Slashdot Mirror
DVD Authoring In Linux
leming writes "This article in Linux Journal explains low-cost DVD authoring available in Linux. Traditionally, DVD authoring has been an expensive affair. Full-featured professional applications can cost thousands of dollars, while cheaper products, such as Apple's iDVD, have arbitrary restrictions that significantly reduce their usefulness. A new open-source effort, dvdauthor, is bringing the possibility of low-cost, professional-grade DVD authoring to Linux. Although it doesn't yet support all the features of the DVD specification, development is proceeding at a fast pace, and new features are being added with each release."
Now if I can just get the woman to let me buy a DVD burner and a nice DV camcorder I can get started on that professional homemade pron I've been waiting so long to make.
When I tell an object to delete this, am I killing it or telling it to kill me?
And for open source in general, if we can provide good open source solutions to things that until now cost alot of money, it is one of the strongest ways of supporting open source, providing however that the product really does do what is needed, and does it well (which in most cases i've seen with similar projects so far, was done extremely well, sometimes even better than the commercial product).
Trick out home videos with a fun, featureful menu system that viewers can navigate from a regular DVD player.
Traditionally, DVD authoring has been an expensive affair. Full-featured professional applications can cost thousands of dollars, while cheaper products, such as Apple's iDVD, have arbitrary restrictions that significantly reduce their usefulness. A new open-source effort, dvdauthor, is bringing the possibility of low-cost, professional-grade DVD authoring to Linux. Although it doesn't yet support all the features of the DVD specification, development is proceeding at a fast pace, and new features are being added with each release. Together with a more established open-source toolkit, mjpegtools, this article explains how to construct a relatively complex DVD application, a photo album, with dvdauthor. We also illustrate the various features that dvdauthor currently supports and how to use open-source tools to construct a DVD-R that can play on almost every DVD player.
How a DVD Works (Quick Version)
A DVD is comprised of one or more video title sets (VTSes), which contain video information in the form of MPEG-2 video streams. Each disc can have up to 99 VTSes, and each title set can be subdivided further into as many as 99 chapters, allowing DVD players to jump to a certain point within the video stream. Within each VTS, a DVD can have up to eight different audio tracks and 32 subtitle tracks that the viewer can switch between at will. A menu system can be included within a title set, allowing the viewer to select between the different subtitle and audio tracks. An optional top-level menu, known as the video manager menu (VMGM), is used to navigate between the different title sets. One VTS may contain a feature film and another may contain a documentary on the film, and the VMGM allows viewers to select which one they want to watch.
The DVD format doesn't eliminate the differences between the two competing broadcasting formats, NTSC (primarily used in America) and PAL (the standard in Europe and Japan). I live in Britain, so the frame information and resolution details used in this article are for a PAL system, but I point out the differences you need to be aware of when they appear and offer appropriate settings for an NTSC disc.
The DVD specification includes advanced features, such as the concept of region coding, the possibility of viewing different angles of a video stream and simple computations using built-in registers provided by a DVD player. I don't know much about these features, and they aren't discussed in this article. The dvdauthor mailing list is a good source for further information.
Planning
Before we rush headlong into creating menus, subtitling and multiplexing, it's a good idea to sketch out the structure of the DVD with paper and pencil. Proprietary DVD tools offer GUI systems for creating this type of structure, but no such tools are available yet for DVD production on Linux. As you'll soon see, the command-line tools have a lot of different options, so having your ideas on paper is preferable to trying to keep everything in your head.
The DVD application I'm creating is a photo album, using pictures that I took while studying abroad at UNC-Chapel Hill this past year. For simplicity's sake, I have only six photos in each category. On paper, I decide that the main menu (the VMGM unit) should have five buttons, four of which are simple text buttons (one for each different photo category), plus a secret link unlocking extra pictures (secret extra features are a common occurrence in commercial DVDs) and a music track playing in the background. The four regular buttons link to one of four menus, one for each different section. The menu system for each section consists of two menus and an audio track, with selectable preview images of the slideshow, a button to move onto the next set of preview images and two buttons that allow the viewer to watch the complete slideshow or go back to the main menu. To keep things simple, the photo slideshow s
(And I'll preface this with a comment that I've only lightly skimmed the article.)
I've seen and heard, over and over, ad nauseum, programmers and OSS advocates continually ask me why I need more or say they don't see why people need all the extra UI features, since something like this can do it all. I'm glad to see this on Linux, the problem is that it still needs a good GUI before it'll get used by many people for DVD authoring (it might be used for the actual production process, but not as much for authoring).
And, as I go into this, we reach one of the primary problems with OSS. If you're a computer person, a program like this is easy to use, since you're used to command line switches and piecing everything together by adding in images from Gimp and pulling in fragments from a number of programs. On the other hand, if you're a video producer, your focus is on CREATING THE VIDEO and PUTTING YOUR VISION on DVD. It is NOT on using the tools, figuring out the command line arguments, or other activites. While I've spent several years working as a programmer, my primary function is writing screenplays. Until OpenOffice came out, there was no word processor on Linux where I could run the program and focus ONLY on writing my screenplays (part of that is the need for macros to handle margin changes). Programmers and other computer people would say, "But you can do this with all the OSS word processors," and I'd say, "I can, but that means I have to spend my time thinking about HOW I'm doing my work, instead of focusing on the work itself."
I love OSS, I love Linux. I know GUIs are a pain (I don't exactly love writing them myself), but for OSS to be used by professionals, like writers or DVD producers, the software has to be so easy to use that the end user can focus on their work instead of focusing on figuring out the software.
Don't get me wrong -- it's great we're this far along. If I had time, I'd love to write a GUI wrapper, but I don't, and it'll be a long time before there is a GUI that makes this software accessible to DVD producers who just want to run the program and use it as a tool to help them realize their vision.
Finally, an article that outlines, with examples, how to create a dvd under linux. not very in depth, but enough to get you going. sure, there are alot of tools out there for making/burning dvd's under linux. i've been searching the net for months trying to find info on how to create dvd's. in the process i've found many different projects mature enough to use. yet none seemed to provide enough information to actually produce a working dvd. the few that did explain how the program worked failed to provide examples.
maybe i was looking in the wrong places or for the wrong thing. but to find nothing helpful enough is odd, to say the least. i for one feel this area has been overlooked as far docs, howto's, guides, and tutorials go. i'm no newbie, but i no longer have the time or the money to fool around with it till i get it right.
of course, if anyone has links to other articles, faq's, tutorial's, howto's, etc.. please let me know!
A DVD "is comprised of" ...
I wince every time I hear that.
OK everyone, the parts comprise the whole, the whole is composed of the parts. Got it?
and while we're at it...
I thought MPEG-2 video compression and AC3 audio compression were patented in major portions of Regions 1 and 2.
Will I retire or break 10K?
This is just the leg-up I was looking for! Thankyou Slashdot! Now off I go to get my hack on. How much you wanna bet that in six months there'll be a half dozen GUI front ends for DVD authoring? Who's going to go try and make one ... show of hands!
[signature]
All you need is $20 worth of crack and you can get anything you want.
Karma: The shiznight, mostly because I am the Drizzle.
I thought the same thing until I started using Amanda. We were this close (||) to buying a whole Arkeia setup to backup about a dozen workstations to a VXA-1 tape drive when I decided to give Amanda a try, just to see if it could save us ~$1200. I read about the tape spanning thing and said, "Wow, that sucks. This will never work." But then I tried it, and all I have to say is, "Damn, amanda is smart."
Basically, once you get your full backups done, Amanda dynamically schedules incremental backups to most efficiently use tape and protect your data. With judicious selection rules about what constitutes a "disk" (e.g., a single disk must fit on 1 tape), and a properly tuned set up, the no tape spanning thing is not a problem. We are currently backing up a total of 300GB (180GB compressed) of data from 24 "disks" on 12 workstations onto a set of 20 33GB VXA-1 tapes with the ability to roll back to any state in the last 16 days.
That said, tape spanning would definitely rock in amanda.
Just as Director has Lingo, which is based largely on Smalltalk, the software should be fully extensible with a robust programming language and or a XML based scripting language. Incorporate a little video. Bam! Flash killer. If we start working now we can even beat Sparkle to market.
Aww, it doesn't say how to add CSS protection.
"Believe me!" -- Donald Trump
Y'know, it is starting to get on my nerves, this endless churning agon regarding graphical interfaces and Linux, or any other software for that matter. I, and frankly many others of my acquaintance, find a text interface both more functional and more intuitive than menus and buttons. C'mon, we've had a thousand years to get comfortable with text-based interfaces. A graph is a graph is a character, eh! I think with words by habit; I remember text and its meaning much more readily that the location of the farking widgit in the Husker Du of menus most GUI programs make available.
Here is my Dad when I introduced him to his new Ibook w/Aqua GUI: Do I click the word or the picture?
Here, developers, is a friggin clue: GUI's are no more fun to learn and use than they are to write!!! Unless I am doing something with a picture, pictures are not to the point. Somewhere in the mists of time, some arrogant HCI genius decided that unlike L33t programmers, normal Joe Sixpacks thought with pictures instead of words and its been downhill ever since. When is it ever easier to manually open and fiddle with more than two objects when a script can do it?!? When!?!
Has it not been amply demonstrated that the mouse is the most anti-ergonomical device in the modern office?!?
Ahrrrrrrrgh!
Thank you, and I apologize for the inconvenience.
illegitimii non ingravare
It feels disorienting to agree with nathanh, but nevertheless... "Hear, hear!" (-:
Got time? Spend some of it coding or testing
Take it outside, god boy.
...ripping existing DVDs, nuking the region coding, and writing them back. Anyone got any favourite words on that topic?
Got time? Spend some of it coding or testing
/ME waves from Region 4.
Got time? Spend some of it coding or testing
OK, so what a writer really needs is a generic text editor - formatting the output should not be part of the writing process (unless you are doing an ASCII art type thing). The writer should be worried about paragraph breaks, chapter breaks, etc.
A text editor is a text editor. If you are authoring long works, you want a document processing system - things like LaTEX - things that separate formatting and output from the creative input.
You claim you needed OO with a GUI and support for macros to change margins. What you really needed OO for was a tool that allows you to procrastinate while looking like you are doing something.
The writer should write, the editor should edit, and someone in the publishing side should handle the layout. (Yes, I understand that all these functions may be one person, but the idea is valid and the tasks should be separate.)
. there used to be a sig here.....
My Dear Hermit,
I had intended merely a throw-away rant. I appreciate your detailed response and feel you deserve a more considered explanation of my views. That your reply has been moderated indicates someone besides us is enjoying this thread so I will take a moment to expand on my remarks and your rejoinder.
I should like to correct your prejudice as to my professional situation. I am a writer by training, a poet, and a film-maker by avocation. It is true that I am making my mortgage doing XML/XSLT plumbing with Java, but I have no formal training in computer science.
I should also point out that my rant overstated my position somewhat. I don't believe that GUIs should be outlawed. I contend that most GUIs are cluttered, mal-organized and require more training to use effectively than a comparable CLI tool would. I feel that the proper balance between graphical and textual tools in HCI has not yet been struck. I contend that the current Linux toolset is closer to a proper, natural balance of these assets in addressing general purpose computing than either Mac or MS at this time. I take the recent introduction of a CLI for Apple machines and the announced return of CLI tools for Windows to corroborate my view. Your mention of *text-based* GUI tools is quite apropos here. The expanding directory tree is one of the few really successful graphical representations we've managed.
I take a more nuanced view of the computer as a tool and its relationship to the user. Central to my thinking is an impression that the popularity of graphical interfaces is not a result of their intuitive ease, but consequent to a particular historical moment in the development of our relationship to the machine. Users did not charge from DOS to Windows. The market dramatically expanded at that moment as people who formerly resisted the notion that computers were useful to secretaries, truckers, actuaries and writers were suddenly convinced that computers might be not only useful, but something crucial and not to be missed. We have a generation now who have conformed their working styles to the demands the machines made of them at that moment and now do find them *intuitive*.
That is, to say that GUIs are good because they are popular is like saying Microsoft is good because its popular. You yourself call MS Office *crap*. In reality, either are popular because there is no practical alternative. And these two developments are closely related and a direct result of this flood of naieve users coincident with the release of Win95 and the development of the internet. The internal dialogue goes something like this: I don't know computers and don't want to know. I want something easy to use and I need to be able to open the Lotus and Word files people send me. Whether a mouse and graphical menus were, in point of fact, easy to use, did not enter into it. Users had no context within which to make an informed decision.
As to *most people* requiring a GUI to work with computers easily, I believe that remains an open question. I have myself overseen the transition of a 100 person compositing department from CLI tools to Quark at a halving of productivity. The rationale for this disastrous move was the *ease* of training new personnel. The ease of training existing personnel to the GUI interface was not considered. In fact they have been efficiently bringing more people on quickly, which is a good thing considering that the department nearly doubled in size without additional business to offset the expense. I left the company over the disaster.
You mention cameras and the example is fortuitous for several reasons. They clearly point the way forward for the penetration of computers in daily life. The embedded processor in your camera has, as its interface, the device itself. Its functions are so closely aligned to the use of the machine that traditional HCI concerns are moot. I think that this type of computing interface will be by far the most common going forward.
illegitimii non ingravare
Now, not only will I have to suffer my friends and families holiday videos, but there'll also be the 'behind the scenes' footage, deleted scenes, and the dreaded 'Directors commentry'.
Posted from my lappie, sitting under a nice shady tree... is that good enough?
Summer over here now. 26degC in the kitchen at 08:56 and rising. I love a sunburnt country. (-:
Got time? Spend some of it coding or testing
...pointless comments are acceptable. (-:
Got time? Spend some of it coding or testing
See you in a couple of weekends, I'll be over there for 7th & 8th Dec. I can bring samples of new limestone formed in the last few decades, if you like. (-:
Got time? Spend some of it coding or testing
What a beautiful thread that was Grandpa.
Is that what the internet used to be like?
Sure was son, but now the internet is full of short, poorly composed, shiney thing, ADHD friendly crap. Just like this POST we are part of right here son.
This is where I keep my clever quotes "" Yup I only got a pair, so I better not waste em!
Though not a videographer or even a professional photographer, I do love photography and am thinking of going into photojournalism, and just as those professional photographers I did know, I prefer to use manual controls rather than have it all automated. Not that I don't use the autmated settings, because I do use them, but for the most part using manual controls allows the most flexibility and creativity even if they aren't the fastest.
You taught learning disabled students? Though to my knowledge none of mine had LDs, I used to tutor and between myself and tutoring others know how different people prefer to learn in different ways. Previously my own prefered style was to do it as much as I could on my own. I'd first experiment or "play around", when I came to a point where I was stuck and couldn't figure it out on my own I'd rtfm and as a last resort I'd ask for help. But now I suffer from a TBI, Traumatic Brain Injury, and need someone there almost from the beginning. This really gets to me, because though I love to help others I hate needing or having to ask for help.
Should there be a Law?