Welcome to Ottawa. I hope that you move-in is going smoothly. Better you than me :-) ; my job has forced too many moves on me over the decades. I realize that the main thread on this topic was a month ago, but I hope that some comments on organizing your music collection might still be useful. I hope others don't mind my putting this message on the OCLUG list even though its Linux content is limited. I have organized these comments under several headings: 1. Are you comfortable having several versions of your collection on your mega-storage disk? 2. How will you actually play your music to listen to it? 3. Have you tested to see how acute your hearing is? 4. How will you rip and/or encode your music? 5. How will you organize and/or tag your music? 1. Are you comfortable having several versions of your collection on your mega-storage disk? The initial description of your needs implies that you probably want at least two copies of your collection on your mega-storage disk. One would be a straight archive of your collection. Someone suggested the "dd" command to rip all the CDs onto the hard disk. From there, you could mount the .iso as a loop device to encode your tracks. Are you certain that this is a sound principle? I am not expert in these matters, but it is my understanding that the principles of error correction on an audio CD and on a conventional iso9660 data CD are quite different. I am not convinced that you would end up with sonically identical CDs if you first copied an audio CD with the dd command and then burnt the CD to disk as opposed to using a respected CD ripper and then burnt the output to a new CD. One CD ripping application is so dominant in the *nix world that I have no idea if there exist any others. It is, of course, cdparanoia. But, heresy though it is to say this on a Linux list, there appears to be broad consensus even among Linux-literate audiophiles that the absolute best CD ripping application available is ExactAudioCopy (www.exactaudiocopy.de) and it runs on "that other operating system (tm)." Another benefit of EAC is that it has a .dll plug-in called AccurateRip which compares the rip of your CD to that of previous users and confirms that yours is error-free (or not). The AccurateRip database is huge and frequently covers even much of my truly obscure "classical" music. The conclusion that I am nudging you towards here is to generate your "archival collection" using a CD ripper. It will generate .wav files of comparable size to the original CD. If disk space is a priority, .wav files compress well with standard *nix compression utilities. 2. How will you actually play your music to listen to it? The point of this question is to consider what quality of audio output you wish to achieve. At one extreme is listening to an audio CD on a dedicated CD player played through conventional stereo equipment into good speakers. At the other extreme is plugging headphones and/or speakers into the audio jack on your computer. Remember that the digital signal processors and audio amplifiers on most computer motherboards --- and even conventional "sound cards --- cost only pennies. They are cheap and nasty. And the electrical environment inside any computer case is, by definition, hostile to the production of a clean audio signal. As with anything in the audio field, you can spend anywhere from pennies to thousands of dollars to deal with this issue. Can you, as an individual, hear the difference? Personally, I have settled on the "squeezebox" series of products originally built by a California startup called Slimdevices and later bought out by Logitech. Squeezebox products are boxes which convert network data into audio output. Logitech has continued development of the hardware and appears to devote adequate staff to maintain the software and to respond to user questions and feedback. Even Logitech has a product, called the Transporter, which is far more expensive than I am comfortable paying but which is highly regarded. And other companies have even more expensive solutions still. 3. Have you tested to see how acute your hearing is? This question is relevant to what format you will listen to the music which is stored on your computer. It is highly personal. In my personal case, I am 60 years old and most men my age have long ago suffered significant high-frequency loss. In addition, I have a diagnosed hearing loss (and a prescription for >$7,000 of hearing aids to fix it). My now 86 year-old musically-trained ex-electrical engineer father jokes that, even in his own late 50s, he could no longer hear the hiss on his cassette tapes. With a friend's help, perhaps you could set up some blind A/B tests using various sources for the same musical material: e.g. a conventional stereo into headphones, a computer into the same headphones, a computer audio line out directly into a conventional stereo, an intermediate device (such as a Squeezebox) into headphones and/or a conventional stereo, different computer encodings, e.g. lossless, such as flac and/or lossy, such as mp3 or ogg. At what point do you, as the owner of this music collection, start hearing differences in quality. You would be surprised how many people do not hear differences between the theoretically "best" and "worst" alternatives in this list. Do you have views on patent encumbrance? mp3 is the dominant lossy codec, but patent encumbered. ogg is not only free (in the RMS sense of the term) but also reputedly produces higher-quality sonic output from the same size lossy file. Can *_you_* hear that difference? If you want to carry your music on a portable device, remember that the device which dominates the market supports mp3, and not flac or ogg. If you jailbreak the device, you can, however, make it support flac and ogg. 4. How will you rip and/or encode your music? In point 1 above, I suggested that the best-regarded CD ripper was not written for Linux. But, in my view, the crucial criterion is not so much the ripping component as the tagging which is supported. (See point 5 below.) The downside of the freedb database is that along with its openness is a complete lack of consistency or quality control. As a result, I avoid freedb wherever possible and look for applications which use the musicbrainz.org database. Musicbrainz enforces a higher level of consistency and overall quality; this is not to say that there are not some hopelessly miscatalogued and misstagged CDs in there, but far fewer than in freedb. Musicbrainz also operates on a user contribution principle easily understood by the FLOSS world. It is appropriate to contribute new material and to improve existing material. It can be *_very time-consuming_* to do so. Google searches suggest that musicbrainz gateways to the most popular ripper / encoder applications are, ahem, problematic. A skilful coder can often massage them into working. There are two alternatives and I use both: a) with installation of the right packages, KDE offers a choice of freedb and/or musicbrainz source. If you choose both, it will give precedence to musicbrainz, which is the right thing to do. My favourite KDE ripper / encoder is audex; b) if you don't use KDE and cannot massage your encoder to use musicbrainz, then musicbrainz has a very powerful cross-platform tagger / file renamer by the name of "picard." The application is not intuitive and neither is its documentation. But effort spent in learing it is repaid with well-tagged and well-named music files. 5. How will you organize and/or tag your music? This question flows out of the previous one. Maintaining a collection of thousand or more CDs, e.g. several thousand separate tracks, will give one rapid respect for librarianship and the fact that most professional librarians have university degrees in library science. If you ever want to find your music and/or to choose what you want to listen to and to know something about it, be prepared to devote hundreds of hours to collection maintenance. The first question is your basic collection structure. I use the filesystem for my basic structure. Over 90% of my collection is "classical." The "classical" music is therefore organized into five top-level directories which make sense to me. The "Composers" directory has almost 100 directories by name; the other four directories, for various forms of collections, are organized by musical period and by theme and have far fewer second-level directories. I have one more top-level directory for jazz and two others for "everything else." I recommend highly using the filesystem for your basic structure. From there, organize in a way that best reflects the types of music in your collection and the numbers of CDs in each. With a collection this size, a rigidly-enforced consistent directory structure, filenaming pattern and tagging structure will make pay for itself a thousand times over in speed of finding exactly which disk or track you want to listen to. For me, it is a question of finding different performances of the same classical work; for others, it could be finding different releases of the same song by one artist or of finding "covers" of a particular song. And so on and so on. I am sorry for the length of this message, but I hope that it prompts you to consider some of the long-term issues of organizing and maintaining a large music collection. Best wishes Bruce Miller -- Bruce Miller, Ottawa, Ontario, Canada bruce [ at ] brmiller [ dot ] ca; (613) 745-1151 In archaeology you uncover the unknown. In diplomacy you cover the known. attributed to Thomas Pickering, retired US diplomat, born 1931