail archives, as clicking on the next in thread link is always fast and snappy.
What interests me is the fact that in these discussions about Theora being an old and antiquated codec, nobody seems to know about Dirac, which is a modern video codec quite comparable to H.264 developed by the BBC.
Dirac is specifically designed to be free in the sense we love, and they have specifically checked to make sure it doesn't violate any patents, etc.
It is supported in recent versions of FFMPEG, and since VLC 0.9.2. Support for it is maturing quite fast, and I don't understand why Mozilla didn't include support for it in their HTML5 video implementation.
Since Opera implements <video> with GStreamer, it should already support Dirac if you have the support installed.
Seriously man, what exactly is the fascination with twitter? Everyone is on twitter and if you want to hang with the "cool kids", OMG, you must be on twitter!
Even so, it is cool that such inertia exists.
I would like to find out what generates such inertia, and apply it to something worthwhile, like free and open source software. Wouldn't it be cool every other week to hear about which new celebrity has switched to a GNOME desktop, or has chosen to release their new album exclusively on Ogg Vorbis?
Didn't think so.
Sounds pretty stupid, and that's probably why there's so little traction with regards to FOSS among mainstream. Firefox is probably the most mainstream open source software there is -- the rest of the community could learn a thing or two from Firefox and Twitter.
Nope, sorry. NetworkManager doesn't count.
Firstly, it's per-user specific, so not only will other users on your system get a dynamic IP still, if you are not logged on, you won't have a network connection at all.
And yes, there is an "Available to all users" checkbox, which in theory should do what the G.P. wants, but in practice, it's broken. Methinks some PolicyKit rules are screwed, but it's never worked for me on the several systems I've tried it on -- I've been in the same boat.
My solution was to set the static IPs on the DHCP server instead, so the client wouldn't have to worry. But that's not feasible for everyone.
It can't be done with Windows (at least not to my knowledge), but multiseat on Linux these days is a cinch. Google has tons of resources on the topic -- basically it involves a bit of xorg.conf hacking, and then Bob's your uncle.
I myself have done it before on an amd64 dual-core 2.2GHz system with two video cards, a GeForce 7600GT on PCI-e, and a GeForce 6200 on plain PCI. Worked beautifully. I could multiplayer FlightGear by running one instance on each seat. Each user can log on and off independently with their own keyboard and mouse.
This is a (blurry and fuzzy) picture of my setup (1280x1024 JPG). You can see glxgears running on each screen -- handled by the same computer. Cool thing about using two video cards is that each user gets his own GPU -- running two FPSes simultaneously (I tested Nexuiz) had absolutely zero slowdown.
What? Letting some company dictate what you can and cannot do with a device you legitimately bought is a "good thing"? A "good thing" because you can't see "what possible advantage is there to using such a device as a server"?
As long as you are not breaking the law, Apple has no right to block you using such applications. (And even if you are breaking the law, it's law enforcement's problem, not Apple's.)
I myself do see an advantage in having a web server on an iPhone, if only over WiFi, and not 3G. It could be an interface for, say, transferring files, or maybe some diagnostics.
I do realise that Apple doesn't have any obligation to endorse apps in its app store. But because of the DRM, isn't it impossible to install apps from any other source? (Correct me if I'm wrong. My whole post depends on that flimsy piece of memory.)
Yeah, except it confuses integers and floating points on PowerPC. Try running Lavacap on x86 under DOSBox, and it'll run fine: on the first level, the score requirement is 360 points.
Try the same on the PowerPC version of DOSBox, and it'll tell you that your score needs to be 359.99999999999...
This has everything to do with DRM. If there was no DRM, then when the guy got banned from the Amazon store, he would still have been able to read the ebooks that he'd already bought (that he could have backed up himself). But he lost them too!
Any negative interactions?
I hope that HTTPS can cache like HTTP does.
Running end-to-end encryption would certainly prevent proxies from stashing away frequently accessed objects.
Your web browser can cache the data in its own cache, but you are correct in guessing that a proxy cannot transparently cache the data.
However, recently, I was looking at Riverbed Steelhead, which claims to be able to cache SSL-encrypted data. I'm guessing it would work similar to a replay attack -- you don't know what the data means, but you still know what its encrypted form looks like, so you can still cache it. Might be worth a look.
Not if you're in a locked down enterprise that doesn't allow you to just willy-nilly fire up a web server.
Though I know it's probably redundant to say so by now, but you do need to have access to your online account to be able to actually use the backup (or be online in order to go into "offline" mode).
Otherwise, you get stuck at a login dialog. Extremely frustrating for LAN parties without Internet access.
Wrong. [...] Truly, NAT in IPv6 is, from a technical perspective, no more difficult than it has been with IPv4.
Oh, absolutely. Not saying it can't be done, or not even that it hasn't. AFAIK there's even an RFC for IPv6 NAT. However, I don't think anybody's implemented it yet, because nobody likes it. ;)
There may still be uses for NAT (like routing a subnet when you don't have the ability to configure routes for your upstream routers, or, as you say, your home appliances), but 99% of the case, your typical NAT'ed network can logically be replaced with an IPv6 /64 subnet.
For those that want IPv6 and NAT, you can do that too.
Not 100% sure what you mean by that.
You can do IPv4+NAT and IPv6 on the same network, but there is no such thing as NAT for IPv6. (Well, there might be a specification for it, but nobody's implemented it.)
In XP, it's an add-on protocol that you must specifically install though it's on the install disk.
Not that it really matters, but I should point out that you don't need the install disc to enable IPv6. It just needs enabling.
The fact that it's not enabled by default doesn't really matter. If a consumer ISP today wanted to offer a residential DSL plan with IPv6, they'd make enabling IPv6 part of the "setup process" (along with anti-virus trialware crap) on the install disc that ISPs tend to give out these days.
No citation provided, but I saw one running Windows 98 in the touristy district of the Spanish city of Santiago de Compostela (right near the cathedral).
I wasn't game enough to trust my debit card with it, but a passerby used it, and boy was it slow. You could see the individual images redrawing on the screen. It's been so long since it was last updated that the CRT monitor has the text burnt into its screen. (Although I thought modern CRTs were supposed to be immune to burn-in.)
I'm referring to patents that cover the codecs inside the MP4. I guess you'd be able to use Theora/Dirac + Vorbis inside an MP4, but saying "MP4" implies DivX or MPEG video. Hence the patents.
No, it won't. You're getting confused with the <object> tag.
Okay, I haven't tested this to back up my statement, so I am as bad as you, but AFAIK Flash won't play FLV videos directly. If it did, then nobody would be making all these .swf applets to do the job, such as the JW FLV Player.
To play a Flash video in Flash, you must have a .swf, and either hardcode the .flv URL in the .swf, or pass it as a parameter to the object.
I can pass a .flv to the <object> tag, tell it to use the "video/flv" mime-type to tell it to play the .flv directly. But guess what: it doesn't work! (Well, actually, the VLC nsplugin registers itself for the video/flv mime-type, so if you have the VLC plugin installed, it'll work -- but that's not using Flash!)
Go to about:plugins if you use an nsplugin-enabled browser. Look at the mime-types Flash registers itself for:
You see? It only handles Shockwave Flash files, a.k.a. .swf files. Not .flv videos directly.
That is what the G.G.P. was getting at. With the <video> tag, you can upload a video to your web server, reference it in the HTML, and be done with it.
Flash simply doesn't offer that.
The whole concept of "renting" digital content is absurd.
When you go to Video Ezy and borrow a DVD or VHS (DVDs don't really count as "digital" as they physically leave the store), you're physically taking it from the store into your home, where other customers cannot access it (unless Video Ezy has a couple in stock). Thus it makes sense to return it.
Renting online content does not make sense, because of the simple fact that you are getting a copy of the information. If you keep your copy of the video for 50 years, it doesn't stop other customers from purchasing and watching it at the same time.
Renting is obsolete. These days, online distribution costs are so cheap, you can offer a movie that you can keep for life for <$5 a pop, and still be making a profit.
Great, so all I need to do now is pay my royalties to some big corporation to watch it. Wow, that's so much better than Flash.
The <video> tag doesn't work like that. You reference a video file directly, such as: <video src="http://videos.example.com/the_video.ogg"></video>
Your browser displays the UI for the video -- Flash isn't involved at all. (Unless Adobe made Flash interpret the <video> tag while running in IE, which would be cool, but at the same time contradictory to their lock-in philosophy.)
Grab a copy of the latest Firefox 3.1 beta and start playing with it. Without Flash installed.
FWIW, KDE 4.2 does this. (Inverts the colour of a taskbar button when an app sets its 'URGENT' hint.)