Why Hasn't Apple Released Quicktime For UNIX?
"There are a couple of solutions to this. The Open-Source xanim software already contains support for the QuickTime file format, and has an API for importing CODECs in binary format. If Apple doesn't want to release the source code to the Sorenson CODEC (which is understandable given the competitive market they're in at the moment), and if they don't want the cost and headache of supporting an additional product for the last 15% of so of the market, they can release the Sorenson CODEC in binary form using that API and let the Open Source process go the rest of the way towards enabling Linux to play QuickTime 4.
There's a better solution, though: Darwin is based on BSD UNIX; Apple is now full of UNIX developers. You can't convince me that there isn't already a version of the QT player with an Xwindows GUI compiled and floating around inside Apple; The absence of a UNIX player in that kind of development environment just doesn't make sense. So how close is it to release?
Let's find out: What pressure can we apply to Apple to let them know that there's sufficient demand for them to release the software?"
"Elementary, my dear Watson. We simply write our own CODEC, open source of course, and include a copy of the decoder in Netscape. We'll use the top of the line audio CODECS, such as MP4, and derive a new, high-quality video CODEC. Our system will become the de-facto standard, because most people will have a copy. After that, Apple will simply have no choice but to release the source for Quicktime, or risk watching it die."
"But Holmes! Video CODECS are so hard to develop!"
"We've some of the premiere mathematicians from a wide range of Universities and corporations, who could assist us in our quest. None of these corporationshas a vested interest in paying someone else for something they could obtain for essentially nothing. Besides! It would be a good mental exercise for the community."
"Where do you suggest they start, though?"
"Oh, Watson, Watson! Think, man! You're trying to compress natural images. Natural images can be defined as regions of similar colour, and regions of shade. By reducing the image to a grey-scale picture, a set of colour maps, a set of brightness maps, a set of contrast maps and an index of where the maps are overlayed, you've vastly simplified the image. Each of the maps can be compressed using delta shift encoding, because they'll all be very similar. You should be able to run-length what's left. The grey-scale picture could probably be compressed in any number of ways, including MPEG, or some lossless format. A lot of the more important information will already have been extracted, so it doesn't matter if you lose a bit. The locations of where the maps are overlayed won't take any serious amount of space, so that can be ignored."
"Wonderful, Homes! I didn't understand a word of it."
"That's quite alright, Watson. I didn't, either. Computers won't be invented for another 75 years, and computer graphics of this kind will take another 50 years to become of serious interest to the average person. Even with my great intellect, I cannot deduce the requirements of technology 125 years in the future."
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
QUICKTIME IS NOT A CODEC.
QUICKTIME IS NOT A CODEC.
Go write that on the chalkboard five thousand times. If I see one more post mistaking QuickTime for a codec....
Now, what is QuickTime, since it's not a codec? The answer is that it's a media layer. Video is what it's famous for, but that's only one part of it. It also does sound, images, and a fair amount of other cool stuff.
Apple could Open-Source this with little trouble, actually. It actually has to a small extent; the open MPEG4 standard has a file format based on QuickTIme's. What gives QuickTime its "IP value" are the codecs involved. I'd like to see those Open-Sourced, but let's be realistic; Apple isn't do that. But they could still release these as binary plug-ins (they already do this with the Intel codecs on MacOS and Windows, so it's certainly possible and wouldn't even disrupt the current QuickTime structure). One would think they might be convinced to even Open-Source some of their older codecs, like the Animation and Video ones, though that isn't as certain.
An Open-Source QuickTime would be a Very Good Thing for Unix/Linux/whatever. It would provide the DE's (Gnome, KDE, and whatever else might come up in the future) with a very versatile and proven media layer. It would also help to create an open standard for such things.
Oh, and for those who gripe about the QT4 interface, three points...
- Yes, the current interface does suck. Talk about form without function.
- The OSX version is much-improved; check out this page for a screenshot of the OSX player (basically, QT4 minus the fluff and with the volume thumbwheel replaced by a slider again).
- QuickTime is only an API; it's not hard to code a new player. If you use QuickTime's default controls, in fact, you have the old QT3 interface.
But anyway, that's enough of my rant. There are really only a few more things I'd like to see in QuickTime anyway...- MPEG2 and MPEG4 support. There's particularly no excuse not to have MPEG4, seeing as it's based off of QuickTime's file format for crying out loud.
- RealMedia support. Real's servers use RTSP, which QuickTime already supports, so it's just a matter of licensing the codec.
- MPEG (and hopefully MPEG2/MPEG4) export. As it is, the only reason I can think of for it not being in there would be if Astarte and HEURIS are paying Apple not to put it in, so they can keep charging obscene prices for their own encoders ($400 for Astarte's M.PACK). Still, there's always Movie2MPEG (an OSS encoder for MacOS). It's very slow but it gets the job done.
Those, plus the Open-Sourcing of at least the layer (if not the codecs) would just be icing on the cake.Don't think so. From the xanim homepage:
I have contacted Sorenson about licensing their codec. They responded that Apple won't allow them to license it to others.
It's currently fashionable to think of Apple as the anti-Microsoft, and thus the enemy of my enemy is my friend. Unfortunately, I see little evidence that that's the case.
We do this sort of software well. What is holding us back here seems to be patent and trade-secret issues.
Thanks
Bruce
Bruce Perens.
As I understand it, Apple and Sorenson have an exclusive license agreement regarding the Sorenson codec. If this is the case (which makes sense, otherwise we'd see somebody else with the codec by now) I don't see how petitioning will do anything. Apple and Sorenson are bound by their contracts. And I'm sure Apple will pay any amount to keep that exclusive agreement.
_________________
rooooar
I think there is a bit of confusion here. Asking Apple to port QuickTime to UNIX is **NOT** the same as asking them to Open Source it.
Let's get real. Expecting Apple to Open Source QuickTime is pretty far fetched. WHY Would a company want to OpenSource such a lucrative revenue stream (particularly, one that lets them make money from Windows users; ie, ones that already tied into the Apple platform.)
It IS reasonable to expect them to PORT it to UNIX. It's simply more money for them to make by increasing their market share.
The fact is, Porting to UNIX != Open Sourcing. You can port something to UNIX without opening the source. See Q3A and the other Loki Games efforts if you don't believe me.
Note that this does not take into account the ideological reasons for opening the source to QuickTime. I'm sure there are some benefits to be had by doing so (some have noted the poor interface for the player could be improved.) I'm jsut speaking to the practical considerations, as well as to the mistaken concept that Porting = Open Sourcing.
In short, expect them to Port QuickTime, not Open Source it. That sort of confusion has to stop.
At the beginning, the complaint was simply "Macs suck." Then Apple bought NeXT, started bringing itself back from the "dead," etc. The /. response? "They're not Open Source!"
Then Apple Open Sourced Darwin, the basis for Mac OS X. They also brought out related projects, including the Quicktime Streaming Server. This time, the complaint was "The License is Bad!"
So they revised the APSL, satisfying some but not others. So people started complaining that the Quicktime client wasn't available for Linux. And that's where we are right now.
So what's the response going to be if there is a Quicktime for Linux? That it or some of its codecs (which Apple mostly doesn't own) aren't Open Source? That the license still sucks? That [insert rant here]? Will people finally respect the company? Or is the truth that they just don't like the name "Apple?"
There are those who think that complaining will get them far in life. I'm not saying that everyone should just be satisfied and never ask for more than they have (although it is good to count your blessings), but Apple's not the type of company that can be "pushed around." Neither are its customers who are still somewhat defensive after years of being bashed by the rest of the computing world. So tone it down a bit, I have far more respect for a debate than a shouting match.
By the way, I think it's interesting to note that the Quicktime format is the basis for MPEG-4, which like all the other MPEGs is a standard for everyone to adopt. At the very least, MPEG-4 players for Linux won't even be a question.
-Rafi Remove the Spanish to email me.
...
.movs, and streaming Quicktime in real time.
IIRC, Apple does not own ALL of Quicktime. The Sorensen codec (the most important one, BTW), is a private, propietary codec that Apple LICENSES from the Sorenson people.
Sorenson is the codec that gives us the astounding quality of the recent Star Wars trailer
Quicktime without Sorenson? May as well settle for realplayer... that'd be going back to Qutcktime version 2 or something.
We need to petition Sorenson to open the codec, *NOT* Apple! If Apple DID release the QT source, they'd either have to exclude Sorenson (a waste, really), OR they'd expose themselves to tremendous lawsuits (don't count on them being this stupid). This will not change until Sorenson changes the terms of THEIR license.
So, until Sorenson opens up.... don't count on Quicktime for Linux, Unix, or anything else.
BTW:
All of the above applies ONLY to the QT MoviePlayer. Quicktime as a whole constitutes a LOT more than just the player. Read the developer guides sometime. It's really quite nifty.
I doubt, even if Sorenson DOES open their codec, that we'll EVER see ALL of Quicktime open sourced. In doing so, Apple would essentially be handing a tremendous amout of technology to gates and his cronies, free of charge. I doubt that'll ever happen.
john
Imagine all the people...
Yes. MacOS X (pronounced "ten" not "ex") is based on BSD running on top of the Mach microkernel. Apple's first major experiment with Mach was MkLinux, which Apple has since abandoned. The Mach+BSD part is called Darwin. It's free an open source. Darwin is a complete OS in its own right, and can run by itself. It has no GUI, though John Carmack has written a port of XFree to work with it. The rest of MacOS X is closed-source. It includes Quartz, which is a way-cool completely vector-based display engine for the GUI. It has Cocoa, a native Java API, Carbon, really left over from Rhapsody, which allows classic apps to quickly take advantage of OS X features. The classic MacOS 9 environment can boot in a window as a single process, and you can run classic apps side by side with OS X apps (rootless, as it were.) More info about the higher-level stuff is available at Apple's MacOS X site.
Huh? Apple has always had a stand-alone installer available. Qucktime 4.1 for Windows and MacOS is available here:
http://www.apple.com/quicktime/dow nload/support/
You still have to fill out some information first, but it'll let you download the full, "real" installer (one that doesn't require an internet connection to install). This download is web-based, however... I'm not sure if they have it on their FTP site, I'll look into it.
By not releasing Quicktime for Linux, Apple
has (probably unintentionally) controlled
the method by which people can access their
product.
Along the lines of the earlier copyright
article (and 'fair use' laws), I say we
re-define what it means to be a product.
Today, a product or service
is typically composed of:
1 - Something that is desired by the consumer
2 - Resrictions on how that 'something' may
be used.
If we, as consumers, can't get companies to
stop including #2 as part of the 'product',
then we should at least be able to
place restrictions on how the companies use
our money (half tongue-in-cheek). Since customers
usually don't know what the restrictions will
be until they PURCHASE the product (a major
problem, imho), we should likewise be
able to place 'money usage' restrictions on
the companies from which we buy products.
Imagine this scenario:
- Bob buys an electronic book from
Barnes & Noble, along with the restriction
that he can only view it with certain
software and cannot print it out and
share it with his friends.
- Bob enters his 'encoded' credit card
information to pay for the book.
Barnes & Noble will be allowed to use
the money obtained ONLY in ways that
Bob has specifically encoded into
his credit card info.
Actually, refining this idea further
could be very interesting...hmmmm...