Contact Us

Email the Academic Torrents team at

People

Mission

Cite

If you need to cite Academic Torrents please use the following papers:

• Lo, Henry Z. and Cohen, Joseph P., (2016). Academic Torrents: Scalable Data Distribution. Neural Information Processing Systems 2015 Challenges in Machine Learning (CiML) Workshop. http://arxiv.org/abs/1603.04395
• Cohen, Joseph P. and Lo, Henry Z., (2014). Academic Torrents: A Community-Maintained Distributed Repository (p. 2:1–2:2). New York, NY, USA: ACM. http://doi.org/10.1145/2616498.2616528

Legal

Please submit DMCA requests using this Google form
We have never received a DMCA request.
We have never received an order to turn over user information.

Downloading

We recommend the Transmission torrent client which supports all major operating systems:
http://www.transmissionbt.com/
http://sourceforge.net/projects/trqtw/

For downloading from the command line you can use ctorrent locally or when connected over ssh to a server. To install ctorrent run:

$ sudo apt-get install ctorrent  ## for Ubuntu/Debian
$ sudo yum install ctorrent      ## for Fedora/Red Hat
$ brew install ctorrent          ## for Mac OSX

To use ctorrent you first need to download the .torrent file of the file/files that you want. You can do this using wget from the command line:

$ wget http://academictorrents.com/download/30ac2ef27829b1b5a7d0644097f55f335ca5241b.torrent

Then specify this .torrent when running ctorrent. The files will be downloaded as they are displayed on the technical tab of the details page.

$ ctorrent 30ac2ef27829b1b5a7d0644097f55f335ca5241b.torrent

Uploading

The following video details the upload process for a researcher sharing data. There are many different ways to put data on the site. Please contact us if you are unsure how to share your data. The upload form is here: http://academictorrents.com/upload.php

Donating Bandwidth

We would like to avoid the blind mirroring of all data. We provide RSS feeds for subsets of all the torrents grouped into collections.

Per collection RSS feeds allow you to mirror only collections you care about. Each collection has it's own unique RSS feed. You can add the RSS feeds found on this page: http://academictorrents.com/collections.php You can mirror these RSS feeds automatically with a BitTorrent client such as uTorrent or qBitTorrent. If you would like a more basic way of downloading a collection you can use this small script:

for i in `curl http://academictorrents.com/collection/nasa-datasets.xml | grep "<link>" | grep "download/" | sed 's/<link>//g' | sed 's/<\/link>//g'`; do
echo $i;
done

Here is an example to add an entire colllection by the URL:

## add2transmission.sh
xml=$1
auth="username:password"


for i in `curl $xml | grep "<link>" | grep "download/" | sed 's/<link>//g' | sed 's/<\/link>//g'`; do
    echo $i;
    transmission-remote -n $auth -a $i
done

Just run the script with a RSS feed as an argument:

$sh add2transmission.sh https://academictorrents.com/collection/video-lectures.xml

Torrents

For more infomation on how the bittorrent protocol works check out this Forbes article.

You can look inside torrent files with our Torrent Explorer web application: http://academictorrents.com/torrentexplorer/

API

Our API is constantly improving based on feedback from users. Please contact us if you are planning to use the API so that we can make it feit your needs better. The basic api is shown below.

===Entries===
GET /apiv2/entries  -- List all entries 
GET /apiv2/entries?cat=6  -- List entries that are datasets
GET /apiv2/entries?cat=5  -- List entries that are papers
GET /apiv2/entries?cat=6&limit=9999  -- Increase number of results (default 20)
GET /apiv2/entries?cat=5&search=graphs  -- Search for papers with term 'graphs'
GET /apiv2/entry/INFOHASH  -- retreive data about an INFOHASH
POST /apiv2/entry  -- upload an entry
POST /apiv2/entry/INFOHASH  -- modity an entry

===Collections===
POST /apiv2/collection  -- create an collection 
POST /apiv2/collection/collection-name  -- update an collection 
POST /apiv2/collection/collection-name/delete  -- delete an collection 
POST /apiv2/collection/collection-name/add  -- add an item to a collection 
POST /apiv2/collection/collection-name/remove  -- remove an item from a collection 

---

The api fields for uploading a entry to Academic Torrents are shown below. To upload a file you must send a POST request to http://academictorrents.com/apiv2/entry with the following parameters as well as your API Key.

name : "The title of the publication"
authors : "Author1 and Author2"
descr : "@article{,
		title = {The title of the publication},
		author = {Author1 and Author2},
		abstract = {Abstract}
	}"
category : 5
tags : "Tag1, Tag2"
urllist : "http://someurltoafile.com/file.pdf, http://anothermirror.com/file.pdf"
file: data:application/x-bittorrent;base64,ZDg6YW5ub3VuY2UzODp...vMRIjFFguASKUplZQ==

To use the above API Key you can use curl. Curl is a simple way of sending data the server just to test that our API is working. Below is the curl command to test your API key on our server. The -b option sets a cookie. We can use this to pass the API key and isolate it from the post data. The --data tag must be present to instruct curl that this is a POST request and not a GET request.

$ curl http://academictorrents.com/apiv2/test -b "uid=14;pass=71135221ceb8b5279cd15150d2126dcb" --data ""
Welcome to the API test!
You have made a POST request.
API Key Received: uid=14&pass=71135221ceb8b5279cd15150d2126dcb
Your username is joecohen
Everything seems fine, You should be able to use the API

You can also put the API Key directly in the POST request via the --data argument

$ curl http://academictorrents.com/apiv2/test --data "uid=14&pass=71135221ceb8b5279cd15150d2126dcb"
Welcome to the API test!
You have made a POST request.
API Key Received: uid=14&pass=71135221ceb8b5279cd15150d2126dcb
Your username is joecohen
Everything seems fine, You should be able to use the API

You can also put the API Key directly in the GET request.

$ curl http://academictorrents.com/apiv2/test?uid=14&pass=71135221ceb8b5279cd15150d2126dcb"
Welcome to the API test!
You have made a GET request.
API Key Received: uid=14&pass=71135221ceb8b5279cd15150d2126dcb
Your username is joecohen
Everything seems fine, You should be able to use the API

This is a Python API usage example. A larger more complete version is here: https://github.com/henryzlo/academictorrents_uploader

# get base64 of torrent
f = open("file.torrent", 'rb')
b64_torrent = b64encode(f.read())
f.close()

post_params = {
    'uid' : "14",
    'pass' : "71135221ceb8b5279cd15150d2126dcb",
    'name' : "The title of the publication",
    'authors' : "Author1 and Author2",
    'descr' : """@article{,
		title = {The title of the publication},
		author = {Author1 and Author2},
		abstract = {Abstract}
	}""",
    'category' : "6",
    'tags' : "tag1, tag2",
    'urllist' : "http://domainname/backupfileurl.zip",
    'file' : b64_torrent
}

data = urlencode(post_params).encode('utf-8')
req = Request('http://academictorrents.com/apiv2/entry', data)

response = urlopen(req)