moparisthebest/xeps
1
0
Fork 0
mirror of https://github.com/moparisthebest/xeps synced 2024-11-23 01:32:22 -05:00
Code Issues Releases Wiki Activity

definition list

Browse Source 
git-svn-id: file:///home/ksmith/gitmigration/svn/xmpp/trunk@2998 4b5297f7-1745-476d-ba37-a9c6900126ab
This commit is contained in:
Peter Saint-Andre 2009-04-06 03:50:37 +00:00
parent a9aea08693
commit aaa9f5d1f1
1 changed files with 8 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
xep-0148.xml
View File

@ -98,7 +98,7 @@
]]></example>
</section2>
<section2 topic='Messaging Hints' anchor='hints'>
<p>Even smart people say stupid things, and we are all familiar with the experience of having said something stupid (or just average) and realizing later that one could have said something exceedingly clever. To prevent people from saying stupid things and to help users appear as smart as possible, the mod_iq jabberd module provides hints to users regarding what to say at a given point in a conversation, based on the advanced linguistic analysis technologies described under <link url='#impl'>Implementation Guidelines</link> below. A user can ask for a hint by sending the complete message to the server itself, wrapped in a &QUERY; element qualified by the 'jabber:iq:iq' namespace. (While it may be argued that this functionality could be provided client-side, thus saving a roundtrip, it is consistent with the Jabber philosophy of "smart servers, dumb clients" as explained in &xep0134;.)</p>
<p>Even smart people say stupid things, and we are all familiar with the experience of having said something stupid (or just average) and realizing later that one could have said something exceedingly clever. To prevent people from saying stupid things and to help users appear as smart as possible, the mod_iq jabberd module provides hints to users regarding what to say at a given point in a conversation, based on the advanced linguistic analysis technologies described under <link url='#impl'>Implementation Notes</link> below. A user can ask for a hint by sending the complete message to the server itself, wrapped in a &QUERY; element qualified by the 'jabber:iq:iq' namespace. (While it may be argued that this functionality could be provided client-side, thus saving a roundtrip, it is consistent with the Jabber philosophy of "smart servers, dumb clients" as explained in &xep0134;.)</p>
<example caption='Requesting IM IQ Information'><![CDATA[
<iq type='get'
from='kindanormal@example.com/IM'
@ -184,13 +184,13 @@
</section2>
<section2 topic='Analysis Methods' anchor='impl-analysis'>
<p>The mod_iq jabberd module makes use of several Analytical Language Engine (ALE) technologies for determining the intelligence of specific messages and thus also a user's IM IQ (the average of all messages sent). These technologies include the following:</p>
<ul>
<li>Phrasal Objectification of Realtime Threads (PORT) -- this is a parsing technique for breaking conversation threads into meaningful phrases, even across message boundaries.</li>
<li>Bayesian Estimation of Entropic Responses (BEER) -- within information theory, entropy is a measure of the rate of information transfer; this technique uses Bayesian estimation methods to determine whether a given message imparts useful information or not.</li>
<li>Situational Analysis of Kladistic Evolution (SAKE) -- kladistics (also spelled "cladistics"), from the Greek "klados" (meaning "branch"), is the study of grouping things into branches that diverge from a common origin; it is used in biology to study descent from a common ancestor, and also in the study of conversation threads to determine how a conversation would evolve depending on things said (or messages sent) at any point in the conversation flow.</li>
<li>Semantic Correlation and Observation of Truth in Conversation Handling (SCOTCH) -- a person may seem intelligent to the casual observer, but his or her messages may actually not provide deep insights or even track reality in a useful or consistent fashion; this technique builds on early semantic web insights to determine the truth value of a given message within the context of a realtime conversation.</li>
<li>Webs of Intelligent Network Endpoints (WINE) -- any given person can engage in conversations with a large number of interlocutors, yet that person's status as an intelligent network endpoint is influenced by reputational factors across the full web of linguistic interactions, not just with any one person; this technique accounts for such reputational effects to paint a complete picture of the person's perceived intelligence across the network.</li>
</ul>
<dl>
<di><dt>Phrasal Objectification of Realtime Threads (PORT)</dt><dd>This is a parsing technique for breaking conversation threads into meaningful phrases, even across message boundaries.</dd></di>
<di><dt>Bayesian Estimation of Entropic Responses (BEER)</dt><dd>Within information theory, entropy is a measure of the rate of information transfer; this technique uses Bayesian estimation methods to determine whether a given message imparts useful information or not.</dd></di>
<di><dt>Situational Analysis of Kladistic Evolution (SAKE)</dt><dd>Kladistics (also spelled "cladistics"), from the Greek "klados" (meaning "branch"), is the study of grouping things into branches that diverge from a common origin; it is used in biology to study descent from a common ancestor, and also in the study of conversation threads to determine how a conversation would evolve depending on things said (or messages sent) at any point in the conversation flow.</dd></di>
<di><dt>Semantic Correlation and Observation of Truth in Conversation Handling (SCOTCH)</dt><dd>A person may seem intelligent to the casual observer, but his or her messages may actually not provide deep insights or even track reality in a useful or consistent fashion; this technique builds on early semantic web insights to determine the truth value of a given message within the context of a realtime conversation.</dd></di>
<di><dt>Webs of Intelligent Network Endpoints (WINE)</dt><dd>Any given person can engage in conversations with a large number of interlocutors, yet that person's status as an intelligent network endpoint is influenced by reputational factors across the full web of linguistic interactions, not just with any one person; this technique accounts for such reputational effects to paint a complete picture of the person's perceived intelligence across the network.</dd></di>
</dl>
<p>Naturally, because of the powerful and potentially unpredictable effects of these technologies, development of mod_iq was restricted to senior developers on the jabberd team, or at least (for developers in the U.S.) those over the age of 21. <note>See Title 23, Chapter 1, Subchapter 1, Section 158 of the United States federal legal code as enacted by the National Minimum Drinking Age Act of 1984 &lt;<link url='http://www.law.cornell.edu/uscode/23/158.shtml'>http://www.law.cornell.edu/uscode/23/158.shtml</link>&gt;.</note></p>
</section2>
</section1>