Slashdot Mirror
Escient (CDDB company) trying to monopolize market?
4. Grand Conditions and Terms
a. You agree to perform the following, which are conditions of the license grant to you:
i) You shall program and design the Player such that, by methods stated in the CDDB protocol:
(1) For a minimum of 2 seconds or during the time the Escient-CDDB Database or Internet Server is being accessed, whichever time is greater, each copy of the Player shall display the CDDB logo as specified by Escient from time to time;
(2) Such CDDB logo graphics will be provided by Escient. The display of the CDDB logo shall be of a size of at least 20 pixels high by 33 pixels wide. The display shall include the phrase "Accessing CDDBtm" or other similar phrase specified by Escient.
(3) Each copy of the Player shall include on its user interface a readily visible "CDDBtm Link" icon button that, when clicked, will launch the user's web browser and link to Escient(R) CDDBtm Web site at www.cddb.com;
(4) Each copy of the Player shall include on its user interface a reasonably legible mail icon (preferably a letter or mail box icon) that, which clicked, will cause the Player to email CD-related artist, title, track length and track name information that has been entered and/or corrected by the end user to CDDB at the address stated in the CDDB Protocol;
(5) Each copy of the Player when accessing the CDDB Server to obtain Data or when sending Data, shall do so directly, without first accessing any other site;
(6) The Player's functionality with regard to accessing the CDDB Server will fully conform to the specifications of the CDDB Protocol; and
(7) Each copy of the Player will include the Proprietary Legends in the "About Box" and "Help File" of the Player. The Help File on the Player will also provide to the end user an explanation of the CDDB functionality, as specified the CDDB Protocol.
(8) You agree to use the CDDB Database as the exclusive source for CD information whenever the access to such DATA is initiated by a process within your CD player application that reads CD TOC Data and retrieves Data via the Internet that is related to the CD. The term of your exclusive use of the CDDB Database will be for the term of this Agreement.
6. Negative Covenants and Restrictions
a. You further agree as follows:
i) You will not use or exploit the CDDB Database. Data derived from the CDDB Database, End User Data, or the CDDB Server, except as expressly permitted herein.
ii) The Player under this Agreement may permit the End User to aggregate Data only (a) in his or her personal computer (b) only in response to the CDs placed in his or her computer.
iii) You agree that your Player shall not have or enable functionality that uploads or permits the transmission of Data to anyone other than Escient. You agree not to upload, aggregate or collect Data derived from the CDDB Database or End User Data.
iv) You will not use the Data derived from the CDDB Database or End User Data to enable or direct the routing of linking of End Users to content or services accessed on the Internet that are related to particular CD titles, tracks, artists or music categories.
v) You may not use or permit the use of the trademarks, trade names or other designations of Escient except as specified herein, or as may be stated in Escient-supplied trademark use guidelines that Escient may issue from time to time.
What is CDDB, and what is it used for? "Online disc lookup" - whuzzat?
How can a company "own" a format? Sure, a company can own a *trademark* (like IrDA) and charge for use of the trademark; but I've never heard of anyone "owning" a data format. This is bizarre.
Basically, more information is needed. Can someone fill us in?
2 seconds is a damn long time... i don't want pop-up screens for that long... reminds me of 'shareware'...
Knew this was going to happen sooner or later. The writing was on the wall after they removed the downloadable version of the database files.
Oh well, time for a new protocol.
OK, so the cddb owners are getting everyone up in arms again. I think that the OSS/GNU movement is forgetting something... You can't _Always_ get something for nothing guys!!
Escient invented this protocol and put alot of capital (yes, M.O.N.E.Y.) into their standard and site. In doing so they provide a service.
It's not as though they are asking for something for nothing. Their CDDB standard is very useful!!
Perhaps instead of flaming, refusing to comply and generally making a fuss we should try a compromise.
Can we just try to make friends?? I agree that it's unreasonable to start trying to charge a licence fee for data supplied to them by others and when the service had been free previously (althought I can't see where it says "licence fee" anywhere in the letter excerts). But if they've put alot of work into the standard, promiting it, maintaining it and providing an open repository for the information that people have until now been able to use free of charge, then what's the harm in giving them some free advertising for their troubles.
Sorta like saying, "Escient provided this service, thank guys, here's where they are" instead of basicly using them (because before this letter cam out we had a one way deal, we used their service and didn't have to give them diddly).
Oh - and I don't work for Microsoft, Escient, ODJ and don't even dual-boot _MY_ Linux system (before the flames come rolling in).
D.
I'm the one who emailed Hemos with this story -- I'm staying anonymous as I am the author of a cd-player application, and don't want to take the chance that my application will be "prevented" from accessing the CDDB (note: I have no real reason to suspect that they would do this, but I'm not taking any chances).
Someone requested more info on CDDB. CDDB stands for "Compact Disc DataBase". It is a format, originated by the author of xmcd, for storing the disc title/artist/track names of CDs (which is not stored on the disc). CDDB servers make it possible for cd-player applications to connect to them and download the information about the disc that is currently in a user's drive. This is a good thing.
The problem is that the format is "owned" by a company (although I'm not sure what the legal issues surrounding "ownership" of a protocol are). This was fine and dandy, until they released their latest license agreement (an exerpt of which I sent to Hemos and he posted).
What we need to know is:
a) can they do this? do cd-player authors have to sign this license to support CDDB in their applications?
b) if the answer to a) is yes, what is the opinion of the free CDDB servers on this? would they be willing to switch to a new, free format if one was developed?
If anyone still has a copy of CDDB from a while ago, (before the Escient buyout), it should still be anyone's property. Why not convert that data to our new Open-Source format and then get this going?
Oh, and someone will have to develop Win9x, WinNT, and MacOS clients, because the whole point of this is that everyone will use this.
So, who's gonna create the new free standard?
The CDDB protocol itself sucks. If you're going to reinvent it, do it right...I'm all for it, but the problem is, we'd have to rebuild the database from scratch, unless someone happens to have that huge file they used to distribute.
I develop a fairly widely-used player, and I do not like these restrictions. I'm all for creating a new format. Anyone with me?
We need people familiar with socket programming in perl or C..
We also could use servers. If anyone has a copy of that 50M file containing all of the cddb entries, that would be immensely useful too. Just don't tell Escient.
Email me at timg@means.net if you're interested in helping out. I'll set up a mailing list or something if there's enough interest. PLEASE do not spam my box with junk. Interested parties only.
What about a proxy for the CDDB? That could display the Escient logo all the time, use CDDB exclusively, and allow CD Players to connect using a non-CDDB protocol. One might also engineer alternative proxies using the same CD Player protocol interface, but using non-CDDB sources for the data.
Sounds like a plan to me.
Paul.
From reading quoted terms, it looks like most of the actual restrictions only kick in while the data base is being accessed. I don't find those kinds of restrictions too unreasonable. The most objectional thing I saw was mandating the upload feature.
I'd like to see the rest of this letter. I wouldn't be surprised if the rest of it were even more objectionable, but from what's posted it looks like you'd be free to set up your own database using the CDDB format.
Does anyone know where I might be able to find out (or you can just tell me) how much they actually own? I'm looking for details on the format, CDDB serverclient interaction, etc. I was unable to locate any patents on IBM's server.
Thanks.
Dan Moore
pergamon@68k.org
Escient didn't invent the protocol, they bought it from some folks who had very nice terms of use.
I totally agree. While I don't actively use any compact disc database, I suggest those interested say a big goodbye to cddb and a big hello to seedydb, or some equally riduculous ripoff.
Quite frankly, IMHO, this is not a smartest move by the company in question.
OH, sorry, ignore the timg@means.net address and send mail to the first guy.
-Tim (too lazy to set up an account)
Title says it all...if they want my input then I want credit for it too.
CDDB sucks anyways... any software company with 1/64th of a brain can make their own DB format. I'd say that this is a good reason for all companies out there to say "well that's it then, we will not use CDDB for anything." This isnt disturbing in any way.. it just sounds like the death throws of a company that is dying anyways... :-) Let's all boycot CDDB.. I have yet to use anything that uses that DB format anyways, (I dont use winnuts so I tend to be odd according to software people)
the mantra of "Gimmie Gimmie" is that of the mis-managed and almost Chapter 7 business.
There was a project awhile ago called DISCO that was supposed to be an alternative to the CDDB, but I think it kind of died. It was basically the same idea with more CD info and better interaction, I believe. Perhaps this could be a starting point for a new CDDB project? The last address I have for this thing is: http://www.spies.com/~lunatic/disco/
So does my car CD play have to comply with this? I think not. Who cares about this silly toy? insert an audio CD, access the tracks as 1,2,3,4,5,6.... Like any business really needs to keep a database of the latest green day cd's in the cd-jukebox and serve that list to all 120 workstations. Give me a break...
For those who don't know; the CDDB works by
letting people look up information about their
CDs automatically. It's damn cool cause it's huge.
Well, if you look up a new CD and the database
doesn't know about it, you can enter information
about it and then send it to the CDDB. This is
how the whole thing got as big as it is. Here's
the kicker, you can also make changes and submit
THOSE. Perhaps they are going to have someone
look over every submission from here out, but
I doubt it.
Welp, I think it's high time that we all start
changing our CD track information to include
bogus information and random misspellings and
then start sending that information back as
updates to the CDDB. This is a service that could
only have come about with the interest and
dedication of the thousands of people who sent
in album information and we have the power to
take it back.
The open source community has an excellent track record of giving credit where credit is due. The fact that Escient is using the implicit threat of litigation to get free advertising is a clear indication that whatever they did is intellectually trivial.
There are three ways to own bits. One can own a trademark, a copyright, or a patent. A protocol or a format do not qualify for either of the first two. Do they have a patent? If not they should be ignored. If so they should be challenged in court. It won't be easy to find someone in a position to do this, but after all it is free advertising...
Escient is opening up a can of worms by trying to turn this into a proprietary service.
Does anyone remember lyrics.ch? Does anyone really think RIAA will let Escient get away with commercially exploiting the database of RIAA artists and songtitles?
Sure, maybe RIAA's position is weak, but I bet Escient doesn't have the resources to defend itself against an RIAA lawsuit. RIAA has *big money* to throw at this.
In summary: live by the sword, die by the sword.
RIP, Escient!
I just had a quick look at the xmcd source (the original cd-player by the guy who developed CDDB and then sold out). It's source code, including all of the CDDB code on the cd-player end was released under the GPL.
This *should* mean that the format cannot be restricted under another license. In particular, the algorithm for calculating the CDDB disc ID was freely released with no restrictions (other than those in the GPL).
This is good news!
I don't remember the site, and am too lazy to look it up in my 3k-link bookmark file, but there is at least one alternative to CDDB out there. IIRC, the author of the one I'm thinking of writes for the Mac, though his protocol and db format were open.
Perhaps someone else can find the site...
One problem here. They require a logo to be displayed. I don't usually run a program that uses graphics! I CAN'T display that logo! So now what? I have to load X and get all that overhead just to use a database lookup? Not to mention most of what I use CDDB for is SCRIPTED!
Of course, the data was provided by others.. this stinks to high heaven.
Capital?? Have your read that spec? It's SIMPLE! People have stated the database was 50M when it was available for direct download. That's worth about $5 in HD space. And the bandwidth isn't anything huge.
I've allready volunteered to assist in the creation and deployment of a free method of doing this. I'll even offer server space and my DSL line for connections. (Yes, something I pay MONEY for!) I'll even set up Apache and host a page for it, if needed.
ftp.wustl.edu /systems/os2/apps/mmedia/sound/players/cddb.zip (found with ftpsearch)
It's a year old, but even with a years worth of data gone, these 32 MB (compressed) are a good start.
After browsing around Escient's web site, I found that they offer a "TuneBase" line of products for your home audio system. Basically, they are taking the information that they acquired from the CDDB and making money off of it. This is a perfect reason to switch to a new format and boycott the old.
However, there should be a license placed on the new format that will not allow the DB to be sold in any way, and that would prevent the Esceint scum from leaching off our (our, being the people of the Internet) works in the future.
screenshots 1 2
Well it would appear that the difficult part of the code (identifying what CD it is) is GPLed so all we need is a simple storage format (XML seems like a good idea), a simple to implement protocal for communication (say http), a large web server and a suitable old version of the database (i.e. from before CDDB was bought out).
Somehow I don't think this will be a problem for very long.
Oh and if you want a hand its benji@i.am (can't think of a decent nickname).
Did the GPL license cover only the reading and displaying of the data, or does it also cover the storage and retrieval? The the storage and retrieval side wasn't GPL'ed then all you can do is write your own reader. That doesn't mean that the CDDB folks have to give you free access to their database.
Even if the storage and retrieval code is GPL's But that still leaves the server software. Unless the original CDDB server software was GPL'ed, it would have to be rewritten from scratch, clean room style. Certainly doable by numerous people, but not to be underestimated.
There is also the issue of releasing all the player software which has hardcoded access to the CDDB site to point to the newly created site. And of course making all the CDDB users aware of the new alternative. Again, doable, but don't underestimate to effort required.
Jim McCorison
jimmcc@stratcoinc.com
So why does the "license" include provisions for preventing people from using any other database? It's quite simple, you're trying to eliminate competition. You are already an "ugly corporate monster", so when are you quitting?
smb
nfs
appletalk
started out as propritary protocols iirc
Microsoft licensed CDDB software, etc. for Win98's Plus "Enhanced CD" player... Anyone care to wager how many '98 users will continue to user CDDB?
Looks like what they are trying to do is restrict how their database is used. By attempting to treat a freely available protocol as though it is propriatary.
/. are "windowsisms". Considering the system was originally developed under Unix this is taking the piss.
This simply isn't the way The Internet works.
(How long will it be before someone submits an RFC for something like this, though without the need to tunnel the protocol inside HTML. Actually you can probably have a few variations, since lookup
can sensibly be done via UDP.)
Also several of the conditions mentioned on
You cannot own a file format. You cannot own an access method. Otherwise believe me the Microsuck
corporation would have sued everyone who loaded a
word file with another application.
I would suggest the Slashdot readers refuse to use
tools that display the cddb logo. Its simple. Your
app uses cddb, we wont use it.
Use the free cddb stuff and write back confirming
you are dropping support for their database. The
format of the data is open too unless they can
find a patent they own predating the xmcd release
> Did the GPL license cover only the reading and displaying of the data,
> or does it also cover the storage and retrieval?
Depends what you mean. The GPL'd xmcd code does local storage of entries. cddb.com used to have the server code online (but it was removed). I wonder if that was ever released under the GPL? Anybody know?
> That doesn't mean that the CDDB folks have to give you free access to their
> database.
No, but most of the servers were not originally run by them.
> Unless the original CDDB server software was GPL'ed,
> it would have to be rewritten from scratch,
> clean room style.
We're not talking about a complicated protocol here. Implementing the server only from information contained in the client code would be trivial.
> There is also the issue of releasing all the player software
> which has hardcoded access to the CDDB site
Any well-written cd-player should not be hard-coding the cddb server anyway. If I release my cd-player with CDDB capabilities (ignoring the CDDB license) and let the *user* configure it, there's nothing to keep them from pointing it at a commercial server.
You can email them at support@cddb.com with "comments or suggestions".
I've created a majordomo list to discuss development of an alternative protocol.
Send "subscribe bddc" in the body of a message
to waste@waste.org.
>
The CDDB team is comprised of real people working very hard to make the service the best of its kind on the Internet. We are also working very hard to keep two very important attributes of the CDDB service in place. These include:
1) The CDDB service is free for end users worldwide.
2) Access to the CDDB service is provided royalty-free for developers worldwide.
Our standard Player Agreement allows us to continue to provide CDDB free for everyone. There are very real costs involved in doing this. Our requirement for brand recognition and a CDDB click-thru helps provide us the means by which to continue to offer royalty-free access to CDDB for developers.
As General Manager for CDDB, I take a personal interest in addressing concerns from the developer community. We certainly hope that developers will understand that the Agreement actually protects their access to our service. Please feel free to email me directly at jk@cddb.com if I can be of assistance.
Best regards, jk
Jim Kinney
General Manager
CDDB, Inc.
You can probably copyright a *database*. But does that then give the right to demand that anyone who writes a client must provide you with advertising? How about the clause that they must query your database before any other? Then they demand that the client program provides a preferential update service to them. The latter is certainly anti competative.
Wonder what the record companies will have to say too....
business? All businesses suck. Who cares about businesses? destroy them all. I want my cddb to be free (speech free). When's that star trek society supposed to hit? soon I hope. friggin businesses and businesspeople. messin up my computer with their damn marketing and money and figgeldy foggle. wobbo wooboo woooo!
What is the licence of the original xmcd? If this is GPL then issues of creating a new ID algorithm don't arise.
Also there are (old) copies of the database freely available.
One possible option would be to create a new format and protocol. At the same time offering proxies which understand the current CDDB protocol.
Actually, I'm not offended by this license agreement. They didn't ask for money per player, they only try to ensure the continued value of CDDB. When we all entered our CD info, hoping to share it with the community, we did so hoping that the value of what we submitted would be sufficient to maintain the servers of it. That is what we are seeing now: our investment is self-sustainable. The revenues from the value stored in CDDB has become sufficient to pay for the servers and staff necessary to maintain it, and thereby ensures that we all will not lose access to the Data we entered. If this were not the case, Steve surely would have given this up for a more lucrative profession by now, and we'ld all be left with nothing.
As for some peoples suggestions about competing and/or sabotaging CDDB: SAbotage would increase the cost of ownership of the CDDB, and make it much more difficult to maintain and serve. And competition would reduce its intrinsic value. In both case, if the equation of income from value cost of maintenance is broken, then CDDB will eventually vanish.
I sugesst we support CDDB in making this data freely available to all of us. Its a small enough price to pay to have a self-supporting data service with which to share our CD TOC data.
I seem to remember there being an amiga pseudostandard for this - whether there was ever a full online database, I have no idea, but some german websites had index files, which were trivial to download (I don't know if any amiga cd player programs downloaded them automagically), and amiga magazines sometimes had audio cd indexes in AmiCDPlayer format on their cover CDs-
The file format which was basically a frame of ANSI text with the CDID number of the cd attached - easy to serve over existing protocols, no need to invent a new one.
All we'd really need is a server that responded to a message containing the CDID no. ( which has nowt to do with these money-grabbers, and is unique) with, basically, a list of tracks separated by newlines. Hell, we've got database server programs coming out our ears in Freeland, anyway...
The only way I know of, on how they could claim rights on the protocol is by copyright on the standards document (which will not include a right to restrict access to the idea behind it) or, which might be much more inconvienent, by a patent (which restricts the right to duplicate the idea).
From the excerpt it seems, they do not own such a patent.
They can then (and do) claim rights on the use of the CDDB(tm) trademark. Why not?
The solution to this problem is to write a Free implementation of the database server.
I think there're not far away from collecting royalties. So let's hurry up!
Thank you.
Jim Kinney
General Manager
CDDB, Inc.
jk@cddb.com
That's nice and all, but the exclusion clauses in the license agreement (that you can't let the user look up data in or submit data to any other database if you use CDDB) is completely unacceptable. That's not just establishing brand identity - that's trying to use the mass of data freely donated by end users world wide to lock out any future competition.
The graphic thing I don't really care about but (a) it's not compatible with many free software licenses, and (b) you could have tried asking nicely before trying this license bullshit.
straight off of http://metalab.unc.edu/tkan/xmcd
"Xmcd is FREE software. It is released under the GNU General Public License and is thus offered to you free-of-charge, with the specific terms listed in the GPL."
By this evening I'll have a database format up that uses mysql and html or XML (Or both.) Wonder how long it'll take to populate...
Use XML... it's more flexible.
Thanks for the astroturfing job you did.
Really, it was believable.. yes siree..
You get a service for nothing...nada, yet you critize a company for trying to make the service self sustaining!
How do you think slashdot operates? People just donate money to keep slashdot alive...No. There is advertising at the top of the page and click throughs to cdnow.
The point is is that it is hard to make a service that is 100% free. It takes time and resources to provide a service. People need to eat. Bills need to be paid.
Also, I get tired of the "they used our data" etc., I have submitted data too, but I got a hell of alot more data out of it, that is why the service is so useful afterall.
If you have problems with the license agreement I would suggest addressing those in a rational manner. You may find that they (Escient) are a reasonable company. Threats of sabatoge and constant flaming makes everyone here look like a bunch of clueless high school students. It's no wonder that we, the OSS community, have a hard time being taken seriously. Nuff Said.
Hi, I'm developing a CD player for the BeOS, and I was wondering, if I were simply to make the program look up CD information through a seperate module, and then had a CDDB module, wouldn't that module be a program by itself? So then I could include a module for any CD information protocol I wanted without violating the contract?
well, it doesn't say continuously for 2 seconds, only from 'time to time' :)
First the IMDB and then the CDDB went commercial, using data I sent them based on the notion that it was a community effort. I haven't sent a single bit to either since their respective "conversions". I'd love to see a graph of their submission rates before and after.
Prong you, Mr. CDDB Man. Take your database and go rot. Like all such damage, we shall route around you.
The cddb servers are run by volunteers, so the current version is out there, too. And while Enscient may own the format, or the protocol, they sure as hell don't own the information. At worst, the record companies own it all (I've seen cddb entries with full lyrics and even liner notes, more than lyrics.ch ever put up), and at best each entry is owned by whoever wrote it. It only takes one viral GPL'd entry to make the whole db forkable...
The source to the cddb server cddbd-1.3.1 is available and released under the GPL. If the current cddb server that Escient is using is derived from this source code they have the obligation to release their source code. I think it is pretty obvious that it is derived from the original GPL'ed server.
Here is an interesting quote from the cddb.howto : While we don't have anything against commercial software per se. we take a dim view toward having someone profitting from the use of the CDDB and servers.
How many CDDB users know that their e-mail addresses are sent to Escient whenever they look up CDs? It's in the protocol specs. Knowing what CDs you listen to is something that marketers would pay good money for. Now that CDDB has gone commercial, will the temptation to sell your e-mail address be too much to overcome?
Are you, like, just stupid, or what? You are totally not getting the clue here, Jim. You hosed this! Now fix it and move on, or you'll find yourself holding a handful of worthless crud while the world moves past you.
http://www.cddb.com/ftp/cddb-docs/ cddb_howto.txt
I've just done that. Go check out the CD Index Project home page.
Or contact me at rob@goodnoise.com
One another user-maintained CD database on the net is Mind on Music. That might be a willing site to kick-start a free CDDB alternative database.
I've got this up and running already -- check out The CD Index project home page.
That algorithm is given out, in plain C code, on CDDB's web site, presumably still under the GPL license, although it makes no mention of ownership. How many GPL-infringing, proprietary CD players are out there?
So does anyone out there have the extra bandwidth to set up a free cddb server?
If you do, do a ftpsearch for xmcd-cddb.tar.gz to get your server started. That is the data file with the data that we all submitted. The latest I found was from Nov. 1997 and it's 34.2 megs. Once a free server with enough bandwith is running, I'm sure we could get it up to date fairly quickly. If we can bring down web servers with the slashdot effect we should be able to fill up a music database with no problem!
An email address is only sent in if you try to update the database, not just access it. They want somewhere to bounce errors in entries.
In both cases, the command must be of the following form: cmd=server+command&hello=joe+my.host.com+clientnam e+version&proto=1
This is a requirement for all commands via http, not just submissions.
Find a country in which this sort of thing cant
be prohibited via a license agreement, i.e. one
with a decent 'fair use' law, and then get that
to handle all the connections.
We (FSC) could also probably organise a mass
cataloging of CD's, provided we could find a
location to host the database.
Find a country in which this sort of thing cant
be prohibited via a license agreement, i.e. one
with a decent 'fair use' law, and then get that
to handle all the connections.
We (FSC) could also probably organise a mass
cataloging of CD's, provided we could find a
location to host the database.
OpenCDDP (Open CD Description Protocol) would
be a good suggestion for a name -- anybody???
I just found cddb-1.4b39.tar.gz on my machine at home. This is the CDDB -server- code WHICH *IS* GPL!!! If anyone wants a copy of the tar.gz file, email me at mark@chalice.gen.nz.
Email it to me at james@rtweb.net, please; I'll post it under http://www.rtweb.net/theorb/freesw/. Thank you.
Hmm, the construction and ideas presented here look very similar...perhaps the poster is the same person who originally posted the ``Hello from CDDB'' message? Who can say. Whoever it is has hidden behind the Anonymous Coward monicker much as I am doing now. Sometimes being an AC has it's advantages, huh?
Face it, the existing CDDB database is practically *useless* in any new effort. Why? The 8-byte DISCID isn't nearly unique. I can put in "Rhapsody in Blue" and it comes up with two completely different CD suggestions. It'd be useful to include multiple views into the data and fuzzy matching. CDDB sucks for mp3, because you have to know things like intra-track offsets. Why not something that let you find CDs based on run-length to the frame resolution, run length, and other metrics? A database? GPL'd?
I'm also interested in this, although I think that, if people are interested, now's the time to talk about protocol, so we don't have six packages doing things six different ways. Whatever we do, it should keep TONS of data about these CDs -- frame offsets and durations, at the *very* least.
I'm interested in starting a mailing list, if someone else hasn't. Mail silly@buttsoft.com if you're interested, or if you've already started one.
I wrote a script using CDDB not long ago (perl and perl ioctls to get CD information). The format is plain insane. The CDDB ID is an 8 byte number, but you create 4 bytes of that number by summing the digits of the track lengths!
Obviously this might have worked for a single user system, but when you have literally 10s of 1000s of CDs you quickly start getting duplicates. Adding to the problem is that the format is difficult to get right, so many players submit *bad* IDs. CDDB uses "fuzzy" searches now, and it shows.
I know formats are always easier to design in retrospect, but you certainly can design CDDB far better than its current state. This silly license provides the perfect opportunity to dump the CDDB format, and the existing database, and start work on a truly free database with a decent protocol.
My thoughts, just implementing a client to access CDDB, is to use sectors instead of seconds, to preserve *all* information for *every* track (not combine them into a hashed form, if the database server wants to do it for speed then that's their concern), and to clean the record format (dbase type record, not musical record) to store extended info with well defined tags, not just as EXT.
And if the dbase record could look more like HTML, instead of the current KEYn=VALUE nonsense, then it'd be a lot easier to parse as well.
Seems that didn't work as I expected...
:-)
While I sort out exactly why I got to be an
Anonymous Coward, I'm viktor@dtek.chalmers.se...
I think I can assist with the machine part. :-)
:-)
I have a webserver with lots of extra power
(Sun Ultra-2, 2 CPUs). We are a swedish
university site, with plenty of bandwidth
as well.
We need a good format, and a simple program
to read the data of a disc and upload it. Does
anyone have a clue as to how you could go about
uploading data in a reasonably secure way?
We should probably ask the CD-player writers
what features they might lack in the current
CDDB as well, since added value always makes
it easier to convince them to switch...
from:- ------------------------------------------ --------
- ------------------------------------------ --------
http://www.cddb.com/info/info.html
-----------------------------------------------
General Information
CDDB was created for the enjoyment of Internet users everywhere. Use of the servers is free for private, non-commercial purposes. By using CDDB, users agree to the following:
Use of CDDB is solely at the risk of the user. This service is offered without any warranty, as CDDB cannot warrant the quality or reliability of service you may experience using CDDB. In no event will CDDB be liable for any damages incurred as a consequence of using CDDB. We reserve the right to refuse service to anyone at our discretion. Many of our system resources are donated, and their abuse will not be tolerated.
Database entries submitted to CDDB become the property of the CDDB project. Users of CDDB are granted permission only for private use of CDDB services and data. Repackaging or redistribution of CDDB data is prohibited. If you are a developer and wish to make use of CDDB in your software application, please complete our licensing request form. Once we have recieved your completed request form we will forward you the appropriate License Agreement via email.
If you would like to receive periodic mailings pertaining to CDDB, you can join the CDDB announce mailing list by sending email to announce-subscribe@cddb.com.
-----------------------------------------------
If you want to fsck someone at cddb go for the CEO .. email him saj@escient.com
Oops..
.. email him saj@escient.com
If you want to fsck someone at cddb go for the CEO
IANAL, but I'm quite sure that according to US copyright law you CAN NOT copyrite a complitation database (I know because I won a suit over this). Basicly if you have a database of publicly available data it can NOT be protected by copyright.
It would be quite easy to take a legit XMCD (CDDB) client, and make a moddified kernel driver to fake rapidly swapping disks with every possible CDDB id, then you make a program to convert the XMCD cache to your new database. (The reason you couldn't just write a CDDB suck client is because they only permit registered clients, which they might have some legal grounds to do).
I perdected this when they orignally went commercial, and they PROMISED (on this forum none the less) that they would NEVER charge for or restrict non-commercial use. It's obvious now what kind of liers they are and how much they care about us, the ones that made them their fortune. It's time we take our data back!
Last I checked, cddb.org wasn't distributing the source to the server anymore. I emailed them, and they stated that the db had gotten too large, so therefore, they weren't distributing the server.
I thought it seemed a bit fishy...
Licensing misconceptions seem like a common thread among Slashdot posters. It is important to remember that it is not illegal to "violate" the GPL. It is illegal to violate copyright. Since Escient is the company formed from the CDDB guys, and they wrote XMCD and presumably the code on their web site, the fact that the code was released under the GPL is irrelevant. They own the copyright and they can do whatever they want with it, including license it under other terms if they see fit.
XML would work nicely here, but also the protocol shouldn't retrieve all data at once. Most players don't do anything with the EXT data anyway.
The xmcd folks sold the database itself, and perhaps the server software, but xmcd itself is still GPL'd and unconnected with Escient. And AFAIK, Escient doesn't have any xmcd people on staff, they just gave over a chunk of change for the database.
I am quite angry about what CDDB has done. Like so many other people, I feel betrayed. You say that "CDDB will remain free for the end user". Well, that's very big of you, but you really didn't have a choice, did you? It's not as if you could have actually charged users anything, and I'm sure if you could have, you would have. And, in fact, it *isn't* free for the end user. I won't be free to choose any CD player I want. I won't be free from your dumb 2-second splash screen. In short, I won't be free from *you*.
- -----------------------
As for access to the CDDB being free for developers - well, it isn't *really* free, now, is it? They have to agree to be bound by you to all
kinds of restrictive licensing agreements. They won't be free to design whatever player they want since they have to abide by your ridiculous logo placements and other nonsense. I don't call that "free", I call it "without monetary compensation".
The thing that galls me and everyone else more than anything is that this is *our* data. *WE* created the CDDB. If it weren't for thousands
of people like me taking our valuable time to type in the data, you would have nothing. And now you are taking our hard work and trying to make a quick buck off it. That is what disgusts us - that you have perverted something that people did out of the goodness of their hearts,
as a favor to other people, not to line your pockets. If I send you a list of the albums I typed in, will you remove them from your database?
Since you will now be profiting from my work,can I expect a check from you for my time?
You better believe I am *never* submitting another album to CDDB, ever.
You plan will not work. We will defeat you. CDDB is doomed. Already work is starting on a truly FREE replacement format. And since it's so
simple, I expect it will be done within a couple weeks. You have shot yourself in the foot. If the RIAA doesn't get you first, you will die of starvation, because you have made enemies of all the people who write the software, and without us, you got nothing.
--
-----------------------------------------------
Doug Linder
dglinder@datacomm.ch
Escient's statement about submitted entries becoming their property is just so much bullshit legal mumbo-jumbo. The supremem court has already decided that publicly-available DATA (such as Cd tracks and titles) can't be "owned". Howevere, the *format* they are stored in can. For example, a name and telephone number can not be owned. So while I couldn't legally Xerox the phone book, I *could* legally scan all the information in, OCR it, and print it out in a different format.
In just the same way, it would be perfectly leagal to suck all the information out of the CDDB and import it into another database.
Escient is just trying to scare people with the Big Bad Legal Words. Ooooohhh! I'm not scared.
Doug Linder
dglinder@datacomm.ch
Another example of someone trying to make "easy money" cashing in on others work and something that is free. This reminds me of the early days of the Net, when people were trying to make money by charging for access to dictionaries.
What the hell are they going to be patenting? There's nothing to this application that doesn't have prior art, probably dating back to the 1950's. At least 3 people have started up mailing lists to discuss coding a replacement protocol, something that could be done with MySQL and Perl in under an hour. And whatever we finally settle on as our replacement protocol will be MUCH better than the CDDB protocol.
I think all open source projects should IMMEDIATELY drop CDDB support. Lets put Jim out of a job!
The Motif CD player xmcd, the database format cddb and the server software cddbd are protected by the GNU General Public License and in part by the GNU Library Public License.
See files, e.g. in xmcd-2.0.tar.gz and cddbd-1.3.1.tar.gz. The database and protocol are specified here.
At first the cddb format was created by Ti Kan in order to store CD information for xmcd in a local database. The user entered the data and stored them on hard disc. The user also had the choice to send these data to Ti Kan via e-mail.
Later, the cddb database service was run on a line of servers across the internet, based on the cddbd software by Ti Kan and Steve Scherf. CD data could by requested or sent online. The public database grew faster and faster. Snapshots of the cddb database were available on public ftp-servers. Then other CD players adopted the cddb mechanism and both provided features to retrieve and upload cddb data.
The database grew very fast and download from ftp-server got impractical. This was about the time (1997) the cddb.com thing started.
>A bit of unsolicited advice. Remeber that we live >in a free market society. Try not to immediately >demonize all profit making ventures that don't >completely agree with your personal code. As >Dennis Leary would say, ;)
>>"This is America. You have the right to be stupid and wrong."
THIS IS NOT AMERICA
Greetings from Europe.
There's a lot of prior art in the field, so any serious patent office wouldn't award any patent on this. Then again, no serious patent office awards software patents at all. Somebody in the states had better get in contact with the patent office and cite a number of prior claims, since while we can always route around a patent by distributing primarily from non-US sites, having a bddc reimplementation being barred from the US could make things more difficult anyway.
Either way, with the original code being under GPL, the database being copyrighted by a) the actual owners of the copyrights to the music albums and b) the submitters of the titles and c) Escient, they're rather far down the road to delusion if they actually believe they have a legal leg to stand on.
The best outcome Escient can hope for unless they change their stance on this is being reduced to an unused database and an unused protocol. If they try to make something of it, the ever cursed RIAA will pretty much stomp on them, and they'll be without the public image protection of the OSS community (dont expect any help from the people you're trying to stab in the back).
So, come on, Escient. Give it up. Revert your position and put the client, server and database under GPL again. Offer guarantees that this will not change. Offer to allow mirrors of the data and code. Make your money by improving the protocol and code, reap the rewards of an improved product, by selling non-GPL commercial licenses to proprietary software developers. Because between estranged customers who'll kick the legs out from under you and a litigous RIAA who'll sue if they're not worried about an uproar, you have little to gain.
Note that 1.3.1 is not the most current free version; 1.4b39 appears to be. Use ftpsearch.
Actually, I believe the "from time to time" means that Escient will, from time to time, specify a new logo to be displayed. For 2 seconds. Continuously. Or for however long it takes to access the CDDB server. Whichever is longer.
I'm not sure about the legality, but this seems VERY unethical. All the data was entered by end-users, under the guise of a freely distributable forum. Not to mention that the database originated with a free software program, XMCD.
booch
I've done a lot of network programming, including a variety of protocols. I can honestly say that CDDB was the dumbest, nost bloated networking protocol I've worked with, yet it's still such a great idea. I think we need to develop an open alternative that provides similar functionality, yet is owned by no one. Think what it'd be like if someone owned the rights to the telnet protocol!
Yes, Virginia, there really is a CowboyNeal.
If this is true, then why do they mandate that you can't use the data (which is public information, much like telephone numbers) for any other purpose, can't retransmit, can't use a proxy, etc.
Furthermore, they're mandating what every CDDB enabled CD Player must look like. The only way I use the CDDB is through a script, and so these restrictions are completely meaningless to me.
Also, why do they mandate that the CD Player can't use any other source of information? I think this is the clause that worries the most people, because it raises the barrier to entry for any competitor so high, that Escient is creating a monopoly on the CD database market.
Escient doesn't really own any part of the CDDB except perhaps the servers that it resides on. The data is public information, and the format, protocol originated in a GPLed program. I say we start a new project and shut these guys down.
Heh.. They must be from the Windoze world -- that's the first thing I started wondering about..
Perhaps Mozilla's Open Directory or whatever should be put up to the task of having a new CD database.. You could 'integrate' it with the directory system and allow users to go to a page about the artist that made the CD or something..
just a thought..
- A.P.
--
"One World, One Web, One Program" - Microsoft Promotional Ad
"Remember when the U.S. had a drug problem, and then we declared a War On Drugs, and now you can't buy drugs anymore?"
- A.P.
--
"One World, One Web, One Program" - Microsoft Promotional Ad
"Remember when the U.S. had a drug problem, and then we declared a War On Drugs, and now you can't buy drugs anymore?"
This just goes to show that there are still stupid people out there. CDDB was great because it was free and open. I guess now someone will have to create a new CD database.
---------------
Do not discount the fact that you have free will.
I would be more than willing to donate some of my resources (server space and bandwidth) to a free CDDB-like service. I'd even be happy to help write a server and come up with a protocol. Something similar to POP3 or IMAP for the network interface (but not too chatty), and a regular old database on the back end, maybe MySQL or PostgresSQL (or even DBM). It'll probably take longer to install the database than it will take to code up a prototype daemon (which I'd do in Perl or Java, then rewrite in C). Heck, we could probably submit the protocol as an RFC, if we wanted to.
My only problems are disk space and available bandwidth. I think I'm down to around 3/4-gig free on my file server, and I only have a 128Kbps ISDN into my network. But still...
Any takers? I'm game. I'm certain we could hammer something out in a few days, maybe even have a working prototype by the end of the week.
Rev. Dr. Xenophon Fenderson, the Carbon(d)ated, KSC, DEATH, SubGenius, mhm21x16
I was dubious about the Escient takeover from the start, particularly for the way it put the actual database off-limits. There's now no UK mirror, because the one using an academic server has been taken down. And slowly but surely, Escient is squeezing those who make use of its resources: resources that were freely donated.
Time to fork the tree, folks.
It'd be good to put together a protocol that integrates the track-listings of CDDB and the fields of ID3, so that you can keep together your list of MP3 and CDs...
Same here.. I wrote some software to read data from their database, bit hard to put a logo up when its used tho!
The copy of their demon software I have is GPL'd though, so we could use that as a starting base..
I'd personally like to see some changes though, because the data format is pretty awful at the moment..
OK, straight question --
The Id numbers on the CD's are unique right?
Why do we need to map them to some unique ID
number -- a database could easily use the ID
as a key.
John_Chalisque
Basically, we need to look at the license provisions of previous licenses granted for software to use the CDDB database.
If you have a CDDB program, and a license to use
the CDDB data without the terms prohibiting archiving, redistribution etc. then you could probably use the software, archive data and retransmit it to a new CD database.
In any case, what is needed is SWIFT action -- get a project underway, and get a simple format out of the door -- it needn;t make the service easy to use at first (give it a few weeks to sort that out -- it should allow users to submit CD details, and have an OpenContent license drawn up (or cloned) so that people know that their work wont be abused and exploited like escient are doing.)
Maybe someone could get Rob to host one temporarily bolted onto the side of slashdot, using a separate MySQL database?
(i.e. http://cd.slashdot.org)
A CD database server placed here with an explanation as to why not to use CDDB should get a lot of attention in the FSC -- and that would make a very good start.
John_Chalisque
I'm not sure if they can do this or not. AFAIK, there's no way they can restrict a protocol. Once it's documented, anyone can write an implementation of it. (This probably wouldn't hold if they had a patent (ick), but I don't think they do.) It seems that they're trying to restrict access to their database. That database, however, is composed of entries voluntarily submitted by users of CDDB-aware CD players. Can they claim ownership of data collected from such a wide array of sources?
For the people re-creating the database--please take a look at workman. The reason I don't use CDDB is that workman is far more flexible that the CDDB protocol allows. I can set separate authors for different tracks, which is a lot nicer than having a CD of songs performed by "Various Authors". I can also tell it about musical pieces that span multiple tracks, as is common with classical CDs. I can throw a classical CD in the drive, tell it to play the pieces in random order, and workman will do the Right Thing.
--Phil (I don't develop workman, I just really, really like it.)
355/113 -- Not the famous irrational number PI, but an incredible simulation!
CD-LDAP? Cool, I like LDAP. Do you have a URL for the CD-LDAP Project (if there is one?)
Need a mirror server?
Skye
"Without music life would be a mistake" - Nietzsche
This service is really simple to implement. why can't someone write a RFC suggesting it? There could be a network of servers, etc..
Or is it patented?
It's a large database of CD's on the internet. Stick your new CD in the computer, load a player (most are free/cheap) and it goes to the database and finds the artist/track names, and sometimes lyrics, based on the CD's serial number. It works great.
Er, www.cddb.org. Go get a player, less your already using linux, then I think the cdplayer in X does it.
Do not meddle in the affairs of sysadmins, for they are subtle, and quick to anger.
We're Escient, and we've purchased this wonderful user supported database of CD contents and other info and well...
We'd like to kill it.
We mean really rip it's heart out and leave it quivering on the floor as it gives out its death rasp.
If you would like to help us, just agree to our new license.
OK, Steve, either you're a) lying, and don't work for Escient, b) mis- or uninformed about the restrictions this places on developers and users alike, or c) stupid.
Since I have no reason to believe you don't work for who you say you do I'll have to discount a.
Since it's obvious that Escient wants to have their database as the exclusive source of CD information on the net, and should have gone to great pains to educate their employees as to their intentions before turning this loose on the world, and you would have had to fail to read the relevant portions of the license extracted here before you posted you reply, I will have to discount b as well.
Which leaves c.
Now, I doubt you're really an imbecile, since I haven't seen you at our meetings.
Please re-read the license, or at least the relevant portions that are here. We want to have the option of choice, the new license removes that from us. That's what we're all pissed off about.
I second the call for a free CDDB-a-like which is protected from potential future exploitation by a GPL-like licence. I am happy to come up with an interface spec (email ianc@dai.ed.ac.uk) if anyone has a nice server and some programming time!
Alternatively, would it be possible to use some kind of clean-room method to "re-discover" the CDDB interface protocol, and circumvent these idiots.
--
I work for a record company. I've written to Escient and told them to remove all of our intellectual property from their database until they are prepared to compete based on the quality of their service rather than restrictive licensing techniques.
I've also sent copies of my correspondence to other independent record labels, suggesting a similar course of action. The 800 pound gorilla in this game is RIAA-- if they lean on Escient, then Escient will crack. But we all hate RIAA, right??
Intellectual property can be used for good as well as for evil, kids. Think about it.
If we design a new protocol, we better make it distributed. Like the IRC. This way there's no one person that needs to donate *all* the bandwidth needed for the project, and there could be local mirrors. So there needs to be a client (player) - server protocol, and also a server-to-server protocol for synchronization.
you know when you hit the track list on your cd player on your computer, and the names of the songs appear - along with the artist and album title? Well, none of that data is actually stored on the cd. All of that is gotten from a CDDB. There is a number (or some similar defining characteristic) on your cd that CDDB looks up for you and retrieves that info.
Joseph Elwell.
I can't believe these fucking pricks. I say we should write scripts that send them fake/random data so as to destroy the usefulness/reliability of their database in protest. If we cause alot of trouble for them, maybe they'll reconsider these annoying restrictions.
--
-- Knowledge shared is power lost. -- Aleister Crowley
That's fair enough. The proxy only has to work on localhost which would remain on the user's personal computer. The command send from a player to the proxy could be 'tell me about the cd in /dev/cdrom now' which satisfies ii A and II b.
As for iii, it's arguable that the data is not being transmitted - it remains within the one computer. The data certainly isn't upoaded, aggregated, or collected. Though that would seem to prevent caching.
Paul.
Why? Take a good look at section 6, paragraph a, subparagraph 1, the first sentence:
You will not use or exploit the CDDB database...
Now, correct me if I'm wrong, but this essentially makes it illegal to access the database, right? After all, when one accesses the database, one is using it.
Even if it's not quite that way, it makes a great reason for a boycott of CDDB. That'll shut these profiteering bastards up.
Come to think of it, I've never even seen the need to use that database, anyway.
I knew the guy who was implementing a similar project, called DISCO. The author is a Mac guy, and may be the person of whom you're thinking. I don't know the current state of development, but the web page is still up.
--Gus
So we should form a working group and submit the :)
first drafts then
-- Jochen
I wonder if they're going to have an ASCII-art logo... what if my cdplayer application isn't graphical?
CDDB isn't that complex, and what its doing isn't that complex. I assume they don't have a patent on it, someone should just duplicate the service...
This is exactly the same problem I had with DIVX, except in that case the technology keeps you from using another DIVX provider if one is available, rather than this questionable license agreement.
When last I looked, CDDB was a simple text format describing a CD's title and track info.
Also, the trademarks cannot be too general and are granted for use of terms used in a specific industry or market. That's why you cannot call your new computer company "Apple Computer," or most variations that could cause confusion in the marketplace, but "Apple Bank" (here in NY) can operate without infringing on Apple Computer's mark.
So, in a nutshell what these sleazy weasels are attempting is a simple land-grab.
The OSS community should start "outting" outfits like ths with press releases that (in a politely worded fashion!) let the press know exactly what kind of liars and cheats these companies are and how they are attempting to steal the work and ideas of others.
Its time to take back intellectual property from the intellectually disingenuous...
Unfortunately, That isn't the case. The CDDB evolved from an open source project I believe. They used to distribute just a flat file of all entered titles, and then moved on to developing the CDDB Protocol. Reverse engineering would be simple, you can do lookups with telnet. As a matter of fact, I have a perl script I wrote to do just that. They base the uniqe value for lookups on track length I believe. I have some C code to generate that for Windows (adopted from the example they (used to?) give out on their page. I seem to remember, that all the servers are run by volunteers as well. They may own the protocol, but i doubt they own the information. It wouldn't be terribly hard to write a script to rip down all the information and populate it into a "new" database.
P.S. Any misspellings or faults of grammar you think you detect are mearly transmition errors, and probably your fault a
First of all, you make us all look bad. My opinion is most people are upset their license prevents you from linking to "other" CD databases. Not that anything else is even close to cddb.
Second, you could get yourself into legal trouble. Threats won't sway ANYONE.
Thirdly, supposing the above were not true, would you be willing to "clean up" CDDB's database which you took part in breaking? I thought not...
The GPL, LGPL, QPL, NPL, MPL, X, and BSD licenses are examples of licenses guaranteeing freedom. The GPL, additionally, is copyleft, so freedom is forced. If you don't understand the differences between these licenses, READ THEM instead of posting ignorant comments.
A new free CDDB-like format should be developed, and a standard library under LGPL (or some other non-viral license) should be distributed making it extremely easy for current CD player authors to replace their current CDDB support with a new format.
When will these companies realize that this "bait-and-switch" tactic just isn't going to work in the software community? We'll just go to something else. Of course, what would be nice is if we didn't start using their crap in the first place, unless freedom was guaranteed.
Incidentally, isn't mandating exclusive-use agreements illegal under trust laws? I honestly don't know, so it would be helpful if someone with greater legal knowledge in this area could respond...
Okay, so here are some defined goals as to what we need:
If anyone wants to discuss this with me personally, remove the SPAM from my email address (krose@Stheory.Plcs.Amit.Medu). Things like this really piss me off, and they should piss you off too.
--
Kyle R. Rose, MIT LCS
[ home ]
..
--
Kyle R. Rose, MIT LCS
[ home ]
\{*GPL\}\subseteq copyleft, not viral.
Not enough caffeine this morning. D'OH! Naturally this kind of mistake would occur when I was bitching at others to post correctly about licensing issues.
--
Kyle R. Rose, MIT LCS
[ home ]
Two lists have been created:
freecddb-developer and freecddb-announce
@bigred.lcs.mit.edu. Send mail to majordomo@bigred.lcs.mit.edu with "subscribe " in the body.
--
Kyle R. Rose, MIT LCS
[ home ]
...as those are © the respective authors. They can own a copyright to the anthology of titles, but considering it was a community submission effort that created it, and there was no restrictive license during most of its lifetime, I really doubt it would stand up in court.
I'm currently trying to coordinate a development effort for a new free protocol. Please send an email with "subscribe freecddb-developer " in the body to "majordomo@bigred.lcs.mit.edu" if you are interested in development. Do not send subscription requests to the list, as majordomo will bounce them. (In fact, majordomo seems to stupidly bounce any message containing \bsubscribe\b. Duh.)
And no quips about majordomo being non-free, either... =)
--
Kyle R. Rose, MIT LCS
[ home ]
> What is CDDB, and what is it used for?
CDDB is a database that programs can access over the internet to get the title of a cd, the track names and such using a serial number (or something related) that differs amung different cd titles. Programs like the Windows 98 PLUS cdrom player, or there is one called (spelling not included -- its close to it anyways) Quatrosum CD player do this. Music Match 3 (also for windows) will take the track names right before you copy the track to mp3 from the server. It just makes it easier for you to get the cd name and track names instead of having to enter them yourself.
>How can a company "own" a format? Sure, a company can own a *trademark* (like IrDA) and charge for use of the trademark; but I've never heard of anyone "owning" a data format. This is bizarre.
Maybe theyre talking about owning the system that it uses to gather whatever it uses to reference the cd names. Dunno.
Myconid@sover.net
Stan "Myconid" Brinkerhoff
SB.
Hey Casady,
I say go for it. Design a protocol and write a simple app for your platform. How hard could it be? You could probably get pretty far in one afternoon. Other people will write the apps for other platforms, if the protocol's simple.
James
james@jmarshall.com
From what I've seen of the CDDB protocol and features, they are shite. Personally, we'd be better off as a community creating a brand new, well featured and free system for everyone to use, and let them screw with CDDB as much as they like. Just my two cents.
Check your facts. Escient did NOT invent this protocol. They bought it - it used to be free and open. The guys who invented it sold it, unfortunately, and now Escient wants to exercise some control.
I'm not familiar with the cddb data format... but it does seem that if a new format was developed, the new database could be populated quickly from the data stored on users' hard drives... if there's any cddb/escient specific data, you'd have to ignore that, but it seems to me that disk ID, name, tracks, and track names would be free for use. I mean, how could you tell the difference between a user re-entering the data for the new format, and simply retrieving it from their local database subset?
How long 'til the music label folks sue to get a piece of this action? They probably own the rights to the band and song names, eh? >:(
From the web site: "DISCO is technology for sale." That might get us right back into the situation we're in.
Hey, I tried to send you mail but your address (bigboss@my(dot)dejanews.com) bounces.
Let's establish an Open CD Base then. It's
unbelievable (although we saw ago that CDDB
became comercialized, which was as bad as this,
IMHO), since what do they think they got their
data from? I entered dozends of CDs to their
database before it became comercial. Will they
pay me?
.sig: SEGV
I have absolutely no problem with a company setting up (maybe expensive) servers to serve the data to people, but as long as they don't own the data, it should be allowed for everyone to grab the whole bunch of data and make his/her own CDDB server.
Yes, such a huge collection of CD data is money. But it cannot be anybody's property.
(Actually, I think it can, but only of artists or record companies, but not of somebody who just gets it from others.)
.sig: SEGV
--
--
I have no sig.
Catch some of the earlier posts.. there is another guy who mentioned it first.. we should work together to form this thing.
As for resources. I have a gig or two available (since the last known size I'm aware of on that database is 50Meg it's not a big problem). I also have 256K DSL I'm willing to use for this. I sugest we think up a way to distribute the load so that one server doesn't have to carry the entire burden.
If all they wanted was to be sure credit is given to the CDDB when lookups are performed the license would have said something simple. Such as: Durring all lookups of CDDB information the user will be notified in a conspicous manner that CDDB is the source of Album/Title information.
Restricting re-distribution of data they don't own as well as the other things people have allready commented on is what has people upset.
Even if, for the sake of argument, it has not yet become an "ugly corporate monster", the writing is on the wall, so to speak. A license like this one tells OSS people that this company plans to become an "ugly corporate monster" and this bothers us becuase WE created that database! It came, at least in part, from user submissions.
Also, those players and utilities that support lookups are ALLREADY telling users about it! The only ones I've used that support it make a big deal out of it because it's so nice! Not to mention people like myself that use it mostly in scripts, we can't display logos or hyperlinks to your WWW site! You do remember text mode, don't you?
That is why there is a call right now to produce a free, open spec that anyone can use in any way they want. No companies restricting the uses or forcing people to display logos and such. I, and many others, find such restrictions horrid and refuse to support those who would use them.
You may want to look in on the other side of the fence and try to see why people are upset.
Travis
Oops.... I typed it wrong on SlashDot. I'll get ti corrected. It's bigboss@my-dejanews.com. I notice I had an underscore in there.. if that bounces for some reason use bigboss@xmission.com.
Although the vast majority of CDDB players are for Windows, CDDB got it's start about 3 years ago with Ti Kan's xmcd player. He invented the format. xmcd is the X11 Motif Cd Player. Definatey a UNIX program.
:(
It's sad that the corporate World has Win-ified such a good system. Your probably right though. This could mean the death of CDDB.
-Alan
Heh. The cddb software I'm working on HAS no gui. Heh.
I agree that it's time to start a new db protocol. CDDB has been getting a bit too weird lately. The problem with starting a new DB is that the cddb license prohibits us from building a new DB using data from theirs. Starting a new one would be very, very slow. Plus, there are issues with patents owned by Escelent. Are there any? Even a new ID calculation method was developed, would it still be in violation of Escelent's intellectual property? I'm interested in starting such a project... it wouldn't be that difficult. Anyone interested?
-- Minds are like parachutes... they work best when open.
I'm not sure. I _KNOW_ from the license, that we can't write software that checks a new database, then checks CDDB if the entry isn't found, then puts that entry into the new db. What MIGHT be legitimate, is an application that does nothing but connect to the cddb database, then stores that info locally, then a seaparate app to upload that data to the new DB.
The more I think about this, the easier it sounds, and the more practical it seems. Their protocol is simple. A clone, and a much BETTER clone, would be simple. The only trick is coming up with a ID algo that generates unique IDs, and is nothing like the CDDB algo.
-- Minds are like parachutes... they work best when open.
OK, so as far as I know, the algo. is free. If that's the case, and I/we can get our hands on the most recent copy of the database that was publicly released, I don't see any reason why a new database couldn't be started. Am I missing something? As far as I can tell, the database shouldn't be more than a few hundred megs...
-- Minds are like parachutes... they work best when open.
There are a few problems : One, not all CDs that are identical have the same stamped ID. Example : Pearl Jam Verses. Not all Stamped IDs are unique. Example : White Zombie Astrocreep 200 and other delusions of the electric head and a crappy bach remix I bought a while back. The discid algo that xmcd and others use is guaranteed to be unique, or at least, pretty damn close. The search engine that CDDB uses checks not only the discid, but the frame length of every track. It can generate "fuzzy" matches, which the user then chooses from.
-- Minds are like parachutes... they work best when open.
Just because the code is available doesn't mean it's free (like beer, or like OSS,) The fact that it was once GPL'd, though, means two things. One, it's free (like beer, and like OSS,) and two, Escient is violating the GPL whenever it derives a licensing fee from commercial players (as are the authors of the commercial players.)
-- Minds are like parachutes... they work best when open.
CDDB is used so people dont have to type in the song names for their CDs, it looks up the times in the database and then usually asks the user to pick from a number of titles and assigns the correct song name to the tracks. its pretty trivial, but alot of people use it.
Ugh. Read the license again. :( It says basically, you can do lookups only in response to a CD in your machine's drive, and then only upload data back to Escient. (i.e. No lookups on a proxy request, no data relayed to anyone but Escient.)
ii) The Player under this Agreement may permit the End User to aggregate Data only (a) in his or her
personal computer (b) only in response to the CDs placed in his or her computer.
iii) You agree that your Player shall not have or enable functionality that uploads or permits the
transmission of Data to anyone other than Escient. You agree not to upload, aggregate or collect Data
derived from the CDDB Database or End User Data.
This does not address one of the problems with the new licence, viz that the logo etc is required to be shown and this is not possible for a text-mode player. So does this mean that only graphical players are allowed under the new licence?
Under the GPL, they don't have to release
source code unless they distribute it.
I will not "pop up" any damn window in a background job from a command-line utility. they don't provide the service, they require going through them exclusively. No compromise.
byebye cddb, hello cd-ldap
I've finally had it: until slashdot gets article moderation, I am not coming back.
Hey cool, I'm glad you're here.
Allow me to introduce my concept of the new Open CD Database. Three letters for you: RDF. I'd really love to see you try to sue me for all the zero money it makes and the couple grand that is my entire net worth. Hope your lawyers work for cheap.
Scum.
I've finally had it: until slashdot gets article moderation, I am not coming back.
it's something i made up. LDAP as access method, RDF for the format. Seems a little odd, but LDAP's just another access protocol like HTTP, really.
I've finally had it: until slashdot gets article moderation, I am not coming back.
It's too late now. Remember ARC? I didn't think so.
I've finally had it: until slashdot gets article moderation, I am not coming back.
And it's being done:
http://www.freeamp.org/cdindex
They're hosted by GoodNoise -- GoodNoise might want to put real sponsorship this if they get mention on *voluntary* placement of blurbs.
This could be the quickest demise of a format in history.
I've finally had it: until slashdot gets article moderation, I am not coming back.
Sure we hate RIAA. That's why we want to set them and Escient in the ring for a good old fashioned cockfight. Set them against each other nice and good.
I've finally had it: until slashdot gets article moderation, I am not coming back.
Can someone setup a mailing list or a web site to get people willing to work on an alternative protocol/database together?
as for dirtying up cddb's database, I don't think it's such a bad thing either; at least it would drive the point home that, when you run a service that depends on its users, you don't want to piss them off, and that the users and the app developpers don't quite belong to two separate, unrelated works. and you could always make it easy (but not trivial) to clean up by including a fixed string in all your bogus entries.
I don't have a lot of resources to donate (I'm over a 56k connection, and I use all of that as it is), but I would be more than willing to enter all of my cd's (130+) into a new database. Hehe. Probably about 5 or 6 a day, however. Not all at once. ;-P
Combine that with an old copy of the cddb, we could probably come up with a sizeable amount of data. *shrug* That would probably be plenty to motivate people to move over to a new standard.
This sig is false.
Why don't you give me back all the disks I typed into "your" database then?
Yeah, I got a better idea (Jim Kinney, are you listening)? If you're so damn worried about recouping the money for the servers and staff, etc etc etc ad nauseum, why not turn the database back into the public domain (like it was to begin with) and I'm sure you'll get tons of people willing to maintain distributed copies.
A place we might be able to find a ready-made database of albums and tracks in the WWW music database at www.roadkill.com. Lotsa albums, track info, etc. Or, if anyone has older copies of the DB. Or, if somebody (like, say, end users) downloads the full DB somehow (if they haven't been presented with the terms, and are acting under the terms that came with their app, who's to stop 'em?)
Coming from a pre-GNU time when shared "open" source code came on DECUS tapes, this CDDB issue sure smells like the initial stages of the ARC/ZIP conflict.
Who uses ARC to package original content anymore? ARC won their lawsuit hoping to hijack the market. ZIP won the market with better support and technology.
It would be nice if everyone could get along and be happy--however, it seems like a gaggle of lawyers have already piled onto a CDDB litigation train that is starting to pull out of the station.
United States
What do these people think they are going to accomplish, other than forcing someone to either reverse engineer their client, or write an entirely free CD database client and server. I can't imagine it's all that difficult.
-josh
I think cddb.com has every right to make a god damn good living. They don't make their money off of software "sales" so I don't think that they're required to distribute any mods per GPL. Basically, they hold the card - the data. They provide the online links, service, and maintain the database!! Go ahead and start another cddb from cddbd. They arn't saying don't. (And infact competition ain't so bad afterall).
People would be ALOT happier if they removed the exclusive lookup license bit. And they should include a provision for non-graphical programs and scripts. Other than that, we are just reciprocating links just as if it were web pages.
Also, they are not making money off the sales of cddb software, just the data. So, the gpl arg is pretty dumb. However, if they ever get a partner which they license the software too, they have GOT to release the code.
So change the license already, are people in agreement? (It's alot worse than what we pay for in eyeball time on excite.com ANY day of the week!)
Panaflex, the incredible thinking cube.
I said no... but I missed and it came out yes.
I'd rather not. I'm assuming you work for Escient. Your company is trying to profit off of other people's work. I really can't see much of anything you did do. Somone else came up with the idea and implimented it. Thousands of users populated the database. You bought it. Now you are trying to bully the writers of aplications which use CDDB. I really don't see how you didn't expect that this would cause a revolt against you.
-matt
In any case, this announcement is going over like a lead balloon; no serious company will put up with these restrictions or pay these CCDB ``owners'' licencing fees.
--jon. Postel is dead. May we all mourn his, and our, loss.
Now I've been feverishly adding CD tables - of - contents over the weekend, and now I find that the CDDB owners are scum.
So, we've got to scramble, come up with an alternative free standard, and convince all the freeware and shareware authors to support it instead of succumbing to the draconian licensing terms of Escient.
And we need some sort of upgrade utility that will read the cache of CDs on everyone's machines and load them to the new databases, so that our database gets populated quickly. Or would grabbing all that data violate a previous license with CDDB?
Until now, I've been a consumer rather than contributer in the open source movement; this may get me off my ass.
Bravery, Kindness, Clarity, Honesty, Compassion, Generosity
...Nothing interesting here. Just move along...
Anyway, it's interesting to hear about his hometown. And that is one scary motto!
Bravery, Kindness, Clarity, Honesty, Compassion, Generosity
...Nothing interesting here. Just move along...
Bravery, Kindness, Clarity, Honesty, Compassion, Generosity
...Nothing interesting here. Just move along...
It says I have to connect directly to the CDDB server. Does this mean I have to dial in directly? :-) I can't use the firewall at work, can't use my masquerading linux box at home. Can't use the routers inbetween here and there.
Also what counts as a "personal" computer. Is my work machine a "personal" computer...even though it's not owned by me and it is technically a "workstation"?
What if a player doesn't have an "About Box" a "Help File" or a GUI interface to display the icon? What if they don't provide an icon in a format that you can read?
Is it your responsibility to write your program such that it can read an icon in a format they provide or do they have to provide the icon in a format you can read? I mean, what if I use PNG and they don't give me a PNG?
If Escient doesn't provide an icon in a format you can use then have they failed to live up to their side of the license? Can you sue them? Use the the database however you want (e.g. copy and redistribute it)?
Just curious about these things....
Escient can't restrict use of the protocol. Its been published in many forms for years, and they can't make a license retroactively apply to it. The only way to do that is make incompatible changes to the protocol (which is not out of the question).
Their real value is that they have lots of information about lots of CDs. Redesigning the protocol is easy, but pointless without the CD info.
Escient's real option, therefore, is to say something like "CD data free to users using a licensed player", but that's impossible to enforce. Their emphasis on players also ignores lots of other uses for the information.
The real question about this license is "why would you agree to it?" It only makes you give things up without giving you anything in return (the exclusivity clause is particularly onerous, and probably not legal). There's no reason to consider agreeing to it.
Where's the full text again?
J
Bah-loney. After this post, there's no question that I'm ditching CDDB for the new services and encouraging others to do the same.
RinkRat
When I get an MPEG player into my car, I'm planning on using an object database to store the music information. The database will allow me to store arbitrary fields with each recording and piece.
This way, I can make up a new category for music and rate everything on a 4 or 8 bit scale of how well it matches that category. Plus, I can store comments on the stuff, prefered track orderings (force several tracks to be played in a given order), volume and equilizer settings, etc . . .
Of course, a database like this will include many fields that different people will customize differently, but it will also include common info, like who made the music (multiple artists should/can be supported). It'll be far more useful than CDDB.
Now I just need a free object database and some time . . .
"Luncheon meats make the sawdust in your stomach explode."
3 reasons:
1. Folks, if we're going to copy their databases and put them in the new-improved versions, then let's not fill them with bogus data!
2. We don't have to thrash cddb's databases (even if they are full of our info) because they will be rendered moot and unusable very soon.
3. If we thrash them, they'll thrash us back. Really. It won't take rocket scientists to screw up our bddc2 db's. Nobody wins a pissing contest, everybody just ends up stinking.
hanzie.
********* sig: If you don't like the law, get filthy stinking rich, and buy a better one.
The only cddb-enabled program I use is the bash script ripenc. I think the author will have some trouble fitting a splash screen and a mail icon into it :)
Design into the system a mirroring system, and code to find the closest host (net time wise). Have servers be forward requests they don't know to higher up servers like what is done for DNS queries. Another thing to design into it is a language selection feature. This is for internationalization. One would be able to download titles in the language of their choice.
I was initially thinking a HTML/CGI front end would have some advantages. Use existing running WWW servers, but dedicated servers would be fastest for processing requests.
One thing to think about is making sure your data is unique. This is to cover your ass copyright wise in case they try to assert copyright. Adding new fields not included in their data, and the internationalization may both help with this.
From the beginning have a copyleft on the DB contents.
Did anyone else notice how Conditions and Terms, section 4, more or less precludes the possibility of accessing the database through a proxy or firewall?
Additionally, unless I'm reading this incorrectly, they're only saying that you can't access another database while you're accessing theirs. So all you have to do is make sure that your database queries aren't concurrent.
the Fork
While this doesn't mean that we'll see CDDB die overnight, it would die eventually.
I'm glad they've decided to do this. We've needed a new protocol for a while. They'll be routed around quick fast and in a hurry.
FYI, Escient is located in Carmel, Indiana. This is a very wealthy community full of people who believe in equality as long they can be a little more equal.
Escient's catch phrase on their home page says "We make technology behave."
Jason Dufair
"Those who know don't have the words to tell
Jason Dufair
"Those who know don't have the words to tell
and the ones with the words don't know too w
I've had a brief discussion with Jim at Escient. Here is the gist:
JK>>Anyone could set up a CDDB-like server or service. The issue is that "CDDB" is a trademark of CDDB, Inc. Versions of the CDDB server code have been copyrighted now for over a year. If someone wished to compete with our service, that's OK. Certainly, any application that advertised "CDDB" submissions and routed the submissions anywhere else than our servers would be encroaching on our trademark.
JK>> The protocol is Open Source just like HTTP. What we hold copyrights to is the server code and the aggregation of data.
JK>> Patents will be announced later this year. The CDDB name and logo are trademarked. We hold copyrights to the server code and that database.
It appears they are mainly concerned with the use of the actualy "CDDB" term. Since the name doesn't have a huge amount of brand recognition, it would behoove us to come up with another name, use the GPL'ed protocol and brand it like heck. In the interest of using what we have while building a new protocol, I suggest we simply call the service something else and set up servers elsewhere.
Jason Dufair
"Those who know don't have the words to tell
Jason Dufair
"Those who know don't have the words to tell
and the ones with the words don't know too w
>The day CDDB turns into some ugly corporate monster is the day I'm gone.
...
Check your calendar, Steve.
>CDDB is *free* for users, and will remain *free*.
In the sense of "free beer," not "free speech." (Yuchh, I hate quoting RMS.-) 6.a.ii specificaly requires player authors to restrict the actions of the player's users.
>The licensing agreement mostly seeks to ensure that CDDB is given credit for the fact that a CD lookup is being performed
Then why does 4.a.i.8 prohibit CDDB players from using any other Internet-based databases?
Stupid job ads, weird spam, occasional insight at
Gotta love that. Thought this database thing was for CD info, not graphic display.
I guess this means I always have to have a GUI running to access their CDDB. Yeah, right!
--The more you know, the less you know.
You ridiculous person.
Xmcd was and IS GPL'd. The information in your database was provided FREE to that repository by users.
Setting it up to be a money making proposition with intrusive advertising is a slap in the face to all the people who uploaded data.
I'm strongly urging ALL the cd player authors to get together and write a competing GOOD standard and leave your fledgling idiot idea in the dust.
-- Perl Hack, Web Hack, SQL Hack, Guitar Hack
I think there was some (admittedly hazy) licensing of the CDDB protocol which says you agree that any data you submit becomes property of the CDDB people / person. Surely somebody must have smelt a rat at this stage if they did make their commercial intentions clear earlier?
:-)
Still, I think whatever their intentions they've screwed up their PR something chronic, and I hope they get rewarded with a better, faster, free protocol and network of enthusiast-run CDDB-alike servers
Matthew @ Bytemark Hosting
They're stupid. We've got the power,
the technology and the people to design,
implement a new format and populate databases.
But, it was the best of the given options. It is:
a) distibuted
b) free
c) open source (anyone can add cddb to their player/whatever)
Possibly the main reason they are doing something like this is that commerical products are starting to use it. Xing AudioCatalyst has CDDB lookup for example..
- Give a man a fire and he's warm for a day, but set him on fire and he's warm for the rest of his life.
Can they do this? Yes, I'm pretty sure they can. Let me first say that I'm not a lawyer. I haven't been able to ask our resident intellectual property expert yet either. However all historical facts I can see point in their favor. Microsoft, for example, has come up with quite a number of "standard protocols" that they sucessfully hold the rights to, although I can't recall any of their names at the moment. Also I believe that Novell holds the rights to IPX.
As I recall from the previous version of this particular agreement, they did explicitly reserve the rights to the protocol and said they could change it if such pleased them. I guess it did. It sounds very similar to Miribillis and ICQ.
And yes. I would gladly support a free alternative. I personally believe that protocols should be commodities, not revenue sources.
Now this begs the question, will they actually enforce this? Well... I'll state my opinion on that in a seperate posting.
I realize what Escient is doing offends the diehard GPL zealots out there. As I have said before, I believe that protocols should be free.
;)
Having said that, let me get down business (pun intended).
I would be shocked and amazed if the folks over at Escient woke up a couple weeks ago and said, "Ya know, now that everyone is using our protocol, let's reel them in and make them pay!!! Fools!! All of them!!" Not a very likely scenario.
What probably *did* happen was something more along the line of a high level management discussion that revolved around the realization that people were using thier product and they could use that as a tool to build a company brand identity. Because branding is big business. Think facial tissues and Kleenex. Everyone in business wants to be like that.
The big problem is, they timed this all wrong. Someone at that company should get fired for releasing that updated licensing agreement right after the linux/gnu media frenzy kicked up. Shesh. Seems silly to me, but then again I'm on the other side of the fence.
A bit of unsolicited advice. Remeber that we live in a free market society. Try not to immediately demonize all profit making ventures that don't completely agree with your personal code. As Dennis Leary would say, "This is America. You have the right to be stupid and wrong."
You can telnet freecddb.freecddb.org port 888 (212.86.129.8) and check it out.
Submissions will not be updated automatically by now.
No cgi availible yet, so if anyone has a pointer to where I can find "the" cgi-script, please let me know.
If there are any efforts to create a free database, please contact me.
Guys, you need to calm down. I've been with the CDDB server project since the beginning, and I still am. I'm not going to entertain the sort of uninformed spazzing seen here, but I will say one thing. CDDB is *free* for users, and will remain *free*. The licensing agreement mostly seeks to ensure that CDDB is given credit for the fact that a CD lookup is being performed, and believe it or not that's the most important thing to Escient.
So stop fussing! The day CDDB turns into some ugly corporate monster is the day I'm gone.
Steve Scherf
CDDB