tag:blogger.com,1999:blog-6569681.post4242715896189712463..comments2024-03-24T10:38:16.997-07:00Comments on Geeking with Greg: Microsoft and intelligent agents on the desktopGreg Lindenhttp://www.blogger.com/profile/09216403000599463072noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-6569681.post-46645607819099711372007-12-18T09:01:00.000-08:002007-12-18T09:01:00.000-08:00For years, I've been hearing about "Intelligent Ag...For years, I've been hearing about "Intelligent Agents" that will save our collective brain power by searching the "Semantic Web" and only showing us what is important to us.<BR/><BR/>There is nothing special about parallel processing in this context. The problem is that the data is not understandable and the cost of putting structure behind that data to make it understandable is just too high.<BR/><BR/>The Semantic Web looks great in the Ivory Tower, but in the real world, it is a long way off.Travis Jensenhttps://www.blogger.com/profile/08454148260119349712noreply@blogger.comtag:blogger.com,1999:blog-6569681.post-49281706494958633992007-12-17T15:33:00.000-08:002007-12-17T15:33:00.000-08:00There is something a little weird here: overnight ...There is something a little weird here: overnight for processing an inbox? Let's say you had 100 messages, what could it be doing that would take 5-10 hours to process to help you the next day? The problem (at least as I see it) is not an issue of crunching, but an issue of algorithms: smart algorithms. The whole AI-is-data-crunching approach is misleading.Anonymousnoreply@blogger.com