To merge, or not to merge

I’ve done a few merges, but I’m not sure if I should do this one, so I’m here for advice.

This release has very little info on it. What is there matches one (and only one) of the more complete releases, with one exception. The one exception is that the release has a disc ID attached. This ID doesn’t exist in any of the other releases, but the release I’m intending as the merge target already has 16 disc IDs.

Is it best to go ahead and merge, or leave the release alone?

If we can’t confirm where a DiscID is from or what release it belongs to (or if it’s a duplicate release or not), it has no value, and I would just remove it and then merge the releases. You could leave it and merge as well, but I like concrete data.

In any case I have left a comment asking the person who submitted the ID if they can confirm:

Now you can leave yourself a reminder to check back on that edit in a week or two - if still no reply, merge away imo :slight_smile:

1 Like

Ok, thanks. The originator of the release hasn’t shown any activity in 5 years, and they only have something like 130 edits.

1 Like

I used to think ‘why bother’, but soo often I’ll get a message back from an editor that did like 1 edit 10 years ago :sweat_smile:

1 Like

Please don’t merge this well identified release into the catch all pre-NGS 16 disc ID release.

It is the same linuxrocks editor who added the release, its catalogue number, its barcode and the disc ID, so it’s pretty well identified, unlike the catch all.

1 Like

I don’t think we should have duplicate releases just to store DiscID’s? Is that something we do?

Better to remove the DiscIDs so they can be added correctly imo

A release is well identified by a unique editor with CD in hand but you suggest to merge their release away into a bogus catch-all NGS migrated release bin, and remove their disc ID?

That’s literally erasing a good release, for the reason that there is a junk release, already existing.

What info did I miss?

It’s not a duplicate.
The disc ID identified in this case a specific physical edition.
The 16 disc ID release is just a junk release resulted from NGS migration that does not match a specific edition.

It has already been added correctly.
Disc in hand. Catalogue number, barcode disc ID, tracklist.

Did I miss what data was not correct? it seems.

If anything has to be removed and wait for better addition, it’s the junk release.
But we can keep it for taggers and add a “catch all when you don’t know which edition” kind of disambiguation comment to it.


I agree with @jesus2099. Those pre-NGS releases with 16 discIDs are a technical mess. Looking at that one there are 30 second differences on some of those DiscIDs! There are also some of these kinds of releases with a merged heap of contradicting artwork.

When we have a nice release like this five year old one with a single DiscID we know for certain that the DiscID is real. All the data is clearly validated in one go. There is no reason to doubt that editor.

There is never a situation when a release would ever have 16 different DiscIDs. It is rare enough for two different discIDs on a release.

If you must merge them, then at least put that good DiscID into the annotation so we know which is the REAL DiscID for that release.

I would also suggest that as the artwork has been attached to the suggested target for more than five years, then the person who made the new release did it due to having a different looking edition. They just didn’t get time to return to add their alternate artwork.

For example, notice that the new one doesn’t have the DIDX cat no listed. I’m looking at 11 different releases at Discogs with this barcode and they don’t all have the DIDX code, so I would not merge this on the basis of different cat numbers. (For example: They may have had this CD in their hands : No DIDX cat no on CD All of the CD images I can look at on Discogs of this release are missing the DIDX code that appears on the old MB release here.strong text


surely keep the real DiscID as a discid. Why hide it in the annotation?

What I mean is when there are 16 DiscIDs like that on a release there is no way to know which are the good ones. So in those examples, adding it to the annotation allows someone to make a better decision as to the REAL one among the bogus ones.

This is why the new release is the better one as it is certain to be that DiscID. It is real and verified, not scraped from some junk “CD Stub” or other unverified data from a totally different release.

Notice those same 16 discIDs are attached to all three of the oldest releases listed. Due to the way the pre-NGS data was spread around it means this good DiscID is attached to releases that is may not actually represent.


No, if the other release is a duplicate that only exists to store unconfirmed discIDs then I suggest to remove those IDs, and then merge. Keep the good ID/IDs, of course.

1 Like

Ok, I will not be merging the release. Clearly there is enough reason to suspect they are not the same. I wasn’t aware of the reason for the proposed target having so many DiscIDs. I think I was mostly influenced by the lack of so much, like status, etc., but that’s apparently been taken care of since I posted this.

Thanks for the info. I’m glad I asked.


My initial post was only concerned with whether the releases are the same or not, not disc ID’s, which I think is my mistake, thanks friends.

All the disc IDs on the identical release are pre-NGS and are attached to multiple other releases - I’ve put in edits to remove them so that we can merge the releases (assuming that editor doesn’t reply and point out some difference in releases).

Hopefully that’s okay with everyone. We will have a :star: gold standard :star: merge after this!

Why are you trying to merge two releases with different catalogue numbers? The newer release does not have a DIDX number listed. There are two cat numbers on the spine of the older release. And the new release only lists one. Seems to my like they have a different edition.

1 Like

Out of 10+ Discogs releases with this catalogue number, they are always paired with the DIDX number on the spine or on the back (and Discogs also often doesn’t list it in the cat no. field).

I can hold off the merge if people disagree. But the disc ID’s can still go from that other release without any harm caused imo!

My only problem is the doubt. Why would an editor add a new release if they didn’t have a difference to the one in that older Release?

Look at the CD image. That CD image shows both cat nos. Now follow the Discogs link on that old Release and you see a CD image without a DIDX number on it. Was this what the new editor was trying to show?

Meanwhile… notice on that old release the artwork does not match the Discogs link. :laughing: (The Discog link needs removing\updating to match the current artwork) This is kinda the point - it is the older Release that is the biggest mess. We can not know if the newer release is the same DiscID as the older release due to lack of images on the newer release.

I am not voting in this one as I don’t know what is right or wrong here. All I can see is an older release with contradictory data attached. If they are going to be merged, then update the Discogs link as well to match the artwork.


Hmm, you’re right, maybe it’s better not to merge them.

But we still end up with a problem that bugs me… we know there’s ‘good’ data on that older release, but have no way of ever being able to identify the specific release, down to the back cover details. It will sit there for all of time, nobody ever knowing if it’s a duplicate, or if scans can be attached, and so on. I will lie awake at night* :cry:

*I won’t, I just remembered that I don’t care that much about Jimmy Buffett’s Greatest Hit(s) :stuck_out_tongue:


This release has a unique discID added that is not assigned to any other release. Unless you can verify that this disc ID matches one of the other existing releases and all other info matches as well it shouldn’t get merged.


I also have no idea who this is… :grin: but if it was someone I cared about the main thing I’d do is fix that Discogs link to make it one consistent Release. Then we wait for someone to come along and adopt the new release with some artwork.


I always wondered why the title is “Songs You Know by Heart.” Heart never did any of these songs. :laughing: