Astiga Roadmap - 2025
It's time to update the roadmap for Astiga development for 2025!
This is our chance to signal what broad areas we're going to be working on this year. As ever, we can't be too specific as to features or timescales, or over committal to any one thing for fear of changing priorities, but this is our way of at least communicating how we are steering the service.
Indeed, this past year has been a good lesson as to why we avoid being too committal. We've been hit by a number of unexpected requirements and have been finding a growing trend in integration with other cloud services is an increased burden on verification. For example, Google now require a month-long assessment process, yearly, which also costs us a significant amount of money (the assessments must now be performed by outside independent assessors). Hopefully this helps build credibility, but it's the kind of thing that can come unexpectantly and throw us off-course!
Last year's roadmap is here: https://community.asti.ga/discussion/662/astiga-roadmap-2024
How do we decide what to focus on? Well, basically it's a combination of ensuring the service can support itself, financially, and delivering those features that are requested by our current users.
This year, we're going to concentrate on broadly the same topics, looking to make up for ground we lost last year... Let's separate it to four main areas we're looking to improve: library browsing, playlist handling, our mobile apps and new integrations.
Library enhancement
A large part of the usability of a music player comes from the nature of the library. If you have a library that is not well organised it is difficult to use the service, no matter how slick the service itself is.
This doesn't apply to all of you of course... if anything, Astiga users tend to have tidier libraries than most. However, a large proportion of our users don't have sparkling libraries and it's a big turn off when "it doesn't look like Spotify".
In the background we've been working on combining Astiga's sync process with bliss to correct and complete library data. The first steps are currently in alpha testing with some of our users, and is currently limited to fixing album names, multi disc organisation (no more "All Things Must Pass Disc II" style names) and adding artist biographies and images.
We're hoping to open this up to more people for testing soon, and also start to expand the features; we've been getting requests for genre consolidation, track name fixing and more. This feature is disableable so you can go back to using your tags exclusively.
Playlists
Playlists are important to a lot of our users, and there're a number of areas we want to improve:
- Performance of large playlists on mobile apps.
- User-led fixing of playlists and orphaned/missing tracks.
Apps
This one has been blocked by access to skilled resource. If you know any good mobile developers who'd like to work on Astiga, let us know!
Broadly we want to (and these will be our focus in these areas):
- Improve the UX of the Android version
- Launch an iOS app
Integrations
We want to ensure Astiga is available for you to be used on all your devices. We're currently in the early stages of testing for integration into the Sonos platform and there are other ideas that have been put to me: Android TV, WiiM, Alexa, Home Assistant, Roku and more.
Outside of playback, we could also integrate at a storage level, providing more convenient methods for integrating self-stored NAS drives with Astiga.
So those are the main areas we're working on. Feel free to drop us a line, here or by email, if you have something to say about these!
Comments
Thanks. Will we be able to disable these corrections via Bliss? I have a lot of obscure music and bespoke album names. It has taken me decades to get my library perfectly how I want it and the last thing I want is it to be undone. I do not want any corrections to my library by a generic online service (whether you own it or not).
I'm really very concerned indeed about this. I have my own genres, album names, sometimes even artists where this is needed. None of these online services know my genres like I do, they have generic tags - it's in part why I built my own library and use Astiga.
It needs to be opt-in NOT opt-out. I cannot stress this enough. If my library is modified in any way without my prior approval, then I will end my subscription, I take it that seriously. Frankly, I don't see why it is any of Astiga or Bliss's business how I structrure my library.
This already appears to be happening. I am perturbed that some kind of update is clearly going on without my persmission as during the current sync artists are being reported as 'enhanced'. I didn't agree to this and don't want it. I have not opted into anything or being approached about this.
Dan, I have contacted you. I need to opt out of anything that might modify my library and I can see this 'enhanced artist' addition going on already in my latest sync.
Hi @Coyley, viewing library enhancement is opt-in.
Library enhancement is implemented as an additional 'layer' on top of the existing library structure. If you have library enhancement disabled then you see your original library as you always have done. I can see yours is currently disabled.
This already appears to be happening. I am perturbed that some kind of update is clearly going on without my persmission as during the current sync artists are being reported as 'enhanced'. I didn't agree to this and don't want it. I have not opted into anything or being approached about this. Why is this happening?
Because sync is required to build the library _before it is viewed_, the library enhancement process also needs to be run in advance, in the same way.
However, as described above, the outcome of the library enhancement sync is not displayed unless you click the button in your settings (this is only visible to a few users at the moment).
I guess we need to improve the sync logs in some way, to make it clear that the library isn't changed.
Thanks Dan. I appreciate the prompt and reassuring response.
I was completely unaware that I had this disabled. It really does need to be opt-in. Please can there be a post on this, clarifying where this is set? I'm lucky that it was disabled, I am unaware of it at all.
Is it adding another meta-tag at library level or modifying the tags of the music? Can the user override it or it is on/off? If it has run, is there a rollback? These things need some clarity - prior to rolling out.
I back-up from my online library, so tags are modified, without knowing it I could lose masters I have tagged a particular way.
For a detailed library, refining it to Bliss based tags doesn't help. I have about twenty different classifications of Soul music for example, based on sub-genres, locations etc. These really matter - New Orleans Soul from 1960s is incredibly different to Soulful House from recent decades for example. Key sub-genres such as 'Deep Soul' or 'Beat Ballads' do not exist in these services, but to a fan of the music, matter greatly. It is the same for such as Rock Music - Prog, Indie, Thrash, Power Ballads, Psychedelic, Garage, Grunge, Stoner etc all being very separate from each other. These general tagging services wouldn't pick up any of that expert knowledge.
I still don't understand why artist enhancement is being shown within the sync of my library. What I mean is, I synced one directory and my whole library is now showing as trying to enhance every artist in the library. With 100K tracks that's been running a while. I don't understand how syncing one folder can initiate this for the whole library anyway.
I have 100K tracks in my Astiga library as you know, and easily have that again I can add over time now that I have expanded my storage further. So I'm keen to ensure the work I've done on the library is preserved.
Thanks
Mark
I think what has happened here is that, as I'm about to invite a few more frequent listeners to this, I'd enabled library enhancement on a number of accounts, yours included. Knowing it's non-destructive, I wanted to do this before sending the invite so that sync has the time to populate the data.
I was completely unaware that I had this disabled. It really does need to be opt-in. Please can there be a post on this, clarifying where this is set?
It's on the settings page at https://play.asti.ga/settings:
That screenshot shows it enabled - in your case the slider should be slid to the left.
I still don't understand why artist enhancement is being shown within the sync of my library. What I mean is, I synced one directory and my whole library is now showing as trying to enhance every artist in the library. With 100K tracks that's been running a while. I don't understand how syncing one folder can initiate this for the whole library anyway.
Oh, ok, yes, we should only try to enhance those library elements involved in the folders you chose. I'll add that as a ticket.
So, given the above I turned off the library enhancement facility for your account. The next sync shouldn't run the enhancement. Also you won't see the above slider now (unless the page is cached).
All this said it would've been interesting to hear what you find with it enabled! Currently only artist bios/images are added, and album name and split multi discs are fixed.
Thanks Dan. I don't see the Artist Enhancement slider now as you say.
I wouldn't mind artist bios/images being added so long as the artist names themselves aren't modified - I've had enough problems with various apps trying to clean up artist names over the years. Also, I already have a root image in every main artist folder and that would be my preferred one. Again, this has taken a long time to do.
The issue of artist renaming is that it often fragments them. An artist like Brian Eno has collaborated with dozens of artists but he is my primary interest so I will often tag the album artist as Brian Eno to collate his music together, even if there are others involved. By going to the tags in Bliss, that will likely be undone. The issue also arises with 'featured artists' where a singer works with a musical artist, you then end up with many variants that create further fragmentation, when the core interest might be in the musical artist not the featured singers who are only on one or two songs. That's fine in a casual library, but not in one where you're collecting a particular set of artists.
From the above new enhancements, in particular I would not want the album name or multi-disc aspect fixed. For example, I put the artist name at the start of the album. I really don't get why services don't. Seeing an album view A-Z is rubbish without it, it's otherwise not really very useful. If I want to see albums A-Z - I want to see albums by The Beatles (as an example) together. When I view by album in streaming services they just look a random jumble to me, accidentally favouring those at the top who might start with A. So tagging as I have gets that how I prefer it, which the enhancement you describe will undo.
I have myself split muti-disc albums into separate parts in the album naming, to suit the way I prefer it rather than using the Disc tag which is inconsistently recognised across services and devices. I woudn't want that undone either.
That kind of thing is why I don't want album names changing. A good intention could undo all that work.
I think Adding 'rich data' can be helpful, but modifying data or how it is shown is another thing entirely for me.
For the reasons quoted above and others I'm happy to explore, it's really important that the enhancements don't ever become the default. It's also why I am repeating that it must be opt-in, not opt-out. You could even have a 'Collector' and a 'Listener' type delineation to meet both needs.
Serious music collectors have taken time to get their libraries how they want them and it's why they aren't on streaming services. I can appreciate the benefits of the changes for others, but it's absolutely vital that nothing is changed in my library. It's why I'm here and very happy to pay for a service that let's me define my library how I want it, not just at generic streaming services.
From the above new enhancements, in particular I would not want the album name or multi-disc aspect fixed. For example, I put the artist name at the start of the album.
I'm hearing you about wanting to preserve your hard work, and as above this feature is non-destructive and opt-in, so that should be moot.
I would say the whole point of the bliss project is to make these things configurable. There, there's an optional "canonical artist" rule to override release level artist aliases, for example. We've had a long standing feature for "derived tags" where tags such as album name can be formatted according to other data - so in your example that might be "[artist name] - [album name]" or whatever. You can even change the capitalisation of tags if you want.
All this said, it might be worth considering that this goes beyond "tags" as we traditionally know them. If the metadata is stored in a way that better reflects the complexity of musical releases, we can go further than traditional tagging and give a richer experience. Your Brian Eno page will still show all the albums he's the main artist of, the songs he's involved in, other albums he's contributed to and more. You shouldn't have to choose who an album is assigned to and live with that. That's the vision anyway.
And as I've said, the default is still there to use your existing tags.
I have myself split muti-disc albums into separate parts in the album naming, to suit the way I prefer it rather than using the Disc tag which is inconsistently recognised across services and devices. I woudn't want that undone either.
Just to be clear: that wouldn't be undone either, because Astiga doesn't write back on your files, so if you point another app at your files it has no clue what Astiga has stored.
Again, this is something bliss is used for - doing things like track number formatting to get around disc number support in less capable players, such as old car stereos etc.
it's really important that the enhancements don't ever become the default
Depending on your definition of "default", this might be something I disagree with. This feature is exactly intended to make startup with the service easier by presenting your library more accurately; badly organised libraries are far, far, far more common than well organised ones. However, I agree it might be best not defaulting existing users to library enhancement, because it might be rather surprising. 🙂
Yes apart from its other failings such as not using Genres at all (what?), the modifications made automatically to the way the library was presented were why I left Plex. Server level changes would often overwrite configurations I made. Whether at the tag level or not, what's seen is the reality to the user and it only really worked at a superficial level for casual users. It was infuriating for me and I was constantly trying to adjust what it showed, until I gave up. I also find most online bios pulled through are poor and often bring back the wrong artist when you get beyond chart artists, regardless of which service is used.
Your point about going beyond tags only works if I assume I'm on Astiga forever, which is nice to think but technology history indicates it is unlikely. If I have to or chose to move my library, it will always end up back at the tags. I have been forced to move my library off withdrawn services including Amazon Music, Google Music Play, left Plex and Subsonic all in the last ten years.
While that might not be optimal about tags, it is the basic continuity position. I also use my music files across Astiga, a master library with far more music included over network and a subset on a hardware music player for travel. So Bliss may be wonderful,but it wouldn't work across all those sources. I appreciate I maybe an edge case and appreciate the work on it.
Anyway, thanks for explaining.
Understood, and thanks for explaining!
I would love to hear more about the app "UX improvements" when that part of the roadmap starts getting fleshed out into specific features.
My personal feeling is that the user experience in the Astiga Android app has always been superior to the browser and if anything could mostly use optimization and bug fixing so that navigation is faster and getting from current queue items to related albums/artists can happen with less touchpoints.
"UX improvements" is very vague, so I'd love to hear if this is being driven by community engagement and user feedback or simply a desire to align user experience with other popular music services. I would much rather have a fast, well organized UI informed by how users hosting their own large libraries like to find their music, rather than an attempt to look like Popular Streaming App because large user base =/= ideal user experience imo.
Looking forward to the year ahead, thanks for always keeping us up on what's happening.
For the app UX improvements I wasn't thinking much that is revolutionary. Here are some headline things I've thought of:
There are also other things such as the playlist performance issues which I'm not classing as UX here.
Love the idea of offline mode being a first-class citizen. The big list is pretty clunky. Playlists getting promoted out of the hamburger menu would also make sense given the feedback around playlisting.
Personally I've never found the home page confusing and I like that all the different navigation options are right there, easy to access, as soon as I open the app. I do gravitate to Recently Added albums, Starred albums, and albums/artists alphabetically though. Starred tends to be pretty slow to load in compared to other tabs but I have grown to use it more and more as my library has grown. If the home page defaulted to assuming I want one particular navigation view I feel like I'd start noticing all the times I have to navigate away from the home page so I can access the category I want, but maybe I've just shaped my habits around how it's been since I adopted the platform. Open to that changing if the underlying navigation still works efficiently.
Unsolicited user feedback below~
I would use Recently played and Most played if they were tweaked. I can't tell if Recently played is rendering a predetermined number of albums or over a predetermined lookback period, but if I play an album and it's already showing up in Recently played it doesn't get moved to the top of the list and I may have to scroll a while to see it. If it was updating to actually put the most recent thing I listened to at the top I would use it much more often. Similarly, Most played is a non-configurable lookback period. If I could set it to Most played in a period (30d/60d/90d/365d/all history, or just default to all history) I would use it a lot more and would appreciate the insight into my own listening habits.
It's possible I've made a ticket for this in the past, but I just wanna plug the "show artist/album" links one more time. They are super nice quick links and I wish they showed up on the now playing screen so I could go directly to the current artist or album playing. I know that if I go to the current queue list view and click on the ellipses by a song I can go to album or artist.
I also like the search feature when I'm looking for a specific album/artist I want to listen to and don't wanna scroll through the alphabetical list, but the search has a bug where it won't hide the keyboard once the search query has been sent, so I have to lock my phone screen to get the bottom half of the UI back.
Lol after literally 20 seconds of testing that I failed to do before sending my comment I found that the keyboard issue is because I use Microsoft Swiftkey and not the native Google keyboard on my android. So I guess still a bug but more uniquely my problem.
Thanks for that @buppi , those are also like the smaller UX improvements I was thinking off. I'll put a reference to this in the "big list".
@buppi ...
I can't tell if Recently played is rendering a predetermined number of albums or over a predetermined lookback period, but if I play an album and it's already showing up in Recently played it doesn't get moved to the top of the list and I may have to scroll a while to see it.
I think by default this list is cached. If you scroll to the top of the list then pull down to reload it, does it go to the top of the list then?
Seems like enhancement is affecting my auto sync. I added a load of files to my drive yesterday but not one has been added to my library. I didn't have this issue before I enabled enhancement. I a going to switch enhancement off and see if that changes things.
Wow. And there you go. As soon as I switched enhancement off, I could see the new files!
Now I need to check I haven't been stupid, so I will turn it back on, add some files, and report back
New files added. Overnight scan happened.
No new files in library on the app. But online I can see them. In fact, the Recently Added parts are totally different.
Hi - the switch affects both Web and app. Remember the app is cached though, so you need to pull down from the top of a list to refresh it.
Can you get in touch with info@ast.ga and let us know which files weren't showing?
Thanks for having it enabled though!
Library enhancement should only add artist info in the context of the initial folder.
This should now be fixed.
We've also improved the artist info layout, following feedback, and tried to make the sync log wording more representative of the fact it's not destructive.
@kalowski I'd love to look at your issue - if you can provide the name of files I'll take a look. We do know of an issue where, if there's an error while inspecting a file, we don't add the original data to the enhanced library, which we should do as a fallback. We'll implement that soon.
What's the best way to provide the file name? Just on here? Are you able to "look" at my app in some way. Thanks, @gravelld
I can see your library entries (the top level metadata you see under the "Details" link on the web app) from the path name you give me. From that I can work out what went wrong during sync, I hope. Send the path to info@asti.ga, but only if you like.