[ art / civ / cult / cyb / diy / drg / feels / layer / lit / λ / q / r / sci / sec / tech / w / zzz ] archive provided by lainchan.jp

lainchan archive - /cult/ - 7039

File: 1490572932682.png (80.9 KB, 300x207, Warez_Depot.jpg)


Would anybody be interested in starting a music archiving group?

I've created a rough rule set based off of the scene to tag and name music however it's adapted to be focused on consistent and easy archiving rather than preing/racing. The idea is to create a p2p archive of music tagged with a set of easy to follow standards which group members can use to contribute releases towards while also quickly expanding their own collections.

I have a fairly large archive of music to get things going (it's constantly growing but it's currently 1.29tb of flac and 28gb of mp3 and upload around 200-300gb a week on slsk)

I've got a few other ideas regarding the project if anybody is interested.


bumping for this lainon
what do you mean post was too empty?


I would upload all my soykaf if the network it was hosted on was private and did not have a single point of failure. Something like an onion, GNUnet, MaidSafe, Morph.is, IPFS, etc. that would create a resilient archive and give strong anonymity to an uploader. Sitting on 5TB of music content, majority of that MP3s.


File: 1490641567044.png (19.02 KB, 200x94, X4Omi38.gif)

Thought I would post some more about the projects goals as I'm going ahead in implementing the naming structure to my archive over the next week or so.

The main aim of the group would be primarily to archive and share music both internally and to anybody wanting to download or share on whatever p2p or cryptonet network they like. We wont be releasing to private/public trackers or actively encouraging sharing to a wider audience but rather to other hobbyists, datahoarders, archivists and music lovers.

Because the focus is firmly on archiving and not releasing it's important to maintain a high level of quality however we also need to be realistic, it's just not practical to spend hours a day tagging and naming music. I know this from spending many hours a day tagging and renaming music. So this is where the rule set comes into play, instead of wasting time worrying about rules (or more importantly if they're being broken) on whether subordinating conjunctions or prepositions need capitalizing a comprise is made in that all words are capitalized unless stated explicitly by the release info. This creates a standard which is easy to follow and more importantly easy to automate.

A quick run down of the current process looks something like this;

*The first letter of all words in the tracks metadata fields are capitalized
*copy is made of original album art file
*copied album art is resized to 500x500 pixels and embedded into the track
*files are hashed using md5
*directory is renamed according to naming standards
*filenames for all files in the directory are then renamed according to naming standards

Essentially what you're left with is something which looks like this;


I can provide a more detailed walkthrough of the process and tools used but all of which can easily be achieved using batch renaming software and MP3Tag actions etc.

If anybody is actually interested in contributing, has any suggestions or questions about getting involved in sharing/archiving or just wants to leech then reply here or if you would prefer you can email me at flacjacket@420blaze.it

I have plenty of other ideas I'm thinking over at the moment but I thought it may be useful to give a more detailed overview of everything so far.

Sounds good but I'm trying to avoid mp3 whenever possible, storage is so cheap these days it seems silly not to favor flac (as well as being opensource) besides when it comes to lossy formats if you have flac you have them all. That is one hell of an impressive collection though and I accept that sometimes it's just not possible to find a release in lossless so I would be interested in seeing what you have.


this sounds really interesting especially after the death of what.cd please keep us updated


I think this is an awesome idea but the rules at >>7053 are giving me pause

It's not that something like this isn't well-intentioned (and downright needed), but for some reason these rules about capitalization are sp00king me a lil bit


Why don't you guys use soulseek?


Simple, it's proprietary.

I had no idea that happened. That is like kicking dead whales down the beach.


File: 1490763604064.png (34.42 KB, 181x200, rms.jpg)

>using nonfree software at all ever for any reason
there is some lenience given to those actively working to liberate things
like, if you're temporarily using soulseek for the sole purpose of getting everything there off and into freedom respecting channels, power to ya
but otherwise, come on anon, step up


I'd be willing to contribute, don't know if I have anything not already widely available, as I didn't rip anything my self, probably not. But I have just about 200GB of music.


File: 1490788423203-0.png (128.42 KB, 200x114, bbs_ad_cte.png)

File: 1490788423203-1.png (25.29 KB, 128x128, index.txt)

To be honest it was the raid of wcd which prompted me to get a bit more serious about this.

I completely get where you're coming from, I've been involved in various filesharing p2p networks, trackers, communities and another scene since the early 00s so I fully appreciate that what I'm suggesting is absolute heresy to some however there is a rationale behind it.

I was very active in wcd (and oink before that) and prior to the raids I had nearly 2tb in buffer. The buffer I couldn't care less about however when I think about the sheer amount of collages and releases I had bookmarked but never got around to archiving/listening to because I spent so long tagging the music I did have I feel a bit sick. I had/have a lot of free time but even I have to draw a line somewhere, it's just not practical to be spending so long worrying about capitalizing or whatever, especially under a rule set which most people got wrong anyway. I came to the realization that the three most important things is the music itself, having the correct release info and having it well backed up/distrubuted. All things I think this project aims to address and lets be brutally honest with ourselves; having 'Of' instead of 'of' doesn't make the music sound any different.

This is such a heated topic that some people will just never agree to this which is cool but something I think we can all agree on is that the state of naming/tagging in the p2p music world is dreadful. As long as they're willing to archive/share the original directory and it passes checksumming then they can tag for their personal collections as they see fit. The aim is to have a standardized and distributed archive everybody can pull from rather than one claiming to have 'perfect' metadata so to speak.

I've been using soulseek since the mailing list days, it's a shame that it isn't opensourced however you have to respect it's longevity and the fact it's still more or less still the go to network for music lovers. I get messages on there from pirate stations, small label owners and the artists themselves which I've never got on other networks. Obviously the more networks the better and hopefully this project can help populate other more freedom respecting/opensource channels but I don't think that others are going to overtake slsk anytime soon.

Great, anybody willing to help is encouraged to get in touch
If this gets off the ground I was thinking about drawing up an application form or whatever people could fill out something with; collection size, Bandcamp collection size, availability to trackers/archives, freespace, recruitment etc

Anyway I've just finished renaming my small current backlog and the A directories of my main collection, thought I would share it here to give people an idea of what things look like.



Why not use http://beets.io/ for automatic tagging? Uses the MusicBrainz library which is about as standardized as it gets right now. Only have to worry about tagging the stuff they don't have, and if that's the case you can do it once and everyone will benefit.


I sent an email with a list of my collection and some information on it.


this; very this.

if you're not going to worry much over formatting, just refer to musicbrainz. less work and greater accuracy, and, like you said, people can use whatever other format for their own collections.


Can we please use either i2p or IPFS for this?


We'll use a single multiterabyte torrent, just to spite you.


epik xd

>the body was west Philadelphia


File: 1491560421217.png (2.98 KB, 120x117, 1470346412596-2.jpg)


and this. However, IPFS is the new eMule with http interface, it will get censored sooner or later.

But MusicBrainz + I2P is going to be awesome.

Yes, put it into a torrent, torrents over I2P are awesome.

Your rules suck. Especially for weaboo soykaf. TLMC (Touhou Loseless Music Collection) is 2TB of soykaf that is out with your rules. Not only the names of the circles there are peculiar, but splitting it into separate tracks would leave you with no torrent peers, and your recording stops being loseless proper, because you loose data between tracks (yes, there are recordings that have cool stuff between tracks).

The proper way to implement this is to wipe all the metadata from the original lossless files, store it in CUE files, rename the files into directory tree based on their hashes (sha256/512 for the win! and you can always hardlink them, if you want to keep your directory organization too), put the result into torrent and distribute over I2P (Vuze and Transmission can use both I2P and clearnet peers, btw). Distribute CUE files in a git repository.


* a lot of peers: wiping metadata is deterministic and reproducible
* easy to get started: get the torrent, run the wiping-hashing-renaming script on your music, recheck using the torrent
* no bikeshedding over metadata, people can fork CUE repo and do whatever they want with it

Other reasons to do it:

* Library Genesis guys do exactly this for books, but they don't wipe the metadata from original files, they just put a DB over the torrent and censor duplicates by hand. Please, look at what they did and read their forums (especially, if you can read Russian, but most of them are in English, so there's no excuse) https://genofond.org/index.php

* There's a tool called "hydrus" http://oxwugzccvk3dk6tj.onion/hydrus/index.html (its 8ch over Tor, because clearnet 8ch is Cloudflared) that does the same for images.


File: 1491603004944.png (14.67 KB, 200x121, 2017-04-07-215811_484x292_scrot.png)

This sounds like a great project. I'd be interested and have music to contribute.
I'm in a bunch of p2p sites so as long as nobody is being anal about releasing on those (especially my own rips) then that'd fine.