= The Public Test Grid =

This page is about the "Test Grid", which is also called the "pubgrid".  The pubgrid has several purposes:

 * to make it easier for people new to tahoe to begin to experiment
 * to enable small-scale trial use of tahoe
 * to help the tahoe community gain experience with grids of heterogenous servers without a pre-existing social organization

The pubgrid also has two critical non-goals

  * The pubgrid is not intended to provide large-scale storage, and it is not intended to be reliable.   Don't store any data in the pubgrid if losing it would cause trouble.
  * Don't view the pubgrid as a free hundreds-of-MB backup service.

There is no public gateway; that was [http://tahoe-lafs.org/pipermail/tahoe-dev/2011-July/006572.html disabled on 2011-07-24]. Note that the introducer furl has changed recently (as of November 2012).


== Cautions ==

The canonical way to access Tahoe-LAFS grids is to run your own client node.  Having one's own node is necessary for the data that should remain private (plaintext, capabilities) to remain on your computers, while storage nodes provide storage of ciphertext.  However, writing to the publically-writeable test directory (below) means that others can see and change your files.

Do not confuse the security properties one gets with tahoe when running a client node with how the public web interface (now disabled) used to behave.

== Publically writeable test directory == 

URI:DIR2:ctmtx2awdo4xt77x5xxaz6nyxm:n5t546ddvd6xlv4v6se6sjympbdbvo7orwizuzl42urm73sxazqa .  Note that anyone may view and change this data.

== How To Connect To The Public Test Grid ==

The test grid is subject to being updated at random times, so compatibility
is likely to break without notice. Also note that many upgrades require all
files and directories to be flushed, especially as storage formats change.

The test grid is currently running an 1.0.0-compatible release (allmydata-tahoe: 1.8.2) (see "My versions" on the welcome page for the current version of the web gateway server).

Set up the code according to [source:docs/quickstart.rst] and [source:docs/running.rst]. This creates a client node in the {{{.tahoe}}} subdirectory of your home directory. Edit the following lines into the {{{.tahoe/tahoe.cfg}}} file:

{{{
#!html
    <table cellpadding="10px" border="1px solid black" border-collapse="collapse">
    <tr>
        <td>
            [node]<br>nickname = <b><i>Another nickname than "None"</i></b> 
        </td>
    </tr>
    <tr>
        <td>
            [client]<br>
            introducer.furl = <b><i>pb://u7w4hdwgw5lfsuw7cnovh4jcoyvlne7o@128.59.153.44:51644/introducer</i></b><br>
        </td>
   </tr>
</table>
}}}



Then run {{{bin/tahoe start}}}.

We created a shared public directory: feel free to use it for experimentation -- once your node is up and listening on port 3456, [http://127.0.0.1:3456/uri/URI:DIR2:ctmtx2awdo4xt77x5xxaz6nyxm:n5t546ddvd6xlv4v6se6sjympbdbvo7orwizuzl42urm73sxazqa/ this URL] should give you access to that directory.

== Social Norms ==

By running a client node, you can store data in the pubgrid.  The storage is provided by people who run storage nodes, and they share disk space and network capacity as a courtesy to the tahoe community in order to help newcomers and promote experimentation.

Norms for the pubgrid are similar to those for some private grids:

 * If you are just trying out tahoe and not contributing a server, only store a small amount of data, perhaps 1-20 MB.  This is perfectly fine; people contribute resources so that new people can experiment.
 * If you are a more serious pubgrid participant, contribute some amount of disk space and run stable servers with public IP addresses (so that clients can connect to them).  Then, only store about half as much in shares (3.3x expansion for 3/10 encoding) as you provide in storage.
 * People storing more than the above small amounts of data, as well as those running servers should be on the [http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev tahoe-dev mailing list].

Suggested server setup:

 * Enable expiration with 1-month lease maximums on your node, to keep it from just filling up.