Search companies vie for the desktop

TalkBack 3 of 3:: « Previous

Thread View
Flat View

Desktop Search is useful on a scale of 1 out of 10, v2: Previously posted at http://news.zdnet.com/5208-9588-0.html?forumID=1&threadID=6970&messageID=143272&start=-1

Desktop search is great, especially when I can index and then search all of my content stores. I've used Enfish Pro for years, as well as DTSearch and sgrep, each having their different strengths. I've also regularly used the MS Indexing/Search functions, the Exchange Search, and the SQL Server search for the desktop and LAN. I also use WebFerretPro for Internet Searches, and occasionally just let the MS browser perform its search function.

What I want, having worked with the issue for about 20 years, is a single engine that extracts all of the words and text symbols from all of my desktop content, my specified LAN/WAN/intranet content, and specified Internet content. I want those words and symbols categorized by their parts of speech (e.g., noun, verb, comma, adverb). I want a sentence-diagram of every sentence's subject, predicate verb, and predicate object identified in noun/verb/noun format as an RDF "triple" assertion, with all adjectives, adverbs, etc., arrayed against the nouns and verbs as properties. I want every sentence assertion diagrammed in directed-labeled-graph (DLG) node-link-node format. I want every paragraph's collection of diagrammed assertions and sentences categorized by the subject(s) of the paragraph. I want every document's or message's collection of paragraphs to be categorized by the document's subject(s). I want a subject tree and subject DLG diagram of all collective document content.

I want the SQL and ISAM table/field/value, or XML/LDAP element/attribute/value triples of all my tabular and hierarcical data sources categorized by data table (entity) or data tree root element name. I want a subject-tree (i.e., data dictionary) and DLG diagrams of all of my data content.

I want the whole collection merged into a single navigable and linked subject tree with DLG diagram, presented as a tree taxonomy of my local and specified distant content. From this same taxonomic collection I want a complete noun=concept and relation=verb DLG concept map, so that when I select a single document or data store, it's matching concepts, and related concepts from other content, is presented as a DLG semantic map overlap/highlight within my whole concept map. I want the attributes that describe my documents and data to be arrayed against their appropriate concept names, and I want the ability to add, modify, search, and navigate other attributes for any concepts and their linking relations, as a data model.

I want an overarching ontology (as a subject network) to graphically (as DLG and as drill-down containers) and in data tree format (e.g., XML/XMLS, RDF/RDFS, OWL) present my entire individual concept map, semantic models, and data models as a knowledge-base instance of that ontology. I want unchanging universal (e.g., UUID/GUID/URI) and container-relative (e.g., ISO/IANA Object Identifier in ASN.1 dot.decimal format) key structures to provide me the means to merge my subject tree (taxonomy) and subject network (ontology) into a larger context group, organization, community, or global taxonomy and ontology.

I want to be automatically informed via software agents when something of relevance to my prioritized subjects of interest occurs that affects me or that I affect, with the ability to turn off any agent responses and the sources that appear to be spam or other unwanted material.

That's all I want for Christmas, this year. I believe that several products that can do all or most of this, from relatively inexpensive (desktop to enteprise) to very expensive (enterprise scale). Such a capability, as a free downloadable product, without spam/advertisements would be nice as well.; Posted by: roy@... Posted on: 12/21/04 You are currently: a Guest | Members login | Terms of Use