<p>Unsolicited bulk email, commonly called "spam", is a widespread problem on today's email network. We want to make sure that unsolicited bulk instant messaging, commonly called "SPIM" (short for "spam over IM"), does not become a widespread problem on the XMPP instant messaging network. To that end, this document specifies a protocol for reporting particular instances of SPIM.</p>
</section1>
<section1topic='Protocol'anchor='proto'>
<section2topic='Reporting a SPIM Stanza'anchor='proto-stanza'>
<p>The protocol for SPIM reporting is quite simple: wrap the complete SPIM stanza in a <spim/> element qualified by the 'http://jabber.org/protocol/spimreport' namespace and send an IQ stanza of type "set" to the entity that shall receive the report. This protocol is intended to be used by recipients of SPIM stanzas and by other entities (e.g., servers or services) that wish to pass around SPIM reports. The following is an example:</p>
<examplecaption='Unsuspecting User Receives SPIM from Evil Bot'><![CDATA[
<presencefrom='makemoney@spimmersheaven.info/bot'
to='unsuspectinguser@example.net'
type='subscribe'>
<status>
You too can be rich! Find out how at
http://spimmersheaven.info/makemoney
Let's chat to make your dreams
come true!
</status>
</presence>
]]></example>
<examplecaption='Unsuspecting User Reports SPIM'><![CDATA[
<p>If the entity that receives the report does not support the SPIM Reporting protocol, it MUST return a &unavailable; error to the reporting entity:</p>
<p>If the entity that receives the report is able to process the report, it MUST return an IQ-result to the reporting entity:</p>
<examplecaption='Report Processed'><![CDATA[
<iqfrom='example.net'
to='unsuspectinguser@example.net'
type='result'
id='report1'/>
]]></example>
</section2>
<section2topic='Reporting a Spimmer'anchor='proto-spimmer'>
<p>If an entity that processes SPIM reports determines that the suspected spimmer is indeed an actual spimmer, it SHOULD report its conclusions to appropriate other entities (e.g., the spimmer's server or dedicated SPIM reporting services). Here again, the protocol is quite simple: wrap the spimmer's JID in a <spimmer/> element qualified by the 'http://jabber.org/protocol/spimreport' namespace and send an IQ stanza of type "set" to the entity that shall receive the report. This protocol SHOULD NOT be used by recipients of SPIM stanzas. The following is an example:</p>
<p>The following rules and guidelines apply to the act of reporting SPIM.</p>
<olstart='1'>
<li><p>The recipient SHOULD NOT report the SPIM stanza to a server or service until it determines that the server or service supports the SPIM Reporting protocol (see the <linkurl='#disco'>Discovering Support</link> section of this document).</p></li>
<li><p>If the recipient's home server (i.e., the server with which it has a registered account or other trust relationship) supports the SPIM Reporting protocol, the recipient SHOULD report the offending stanza to its own server only.</p></li>
<li><p>If the recipient's home server does not support the SPIM Reporting protocol, the recipient SHOULD report the SPIM stanza to one or more dedicated SPIM reporting services if available.</p></li>
<li><p>The recipient MAY report the SPIM stanza to the suspected spimmer's server (but see the <linkurl='#security'>Security Considerations</link> section of this document).</p></li>
<li><p>The recipient SHOULD NOT report the SPIM stanza to the suspected spimmer.</p></li>
<li>Before processing the report, the processor MAY respond to the report by sending a challenge to the sender (e.g., using &xep0158;) in order to make sure that the sender is not sending spurious reports or otherwise abusing the SPIM reporting system.</li>
<li>The processor MUST add the report to a list or database of pending SPIM reports.</li>
<li>If the suspected spimmer is not already on the processor's list of known spimmers, the processor SHOULD use the report in determining whether the suspected spimmer is an actual spimmer (see next section).</li>
<li>The processor MAY report the SPIM stanza to one or more dedicated SPIM reporting services if available.</li>
<li>The processor MAY report the SPIM stanza to other servers it trusts.</li>
<li>The processor MAY report the SPIM stanza to the suspected spimmer's server.</li>
<li>The processor SHOULD NOT report the SPIM stanza to the suspected spimmer.</li>
<p>Not all SPIM reports are valid, and not all JIDs that send stanzas reported as SPIM are spimmers. Care must be taken in correctly determining if a suspected spimmer is an actual spimmer. The following rules apply:</p>
<ol>
<li><p>The processor SHOULD NOT add a suspected spimmer to its list of known spimmers until it has received at least three (3) valid SPIM reports with regard to that suspected spimmer.</p></li>
<li><p>If the processor determines that the suspected spimmer is an actual spimmer, the processor:</p>
<ol>
<li>MUST add that JID to its list of known spimmers.</li>
<li>SHOULD send a spimmer report to the spimmer's server.</li>
<li>MAY send a spimmer report to SPIM reporting services or reputation services.</li>
<li>MAY send a spimmer report to other servers it trusts.</li>
<li>SHOULD NOT send a spimmer report to the spimmer.</li>
<p>In order to discover whether another entity supports SPIM reporting, &xep0030; SHOULD be used. If an entity supports SPIM reporting, it MUST return a feature of "http://jabber.org/protocol/spimreport" in responses to service discovery information requests, as shown in the following examples:</p>
<p>Not all reported SPIM is actual SPIM, and not all reported spimmers are actual spimmers. Processors must take care to ensure that processing of one or a few reports do not result in branding of a legitimate sender as a spimmer, since otherwise the sender could effectively be the subject of a denial of service attack.</p>
<p>Client recipients are discouraged from sending SPIM reports to the servers of suspected spimmers, since the sending server itself could be a rogue domain that could send SPIM from any JID at that domain.</p>
