I do not care for low quality art, but I am about preserving history, and no art means no history or extra history that an editor can extract from a release to increase the quality of that release. I have spent a good deal of time and effort replacing the lower quality art of releases until discogs turned off access to their high quality art.
As to an album art downloader searching ebay, that’s a futile attempt, at best it may get you close but more likely wrong data. Using ebay for research requires you to really understand what you are looking for, it takes knowledge and understanding of your subject, it’s no script, it’s manual, it’s time consuming.
You said the correct phrase, “you investigate”, that’s not automation. You make sure the correct art goes with the correct release. And by the way there is a bit of discog art that does not go with the release. Discogs is far from infallible their editors are human like us and some prone to sloppiness.
When I spoke about quantity I spoke about the 1000 uploads.
When I spoke about bad data, I didn’t mean low resolution, or clumsy photo, this is no problem.
When I say bad data, I mean uploading the images of another edition on a release.
This is quite destructive for MB and wasting everyone’s time and efforts.
If the goal is to have an image on each release group, then we should rather easily add a download edition to the release group, with its cover art from the linked online shop, instead of adding scans to 1000 physical releases without being able to validate them.
Yes, and that’s very good!
But you did not change your workflow when we pointed out that you might likely upload wrong edition scans, because nowhere in your 1000 upload workflow that you inspect your edition in hands.
And all editions don’t always look alike, in details.
My point is more for popular albums, that gets several editions, and internationally. Less so for rare local one-time releases, with only one edition and one art/packaging version.
If we can’t make sure it’s the correct edition, we should not upload.
Your answer to this concern, so far, was:
But it doesn’t say where the 1000 MBID came from, were they carefully selected, with release in hands matching all the details (disc ID, catalogue, barcode, manufacturer, labels, etc.), or not?
But instead of deleting your accounts to create new ones, you could keep the same.
Because it seems you don’t want to leave.
And that’s good! We are not against you, we are for a good database. So we would like to keep its reliability when possible, when editors are open to the discussion, like you created this topic.
Resetting your MB accounts feels a little like if you wanted to always get back to before the discussions started, on your same grounds and position.
My MP3 collection is files downloaded from the internet. Not very important Various Artist compilations for which it wasn’t worth buying physical CDs. I don’t have them in my hand.
I took 1000 MBID from Mp3tag, where the Cover tab was empty.
I have 1000 important physical CDs from which I didn’t make MP3s because I like listening to the original. I entered IFPI and Set track length from them.
That you are make sure that the covers you upload exactly match the release in MusicBrainz. Same release country, barcode, etc.
If you are doing that, I don’t think anyone has a problem.
If you are not, you should not be adding them - the selling point of MusicBrainz is that we have very specific release data, including cover data, compared to what a bot/web scrape finds.
Please just add an edit note. I checked half a dozen at random, and found Discogs images on each one. So just write “Discogs”. Or even “Discogs or Amazon” if you can’t remember exactly.
Seeing artwork with no comments makes some of us nervous as to the source.
“Do I need to add the Discogs source to my 150 edits?”
“No, of course not But try to add an edit note in future edits. And you might add “spine” to this kind of covers you already added”
