as I’ve been obsessing about a song recently, I’ve been thinking about the obsessions functionality that @MusicListenzGood suggested… and also created a ticket for it [LB-1260] add an “Obsessions” feature - MetaBrainz JIRA
warning, semi-random thoughts about the feature incoming
I’m thinking there would be some sort of a threshold for a track to be marked as an obsession, like 4-6 plays within two days or something. after that, could try to guesstimate how long the obsession goes on for, just a couple days or a month. in my mind there’s two important stats to track, the amount of plays (perhaps a per-day average?) and the length of the obsession. the lower limit should probably be two days
a totally hypothetical example might be helpful. say I listen to Excited About Burger say,
13 15 times in a three-day period. that would obviously get marked as an obsession. perhaps it could show up in a weekly or monthly summary, and it could show up on my Year In Music if it’s my longest or most played obsession
I feel like it should only count “obsession” plays during the time period. if I obsess over the same track later in the year, I don’t know if I have a preference whether the two periods are combined or seperated… if seperate, it should mention there were two periods of obsession tho
perhaps it could also be expanded to works? like, say you listen to several Discord remixes over the course of a couple days. might get a tiny bit tricky if we want to include arrangements and other versions tho…
also adding a couple other tickets I made and never posted, lol
It definitely needs to be a relative thing based on how much music a person listens to, I think. For people who listen to less music a smaller threshold and for people who listen 24/7 higher. Or people who are a bit extreme might have a lot of obsessions or other people might not have any. That or if it’s a fixed threshold, only count the top x number of obsessions for a period or something?
yeah, I was thinking about a percentage of listens too, but I didn’t end up typing that bit up, lol… mostly because I’m not sure what might be a good percentage
if it ends up on a weekly or monthly summary, yeah, probably show the top 3 obsessions at most
I don’t know if this should be in redesign or here. But new UI features for adding songs to playlists would be nice. Specifically, if there was an option on the 3 dot menu on songs, and maybe it could show the x most recently modified playlists in a sub menu for quick adding, and another option that brings up a dialog to pick from all your lists. Plex has this, and spotify is also similar.
100% this ^
also, adding the ability to paste the recording URL/MBID into the current “add song” box would be crazy useful~
Any chance listenbrainz starts matching recordings using ISRC? For example the listening companion on https://d.ontun.es/ is able to see what exact MB recordings I’m listening to in real time.
Matching based on track names seems very iffy.
EDIT: also please add a dark theme! It’s so damn bright… and unlike these forums make it switch to dark theme based on your browser preference.
I’m not sure if this has been mentioned, but search feature didn’t show any results:
In addition to importing our last.fm history, it would be nice to be able to import our last.fm Loved tracks and with the date we loved them on last.fm. I just went through listening to all mine so I could easily make sure they are all loved on LB too, but if this feature was added I’d remove them all and import with the original dates. And I see some people have thousands of Loved tracks on last.fm which would be a large task to re-love on LB I feel like.
Even if there was a userscript for now that could do it that would be amazing.
I’ve wanted this feature (add to playlist) for a while !
There’s an issue open for it but I haven’t had time to look into implementing it yet [LB-1046] Add a "Add to Playlist" option to listen dropdown - MetaBrainz JIRA
There’s also an issue for that feature @DontMindMe: LB-564
One part of it is actually already implemented (the API has an endpoint you can call), but we don’t currently have a user-friendly way to use it on the website.
If you are technologically inclined, you can send a POST request by running this command in a command line/terminal:
Of course you’ll need to replace my-listenbrainz-user-token with your own token (which you can find in your profile) and MyLastfmUsername with your real username.
curl --location --request POST 'https://api.listenbrainz.org/1/feedback/import' \
--header 'Content-Type: application/json' \
--header 'Authorization: Bearer my-listenbrainz-user-token' \
Alternatively, you can wait and we will get to adding a button for it in the website
I tried it out, it does conserve the date and time at which you loved the song on LastFM.
ListenBrainz submissions include extra information such as the player, submission method and other data. I’d love to see breakdowns of this on the Global stats page.
I look forward to it, thank you!
(I originally posted this in the redesign thread, but then I realized that it wasn’t really related to the redesign. So I looked for a more appropriate topic, found this one, and am posting it here instead.)
An enhancement that I would love to see is more frequent and consistent ‘Stats’ refreshes.
Currently, the stats refresh rate ranges from once a day to once every three or four days, which is frustratingly inconsistent. I understand that it has to do with limited computing power and resources, especially for a community-driven project like ListenBrainz.
But still, real-time stats updates is a killer feature that Last.fm absolutely nails, and that I miss sorely. Of course, real-time would be way too taxing on the ListenBrainz infrastructure, so maybe twice a day would be a nice compromise/goal?
Delete artists. The listen count should decrease when we remove a listen. Finally, a view to see artists in common between the two users
Here to report another issue: the timezone preferences don’t work for me.
By that I mean that I use a web browser that falsely reports the timezone to be UTC 0 to websites as some kind of resist-fingerprinting feature. (No way to disable it, unfortunately.)
I thought that the ListenBrainz timezone preferences could override this browser-level forced-UTC, but my ‘Listens’ page still shows my listens in UTC 0 time, instead of the timezone that I set.
Have I misunderstood the setting?
sync Spotify playlists and liked songs into listenbrainz
As you would know making playlists in different platforms is a very tedious process. It would be cool if there was a sync service to automatically sync playlists, It also expands this platform as a more social way to do everything by being able to recommend your playlists etc.
I think liked song sync might be a bit complicated - Spotify allows an unlimited amount of liked songs, and some people (myself included) wouldn’t want every song in our Liked Songs library synced with the Love feature we have now. Playlist sync is a good idea though.
I’ve been using last.fm for years and one of the things that I wish it did was quantify listening time instead of the total number of plays. I really don’t care about total number of plays. I want to see that I listened to a certain artist for X hours. Quantifying total number of plays “punishes” or “rewards” certain artists for having longer or shorter tracks, and skews the stats in a way. The artist that consumed more listening time gets “beaten” by the artist with shorter tracks. There’s definitely value to seeing the total number of playbacks, but to me it’s not as useful to me seeing the total amount of time.
Unlike last.fm, it looks like ListenBrainz records specific track length (?), and also has the MusicBrainz database to fall back on in terms of determining play time, so issues where songs of the same title get mixed up shouldn’t be prevalent.
welcome to the forums~
I’d love to see more of an emphasis on play time too~ that said, you can’t always count on track times being there, because not all ListenBrainz listen submitters* support track times (especially on Android and web), and MusicBrainz doesn’t always have track times either. that said, it might be possible to guesstimate track times when the data isn’t there, as long as it knows whether the track was paused or not.
* I feel like we need a word like scrobble, because I never know what to call this kind of thing… listen submitter? if we don’t, I feel like scrobble could become the default, but that’s a Last.fm thing
As someone who listens to dj setlist, I would like to see listening time
I would like to embed ‘now playing’ from listenbrainz on my website. Last.fm and Spotify have this feature.
Hi @DontMindMe !
Sorry for the delay, you can now easily import your LastFM loved tracks from the page https://listenbrainz.org/profile/import/
Currently only available for LastFM, but we will be implementing the same mechanism for LibreFM