|
angrytech posted:So basically Redhat and Canonical have to get together and agree on DRM? If Netflix were to ship a desktop app, they'd just ship their own statically linked libraries and be done with it.
|
#
?
Oct 20, 2012 07:05
|
|
|
|
| # ? Jun 3, 2024 07:38 |
|
|
That wouldn't help in the slightest. You can still attach a debugger to a statically linked binary, and you can still modify the kernel to do whatever you want to arbitrary processes.
|
|
#
?
Oct 20, 2012 07:07
|
|
|
angrytech posted:So basically Redhat and Canonical have to get together and agree on DRM? Not really. It's just that H.264 is the only codec that isn't worthless, but it has the GPL on one side - which prevents inclusion in non-open devices/apps - and patent hell on the other. H.264 has to work on Roku, Boxee, iOS, Android (times 3), Windows Phone, Windows, OS X, Sony TVs, Samsung SmartTVs, Wii, PS3, Xbox 360, and that's just the ones with major market share. On top of that, H.264 is actually a codec platform of sorts, because it's got multiple profiles each which enable different capabilities and target a range of required decoder complexity and encoding efficiency. This means that targeting from the low-end to the high-end is possible with the same infrastructure. For example, 720p encoded at a medium level vs a 1080p encoded at a high level can easily be the exact same bitrate, because the latter uses more CPU-expensive operations or relies on GPU- or CPU-acceleration to achieve better throughput. This is all my own opinon, but... All of the web codecs are, in their current form, doomed to fail. H.264 is just that good. Effective competition will most likely require: * A codec designed with deep consideration for device capabilities * A high quality encoding/decoding implementation, * A permissive non-copyleft license Talk to DarkShikari a bit and you'll quickly understand #1 and #2 are ridiculously loving hard. #3 might be a dealbreaker for both open-source supporters and for the people who are a party to MPEG, which would probably dig some patents out and sue any viable implementation. Ideally the encryption would look something much like HLS, only better specified, and with a better reference implementation for segmenting (Apple's segmenter has its problems and a lot of platforms calculate playtimes strangely), would also exist. That said, I'm not the person to prescribe a solution since I only have privileged access to some who actually deal with 95% of this clusterfuck, rather than having experienced it myself, and only have personal experience with the aforementioned Android issues. EDIT: I kind of got off track. DRM is mostly just an obfuscated way of encrypting content. No fully open solution will really accommodate that. Allowing a standardized way of including - in a pluggable way - a closed-source de-obfuscation method into the video decoders would probably be a decent enough DRM solution. Good enough to use by media companies and claim good-faith DRM'ing, but also possible to use with a completely free solution with open implementations. It would certainly be unpopular as it compromises the principles of open-source, but on the other hand it would effectively be a lure to make open-source more attainable by the rest of the ecosystem. Doctor w-rw-rw- fucked around with this message at 07:35 on Oct 20, 2012 |
|
#
?
Oct 20, 2012 07:24
|
|
|
A good-faith DRM effort isn't sufficient, it has to work. DRM was the issue holding Netflix back from Android.
|
|
#
?
Oct 20, 2012 07:38
|
|
|
pseudorandom name posted:A good-faith DRM effort isn't sufficient, it has to work. Unfortunately, I can't get into any further details other than to say that that certainly was a stumbling block for them, and I understand why. Sorry.
|
|
#
?
Oct 20, 2012 07:50
|
|
|
I would think streaming video is a bit different as far as DRM than having actual downloads though. During streaming you can protect the content on a CDN level so rights to the media are determined before you are accessing it. The content doesn't ever have to reside on the device in a place that is easy to access either, and it wouldn't be in a form that would be easy to just rip out if there even was enough of it. Web clients is a totally different issue. You can't just use HTML5 because there is no DRM. It needs to be wrapped in something. Was just having a discussion about this because of the project I am working on at work these days. Netflix was originally tied to Silverlight probably because they were using MS Playready, but now silverlight is being discontinued. Switching DRM architectures isn't such an easy task either. Probably a lot more work getting the back end all going then doing the clients. Somebody mentioned google nativeclient as an alternative for doing browser based code http://code.google.com/p/nativeclient/ Other than flash I don't know what other options there will be. WIll be interesting to see what happens with the technology.
|
|
#
?
Oct 20, 2012 14:44
|
|
|
Are the iphone and android versions that secure that nobody could obtain the videos from them, though? It seems more likely that nobody cares since the quality isn't that high. Couldn't they just make a linux version that just plays the iphone videos?
|
|
#
?
Oct 20, 2012 15:07
|
|
|
I'm not sure why they care so much - isn't anything on Netflix available from a hundred other places as  already?Or are they just worried someone is going to add 'right-click save as' to the Linux Netflix client?
|
|
#
?
Oct 20, 2012 15:18
|
|
|
Bob Morales posted:I'm not sure why they care so much
|
|
#
?
Oct 20, 2012 15:21
|
|
|
mystes posted:Are the iphone and android versions that secure that nobody could obtain the videos from them, though? It seems more likely that nobody cares since the quality isn't that high. Couldn't they just make a linux version that just plays the iphone videos? With things like airplay and the hdmi out on phones these days you can plug them into TVs so I think HD streaming is available. If you want to watch legal streaming content on linux, Amazon is always an option. They apparently have no problem streaming SD content to linux over the wire.
|
|
#
?
Oct 20, 2012 16:15
|
|
|
Longinus00 posted:With things like airplay and the hdmi out on phones these days you can plug them into TVs so I think HD streaming is available. I was unaware of this. Thanks for the tip. Once again Amazon gets my money for being the least lovely. drat it netflix get with the program.
|
|
#
?
Oct 20, 2012 16:42
|
|
|
pseudorandom name posted:That wouldn't help in the slightest. You can still attach a debugger to a statically linked binary, and you can still modify the kernel to do whatever you want to arbitrary processes. Note that this is equally true on the Windows platform. Windows still allows users to install unsigned kernel modules, and custom directvideo drivers, and attach debuggers, and run DMA. Windows is not inherently more DRM-friendly than Linux; Linux just has an arbitrarily higher bar for "secure DRM" than Windows does. You can justify this by pointing at Linux's more-technical userbase, or how the Linux community "encourages" breaking DRM measures, but personally I think it's just because managers at the DRM-using companies think "if we support Linux we have to go open source".
|
|
#
?
Oct 20, 2012 17:49
|
|
|
Hot Yellow KoolAid posted:I have to edit several files on my school's servers using SSH. The only way I know how to manipulate a text file remotely is using VI or Nano. The problem is that these files are somewhat large and intricate, and I don't want to risk doing them in a terminal text editor. SSH in with X forwarding enabled (and have an X server installed on your workstation, you can even do this in windows, I love working on a windows box and using gvim) and you can open up gvim or whatever graphical editor your heart desires to do your editing in. I prefer this because I like being able to use PgUp and PgDown in my vi instead of Meta key malarkey.
|
|
#
?
Oct 20, 2012 22:26
|
|
|
How am I supposed to access the Windows system tray when running a program via Wine that resides there? Ubuntu 12.10 with Unity and Wine 1.5.15 I've tried googling various combinations of "ubuntu wine no system tray" without any success... edit: Specifically, I'm running Steam. edit2: Nevermind, now it's showing up on the Unity panel. Thermopyle fucked around with this message at 00:42 on Oct 21, 2012 |
|
#
?
Oct 20, 2012 23:15
|
|
|
ShoulderDaemon posted:Note that this is equally true on the Windows platform. Windows still allows users to install unsigned kernel modules, and custom directvideo drivers, and attach debuggers, and run DMA. Windows is not inherently more DRM-friendly than Linux; Linux just has an arbitrarily higher bar for "secure DRM" than Windows does. DRM capable audio, graphics, and video codec drivers get different WHQL signatures, and can make decisions on what they're willing to do based on what other drivers are loaded. Userspace apps can decide not to talk to non-DRM capable drivers at all.
|
|
#
?
Oct 20, 2012 23:23
|
|
|
who here has experience with DAAP servers? trying to make shares compatible with all three OSes. i was considering running FreeNAS in a VM but that seems kind of crazy if there's a good software solution
|
|
#
?
Oct 20, 2012 23:38
|
|
|
angrytech posted:I was unaware of this. Thanks for the tip. Once again Amazon gets my money for being the least lovely. drat it netflix get with the program. Just note that Amazon Prime streaming uses Flash. It'll work in Linux so long as Flash isn't acting up.
|
|
#
?
Oct 20, 2012 23:41
|
|
|
syzygy86 posted:Just note that Amazon Prime streaming uses Flash. It'll work in Linux so long as Flash isn't acting up. Like I said: least lovely.
|
|
#
?
Oct 20, 2012 23:46
|
|
|
syzygy86 posted:Just note that Amazon Prime streaming uses Flash. It'll work in Linux so long as Flash isn't acting up. xbmc has a plugin that will let you watch and should bypass the flash requirement.
|
|
#
?
Oct 20, 2012 23:54
|
|
|
JHVH-1 posted:I would think streaming video is a bit different as far as DRM than having actual downloads though. During streaming you can protect the content on a CDN level so rights to the media are determined before you are accessing it. The content doesn't ever have to reside on the device in a place that is easy to access either, and it wouldn't be in a form that would be easy to just rip out if there even was enough of it. What they're concerned about is stream ripping tools allowing the user to save what they're watching in to their own collection. Since playback of HD content on lightweight HTPCs and mobile devices needs hardware decoding to be useful, they are pretty much required to use standard codecs. This means they have to be able to trust the kernel and drivers to not intercept the content after decryption. I believe some of the SoCs used in mobile devices that have hardware encryption capabilities can pass data from decryption straight to decode and then out to the display without having to go through the CPU or system memory, which would allow for secure DRM even on rooted devices, but that does not apply across the board nor is it available on a PC platform outside of possibly the latest CPUs that have built-in GPUs. Bob Morales posted:I'm not sure why they care so much - isn't anything on Netflix available from a hundred other places as No one ever claimed the content companies were being logical. There are also plenty of people who would never download a pirate movie but would save streams without a second thought. ShoulderDaemon posted:Note that this is equally true on the Windows platform. Windows still allows users to install unsigned kernel modules, and custom directvideo drivers, and attach debuggers, and run DMA. Windows is not inherently more DRM-friendly than Linux; Linux just has an arbitrarily higher bar for "secure DRM" than Windows does. I'm pretty sure DRM-enabling features like Protected Audio Path and HDCP disable themselves when using unsigned drivers for this exact reason. Sure you can put Windows in to a mode that allows the same level of access as you'd have on Linux, but it's a lot harder to hide that fact from the DRM software. osirisisdead posted:SSH in with X forwarding enabled (and have an X server installed on your workstation, you can even do this in windows, I love working on a windows box and using gvim) and you can open up gvim or whatever graphical editor your heart desires to do your editing in. I prefer this because I like being able to use PgUp and PgDown in my vi instead of Meta key malarkey. Expanding a bit on this, it's really easy. Install Xming, run it, then in PuTTY you go under Connection>SSH>X11 and enable X11 forwarding. Point it at localhost:0.0 and you're set to go. Start an X11 app on the remote machine and it magically appears on your screen. If you're not on the same LAN you may want to enable compression as well. Or if you're on a Mac just make sure X11.app is installed and do ssh -X <host>
|
|
#
?
Oct 20, 2012 23:59
|
|
|
ACID POLICE posted:who here has experience with DAAP servers? trying to make shares compatible with all three OSes. i was considering running FreeNAS in a VM but that seems kind of crazy if there's a good software solution Just use UPnP. Trust me.
|
|
#
?
Oct 20, 2012 23:59
|
|
|
Suspicious Dish posted:Just use UPnP. Trust me. I don't think I could use iTunes in OS X then? There are other OS X solutions so maybe I'll just do that by the way, no matter what I do, I can't make wifi work as easily in Slackware as I can in Fedora/Debian. it just won't keep my connection! I know it's an OS issue but I can't figure it out. I like Slackware most and would love to be able to keep it on my laptop for wifi work.
|
|
#
?
Oct 21, 2012 00:16
|
|
|
babies havin rabies posted:I think having the Windows TrueType fonts might help with that. Sadly fonts on Linux are one of those things that will never really get better than mediocre, because the cost and effort to produce high quality fonts exceeds the somewhat limited interest in typography within the community. Also, compatibility is a pretty big issue here. wolrah posted:What they're concerned about is stream ripping tools allowing the user to save what they're watching in to their own collection. Folks are limited in their capacity to rip the entirety of instant streaming services when download rates are limited to playback speed to begin with. At best you could automate the process to "play" stuff 24/7, but at that point, there's other more convenient sources for .Personally I like the YouTube model. There's no explicit support for saving streams but there's a plethora of third party tools that can do it. Download speed is limited to playback rate, so you're not going to rip all of YouTube, but a few videos, sure! Thing is, I regularly save YouTube videos for space-shifting purposes, since I don't have a reliable/fast Internet connection in the places I'm most interested in watching them. I don't archive them either. I'm just, on Sprint. 
|
|
#
?
Oct 21, 2012 19:30
|
|
|
ExcessBLarg! posted:On some level I "get" why this is a concern for them, but it's annoying and somewhat of an absurd fear. Part of it it is that the execs/lawyer that make the rules don't "get it" when it comes to file sharing and DRM
|
|
#
?
Oct 21, 2012 21:46
|
|
|
.
Cyberpunkey Monkey fucked around with this message at 22:51 on Oct 22, 2012 |
|
#
?
Oct 21, 2012 23:06
|
|
|
osirisisdead posted:The data of what you watch, when, for how long, etc. all has asset value to these twisted fucks. Take your idiotic conspiracy theories and  .Can we get back to Linux questions, please?
|
|
#
?
Oct 22, 2012 02:36
|
|
|
osirisisdead posted:The data of what you watch, when, for how long, etc. all has asset value to these twisted fucks. That's why I wrote a script to curl nothing but porn and nature documentaries from netflix's servers. 
|
|
#
?
Oct 22, 2012 02:42
|
|
|
In the midst of this Netflix discussion, some of you might find this cute. http://yro.slashdot.org/story/12/10...+%28Slashdot%29
|
|
#
?
Oct 22, 2012 05:04
|
|
|
does anyone have experience with good font managers, though?
|
|
#
?
Oct 22, 2012 05:50
|
|
|
What is a font manager?
|
|
#
?
Oct 22, 2012 06:03
|
|
|
Suspicious Dish posted:What is a font manager? Something like this but that can handle font verification/organization better, I guess. http://code.google.com/p/font-manager/
|
|
#
?
Oct 22, 2012 06:32
|
|
|
Misogynist posted:In the midst of this Netflix discussion, some of you might find this cute. Yet again, the pirates to it better, faster, and cheaper. This time even Netflix knows it. Cyberpunkey Monkey fucked around with this message at 22:50 on Oct 22, 2012 |
|
#
?
Oct 22, 2012 09:12
|
|
|
I have to agree with osiris, it's a really old concept. Even your local supermarket has a rewards card right? Same thing. e: well not about the pleb part, that's just childish. spankmeister fucked around with this message at 10:36 on Oct 22, 2012 |
|
#
?
Oct 22, 2012 09:37
|
|
|
Put plainly, big data has far more use internally than externally, and there is little incentive (read: pretty much none) to offer user data externally, even anonymized. Video watching is tracked in aggregate as a matter of consequence for revenue purposes (revenue from membership fees is divided among the content owners by time watched) as well as, if the video service supports it, video resuming, so it's not collecting anything it wouldn't otherwise. While Facebook or other social sites might be selling demographics, and bundle you in a little package to be shown ads where you might not want them with data more specific than you'd like, paying members bring in far more revenue than they would make as assets by themselves, so from a business standpoint it doesn't make sense to let personal data leak. The money to be made by collecting data that must be collected anyway - and finding buyers of that data - is not something worth pursuing next to subscription revenue, It has far more use internally for such things as recommendation engines and the online equivalent of reporting ratings to convince ad buyers. The only people who would bite on that data would be the ones trying to target the ads that are already naturally a part of the online video streaming, but they're pretty much not capable of targeting on individual data, so why give it to them? I would prefer not to be provoked by silly and ignorant posts, so I will leave it at this: Don't call me a pleb. That's condescending. I understand the business; I work in it. While it may not be true for the next social online circlejerk site, the members really are the customers, not the product. Thinking everyone is out to sell your data is just about as intelligent, insightful, and accurate as saying the secret service is out to get you. Get over yourself. Your individual data is basically irrelevant for revenue purposes and mostly relevant just for recommendation, personalization, and internal metrics.. [[EDIT: vvvvv I didn't call you an idiot. Your conspiracy theory was silly, ignorant, and idiotic. I apologize for nothing. ]] Now let's move on and stop talking about this crap, and start talking about Linux. Doctor w-rw-rw- fucked around with this message at 21:57 on Oct 22, 2012 |
|
#
?
Oct 22, 2012 10:33
|
|
|
I really don't understand how anyone works with Samba. The way it handles permissions is just terrible. I've tried it with ACLs, without ACLs, with ACLs & extended attributes, etc. I haven't found any way to make it handle permissions in a way that any version of Windows does. I don't understand how something that is supposed to work like Windows and interact with Windows totally doesn't work or interact with Windows correctly. I can apply permissions to a folder. When I create a file in that folder, it inherits the ACL (correct). If I make a folder inside that, it does *not* inherit the ACL correctly. If I alter the POSIX permissions of the file (o+w), suddenly the extended attribute ACLs are changed. I've looked at dozens of settings telling things to inherit ACLs, not inherit ACLs, map ACLs, etc, etc. VFS object "acl_xattr" will write to ACLs and extended attributes (both). You cannot get it to just write to extended attributes, though. If the filesystem lacks ACLs, then acl_xattr cannot allow more than a single user & group on a file. If you have a fileystem with ACLs, then Samba will modify & mangle the POSIX permissions *and* POSIX ACLs to try and mimic the extended attribute ACLs, which screws up a system/user that may only be using the POSIX permissions. Does anyone have a Samba config they are actually *happy* with?
|
|
#
?
Oct 22, 2012 21:14
|
|
|
Xenomorph posted:Samba I've never been happy with Samba. I always end up using WinSCP or some other means of transferring files from a *nix to Windows box. It's easier and painless for me. I'd even prefer installing Cygwin over messing with Samba That's about it, I've tried to get Samba working about five times over the past ten years and every time it's been a frustrating foray in futility. Cyberpunkey Monkey fucked around with this message at 22:50 on Oct 22, 2012 |
|
#
?
Oct 22, 2012 21:39
|
|
|
osirisisdead posted:That's about it, I've tried to get Samba working about five times over the past ten years and every time it's been a frustrating foray in futility. I see Linux recommended as a file server all the time, but I've yet to see it work correctly. I've been searching Google and reading a lot of documentation the past few days. Every example config or recommended setting just doesn't work. One recommendation to tame permissions was to use "create mask" and "directory mask", but those seem to only work when the file system does not use ACLs. Without ACLs, acl_xattr cannot write to extended attributes. If I enable ACLs, I lose the create & directory masks! On our Apple servers, "nt acl support = no" is set to keep the Windows clients from even seeing permissions (due to the inconsistencies with the permissions and local ACLs). We have lots of users accessing the file system via NFS, and they don't need every file set to "read write execute" just because we added someone to the ACL with Windows. We'd remove the permission (chmod o-rwx), but the next time the ACL was updated from Windows, the POSIX permission would be changed again. If acl_xattr writes Windows-friendly ACLs to extended attributes, why does it need to change the POSIX permissions? (I understand that the more-restrictive Linux filesystem permissions will override the Samba permission, but this isn't the case in this situation). I've talked to other admins here, and many seem to avoid file sharing all together on Linux. Local logins on a Windows box, files saved locally, backup every client. If they do use Samba, then it's Samba only (no NFS or local access), so there is no issue with POSIX permissions getting mangled by Samba. I believe Samba on FreeBSD left POSIX permissions alone. It wrote straight to NFSv4-style ACLs, which worked great for Windows permissions and local POSIX access.
|
|
#
?
Oct 22, 2012 22:10
|
|
|
osirisisdead posted:How about you stop calling me silly, ignorant and an idiot. You apologize for starting this whole thing, and then we can move on.
|
|
#
?
Oct 22, 2012 22:43
|
|
|
Star War Sex Parrot posted:Grow up. Sieg Heil! (USER WAS PUT ON PROBATION FOR THIS POST)
|
|
#
?
Oct 22, 2012 22:51
|
|
|
|
| # ? Jun 3, 2024 07:38 |
|
|
You done hosed up now son.
|
|
#
?
Oct 22, 2012 22:53
|
|