Internet Video (youtube and beyond)

So I just read an interesting article over at Wired about YouTube/Google trying to figure out just how to adapt their current systems to allow for advertising, how to handle it, and what to do with the profits.

It's a very interesting topic. I've also questioned just how sites like YouTube were able to make any money off their service...I don't think I've ever noticed a single add on their site, and certainly not on embedded YouTube videos found on other sites... Looking at YouTube now, I notice there are banner ads above the videos now, though I could swear those haven't always been there. Double checking with GoogleVideo (youTube's new owner for those who've been living under rocks), there are still no ads to be see that I can find.

Even though the vast majority content found on these, and similar sites are given to them at zero charge, bandwidth and server space isn't cheap. Neither is your IT, development and maintenance crews. In the case of Google, they've obviously got plenty of money, but up until their recent purchase, was YouTube really capable of supporting the millions upon millions of hits they recieved each day off of simple banner ads?

The major advertising industry wants in. They know that traditional TV is a sinking ship and it's only a matter of time before it all but completely dies. The question on everyone's minds, apparently, is how to ad advertisements to services like YouTube without ruining the user experience that's made it so popular. I've got a couple ideas...as per usual ;)

One idea discussed in the article is either putting ads at the beginning or end of the videos. Apparently it's widely accepted that pre-ads are a bad idea and no one likes them. The problem with post-ads is that there's a very likely chance the user will just stop the video after it's complete and not watch the ad following. Sites such as iFilm and GameSpot have been putting pre-ads before their free videos, and personally, I've never really minded them so much. The only time they really were ever annoying was when I would go to watch 10 different videos and all of them had the exact same ad, or sometimes sites will have poorly written playlists that play the pre-ad, but then stop and never play the actual video....now that's annoying! However as most of the new video sites are all flash powered, I don't think the latter problem will come up too often any more.

As YouTube thrives on having their videos embedded into numerous other sites, the ads that surround their video rarely get seen...although they do have kind of a tricky system where if you want to full screen the video you have to click the embedded video which takes you to the main youTube site, and then there's a couple seconds where you ...might... accidentally glance at an add before you can hit that full screen button. Along with those industrious few of us who take the time to view the pages' source and just put the direct *.swf in our address bar, Adobe is apparently planning on adding a built in, real, fullscreen feature in the next release of Flash... so there goes that...and really, that was kind of under handed I think, to begin with....if it was intentional.

Google Video has attempted many other things, such as having a pay-only digital distribution service for certain videos...and potentially that could make a little money, if anyone ever actually used it. Another feature, many of the major videogame sites use is to allow free, low bandwidth/quality streaming and then require a paid subscription to see the high quality videos. All of this brings up many questions of owner ship, especially when subjects such as DRM get brought into the equation. Is the customer merely renting these videos, or are they actually buying a copy of them. Sure Google's proprietary, DRM-ified version of VLC they ship will let you want purchased videos any time you want, but will those videos transfer to your other machines? Can you watch it under Linux? On your video iPod? no... Sure, they could strictly offer download only services, but high quality standard definition video can take quite a while to download even with a decent broadband connection....just imagine how long it would take to have to redownload/stream every HD video you bought from them... It's not feasable.

This also brings up another issue... I've never posted anything to YouTube myself, so I don't know the specifics, but it would appear they have a limit on resolution and bitrate your video can be posted in, who knows they may do all the transcoding for you automagically. Would a site like YouTube benefit from offering higher quality videos to paid subscribers like the gaming site do? Would it perhaps be considered kosher for them to offer ad-less low quality videos, but have high quality version that have both pre and post ads tacked on? I personally would be fine with that last option. As long as the ads are 30 seconds or less, I could really care less myself. And I think most would probably agree.

There's one more option out there, and that's basically virtual channels. Basically a never ending playlist that would more closely represent traditional TV channels. Basically what you would do is have users or some algorithm group similar videos in a channel. A use could either watch a specific video or just hit random video for that channel, then you would automatically go on to watch another video from that channel after that one finished, and then another and so on. Then random ads could be placed between every so many videos. The closest thing to this idea I've ever seen was the old Yahoo Music Video service (aka Launch)...but I haven't watched that in years since there's still no Linux support. You tube kind of does something like this too I suppose by offereing up suggestions of similar videos afterward, but it requires user interaction, which some may not want to bother with.

If I were Google, I'd probably do a combination of the above. I'd leave regular YouTube usage the exact same way it is now, with the same quality videos. All embedded videos stay as they are, however, if you goto the main site you have the option to watch high quality videos. There you can either opt to have ads tacked on them, or pay a subscription fee to go ad-less. In addition to this offer the virtual TV channel service and you're good to go.

So, now that we have a good source of profit coming in, there comes the need to discuss who gets a cut of that. Google has already started offereing profit sharing with their highest watched videos, which is certainly a step in the right direction. The other problem is that almost every YouTube video out there infringes upon someone's copyright somehow. Just about every video is either going to be a straight rip from a TV show, or a home video with multiple copywritten pieces of music thrown in the mix. And as per usual, everyone wants their royalities. There are a few different ways to go about this. Either Google can sign a massive contract with the big record labels and TV/movie studios that does blanket coverage for any of their content that ends up on the site, or Google can get strict and delete any videos that the author did not get proper concent ahead of time for. The blanket coverage would probably be the easiest...but Google's still reluctant. Why? Well, even though copyright is fairly clear cut in the law books, it's not in the mind of the people. Once a piece of content become wide spread enough, many would argue it partially belongs to everyone. The fans take personal stake in franchises. Combine that with how the average person feels the various industries charge too much for their products, and the situation is no longer so simple.

In this situation, I'm afraid I'm gonna have to say it's better for Google to do the blanket contracts. There are so many videos posted, that it's near impossible to moderate them all, and good luck getting your users to rat out their fellow posters. Really, since Google is just a business, it's not really their place to worry with if copyright laws need to be changed, that's up to the people. And if they're all too lazy and/or apathetic, then they'll just have to deal with the consequences.

However, in light of that it seems that sites like DeviantArt and the ever popular mySpace have an easier time dealing with the issue. On DeviantArt, users are much more likely to report copywritten material, as most of the users take the art they post very seriously. MySpace, now owned by one of the large content producing conglomerations around, News Corp, is very thorough about making sure no one is posting music and videos that they don't have the proper copyright for. I think the number one reason they've been successful here is because so many bands have their own accounts and post their music already. If there wasn't such great support from the bands and their labels to begin with, I'm sure it would be much more rampant.

It's all very interesting to consider. If Google doesn't make the right choices now, they're 1.6 billion dollar purchase may become a major mistake, and someone who gets it right will even surpass the seemingly unstoppable YouTube.


Open Source vs. Proprietary Software, and ESR

It's funny, about a year ago, I wrote a rather lengthy article about how annoyed I was that Eric Raymond, didn't think GPL-style licensing was necessary anymore in favor of more BSD-style, public domain type licensing... and now I find myself seeing his latest statements to be some of the few reasonable opinions out there. Don't get me wrong, I still stand by the fact that GPL and LGPL type licensing is still very much necessary, however now there are new topics to discuss.

His statements lately are that the Linux world needs to wake up and realize that as great as open source development is, we can't pretend like proprietary software doesn't exist. Unfortunately, at the moment, we still need quite a few components in our Linux distros, that have no viable open alternatives currently.

This is the logical middle ground that many Linux users agree with, but few mention because as in all things, it's the extremists who always talk the loudest.

Let me just first say that I love open source and everything it stands for. I wish all the software out there was open source, however that's just not the way it is. One thing to remember is that one of the biggest advantages in the open source world is the concept of choice. However, many forget that choice also entails that people can choose to release proprietary software too. For now, a mixture is the way to go. Now, most of this going to be geared towards GNU/Linux based operating systems, but it can apply in other areas as well.

The Linux community says it wants the common computer user to start using it and other open source software, but their not willing to make the compromises necessary to make it happen. Groups like the Free Software Foundation and Debian have the mindset that you should refuse to use anything that's not based off open source/free software, but that's just not feasable at the moment. Linux can be a very easy to use, friendly, yet very powerful operating system. But, a large portion of the content that people want to use their computers for are based off of closed/proprietary components. There are thousands of audio and video files out there that are only available in a proprietary codec such as Mp3, WindowsMedia or Quicktime. And if you're running Linux you have two options. Either A. use a reverse-engineered codec that might work sometimes, might not others, and almost never at the same quality as the original proprietary one, or B. use the proprietary codecs. For most, the choice is simple... either stay absolutist and for go most of the multimedia content available to you, or enjoy your content and get over it. As the people willing to boycott this content is an extremely small minority, this mindset accomplishes nothing. If there were a considerable number of users willing to do without; say about 15% or more of the entire desktop PC user market, then it might inspire change but right now, it's hurting our cause overall more than it is helping.

When we try to convince current Windows or MacOS users to switch to Linux or any other open source operating system...they're going to want to know that they can continue to watch all their videos and listen to all of their music they currently enjoy. When we have to stop and explain the current complex situations our stubbornness is causing, it makes them question the possibility of switching even more than they would have before.

The same thing goes for certain hardware drivers. If someone has an nVidia or ATI video card, and they want any decent 3D hardware support, then their only current feasible option is to use the proprietary codecs. However, once again, many distributions do not install them by default, thus making a new user feel like Linux is incapable of the same 3D graphical performance they've experienced on their machine in Windows.

These codecs are interesting though in that the manufacturers have gone out of their way to make sure they work on Linux systems and integrate with the kernel properly. These codecs may be proprietary, but they are freely redistributable, so there's nothing holding distros from releasing them with their systems by default. In the case of video codecs it is not so simple. Most of the solutions available to play proprietary multimedia codecs under Linux are in questionable grey areas as far as legality go, so of course no one would want to get into legal troubles just for these codecs. However, there is still another option, the companies that make these codecs are all willing to make them legally available if the distro developers were willing to make the proper arrangements.

A company named Fluendo recently licensed the rights to release a free, legal codec for Mpeg/Mp3 multimedia files for any Linux distro. Any distro that wants to include Mp3 support out of the box now can, yet there are still some who won't because it's a proprietary codec. A few distros like Linspire/Freespire are taking a proactive route on this issue though. They have gone the extra mile to make sure they can legally include support for things like Quicktime, WindowsMedia files and encrypted DVDs. They also include all the proprietary drivers, as well as Java and Flash. Flash, once again...while proprietary is freely distributable, but many distros still choose not to.

Java is a very interesting case as Sun has announced that they plan to make it open source within the course of the next year. This is good because there is a large number of programs developed in it, and things like GCJ just don't cut it when it comes to speed and quality.

It's not surprising to me that E.S.R. is an advisor for Freespire now with their pragmatic middle ground stance. I for one fully intend on giving it a shot in the near future. I've always heard nothing but negative comments from people about Linspire (once called Lindows) from all of the other Linux users I know, but I think it's time to find out for myself. With this realization and the fact they now have a community driven, free edition I think it's time. However, since they've only released an initial offering that doesn't vary much from prior versions of Linspire, I think I will wait for their upcoming 2.0 release this spring.

For now I'll stick with Ubuntu, and continue to jump through hoops just so I can use my computer the way I want to... If only they would take a similar approach maybe they'd be able to make greater strides against their so called "Bug #1".

Update: I just found an Ubuntu derivative called "Linux Mint", which is basically just Ubuntu plus Flash, Sun Java, and all the proprietary codecs (including DVD) built-in by default. This seems great, but most likely it's still in that questionable legal grey area and just being hosted from a country that doesn't care about things as much as places like the US do...


Digg Annoyances

So everyone's claiming Digg is going to overtake Slashdot, but I'm not so convinced... I've read it occationaly for the last couple months. Then the other day I noticed them promoting some new features, and upon clicking the link I was welcomed with a "You must upgrade Flash" notice as I keep seeing all over the web. The problem is that Adobe has still yet to release any version of Flash beyone version 7 for Linux users. So, basically Digg has made it clear to me that Linux/Unix/BSD users are pretty low on their priority list. I'm tired of being spit in the face of by all these sites, and I expected better from the so called tech site that is Digg.com.

Then today, strike 2 happens. I noticed an aricle talking about how AMD is considering open sourcing their ATI drivers and that there were no articles about it on Digg yet, so I made my first submission. Well, oddly the "diggs" came alot slower than I expected for such an important topic, and when I came back to check if it had gotten enough votes to show up on the front page I found that it had, BUT it had been marked as a duplicate by many users... some even felt the need to tell me in the comments, and then one of them put a link to the supposed post that I duplicated. Then I quickly saw that the article with much more diggs was posted over an hour after mine was, yet somehow they shot up and now I was being marked as a duplicate.

Sure, this is something pretty silly to get upset over, but it's also very disheartening to have this happen on my very first submission. Being the guy that I am, I quickly came up with a simple solution. Whenever someone marks an article as a duplicate they should also be required to provide the URL of the article that has been supposedly duplicated. Then a simple script can check the times they were posted and determine who the real dupe is. Then I guess they can even take it one step forward and place all the duplicate reports on the actual dupe instead of the real original. It sounds like a lot of trouble for something so silly, but it would have meant a lot to me if my first post wasn't treated so poorly by something that could be so easily corrected :(

I've submitted a bug report for both of these issues, but I'm not holding my breath that they'll actually do anything... so for now, I'll stick with Slashdot ;)


GPL v3 & DRM

From what I've seen so far, the biggest change to the upcoming version 3 of the GPL is inclusion of anti-DRM clauses. This is the thing that has so many people up in arms about what Stallman and the FSF are doing. To me, this makes no sense. The GNU GPL and LGPL are the most commonly used open source/free software licenses, but this move could change that. The way I see it is that DRM is a feature of the software, and does not belong in such a license regardless. The whole point of the GPL license is for development and to make sure everyone plays fair with their use and contributions. Something like DRM should be a decision made by the developers, not the license their software is distributed under. One of the biggest reasons people like OS/FS is because it offers them choice, and this move is restricting choice. If the Free Software Foundation feels this strongly about the issue they should offer two versions of the GPL v3, a standard GPL with no mention of the issue and then a "GPL-no-DRM" license. Since the GPL license itself is open to change as well, the clauses to prevent DRM from being used in a particular piece of software should be listed as optional, and a developer can include them if they so choose, but the standard license has no business doing this.

My final verdict on the use of DRM is still out. As while I don't usually like it and I think it can be a major pain in the ass, I still can see why content producers would want such technology. Technically the use of serial numbers, which has been done for decades now, is a form of DRM, and I never really minded it. It's just when the DRM system gets in the way of me legally using the things I've bought that it becomes a problem. And software licencing isn't going to fix this, it's going to be up to us convincing the content producers to use better practices.

I read a very interesting article today. I'm now not so sure of my prior thoughts. It seems the DRM that the FSF is worried about isn't necessarily the type placed on media files, but rather for in situations like the TiVo, where they have made it almost impossible to modify their GNU/Linux based software by use of a verification key, but are still technically legal according to the GPL v2 :(


Firefox Media Player

Someone really should make an open source, cross platform, embedded media player for Firefox, or any browser really. Something that works just as reliably as the embedded Windows Media player works in IE (and sometime in FF) or Quicktime or Real Player. Numerous sites these days love to use embedded audio/video players as I guess it makes them feel "safe" from people saving their files or something? Perhaps they just think it's more intuitive or easy to have their videos and audio files playing within a website instead of using an external application? Anyway, regardless of their intentions, tons of sites use this stuff, and it's one of the biggest thorns still left in both Firefox and Linux's sides. It's also one of the most visable problems that an average joe user will run into, and sometimes can make people decide to switch back to whatever proprietary setup they had before. Sure, embedded players work "most of the time" in Firefox for Windows, and there's things like the mplayer-plug-in or the VLC mozilla plugin for Linux users that usually seem to work, but it just never seems to work quite as well as they're intended to. Of course, this is most likely such a problem due to the widespread use of proprietary media formats and streaming protocols, but there's probably more to it than even that.

So, what I think we need, is a general purpose, embedded media player for web browsers. One that's very generalized so that it can work with numerous backends, any operating system, and preferably any web browser, or perhaps at least with any browser built on top of mozilla/gecko. This way Linux/Unix users can use whatever their media player of choice is for a backend, be it gstreamer, xine, vlc, mplayer or whatever and of course it could do the same on both Windows and Mac versions too. And no matter what system you view a webpage on, it'll work the same way.

By unifying the way embedded media is handled across all platforms, it'll probably make it easier for web developers, but more importantly it should make it much easier for users, which intern benefits open source in general as well. I'm not sure if there is already a W3C standard for how to code in embedded media, but there definitely should be if not already.

Most sites these days have recently started using Flash as the media player of choice, as it's available for the most platforms and browsers, but watching video through it has some severe limitations. The first thing anyone will notice is that no matter what system or browser you use, there's no way to truly watch a video in fullscreen, although quicktime also shares this limitation and Google Video has tried to create some sort of solution. For Linux and Unix users, Macromedia doesn't seem to find us that terribly important, as they still have not released Flash 8 yet, even though it's been out for Windows and Mac for well over 6 months now. Not only that, but just about every Linux user I've ever known has a perpetual problem with Flash video and animations getting out of sync from the accompanying audio. In addition to that, there is still no 64bit version of Flash available for 64bit Linux users either, resulting in weird hack arounds to get the 32bit version to run correctly, that no newbie or novice users could ever hope to easily achieve. Sure, flash video seems like a good idea at first, and I see why many sites would think it their best option, but it's just not cutting it.

Now of course, personally, I would much rather watch all video and whatnot through an external application that does not play till I click a link for it specifically, but it appears that my point of view on this may be in the minority...hence why I'm making this proposal ;)

One more suggestion I would make to the developers of this currently fictitious project is to allow for multiple gui configurations... or skins, if you will. I reluctantly say that because it's not in the idea that people have custom GUIs for aesthetic reasons, but because different people will likely want their embedded player to have different control setups, and this could help prevent unnecessary forks and fragmentation. Suppose one user wants a really simple bar across the bottom with a play/pause button, progress bar and perhaps a few other features, while another user wants all the bells and whistles you could imagine, while yet a third user doesn't want any graphical controls available till he right clicks on the player and chooses them through a menu.... are you seeing where I'm going with this? As while the project should ideally support different backends, it at the same time may also want to support various front ends as well...or to use the trendy term, different skins.

As a reference, even though I love VLC, I still use the mplayer plugin for my browser as it has convenient controls available, as while the VLC embedded player doesn't seem to having anything to it besides the ability to go into fullscreen, not even so much as a pause feature. Yet at the same time, I wish mplayer plugin had a volume control and I'm sure many other people have many other ideas, so once again...I feel that abstracting the player's control scheme could be very beneficial to prevent people from possibly forking over something trivial.

Unfortunately, like most of my ideas... I do not currently have the knowledge, time, nor skill to start such a project, so hopefully someone more capable will be inspired by my idea.


OpenDocument Firefox Extention Idea

If it hasn't already been thought of... someone should really think about making an extension for Firefox that would allow you to view OpenDocument files, at very least .ODTs.. it wouldn't have to be able to edit them or anything, although I suppose that could be a plus too, but just the ability to open and view them from within your browser would be really nice and might help the format take off a little more ;)