Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I would have been all over this if I hadn't of deleted my mp3 collection a few years ago and now just use Spotify. I was pretty OCD in terms of managing my mp3 collection, so dumping it all and using a streaming service is a welcome life simplification.


Music is way too important in my life to temporarily rent it from some company that may or may not be around or even have the same content available in a few years.


Well said. I've been meticulously maintaining a single audio collection for years now; sometimes I wonder why I do it. But what you said sums it up nicely.


Google Music has an "upload your own music" feature (up to 10k songs free! That's 500 CDs and for anyone who isn't in the music making profession I can't imagine needing more). It ends up being a sort of dropbox-like mechanism (you can download as many times to your phone, only 3 times to your computer though).

In case of failure of Google I still have a decent amount of music on my phone and I do have the stuff on my phone. In the (much more) likely case of drive failure I can get my music again (I could even circumvent the download limit through a Google->phone->PC thing but that would be a pain).

I think it's important not to overestimate your personal capacity to not lose things compared to companies like Google (especially considering that I am not aware of many companies that fell off the face of the earth in less than 24 hours for you to lose all your data)

edit: google music player on android is pretty annoying though (just slightly better than the stock player from old android). The fact that you need a data connection to play a random mix of your music startles me.


The minute Google converts one of my flacs to 320kbps mp3s the file is lost. I don't want to circumvent anything in order to have access to my data. Why does it startle you? When you don't pay for something you are the product. I would be more surprised if google gave you all this great music and did not expect anything in return.


>The minute Google converts one of my flacs to 320kbps mp3s the file is lost. Only lost if the file is lost on your end too.

>I don't want to circumvent anything in order to have access to my data.

If you're an audiophile/actually doing things with the raw files, then this is not for you, but for most people (read: 99% of people) the diff between 320kbps mp3 and flacs is not there.

> Why does it startle you? When you don't pay for something you are the product. I would be more surprised if google gave you all this great music and did not expect anything in return.

Sure, I don't care about sending my music usage stats(hell I already do it publicly with last.fm) but at least have an option when my data connection is down.


The difference will exist at some point in the future when I want to convert those flacs/mp3s to a new lossy format for some reason (e.g. when the only place you can buy an mp3 player is in an antique shop). Then when my great grandson wants to listen to my music, and you can't even buy a player for whatever that format was.


The link that started this discussion is about an open source CLI based music manager. Beets boasts about its numerous features that work with mpd. We have not been talking about music management for the 99% for a while now.


Your description of Google Music makes it sound awful and completely unusable.

It does take a bit of effort to manage my own music collection, but I feel like I am well protected against data loss by having it synced between a NAS, external server, and Crash Plan. 90% of my listening is done via the external server running Subsonic, using both the web interface and the excellent mobile app iSub. I also try to always pick up a vinyl at live shows since they are big and nice to look at (and occasionally listen to), and usually come with a free mp3 download.

There are no limits on how many times I can access my music files. I can access them any way that I please, from pretty much any device, forever. I couldn't have it any other way with my music collection.


and its more work. Obviously your solution is more durable (mainly due to owning all the pieces), but then again your solution is also more durable than something like Dropbox.

This is a halfway solution for people who are not necessarily good at backups. And I get to listen to the music on the web interface or on my phone (which is basically all I care about). Maybe I'm an awful person, but I greatly prefer this to the USB/SD card shenanigans of yore.


Last time I uploaded my own music, it replaced all of the "explicit" songs with the censored songs, it really tuned me off to the product.


Many people who aren't professional music anything listen to music during working hours. Thats ~ 2000 hours a year. Your 500 cds, ~ 500 hours. I think you sorely lack imagination.


I was kind of the same, but the realization that I was a media hoarder came when I was designing a media server for myself where I could access my media anywhere -- and the thought struck me that I was essentially just rebuilding a lot of the on-demand paid services out there, just with less media.

Been using Google Music ever since.


You can always rent and buy.

I'd be nervous buying DRM'd content from a supplier that may not be around in the future. But renting I could care less - unless I paid for years upfront - but you generally don't do that.


…until Spotify dies or is replaced with something better. I haven't found a single way to manage a music library because I don't have a single music library. It's kind of crappy having it spread across MP3s, FLACs, and various music services :/


Try Clementine, it has both a local library and integration of Google Music, Spotify, Grooveshark, etc.


I always dismissed this because it said amarok in every description. I just noticed it does not depend on kde-runtime. Someone needs to do this for rkward.


Why is Amarok something you would dismiss it on?

Sure, they butchered it up in version 2.0 and above, but I've always felt version 1.4 was probably the best library manager I've used.


I dismiss amarok and rkward for the same reason: a dependency on kde-runtime. This is just a philosophical decision that the kde community made that I disagree with. Why does rkward, a stats program, need to have the phonon-backend-vlc video or libkntlm dependencies? An RDF daemon or kio-audiocd (i don't have a CD drive)? I use awesome as a window manager, so I do not have KDE/GNOME installed. However i can and sometimes will install most "gnome" apps because they do not have a dependency on gnome-core. For comparison:

Clementine vs Amarok

  # apt-get --no-install-recommends install clementine
  ...
  The following NEW packages will be installed:
    clementine libechonest2.1 libftgl2 liblastfm1 libprojectm2 libprotobuf8
    libqjson0 libqt4-opengl libqxt-core0 libqxt-gui0 projectm-data
    ttf-dejavu-core
  0 upgraded, 12 newly installed, 0 to remove and 19 not upgraded.
  Need to get 6,715 kB of archives.
  After this operation, 28.1 MB of additional disk space will be used.

  # apt-get --no-install-recommends install amarok
  ...
  The following NEW packages will be installed:
    amarok amarok-common amarok-utils kate-data katepart kde-runtime
    kde-runtime-data kdelibs-bin kdelibs5-data kdelibs5-plugins kdoctools
    libattica0.4 libdbusmenu-qt2 libdlrestrictions1 libepub0 libfam0 libiodbc2
    libkactivities6 libkatepartinterfaces4 libkcmutils4 libkde3support4
    libkdeclarative5 libkdecore5 libkdesu5 libkdeui5 libkdewebkit5 libkdnssd4
    libkemoticons4 libkfile4 libkhtml5 libkidletime4 libkio5 libkjsapi4
    libkjsembed4 libkmediaplayer4 libknewstuff3-4 libknotifyconfig4 libkntlm4
    libkparts4 libkpty4 libkrosscore4 libktexteditor4 libkxmlrpcclient4
    liblastfm1 libloudmouth1-0 libnepomuk4 libnepomukcore4 libnepomukquery4a
    libnepomukutils4 libnl-route-3-200 libntrack-qt4-1 libntrack0 libphonon4
    libplasma3 libpolkit-qt-1-1 libqca2 libqjson0 libqt4-opengl
    libqt4-qt3support libqtscript4-core libqtscript4-gui libqtscript4-network
    libqtscript4-sql libqtscript4-uitools libqtscript4-xml libsolid4 libsoprano4
    libssh-4 libstreamanalyzer0 libstreams0 libtag-extras1 libthreadweaver4
    libzip2 nepomuk-core-data nepomuk-core-runtime ntrack-module-libnl-0
    oxygen-icon-theme phonon phonon-backend-vlc plasma-scriptengine-javascript
    shared-desktop-ontologies soprano-daemon
  0 upgraded, 82 newly installed, 0 to remove and 19 not upgraded.
  Need to get 75.9 MB of archives.
  After this operation, 202 MB of additional disk space will be used.
  
Gnumeric vs Rkward

  # apt-get --no-install-recommends install gnumeric gnumeric-plugins-extra
  ...
  0 upgraded, 8 newly installed, 0 to remove and 19 not upgraded.
  Need to get 456 kB/7,289 kB of archives.
  After this operation, 34.0 MB of additional disk space will be used.

  # apt-get --no-install-recommends install rkward
  ...
  0 upgraded, 70 newly installed, 0 to remove and 19 not upgraded.
  Need to get 61.1 MB of archives.
  After this operation, 143 MB of additional disk space will be used.


You just found your music management system:)


My great life simplification was to aim for a library I can enjoy when shuffled without having to babysit the "Next" button.

Sadly I was far from that in my most meticulous days where I wouldn't even delete tracks I hated because that would violate the sanctity of the discography. And I didn't want MusicBrainz to doubt my thoroughness.


That's funny.

That's a bit of a holy grail. Because you need the shuffle to track your mood. I found that it really subdues my listening, in that I don't always want some punk music interupting a mellow interlude. So I've gone back to listening to albums, or crafted playlists.

There are album fillers, and there are tunes that are just bad, and they are better gone. Though granted, the holes don't feel quite right.

I'm trying to understand your MusicBrainz remark, are albums with missing tracks unidentifiable?


You raise a good point about mood tracking. I've found that iTunes's genius shuffle does a decent job of shuffling a single genre of music, so you don't generally get harsh interruptions.

The holy grail is most definitely a library (or app) where you click "shuffle all" and enjoy every song it plays. Songza, genius shuffle, 8tracks etc. all try to tackle this, but I still think we're far from that point.


Do you miss any music? Setting aside the Little Sister concerns Spotify is not an option for me because I like concert bootlegs and some obscure folk artists. Have they added the ability to limit the upstream bandwidth? I remember a lot of my gaming friends complaining that the client was the ultimate gun NERF.


Yeah, the Spotify catalogue isn't exhaustive by any means, but its pretty darn good. I miss some more obscure items from my mp3 collection, especially some techno/dance stuff I enjoyed in my early 20s (I shed a tear for that occasionally) - but overall I feel like the benefits of shedding a 30GB collection of files whose metadata I was obsessed by outweigh the negatives of dumping my personal music collection. If Spotify dies I'll just move onto the next service.

If there are a few obscure mp3s you can't live without you can import them into your Spotify client. They'll get auto synced to any mobile devices you have connected and the client installed on too (although not cloud synced to other computers - you'd have to copy them manually).

Spotify does a lot of local caching, so once you've listened to a track once it'll likely be cached so it doesn't hammer your collection too badly. Pretty sure I've continued to get good gaming pings while my gf listens to Spotify in the next room :)

EDIT: I did not know about the P2P stuff. Huh! Interesting ...


I've heard the p2p stuff is going away on the desktop app, but you could easily get around it by using the webplayer or mobile app. AFAIK, there's no way to limit the upload speed in the current desktop implementation (in the app).

The content is generally pretty good but they depend on the artists/labels getting it into aggregators or making it available. Bootlegs won't be there because that would probably piss the bands off.


Terabytes of bootlegs that do not piss the bands off:

https://archive.org/details/etree

Short list of musicians granting permission to have bootlegs hosted on archive.org/etree:

  311                                   Blues Traveler
  Camper Van Beethoven                  Cowboy Junkies
  Cracker                               Dark Star Orchestra
  Death Cab for Cutie                   Derek Trucks Band
  Disco Biscuits                        Donna the Buffalo
  Drive-By Truckers                     Ekoostik Hookah
  Elliott Smith                         Furthur
  Grace Potter and the Nocturnals       Grateful Dead
  Guster                                Hank Williams III
  Hot Buttered Rum                      Jack Johnson
  John Butler Trio                      John Mayer
  Keller Williams                       Little Feat
  Live Music Archive                    Local H
  Lotus                                 Matisyahu
  Matt Nathanson                        Max Creek
  Michael Franti and Spearhead          moe.
  My Morning Jacket                     Of A Revolution
  Perpetual Groove                      Phil Lesh and Friends
  Radiators                             Railroad Earth
  Ratdog                                Ryan Adams
  Smashing Pumpkins                     Sound Tribe Sector 9
  String Cheese Incident                Tea Leaf Green
  The Breakfast                         Umphreys McGee
  Warren Zevon                          Ween
  Yonder Mountain String Band  

My personal favorite is Danny Schmidt, a singer songwriter rom TX/VA: https://archive.org/details/dannyschmidt2007-12-13.sbd.flac

The 4,700+ other artists: https://archive.org/browse.php?collection=etree&field=%2Fmet...

24bit flacs: https://archive.org/search.php?query=%28format%3A%2824Bit%20...


Sorry, I should have clarified that comment before posting it.

Many bands allow audience taping but I'm not aware of m?any that allow for commercial distribution of those recordings. If Spotify attempted to distribute those recordings against the bands' policies then it would probably piss off the bands.

That Danny Schmidt is pretty great stuff. You're right about Spotify's lack of content when it comes to stuff like this-- they have all his studio albums but are kind of weak when it comes to live stuff, they have one album. Archive.org definitely fills in this gap (I use it all the time for Dead shows).


Now I see what you were saying. I think the studio process strips the humanity out of music. I have been working on a feature request / outline of functionality for beets and etree that I need to finally submit to sampsyo. How awesome would this be:

  $ beet etree --get dannyschmidt2007-12-13.sbd.flac
  etree: requesting ogg format
  etree: fingerprints verified
  etree: Success: Downloaded Album: 2007-12-13 - The Cactus Cafe
  $ 

Archive.org has a somewhat documented API that describes all of the "resources." A lot of the shows even come with cover art for those that are into that.

JSON for recording: https://archive.org/metadata/dannyschmidt2007-12-13.sbd.flac


I'm with you; the only thing that makes me keep my MP3s is the fact that the cloud isn't absolutely everywhere. When I run I want a $40 MP3 player and these aren't cloud capable.

Otherwise, where ever I am I have high-speed Internet and I don't bother with my MP3s anymore. I do a lot of Pandora. To play music on the big stereo I just use Google Music and the Nexus 7, which connects to the receiver by Bluetooth.

I used to curate the music, organize the files, connect to machines via ssh to play them back on different speakers, etc etc. Cheap tablets, cheap phones, Internet everywhere, and music services have killed all that for me. I used to have a Linux PC hooked up to the stereo and I would ssh to it to control playback. Those were the days. Now the Nexus 7 is a far superior device in every way.


I've attempted to move all my files into Google PlayMusic and manually correct everything. The main problem is PlayMusic seems to always match my explicit songs to edited version and that drives me crazy. I will have it "fix the match" to only find later that the songs are edited again. So for now I spotify.

I kind of hoping spotify does something similar to beats where $15 gets you accounts for all family members. I'm not sure how that works but it seems to be AT&T only so maybe it's subsidized?


My problem with commercial services like Spotify is most of the music I listen too isn't in any of them. I imported my 20,000 mp3s into iTunes Match and it matched about 20%.


My problem is the recommendation algos. If your music preferences do not fit in nice little boxes the recommendations are useless.


It wouldn't matter though, if once you tried a recommendation, you could vote it in or out of future plays.


There's also the lack of information on which particular edition of an album you're getting.

If it's more than a couple of decades old, there have likely been a couple of terrible remasters in the mean time - probably easier to get your hands on too.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: