Another RDF triple store

After the usual native RDF triple stores such as Kowari (Now defunct), Mulgara, AllegroGraph, OpenLink Virtuoso, here is another one called 4Store by a financial company called Garlik (interesting name). It remains to be seen if 4store, an efficient, scalable and stable RDF database (according to the press release) would overcome the querying speed bottlenecks that other RDF triple stores have faced. In my experience all the native RDF triple stores have boasted about capable of storing Billions of triples and loading them in under few hours. While this is not less important, fast and efficient querying still remains the most desired feature in the databases (RDF or non-RDF). While a few seconds querying time as touted by the stores so far is OK given the short time that the native RDF databases have been around, mission critical and real-life applications require the querying time to be in miliseconds. We'll see how this new RDF store fares on the querying speed front.

My First Real Blog Entry

So i finally published my first blog today (Yes Mr. Tanner was kind enough to let me do it on the office clock) on enterprises using security for semantic content. I am really excited and looking forward to any feedback and discussion.

Semantic Security: Is it a real consideration by Semantic Enterrprises

Semantic is a buzzword today. Every new starup that deals with some kind of information, would like to somehow associate their product to semantics. Be it the semantic annotation of the customer data, semantifying the web pages or semantic search. This enables them to sell it to the funding guys (Hey! Look we aren't building a 20th Century Product. Our product is 100% Semantic Enabled). Whether its looking into data mining, text mining, context matching, or link analysis. I would not be surprised if a company analyzing as simple as the datestamps associated with the data would call their product "Semantic Enabled". Well so far so good. We need this to reach our vision of Web 3.0 in another 10 years!! The real question is: How many of the semantic commercialization companies are researching into the security aspects of the data they deal with. Beyond a proof of concept stage no semantic technology is going to make it too far without blending them with the security methods. They may sell it to the VCs without the security annotations (no pun intended), but it would be hard for them to make it mainstream for usage by the customers and enterprises. After all if semantic technologies provide excitement about the embedded so called human intelligence into the information they also expose the security vulnerabilities. Guess What, now anybody having unauthorized access to customer data, also have access to its semantics. So earlier what may have looked like an XML node in your data, is really a customer SSN. Go figure what happens when you make it easier for the data thieves to easily and semantically understand your private data. It would be interesting to see a number on the Semantic Enterprises actually serious about security.

My First Blog

This is my first blog. MOre to come.

cheers