Comments on: Autonomy & HP – a technology viewpoint http://www.flax.co.uk/blog/2012/11/21/autonomy-hp-a-technology-viewpoint/ Open source & enterprise search Thu, 21 Mar 2013 00:30:25 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: charlie http://www.flax.co.uk/blog/2012/11/21/autonomy-hp-a-technology-viewpoint/comment-page-1/#comment-43301 charlie Tue, 27 Nov 2012 11:19:21 +0000 http://www.flax.co.uk/blog/?p=904#comment-43301 I'm sorry you're incorrect - Both Muscat and the open source Xapian use Bayesian stats and allow you to use whole documents as a query - it's one thing probabilistic engines are particularly good at. I'm not sure what you mean by 'query FAST with documents' in this context - FAST was a different technology. All 'enterprise search' engines will attempt to index 'all' kinds of data on a corporate network and there are multiple ways to do this - file filters or even string extraction. Autonomy did of course buy in KeyView to assist this process, but KeyView isn't unique, there are even lots of open source file filters. There are always going to be binary formats which are difficult to index: video and images are particularly hard, although sometimes you can 'cheat' using subtitles, script text for example. I’m sorry you’re incorrect – Both Muscat and the open source Xapian use Bayesian stats and allow you to use whole documents as a query – it’s one thing probabilistic engines are particularly good at. I’m not sure what you mean by ‘query FAST with documents’ in this context – FAST was a different technology.
All ‘enterprise search’ engines will attempt to index ‘all’ kinds of data on a corporate network and there are multiple ways to do this – file filters or even string extraction. Autonomy did of course buy in KeyView to assist this process, but KeyView isn’t unique, there are even lots of open source file filters. There are always going to be binary formats which are difficult to index: video and images are particularly hard, although sometimes you can ‘cheat’ using subtitles, script text for example.

]]> By: ExAUTN http://www.flax.co.uk/blog/2012/11/21/autonomy-hp-a-technology-viewpoint/comment-page-1/#comment-43162 ExAUTN Mon, 26 Nov 2012 10:15:54 +0000 http://www.flax.co.uk/blog/?p=904#comment-43162 I think you are completely missing the point. Its nothing to do with search ranking algorithms but how you can extract results from the engine and get data into it. If you can't index all types of data, then your engine is next to useless in a corporate environment. If you can't query the engine using an entire document to find similar documents (instead of manual search) then for most end users the system wont be used. Bayesian stats ALLOW you to use entire documents to query. Muscat et al. could never index all data types or query FAST with documents. I think you are completely missing the point.

Its nothing to do with search ranking algorithms but how you can extract results from the engine and get data into it.
If you can’t index all types of data, then your engine is next to useless in a corporate environment. If you can’t query the engine using an entire document to find similar documents (instead of manual search) then for most end users the system wont be used. Bayesian stats ALLOW you to use entire documents to query. Muscat et al. could never index all data types or query FAST with documents.

]]> By: Autonomy accounts and business model was suspect, analysts say | Apple http://www.flax.co.uk/blog/2012/11/21/autonomy-hp-a-technology-viewpoint/comment-page-1/#comment-42767 Autonomy accounts and business model was suspect, analysts say | Apple Fri, 23 Nov 2012 10:02:47 +0000 http://www.flax.co.uk/blog/?p=904#comment-42767 [...] Similar doubts are expressed by Charlie Hull of Flax, an open source search company. “Autonomy’s ability to market its technology has never been in doubt: aggressive and fearless, it painted IDOL as unique and magical, able to understand the meaning of data in multiple forms. However, this has never been true; computers simply don’t understand ‘meaning’ like we do. IDOL’s foundation was just a search engine usingBayesian probabilistic ranking; although most other search technologies use the vector space model there are a few other examples of this approach.” He adds: “It’s important to remember though that Bayesian ranking is only one way to approach a search problem and in many cases, simply unnecessary. It certainly isn’t magic.” [...] [...] Similar doubts are expressed by Charlie Hull of Flax, an open source search company. “Autonomy’s ability to market its technology has never been in doubt: aggressive and fearless, it painted IDOL as unique and magical, able to understand the meaning of data in multiple forms. However, this has never been true; computers simply don’t understand ‘meaning’ like we do. IDOL’s foundation was just a search engine usingBayesian probabilistic ranking; although most other search technologies use the vector space model there are a few other examples of this approach.” He adds: “It’s important to remember though that Bayesian ranking is only one way to approach a search problem and in many cases, simply unnecessary. It certainly isn’t magic.” [...]

]]> By: Autonomy accounts and business model was suspect, analysts say | Tech & Comms News http://www.flax.co.uk/blog/2012/11/21/autonomy-hp-a-technology-viewpoint/comment-page-1/#comment-42683 Autonomy accounts and business model was suspect, analysts say | Tech & Comms News Thu, 22 Nov 2012 16:24:24 +0000 http://www.flax.co.uk/blog/?p=904#comment-42683 [...] Similar doubts are expressed by Charlie Hull of Flax, an open source search company. “Autonomy’s ability to market its technology has never been in doubt: aggressive and fearless, it painted IDOL as unique and magical, able to understand the meaning of data in multiple forms. However, this has never been true; computers simply don’t understand ‘meaning’ like we do. IDOL’s foundation was just a search engine usingBayesian probabilistic ranking; although most other search technologies use the vector space model there are a few other examples of this approach.” He adds: “It’s important to remember though that Bayesian ranking is only one way to approach a search problem and in many cases, simply unnecessary. It certainly isn’t magic.” [...] [...] Similar doubts are expressed by Charlie Hull of Flax, an open source search company. “Autonomy’s ability to market its technology has never been in doubt: aggressive and fearless, it painted IDOL as unique and magical, able to understand the meaning of data in multiple forms. However, this has never been true; computers simply don’t understand ‘meaning’ like we do. IDOL’s foundation was just a search engine usingBayesian probabilistic ranking; although most other search technologies use the vector space model there are a few other examples of this approach.” He adds: “It’s important to remember though that Bayesian ranking is only one way to approach a search problem and in many cases, simply unnecessary. It certainly isn’t magic.” [...]

]]> By: Autonomy accounts and business model was suspect, analysts say | Latest News Channel http://www.flax.co.uk/blog/2012/11/21/autonomy-hp-a-technology-viewpoint/comment-page-1/#comment-42675 Autonomy accounts and business model was suspect, analysts say | Latest News Channel Thu, 22 Nov 2012 15:24:53 +0000 http://www.flax.co.uk/blog/?p=904#comment-42675 [...] Similar doubts are expressed by Charlie Hull of Flax, an open source search company. “Autonomy’s ability to market its technology has never been in doubt: aggressive and fearless, it painted IDOL as unique and magical, able to understand the meaning of data in multiple forms. However, this has never been true; computers simply don’t understand ‘meaning’ like we do. IDOL’s foundation was just a search engine usingBayesian probabilistic ranking; although most other search technologies use the vector space model there are a few other examples of this approach.” He adds: “It’s important to remember though that Bayesian ranking is only one way to approach a search problem and in many cases, simply unnecessary. It certainly isn’t magic.” [...] [...] Similar doubts are expressed by Charlie Hull of Flax, an open source search company. “Autonomy’s ability to market its technology has never been in doubt: aggressive and fearless, it painted IDOL as unique and magical, able to understand the meaning of data in multiple forms. However, this has never been true; computers simply don’t understand ‘meaning’ like we do. IDOL’s foundation was just a search engine usingBayesian probabilistic ranking; although most other search technologies use the vector space model there are a few other examples of this approach.” He adds: “It’s important to remember though that Bayesian ranking is only one way to approach a search problem and in many cases, simply unnecessary. It certainly isn’t magic.” [...]

]]>