<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	>
<channel>
	<title>Comments on: Tello Explained</title>
	<atom:link href="http://saunderslog.com/2006/01/29/tello-explained/feed/" rel="self" type="application/rss+xml" />
	<link>http://saunderslog.com/2006/01/29/tello-explained/</link>
	<description>A daily round table on the tech industry with experts and callers from all over the globe. Join us as we pick apart the news and get to the meat of what\'s happening out there.</description>
	<pubDate>Mon, 08 Sep 2008 06:09:30 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.6</generator>
		<item>
		<title>By: Frank Miller</title>
		<link>http://saunderslog.com/2006/01/29/tello-explained/#comment-3139</link>
		<dc:creator>Frank Miller</dc:creator>
		<pubDate>Mon, 30 Jan 2006 21:41:46 +0000</pubDate>
		<guid isPermaLink="false">http://saunderslog.com/2006/01/29/tello-explained/#comment-3139</guid>
		<description>Well, I can make a guess at that too.  Of course, Alec will probably swoop in to correct all my bad assumptions.

Having looked over Alec's presentation, (which is very good BTW for the audience I think he's targeting) and following his blog, this is what I see.  It appears they are using some scripting languages (probably Ruby or Python) to farm the databases of various applications, e.g. Outlook, Skype, etc.  They appear to be looking for direct information, like your phone numbers and IM handles and such, and indirect information, like your making assumptions about where you are based on your calendar and when you log in and log off over various services.  They then mash all this stuff up and combine it with some kind of call receiving thing (like a SIP proxy server, e.g. Asterisk or SER which I think he has mentioned).  If I had to make a guess, I would say they are building some rules to determine if and where an incoming call gets routed.

The basics are very interesting.  The devil is definitely in the details.  Its basically Artificial Intelligence.  In fact, I've wondered whether some of the rules based languages for expert systems, e.g. Prolog, might be applicable.  As we saw during the AI days, AI is REALLY hard.  The good news here is that they have a very constrained space upon which they are trying to be smart and potentially lots of information that can be used to make good decisions.

We'll see how it turns out.  It has lots of potential

FM</description>
		<content:encoded><![CDATA[<p>Well, I can make a guess at that too.  Of course, Alec will probably swoop in to correct all my bad assumptions.</p>
<p>Having looked over Alec&#8217;s presentation, (which is very good BTW for the audience I think he&#8217;s targeting) and following his blog, this is what I see.  It appears they are using some scripting languages (probably Ruby or Python) to farm the databases of various applications, e.g. Outlook, Skype, etc.  They appear to be looking for direct information, like your phone numbers and IM handles and such, and indirect information, like your making assumptions about where you are based on your calendar and when you log in and log off over various services.  They then mash all this stuff up and combine it with some kind of call receiving thing (like a SIP proxy server, e.g. Asterisk or SER which I think he has mentioned).  If I had to make a guess, I would say they are building some rules to determine if and where an incoming call gets routed.</p>
<p>The basics are very interesting.  The devil is definitely in the details.  Its basically Artificial Intelligence.  In fact, I&#8217;ve wondered whether some of the rules based languages for expert systems, e.g. Prolog, might be applicable.  As we saw during the AI days, AI is REALLY hard.  The good news here is that they have a very constrained space upon which they are trying to be smart and potentially lots of information that can be used to make good decisions.</p>
<p>We&#8217;ll see how it turns out.  It has lots of potential</p>
<p>FM</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Richard Stastny</title>
		<link>http://saunderslog.com/2006/01/29/tello-explained/#comment-3138</link>
		<dc:creator>Richard Stastny</dc:creator>
		<pubDate>Mon, 30 Jan 2006 20:33:53 +0000</pubDate>
		<guid isPermaLink="false">http://saunderslog.com/2006/01/29/tello-explained/#comment-3138</guid>
		<description>Now if somebody could post Relevance Engine explained, please ;-)</description>
		<content:encoded><![CDATA[<p>Now if somebody could post Relevance Engine explained, please ;-)</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Frank Miller</title>
		<link>http://saunderslog.com/2006/01/29/tello-explained/#comment-3137</link>
		<dc:creator>Frank Miller</dc:creator>
		<pubDate>Mon, 30 Jan 2006 15:02:16 +0000</pubDate>
		<guid isPermaLink="false">http://saunderslog.com/2006/01/29/tello-explained/#comment-3137</guid>
		<description>I've been reading over the Tello website and have a few thoughts.  Its looking a lot to me like FWD 2.0.  You take Asterisk, pulver.Communicator, and some IM infrastructure (maybe GAIM as a backend, like what Meebo is doing), you rev them, mash them all together, and you get Tello!

There are of course good and not so good things about it.

The good:

They have added IM as a basic component of the service.  There was mention of video a couple of times, but right now it looks more like Voice and IM are the strong suits.

They seem to be focusing on "interoperability".  If there was ever a word that cried out for specific definition, its "interoperability".  More on this in the Not so good section unfortunately.

The client user interface looks very clean.

The not so good:

It appears to be a silo.  While there has been a lot of talk lately about open standards and avoiding silos, this appears to be going in the silo direction.  You need to use their clients (at least if you take their site at face value).  I don't know for sure but I would imagine that all that presence information is sent to the client in using an XML schema that I have a feeling isn't openly available.  I'll have to do some experiments on connecting with my client and see what I find.

"Interoperability?"  I'm not sure what they're really talking about here.  I think what they are referring to is that I can use the Tello client to access multiple services, either voice or IM, thats one way to look at it.  The question is, does this really mean interoperability.  When you ask an engineer what interoperability means, you'll probably get something like, multiple implementations of the same protocol working with each other.  That's clearly not what they're talking about here.

Lots of smoke.  This may or may not actually be good.  There's this inherent conflict between trying to be open and talk techincal to woo the developers vs. talking high level and dressing everything up in marketspeak to address less technically sophisticated users.  They clearly have gone down the latter path.  IMHO, this path is hard for even Grandma to understand.

All that said, the service does look somewhat interesting.  I'm a little hesitant to call it revolutionary, if its not me too, its not too far ahead of me too.  I wish them the best of luck!

BTW, I agree with Andy, this has nothing to do with what iotum appears to be doing.

FM</description>
		<content:encoded><![CDATA[<p>I&#8217;ve been reading over the Tello website and have a few thoughts.  Its looking a lot to me like FWD 2.0.  You take Asterisk, pulver.Communicator, and some IM infrastructure (maybe GAIM as a backend, like what Meebo is doing), you rev them, mash them all together, and you get Tello!</p>
<p>There are of course good and not so good things about it.</p>
<p>The good:</p>
<p>They have added IM as a basic component of the service.  There was mention of video a couple of times, but right now it looks more like Voice and IM are the strong suits.</p>
<p>They seem to be focusing on &#8220;interoperability&#8221;.  If there was ever a word that cried out for specific definition, its &#8220;interoperability&#8221;.  More on this in the Not so good section unfortunately.</p>
<p>The client user interface looks very clean.</p>
<p>The not so good:</p>
<p>It appears to be a silo.  While there has been a lot of talk lately about open standards and avoiding silos, this appears to be going in the silo direction.  You need to use their clients (at least if you take their site at face value).  I don&#8217;t know for sure but I would imagine that all that presence information is sent to the client in using an XML schema that I have a feeling isn&#8217;t openly available.  I&#8217;ll have to do some experiments on connecting with my client and see what I find.</p>
<p>&#8220;Interoperability?&#8221;  I&#8217;m not sure what they&#8217;re really talking about here.  I think what they are referring to is that I can use the Tello client to access multiple services, either voice or IM, thats one way to look at it.  The question is, does this really mean interoperability.  When you ask an engineer what interoperability means, you&#8217;ll probably get something like, multiple implementations of the same protocol working with each other.  That&#8217;s clearly not what they&#8217;re talking about here.</p>
<p>Lots of smoke.  This may or may not actually be good.  There&#8217;s this inherent conflict between trying to be open and talk techincal to woo the developers vs. talking high level and dressing everything up in marketspeak to address less technically sophisticated users.  They clearly have gone down the latter path.  IMHO, this path is hard for even Grandma to understand.</p>
<p>All that said, the service does look somewhat interesting.  I&#8217;m a little hesitant to call it revolutionary, if its not me too, its not too far ahead of me too.  I wish them the best of luck!</p>
<p>BTW, I agree with Andy, this has nothing to do with what iotum appears to be doing.</p>
<p>FM</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Alec</title>
		<link>http://saunderslog.com/2006/01/29/tello-explained/#comment-3133</link>
		<dc:creator>Alec</dc:creator>
		<pubDate>Mon, 30 Jan 2006 04:25:28 +0000</pubDate>
		<guid isPermaLink="false">http://saunderslog.com/2006/01/29/tello-explained/#comment-3133</guid>
		<description>Fixed.  Thank you!</description>
		<content:encoded><![CDATA[<p>Fixed.  Thank you!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Mathew Ingram</title>
		<link>http://saunderslog.com/2006/01/29/tello-explained/#comment-3132</link>
		<dc:creator>Mathew Ingram</dc:creator>
		<pubDate>Mon, 30 Jan 2006 03:57:08 +0000</pubDate>
		<guid isPermaLink="false">http://saunderslog.com/2006/01/29/tello-explained/#comment-3132</guid>
		<description>Alec -- I went to click on the link to Andy's description of Tello and Iotum, but it's a link to a flickr photo of your Blackberry.  Just thought I'd let you know.</description>
		<content:encoded><![CDATA[<p>Alec &#8212; I went to click on the link to Andy&#8217;s description of Tello and Iotum, but it&#8217;s a link to a flickr photo of your Blackberry.  Just thought I&#8217;d let you know.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
