Hi,
Apart from a definition of “Hidden Track” MB Terminology offers a link to Style Guidelines’ Multiple titles / Splits. According to this multiple titles on one track should be credited “Title 1 / Title 2” (if both titles are known). So it seems reasonable how the last track on Björk’s Vulnicura Strings is credited - “Family / Black Lake (viola organista version)” and the recording is a compilation of two other recordings. On Discogs it’s 9a and 9b, so everything looks fine.
Except there was a discussion about “Hidden Tracks” on this forum with another conclusion. “Artist intent overrides Style Guidelines” and so the “Hidden Track” is not included in the title (hidden) and only named in the recording which is a recording of the two separate tracks. Of Monsters and Men’s Yellow Light + Hidden Track recording was an example. It’s a quite similar case.
Now, probably Family / Black Lake (viola organista version) should be renamed to “Family” (track title only) and relationships for the recording should be changed to recording of. It’s probably no compilation of if the title mentions only the first part (and information would be redundant anyway).
I don’t want to start a new discussion about hidden tracks as there were answers from this earlier discussion:
I’m only seeking some confirmation before I start editing.
I don’t really see a concensus in this other thread to not list hidden song because artist intention was to hide the song. Actually most part of this thread is about how to name those hidden songs with only a few comments at the end mentioning not to list them.
In the cases I have dealt with I have always seen them listed and I don’t see why we shouldn’t. MB is about cataloguing the music. I don’t see how omitting stuff that is on the release helps here, even if the artist hid it a bit.
I’ve seen more hidden tracks named than not. It is especially important if a reissue drops that hidden track. I’ve also seen the hidden track given a name in [square brackets] to show the hidden nature.
I agree with @outsidecontext that MB is a music database and we should be documenting the music. Also when I speed read that thread the more official answers seem to be saying name it. As per guidelines.
I especially don’t think you should remove data that is already there. Please don’t strip the details from the Bjork track as otherwise someone like me would never know there is a hidden track there.
I think, there’s currently no information that there’s a “hidden track”. The track title includes the second recording and it’s a compilation of the to recordings. I don’t see any information that Black Lake (Viola Organista version) is an uncredited recording.
But of course there’s a difference between the two examples. In fact there is another (digital) release with the second Black Lake as a separate track. And there is no significant pause between the two parts.
Contrary to this, there must be a long period of silence after Yellow Light (song durations in addition are about half of the track length!).
Lots of things seem to get talked around in circles on this forum. Loudest voices are rarely the ones to follow. Guidelines and a bit of common sense get us further.
For a long time I’ve had the impression that it was agreed upon not to include the title of a hidden (part of a) track in the track title, just like @ernstlx did. Therefore I did not change track titles (only recording titles) from Printed Title to Printed Title / Hidden Track Title in the past – although I would like to have my files properly tagged with the complete titles
Having read this topic now, everyone seems to agree that the accessibilty of this information overrides the artist intent of hiding the track title – so is it okay to enhance all these track titles that buggled me?
For me it’s actually the first time I read about the idea not to list hidden tracks because of artist intent. I’m not editing that much anymore, so maybe the general view on this changed and I just haven’t noticed. But if that’s the case let me state here that I find it a horrible idea to not list hidden tracks claiming artist intent. Other databases, e.g. Discogs also list them.
Yes, of course the artist’s intention was to hide the track, to make it kind of a surprise to the listener. But for trying then to catalogue such releases I think it is essential to list the hidden parts as well.
There are different ways to hide a track. On analogue media (LPs, tapes etc.) it usually was just some additional content added after a longer pause after the last listed track. So you don’t know about this hidden content you just find it if you keep the medium playing after it already “finished” (or in case of LP you inspect it carefully).
When CDs came around things got more tricky to hide something, as the CD lists the individual tracks. So just slamming an additional track on the end does not work, the listener would immediately see it and can skip to it. There are three common tricks applied:
The hidden content is just appended to the last CD track after a longer period of silence.
After the last official track several empty tracks are added and as the very last one (e.g track 99) the hidden content appears.
The most sneaky method: The hidden content is added as a pregap track before track 1. This is sometimes really hidden, as not all players allow accessing it.
So as I understand it we are talking solely about case 1. here and “artist intent” is claimed here. I find it odd, as the intention of the artist is the same for all three cases. They are just different approaches playing with the technical abilities and restrictions of the medium. So if we claim artist intent for hidden tracks we should do not list the hidden tracks of any of the above 3 cases for CD and also not for analog media. Sorry, but for me that makes no sense at all.
What causes the discussion for case 1 of course is that on MB the tracks on release in case of a CD are tied to the physically encoded tracks on the CD itself. MB lacks the ability to cleanly state that a single CD track is actually divided and contains multiple content. But this is not something that is unique to hidden tracks, and we have a workaround for this by listing the individual parts in the track title.
I don’t want to change any rules, only clarification of existing rules. If it’s crediting the hidden track in the track title, I will do so. But I don’t think it’s a a horrible idea not to list it. There is no information lost - the associated recording will contain all the information: “Title 1 / [silence] / Title 2”, consisting of both separate recordings.
On the other hand, if it’s credited in the title and there is no silence (Björk) it lacks any information, that the second part has come as a surprise. Of course an annotation to the track can be added and scan images would show it, but I thought it’s a sneaky idea to have it only mentioned in the recording name. So it’s still hidden at first glance.
I don’t think it’s possible for case 2 and 3, regardless of artist intent. If it’s a separate track on the CD, it has to be listed. I haven’t got a CD with a pregap track, but it will probably appear as a separate track too. It’s probably not possible for all kinds of hidden tracks not to be listed as a separate title (CD and Vinyl).
I don’t insist especially on hiding it, but there should be at least a track annotation, mentioning the hidden track.
It’s at the beginning of track 1, it is part of track 1, but before the first index.
If you have such a CD, you have to keep pressing the REWIND button, display goes Track 1 -01:21 instead of Track 1 00:03, until you reach the beginning of the track 1’s pregap, then release the REWIND button and the hidden song will play, then when you reach Track 1 00:00after hearing the hidden song fully, you hear the credited track 1’s song.
If you have a CD player (rare) that has the CD track index feature, that can seek to index and display current index, pregap song should be index 0, while the default index that is played by default when you skip to a track or when you initially start the CD, is always index 1.
CD tracks can be subdivided in parts called indexes, from the beginning of the CD invention. It was seriously used on some classical CD for parts. But it was soon considered me user friendly to manage everything with just tracks, as CD did allow to play them in sequence seamlessly without sound gaps anyway.
Index 0 is also used on some tracks for some old concert CD to allow you start the song when you seek to a track or hearing the intro audience or MC when you listen to the CD as a whole.
A regular CD player does not make it more hidden.
In both cases you have to rewind and nothing tells you there is an index 0, as far as I remember I only had one CD with index 0 on first track.
I also had an index aware CD player (not at the same time) but I don’t think it would have shown me there was an index 0 on track 1 as it only displays the currently playing index and you don’t fall on track 1 index 0 without rewinding.
Great, thanks. There’s no EAC for Linux, but I’ve learned now, that the software I already use is capable to do so (cdrdao). Maybe there’s still hope for my hidden pregap tracks - if there are any on my CDs.
That was very, very helpful!
I struggled with cdrdao of which I thought it would do the trick, but it’s that easy: cdparanoia -B and the pregap track is caught in a separate file.
Thanks to @chirlu for this hint!
I created a pregap track CD uploaded the TOC to test.musicbrainz.org. If there’s a hidden content, it shows up in the TOC details. The track 1 offset has to be more than 2 seconds. (https://test.musicbrainz.org/cdtoc/3MUS2lL2oSIKNzHZL4Qv6pbCTd4-). So I would not miss it, if there is something. And with cdparanoia it’s no problem to get it into a file.
I don’t remember what those sectors are, they don’t seem to be CD samples as I don’t get correct time dividing it by 44100.
BTW I didn’t know this name but HTOA (hidden track one audio) is better name than pregap as most CD do have pregap (on each track) while HTOA is specific to track 1. Just a pity that it does not refer to index 0, pregap concept.
For sure I know far less than you know about these things, but I read that the time for 1 sector of 2352 bytes is divided by “44100 x 2 bytes x 2 channels (= 176400 bytes)” for the number of seconds (= 1/75). (so you can have all times in multiples of 1/75s)
… but I haven’t understood how these error correction bytes work
And there always seems to be an additional 150 sectors (2s) offset. It’s the same for my “test recording”. The actual time of the track is exactly 2 seconds shorter than the offset, both the original and the ripped track (cdparanoia).
This is exactly 6:48.373333… or 6:48 and 8/75 seconds. (Though the song is 2s shorter)
