Geeks to English

CNN is run­ning an inter­est­ing arti­cle about infor­ma­tion data stores and the seman­tic web. A very sad thing was the fol­low­ing state­ment about the seman­tic web: One hope­ful jour­nal­ist from the Econ­o­mist asked Berners-Lee to give an exam­ple of how com­pa­nies could make or save money using it, but he didn’t have an answer. This is clearly an illus­tra­tion that often geeks speak in terms that most peo­ple do not under­stand. I am often guilty of it myself (a cur­sory look through recent entries will give you an idea of why I’m mak­ing this state­ment) and I real­ize that we need to do a bet­ter job in explain­ing some of the key con­cepts in new tech­nol­ogy. XML, RDF and other tech­nolo­gies related to the seman­tic web are indeed hard to under­stand when you talk to techies. As a result, they often get dis­missed as too hard. In order to help peo­ple get a bet­ter under­stand­ing, I’ll try to come up with a sim­ple example.

Right now, this page is served to you either in HTML or RSS. Those are two dif­fer­ent lan­guages. One, HTML, is under­stood by your web browser. The way text is bolded, for exam­ple, is that I put in a lit­tle tag that told the browser to bold this word. The browser reads the tag and presents it appro­pri­ately. The other, RSS, is under­stood by what we call RSS read­ers. Those are pro­grams that you use to sub­scribe to a chan­nel. A chan­nel is some­thing that you would receive every day. That way, you don’t have to go and check the site to see if it’s been updated. Your pro­gram goes and gets the infor­ma­tion. RSS is an XML-based lan­guage. What it means is that there is a lot of infor­ma­tion in that chan­nel that is there just for the ben­e­fit of that chan­nel, to allow to present only the newest news to you.

A cou­ple of years ago, Tim Berners-Lee, the man who cre­ated the web, looked at his cre­ation and real­ized that there was a jum­ble of pages and that, in order to make sense of it, we needed to give things a lit­tle more struc­ture. So RDF was born. What it is, basi­cally, is a way to orga­nize the whole web so that com­put­ers could talk to each other with­out humans in between. This has poten­tial uses and here’s an exam­ple I thought up: The smart calendar.

Joe and I are work­ing on a joint project. Joe is in Lon­don and I am in New York. I want to arrange a face to face meet­ing with Joe. Right now, I either call, email, or con­tact Joe in some way and we fig­ure out a time when we can meet face to face, then agree on a city in which to meet, then make the nec­es­sary travel arrange­ments. What if I went to my cal­en­dar, typed in meet face to face with Joe, and my cal­en­dar and Joe’s started dis­cussing when the best time and place would be? My cal­en­dar would check my avail­abil­ity and Joe’s. It then would check if any of us has any travel plans in each other’s city. Based on those, The two cal­en­dars dis­cover that I have a trip to Paris set for next Tues­day to Thurs­day. As a result, the cal­en­dar would rec­om­mend that I go to Lon­don to meet with Joe on Fri­day. If both Joe and I agree to this, my com­puter would then go to the travel reser­va­tion sys­tem, check prices and flight times, book a flight from Paris to Lon­don on Thurs­day Night, can­cel my Fri­day morn­ing flight from Paris to New York, book a flight from Lon­don to New York on Fri­day Night, can­cel my hotel stay in Paris on Thurs­day night, book a Thurs­day night hotel stay in a com­pany approved estab­lish­ment near Joe’s office, and notify Bob (who’s also in Lon­don and with whom I had sched­uled a con­fer­ence call) that we can meet face to face when I’m in Lon­don instead of doing it on the phone.

Usu­ally, this would have taken sev­eral dis­cus­sions, a whole slew of new flight and hotel reser­va­tion changes, and a lot of wasted time. Using a seman­tic engine, all this would be auto­mated. A lot of com­put­ers would have talked together (first mine and Joe’s agreed on time and place, then my com­puter talked to the travel com­pany with which I had my flight and changed that reser­va­tion, then my com­puter talked to a num­ber of air­line com­pa­nies to see who had the best price on a flight from Lon­don to New York, then my com­puter talked to the hotel reser­va­tion sys­tem in Paris and can­celed one night, then my com­puter talked to the com­pany com­puter to see what hotels it approved near Joe’s office. It then talked to com­put­ers in the sev­eral hotels in Lon­don to find a room with my pref­er­ences and within my price range. Hav­ing done so, my com­puter talked to Bob’s com­puter to tell him to change the appoint­ment from a phone con­fer­ence to a face to face meet­ing. Bob’s com­puter talked to a com­puter in his build­ing to book a con­fer­ence room. Hav­ing done all this, the hotel and air­line com­put­ers then talked to my company’s account­ing sys­tems to agree on billing) because they all talked sim­i­lar lan­guages (or could point to a trans­la­tor who would explain how they could. THAT is a prac­ti­cal exam­ple that would save money (find­ing the low­est price on air­lines and hotels, reduc­ing the num­ber of trips) and increase pro­duc­tiv­ity (sav­ing time spent on cer­tain tasks) thus allow­ing me to spend more time on money-making tasks.

All this is still a long way away but if the dream of a seman­tic web is real­ized, it will become reality.