We are concerned primarily with tagging FLAC files. Depending on the genre of the music, it would be desirable to be able to easily switch the configuration of mp3tag so that different items appear in the tag panel and in the grid to the right. A good example of this requirement is when tagging tracks whose content is a medley of tunes rather than a single song or tune. This circumstance arises regularly when working with Celtic files. Many Celtic tracks have a "short name" but actually comprise a series of tunes in a medley, each with different titles and composers. Our current approach to that is to create a MEDLEY tag. When the tag is set to "1", the system reading the tags will "know" to find the following tags and correlate them accordingly. Assume for the moment that there are three distinct tunes in the medley. In this case, we want to assign TTILE1 to the first tune, TITLE2 to name the second tune and TITLE3 for the third. The TITLE tag is used for the "short name." Corresponding to the tune's titles, we would use COMPOSER1 to identify the compser(s) of the first tune and so on, with the composers or songwriters of the other two tunes on the track identified with COMPOSER2 and COMPOSER3 tags.
For tracks with only one song or tune, the standard COMPOSER and TTILE tags suffice.
In the current album I'm working on, some of the tracks have as many as five tunes per track. Configuring mp3tag to accommodate the desired medley tag plus six title and six composer tags is certainly possible but the tag panel and grid would become overpopulated and cluttered for the 95% of the albums we tag that have only one tune or song per track.
In addition, classical albums are another special case wherein the field requirements differ from popular and country music.
Here's one idea: is there a way to install, say, three different config files in the user/.../roaming folder for mp3tag and then launch mp3tag using only one of those configurations at a time? Are we the only ones that would benefit from being able to select from one of several configurations when launching the program? It isn't necessary to switch on the fly between configurations. Closing the app and relaunching it into another configuration would be more than acceptable!
I hope someone has a solution for this. Thanks in advance for suggestions or a solution.
Dennis....aka "d2b"
P.S. It has occurred to me that it may be possible to do this using different "portable installations" with three different thumb drives from which to launch mp3tag. However, we have a number of volunteers working on this project at the radio station and in various homes and individual locations out of town. Keeping everyone in sync is already a problem but we can deal with that by emailing a zip file to them to update their configurations when needed. Providing preprogrammed memory sticks is out of the question.
You can create more than one field of a type if you enter the individual pieces of data in one field (e.g. Composer) separated by \\, e.g. Mozart\\Bach\\Beethoven
The \\ creates a multivalue field upon saving. So it would not be necessary to have more than one input field for (in this case) Composer.
Yes, I know that, but that doesn't address the issue I put forth. For almost all albums, I don't want to have these extra fields types appearing in the tag panel and in the grid to the right: MEDLEY, COMPOSER1, COMPOSER2, COMPOSER3, COMPOSER4, COMPOSER5, TITLE1, TITLE2, TITLE3, TITLE4 and TTITLE5.
It is necessary to have these extra fields in order to 1) include the titles of each of the tunes in the medley in the order they are played, and 2) correlate the composers of each of these tunes to the tune they composed. For example, the TITLE2 tune (the second tune in the medley) is then known to be composed by the individual(s) with the COMPOSER2 tag.
I would like to be able to launch mp3tag using one of several pre-defined configurations. Right now, as I understand it, only one fixed configuration can be used.
Closing mpetag and manually installing a different configuration is of course possible but very clumsy and time consuming. What I'm suggesting is a way to have several configurations in the user/appdata/mp3tag/roaming folder with the ability to select one of them at relaunch. Perhaps this could even be done directly from within the application, which of course would be preferable.
Didn't you get more or less all the answers back then in 2014 when the portable installation was suggested to you? /t/15990/1
The answers stay the same, only the installation is easier now.
What have you done about the workflow since 2014?
For years now, I am using my own naming system that deals with such issues
I use >>+<< to indicate separated songs. Songs, not tracks. So let's say some track originally titled Folk Medley
would become for example a file named Folk Song A + Folk Song B + Traditional Tune X
and / or three separates files named Folk Song A
Folk Song B
Traditional Tune X
depending if there could be separated in a way that would not destroy my [subjective] listening experience
And if that means renumbering tracks- then I do it. If that "Folk Medley" was a track number 1, then "Traditional Tune X" would become the new track number 3, and the original track 2 would become track 4 after the adjustments
[Although If I were to do all my files again, I would rethink that method, as to if "Traditional Tune X" would be better of labeled as track "1.3"; so that it would both indicate the original numbering and what I consider a proper one. Or to choose o system with two separate tag fields; the original one and the second with what I consider to be a proper numbering]
I also use >>/<< to indicate what I truly call a medley. In [around] the provided example, if it was something more like a Folk Song A + Folk Song B + Traditional Tune X + Folk Song A + Folk Song B + Folk Song A + Traditional Tune X + Folk Song A
then I would not indulge myself in counting and repeating titles, bu just consider this as a one new song, entitled Folk Song A / Folk Song B / Traditional Tune X
assuming that this was the order in which the were used
As you already have such detailed information, this such not be a problem. The problem is when you know the song "Folk Song A" and "Traditional Tune X" and also clearly hear that there is something else that you are not able to recall by title ["Folk Song B"] or you suspect that there is something more that you simply do not know
I also use >vs.<< to indicate ingredients [particular songs] of mashups, which are something different than remixes and which are different from medleys [because they do not have new instruments, performers, arrangements etc.]
And on top of that I keep in COMMENTS further and / or more detailed information about each song [given that there is such a need and info available], that tell me if it samples some other song [or a movie], if it is based on some other song [or poem], if it incorporates some other song in a verbatim form. And these entries are also separated, if needed, by >>+<< and / or >>/<< signs, so that there will correspond with titles. It gets the job done, although when putting a lot of info in a file containing multiple songs, the info becomes blurry
[Of course this is not the all there is to this system of mine, but these were the basics to which in time I added some more indicators / rules]
A] Invest in widescreen monitor, with aspect ratio of 21:9. I simply cannot count, how much clicks and time have I saved since switching from a 16:9
B] Find a font that has less space between letters / words. Compare Arial Narrow with Arial for a start
C] Adjust system settings for spaces between elements, the thickness of sliders and so on. Those few pixels my seem not much, but they can result in the ability of adding some thin column, for example with track number, without the need of scrolling horizontally to see it
No
I would also benefit from it, as I ~95% of the time use my main configuration [for music], 4% a completely different [for learning foreign language] and the remaining 1% the main configuration with adjustments made on the fly
And the request has been already made by me [/t/17941/1]
My proposition suggested extended modification to the Utils; there shouldn't be a need for a re-launch of Mp3tag
Long medleys comprising up to five tunes in one file usually have a short name, like "The Clog Medley." We use that for the TITLE tag.
We organize our library using the following directory tree: M:\albumartist\album\album files (tracks)
If all of the individual tunes or songs are identified in the TITLE tag along with the primary aritst's name and the track number, you can and often will easily exceed the maximum path length allowed for a search or a file comparison. That is of course unacceptable.
We also need to match the names of the composers of each tune or song in the medley with their composition.
Therefore we have modified the Tag Panel to add the following fields: COMPOSER1, COMPOSER2, COMPOSER3, COMPOSER4, COMPOSER5, TITLE1, TITLE2, TITLE3, TITLE4, TITLE5 and MEDLEY.
If MEDLEY is given the value "1" meaning that there is more than one tune or song in the track file, we use the "short title" for TITLE. For the COMPOSER field, we use "Various Composers" and then include the appropriate number of TITLEx and COMPOSERx fields to identify the tune or song names and their composers in the order that they appear in the music file.
If the MEDLEY field is not set to "1" then the TTTLE and COMPOSER fields are used normally. In this case, the MEDLEY tag is absent from the file.
For most albums, having these extra tags in the Tag Panel is undesirable due to the clutter and the chance for misuse. It would be nice but not necessary to be able to switch betwen configurations that include them or exclude them. Besides that, there may be other requirements for tagging files that are completely unrelated to popular musi, such as when tagging spoken word files, audo books and even classical music.
Only the the FILENAME sometimes get chopped off at the end
It may seem to be the most logical way
But a very long artist name and / or very very long album title may result in the need of having exceptions to that rule, growing in number with time
Long time ago I identified songs with the sa. me name by adding the album title to them. Very quickly I spotted problems with this approach. But for years now I just throw in all the tracks to the artist folder. And which artist? The one, who is listed as first in ARTIST track [and which one is that is another story]. A variation of this method would be to store songs in artist folders according to whomever is listed as the artist of the album- but that is doomed to fail from the get go, because for example you can have an old song by artist X on an album of composer Y, which happens to be a score of a modern movie
And that is what the Mp3tag is for: not only tagging, but also for searching. I never go to folders when I look for something [except when dealing with errors or inserting to them new files]. I use the filter in Mp3tag and or "Jump to file" option in Winamp
Once again: the TITLE tag is what counts, FILENAME is just a backup name and name for storing on the drive [without ever looking anything by that storage name]
Of among my ~20 000 files I have few that have the exact TITLE and were made by the same artist. And to combat that problem I can use for example >>~1<<, >>~2<<, >>~3<< etc. [or any other normally not used sign that is not an illegal character], and keep such marked number in both FILENAME and TITLE. And also apply the same rule to ARTIST, because there is for example one John Williams who made the "Star Wars" music and a second John Williams, who played the "Cavatina" composed by Stanley Myers that used in "The Deer Hunter" movie. And the Various Artists folder would in time containt hundreads of files with the exact same title
Thanks for explaining your system [but maybe next time you could provide some real life or just made up examples]. I'm always looking for making more precise tags or making the usage of them somewhat easier / coherent
Unfortunately, I see it that your system would become a headache when applied to for example "Tubular Bells" album by Mike Oldfield. The so called Part 1 [the A side of the 1973 vinyl record] was in later years not only re-released, but also mixed from the scratch and completely re-recorded; and released as both 1 track and split into 11 [still continuous] tracks. And the 1973 side B consisted of also supposedly 1 track, that in reality contained 5 compositions of Mike Oldfield plus 1 traditional tune; which was in some other releases fully noted by splitting them into 6 tracks
So how would you treat this? Used ordinary COMPOSER and TITLE field for the A side of the 1973 vinyl, but with this version https://musicbrainz.org/release/ce5ec43f-ce...32-e6d1dfbb8700 tou would have used eleven additional COMPOSER fields, each with the same value of "Mike Oldfield"? For me such duality with one composition and by one artist would be just preposterous. I do not care how it was described back in 1973, I have to problems with "retronyms", with changing the tags provided by the recording company or iTunes [which are sometimes plain wrong, be it by a lack of their knowledge or a simple typo]
So basically you are running two systems, even for the same genre of music, the same artist and one and the same album. Or to be more specific, sometimes your system requires a sub-system, which can give false information to the normal system
I think you should rethink the system
I just had an idea, how I can adjust my system to both my benefit and how it could use your approach. Instead of having
TITLE
Tales Of Mystery And Imagination Opening + A Dream Within A Dream {1987 Mix} + The Raven {1987 Mix}
ARTIST
The Alan Parsons Project feat. Orson Welles + The Alan Parsons Project + The Alan Parsons Project feat. Leonard Whiting
COMPOSER
Alan Parsons, Eric Woolfson *3
COMMENT
INCORPORATES: Edgar Allan Poe: "Marginalia XVI " <text> + ∅ + BASES UPON: Edgar Allan Poe: "The Raven" <poem>It could be slightly changed to
TITLE
Tales Of Mystery And Imagination Opening +2+ A Dream Within A Dream {1987 Mix} +3+ The Raven {1987 Mix}
ARTIST
The Alan Parsons Project feat. Orson Welles +2+ The Alan Parsons Project +3+ The Alan Parsons Project feat. Leonard Whiting
COMPOSER
Alan Parsons, Eric Woolfson +2+ Alan Parsons, Eric Woolfson +3+ Alan Parsons, Eric Woolfson
COMMENT
INCORPORATES: Edgar Allan Poe: "Marginalia XVI " <text> +2+ ∅ +3+ BASES UPON: Edgar Allan Poe: "The Raven" <poem>
It would add more visibility and would be equivalent of your COMPOSER2, COMPOSER3 etc. fields. And the same rule could by applied to the >>/<< marker to distinct the >>/2/<< part of medley, a >>/3/<< tune from the medley etc.
[The empty set sign ∅ indicates non-existence of data, which in my system is different than >>???<< indicating knowledge of lack of correct data]
Or you could use something else that plus signs or multiple them to be more visible or whatever. Or use the actual number of the track. [In that provided real life example of mine, it happens to be that those are 3 first tracks from that album, the 2007 Deluxe Edition version of "Tales Of Mystery And Imagination", originally published in 1976 without that intruduction by Welles. But in such adjusted version of my system I would rather use the number "2", "3", "4' and so on, even when tagging for example the 10th, 11th and 12th track of a given album]
But I guess that the core overall issue here would be the ability to read such 2nd, 3rd and 4th entries from one tag field and then sending them further on-line to the listeners of the radio station, without messing up the info?
Yes. I could switch then between tagging music and adding tags [notes] to my language learning files
Only the the FILENAME sometimes get chopped off at the end
It may seem to be the most logical way
But a very long artist name and / or very very long album title may result in the need of having exceptions to that rule, growing in number with time
Long time ago I identified songs with the sa. me name by adding the album title to them. Very quickly I spotted problems with this approach. But for years now I just throw in all the tracks to the artist folder. And which artist? The one, who is listed as first in ARTIST track [and which one is that is another story]. A variation of this method would be to store songs in artist folders according to whomever is listed as the artist of the album- but that is doomed to fail from the get go, because for example you can have an old song by artist X on an album of composer Y, which happens to be a score of a modern movie
And that is what the Mp3tag is for: not only tagging, but also for searching. I never go to folders when I look for something [except when dealing with errors or inserting to them new files]. I use the filter in Mp3tag and or "Jump to file" option in Winamp
Once again: the TITLE tag is what counts, FILENAME is just a backup name and name for storing on the drive [without ever looking anything by that storage name]
Of among my ~20 000 files I have few that have the exact TITLE and were made by the same artist. And to combat that problem I can use for example >>~1<<, >>~2<<, >>~3<< etc. [or any other normally not used sign that is not an illegal character], and keep such marked number in both FILENAME and TITLE. And also apply the same rule to ARTIST, because there is for example one John Williams who made the "Star Wars" music and a second John Williams, who played the "Cavatina" composed by Stanley Myers that used in "The Deer Hunter" movie. And the Various Artists folder would in time containt hundreads of files with the exact same title
Thanks for explaining your system [but maybe next time you could provide some real life or just made up examples]. I'm always looking for making more precise tags or making the usage of them somewhat easier / coherent
Unfortunately, I see it that your system would become a headache when applied to for example "Tubular Bells" album by Mike Oldfield. The so called Part 1 [the A side of the 1973 vinyl record] was in later years not only re-released, but also mixed from the scratch and completely re-recorded; and released as both 1 track and split into 11 [still continuous] tracks. And the 1973 side B consisted of also supposedly 1 track, that in reality contained 5 compositions of Mike Oldfield plus 1 traditional tune; which was in some other releases fully noted by splitting them into 6 tracks
So how would you treat this? Used ordinary COMPOSER and TITLE field for the A side of the 1973 vinyl, but with this version https://musicbrainz.org/release/ce5ec43f-ce...32-e6d1dfbb8700 tou would have used eleven additional COMPOSER fields, each with the same value of "Mike Oldfield"? For me such duality with one composition and by one artist would be just preposterous. I do not care how it was described back in 1973, I have to problems with "retronyms", with changing the tags provided by the recording company or iTunes [which are sometimes plain wrong, be it by a lack of their knowledge or a simple typo]
So basically you are running two systems, even for the same genre of music, the same artist and one and the same album. Or to be more specific, sometimes your system requires a sub-system, which can give false information to the normal system
I think you should rethink the system
I just had an idea, how I can adjust my system to both my benefit and how it could use your approach. Instead of having
TITLE
Tales Of Mystery And Imagination Opening + A Dream Within A Dream {1987 Mix} + The Raven {1987 Mix}
ARTIST
The Alan Parsons Project feat. Orson Welles + The Alan Parsons Project + The Alan Parsons Project feat. Leonard Whiting
COMPOSER
Alan Parsons, Eric Woolfson *3
COMMENT
INCORPORATES: Edgar Allan Poe: "Marginalia XVI " <text> + ∅ + BASES UPON: Edgar Allan Poe: "The Raven" <poem>It could be slightly changed to
TITLE
Tales Of Mystery And Imagination Opening +2+ A Dream Within A Dream {1987 Mix} +3+ The Raven {1987 Mix}
ARTIST
The Alan Parsons Project feat. Orson Welles +2+ The Alan Parsons Project +3+ The Alan Parsons Project feat. Leonard Whiting
COMPOSER
Alan Parsons, Eric Woolfson +2+ Alan Parsons, Eric Woolfson +3+ Alan Parsons, Eric Woolfson
COMMENT
INCORPORATES: Edgar Allan Poe: "Marginalia XVI " <text> +2+ ∅ +3+ BASES UPON: Edgar Allan Poe: "The Raven" <poem>
It would add more visibility and would be equivalent of your COMPOSER2, COMPOSER3 etc. fields. And the same rule could by applied to the >>/<< marker to distinct the >>/2/<< part of medley, a >>/3/<< tune from the medley etc.
[The empty set sign ∅ indicates non-existence of data, which in my system is different than >>???<< indicating knowledge of lack of correct data]
Or you could use something else that plus signs or multiple them to be more visible or whatever. Or use the actual number of the track. [In that provided real life example of mine, it happens to be that those are 3 first tracks from that album, the 2007 Deluxe Edition version of "Tales Of Mystery And Imagination", originally published in 1976 without that intruduction by Welles. But in such adjusted version of my system I would rather use the number "2", "3", "4' and so on, even when tagging for example the 10th, 11th and 12th track of a given album]
But I guess that the core overall issue here would be the ability to read such 2nd, 3rd and 4th entries from one tag field and then sending them further on-line to the listeners of the radio station, without messing up the info?
Yes. I could switch then between tagging music and adding tags [notes] to my language learning files
Our library organization is formatted like this: M:\genre\albumartist\album\file name where M:\ is the media server on our network.
GENRE (for the time being) is part of the path name but in my opinion should not be needed, as it is also embedded as metadata in each file. Beside that, for much of the music we are interested in, identifying a genre is a individual interpretation and is not a scientific process. ALBUMARTIST is the name of the artist or band that generally appears on the album cover. ALBUM is the generally-accepted name of the release, also on the album cover and on the spine of the album package. ARTIST is the primary artist performing on the track. Unless the album is a compilation, the ARTIST and the ALBUMARTIST are often the same. Example: The Beatles.
Example: M:\Country\Doc Watson\The Best of Doc watson\18-Doc Watson - The Train That Carried My Girl From Town.flac
The structure of this path using the field names instead of their values is M:\ALBUMARTIST\ARTIST\ALBUM\TRACKNUMBER-ARTIST - TITLE.flac
For medleys with four tunes (as a very practical example) comprising what you would like to refer to as the TITLE, the structure would of necessity be like this, using my terminology for the names of each of the tunes in the medley:
It doesn't take a vivid imagination to see that the total path length could very easily exceed 255 characters using this structure. Even the full file name could be too long.
I am not in my home office right now, but I can offer a "real" example of this problem wherein the total number of characters in the full path name far exceeds the normal limits of an operating system, Windows 7 Professional in our situation. I can also show how this problem can be eliminated by including field names (and their associated values) such as MEDLEY, TITLEn and COMPOSERn so that all of the pertinent information is available.
That said, software to extract this information and parse it in a presentable format does not exist at the time, but the exercise to do this would be trivial if the method outlined above were adopted—which is one of the goals of this project on our music industry wish list.
For what it's worth, we only work with commercial album releases in the normal CD format or with lossless downloads (FLAC, WAV, AIFF) if and only if we have access to all of the metadata we require. Even then, the latter method requires more time and effort to import and tag, so we avoid it like the plague unless that's the ONLY way we can acquire a valued artist's release. Most artists appreciate our radio airplay and are ready and willing to send the full release in CD format complete with liner notes and one-sheets.
2] Get rid of ARTIST; from the MEDLEY variations. First you put the artist's name as a folder and then second time in a filename?
3] Get rid of GENRE step. On four counts: of being not precise enough [one song can mix genres], of taking precious space, of being often subjective, of possibly succumbing to a retroactive name change of once upon a time a new genre
4] Haven't you already run into 255 characters limitation problem?
5] The 255 limitation can be worked around by adding at the end of files something like #0000, where the number would differentiate files that have booth too long names and / or names than are the same. It does not solve the issue but it lets you keep the files [as for the metadata to be put online, the data should be red from tags, not just from FILENAME and folder structure]
I agree with you regarding the genre level in the path. It wasn't my idea and I oppose it for the same reasons you stated.
As for inlcuding the artist in the file name, we feel it is important information for those that do not have a search tool other than their eyes. Taking that suggestion even further, why not just give a unique, randomly-selected number as the file name? The name of the artist and the name of the composition are, in our minds, the minimum necessary information needed to facilitate visual identification of the nature of the file.
We solve the 255-character limitation with the MEDLEY tag and the related COMPOSERn and TITLEn tags, using the "short name" of the medley as the TITLE tag. Admittedly, I did run across one track whose title was about 120 characters in length, creating a total path length out of bounds. We just had to abbreviate the TITLE tag.
For compilations in particular, the ARTIST tag is different than the ALBUMARTIST tag. Do not confuse the two tags.
How do you differentiate albums that were released one again, with some changes [and thus a need arose to have them both]? You put in their title the catalog number of a given record company?
I do not; because it happens not to be so uncommon to have a track by artist X on an album released by artist Y
But you wrote
which uses two time ARTIST
As ALBUM ARTIST is more important in identifying a record and ARTIST is more important in identifying a track and at the same time limited space is the issue- why do you keep ARTIST first in form of a folder and then again in the FILENAME? Why do you not get rid off ARTIST from the folder structure? And on top of that repetition [which wastes space of five characters], with the ARTIST in the path, any OST containing songs by various artists will be split into multiple sub-folders, thus making a mess
Reissued albums have a different ORIGYEAR field and a different release date (DATE) field in addition to (possibly) a new or different ALBUMSUBTITLE field and/or COMMENT field. According to the directory structure illustrated below, I guess there would be two versions of the same album in the ALBUMARTIST folder. So far, with 12,500+ albums in our library, this issue has yet to arise in any significant way.
I originally wrote this:
"Our library organization is formatted like this: M:\genre\albumartist\album\file name where M:\ is the media server on our network."
The above is the actual directory tree structure. Any other variation is incorrect.
I always ask my self, "what if" and what are the possibilities that I can think of
If system works for 99% of cased it is 1% short of being a working system. It should work 99.99%. And when those 0.01% unusual cases start to accumulate, then it a clear sign, that the system has to be adjusted. That is what happened multiple times with mine, even though I did all that planning. Some times it took me months to figure out, how to adjust it; and sometimes it came down to doing a sort of revolution, changing things completely [and without Mp3tag that would be an impossible task]