The Semantic Web

This term Semantic Web has been in the vocabulary for a few years now. It simply means “adding more meaning” to the Web. Today’s web started with a document-centric view. The HTTP URI structure links various documents, but it stops there. It does not go deeper into the meaning. From a document-centric world, the semantic web wants to go to a data-centric world, where data and its relevance are linked.

Here is the definition from Wikipedia – The Semantic Web is an evolving development of the World Wide Web in which the meaning (semantics) of information and services on the web is defined, making it possible for the web to understand and satisfy the requests of people and machines to use the web content.^[1][2] It derives from World Wide Web Consortium director Sir Tim Berners-Lee’s vision of the Web as a universal medium for data, information, and knowledge exchange.

Watch this latest video of how Tim Berners-Lee talks about the importance of collaboration for the Web where he also explains briefly the Semantic Web.

Gen Y & CPA

Young folks in the age group of 15-24 are dubbed as Gen Y, who grew up with the internet, social networking, sms, and cell phones. They check Facebook at least 10 times a day. They are messaging, chatting with friends always. This “always-on” generation do not read newspapers nor do they watch TV. It’s all news on the web and Youtube videos if and when they feel like watching something. They play Zynga games like Farmville with their Facebook or MySpace buddies. They share photos and instant news. They watch their personal finance via Mint.com.

They are always in touch. In today’s Wall Street Journal, an incident is described where a high school kid was summoned to the principal’s office for continuously messaging his friends during classroom hours.  As he was being reprimanded, his fingers were pressing buttons over his pant pocket messaging. The article describes the pros and con of such behavior. At work place, such close communication can be beneficial. Right now, I feel, it is a productivity drain. There are cases of sleep deprivation and poor school performance for those “socially hyperactive” kids. There is talk of  “attention scope” as opposed to “attention span”. But current behavior says such hyperactivity is a result of the addictive nature of social networking.

Someone coined the phrase CPA – Continuous Partial Attention to describe the Gen Y kids. They are chatting through multiple windows while listening to iPod and playing games. The brain can only serialize every activity, hence multi-tasking is interesting but can be sub-optimal in terms of effectiveness.

Someone said “put your mind where your hand is”. In other words, focus on that one task you are doing, like driving or eating or reading. Gen Y should once in a while get back to the basic activities like reading a book, or wondering on nature’s splendor.

 

 

Top Ten Technologies for 2010

Every year around this time, Gartner Group hosts its big conference in Florida and invites some big names from our industry to talk about new trends. Gartner Group also presents what will be hot in the following year. Sometimes, it is like stating the obvious.

I remember my days at Oracle when certain Gartner group analyst will spend time listening to my views of database futures. Next thing I know I have paid $2000 conference fee to listen to the same analyst presenting back mostly what I had told him. It’s a good business – this “information selling”.

In fairness though, Gartner Group listens to many people like me from different companies and consolidates the data and presents them as future trends.  So this month at their conference,  Gartner Group said the following ten technology trends for 2010.

1.  Cloud computing – style of computing that characterizes a model in which providers deliver a variety of IT-enabled capabilities to consumers.

2. Advanced Analytics – last year it was BI. So this year it has to be advanced analytics, although it is unclear what exactly that is. It is simulation, prediction, optimization, etc.  Neil Raden wrote a blog on why we don’t need Ph.D’s for doing advanced analytics.

3. Client Computing – new ways of packaging client computing via virtualization.

4. IT for Green – ways IT can enhance your green credentials in terms of energy savings, reduction of paper, reduction of travel, etc.

5. Reshaping the Data Center – cutting operational expenses.

6.  Social Computing – use of social software and social media in the enterprise and participation and integration with externally facing enterprise sponsored and public communities.

7. Security-activity monitoring – going from the old model of putting up a security perimeter fence to the new world of monitoring activities and identifying patterns that would have been missed before).

8. Flash memory – moving up to a new tier in storage echelon.

9. Virtualization for Availability – same old notion of virtualization for better systems availability.

10. Mobile Applications – by year-end 2010, 1.2 billion people will carry handsets capable of rich, mobile applications.

Oracle Open World 2009

Interesting times. Although I did not attend the Oracle Open World 2009 in San Francisco, I watched some of the keynotes in the web. With 37000 attendees and the rain storm, it was a challenge in terms of logistics.

The main highlights are hard to pinpoint, as there were so many sessions, news items, announcements. With $40B worth of acquisitions during last 6 years, one has a hard time keeping up with various products. But one that makes the most noise is “Fusion Applications”. Some people have dubbed it as “ConFusion” before. The main focal point was of course the Oracle, Sun merger that is currently held up by the European Council. But Sun was there in full force – Scott McNealy, James Gosling(father of Java) and many others. There were discussions on NetBeans which was anathema before. Marc Benioff gave a session on SalesForce.com – he is supposed to be competition to Oracle On-Demand and Larry’s other investment NetSuite. But SFDC uses Oracle database for its own operations, so they are partners too. Michael Dell was there extolling the virtue of Oracle technology at its data center. Oracle claimed it runs 20 thousand Dell servers. Again with Sun servers, there is co-opetition (competition + co-operation). Infosys CEO Kris Gopalkrishnan gave a keynote which people twitted as dry and dragging.

The usual buzzwords were there – SOA, Web 2.0, Cloud, SaaS, PaaS, Web Services, BPEL, CEP, etc. Fusion apps are almost ready and will be GA next year, which means much work is yet to be done. The new Data Warehouse appliance Exadata on Sun with fast performance was much bragged about with challenges to IBM to match the run-time performance. Such bravado is good fodder for the media to write about. IBM should be clever to ignore such noise. The image of a barking dog at a calm elephant walking comes to mind.

During my days at Oracle this event used to draw 10 to 15 thousand users. Of course those were the pre-acquisition days at Oracle. Now with all the new users from Peoplesoft, Siebel, BEA, and others, the number gets close to 40 thousand. People come for networking and party. It’s a good escape from the workplace. Oracle does a great job putting this event and this year was no exception. Even the governator Arnold Schwarzenegger made an appearance to add pizzazz.

Back to the “Stack”

It’s funny how cyclic this industry is. Not that long ago, during the early 1990s we discussed how the vertically integrated computer vendors’ days were coming to an end. There were just a handful of vendors like IBM, Digital Equipment, HP, Sun, etc who supplied the entire “stack” – microprocessor,  hardware, operating system, subsystems like DBMS, and applications. The joke used to be -  “you don’t get fired by picking IBM” which meant the CIO made a safe choice buying everything from IBM. Customer support was the best and CIO’s slept well at night.

Then came Moore’s law (microprocessors capacity doubled every 18 months for the same price) and the era of the personal computer started. The stack went through democratization – many players appeared in each layer. You want microprocessors?  Pick from Intel, Sun, AMD, etc. You want hardware? Pick from Compaq, Dell, IBM, HP, etc. You need an operating system? There were a few besides Windows. The advantage of such “horizontal” structure was the lowering of price and provision of  “best of breed” for each layer of the stack. Customers did not feel “locked-in” to one vendor. The downside was the challenge of “systems integration”, as vendors did not provide testing for all permutation and combination.  The task of making every component work with each other fell on the shoulders of the customer. Of course this was a big opportunity for “service providers” like Accenture, Infosys and the like. Someone said for every dollar of software license spent, consulting services cost seven dollars. Customers started complaining of too many vendors to deal with. The rule then became – 8 is better than 20, and 3 is better than 8. The less the number of vendors, the easier it is to manage.

Now we are back full circle. With cloud computing as the new big trend, we are back to centralized computing model (anti-thesis of “lots of little” model of PC’s & Microsoft). Then we see consolidation and a few vendors suggesting to provide the full stack, as an integrated solution. IBM offers that. HP is trying to fill up the stack with components. Oracle which supplied only the subsystem  (database & middleware) and application layers, is moving down to systems hardware and operating system, with its intended acquisition of Sun. Once done, they can claim to provide a full stack – sparc chip, hardware, solaris, dbms, middleware, and application. Here comes great profitability for the vendors with “lock-in”.

This is a true Back to the Future story.

Oracle & MySQL

Much has been said about what Oracle would do with MySQL as part of its impending SUN acquisition. While the clearance has come from the US, now it is held up by the European commission for further investigation. Last week at a Churchill Club event, Ed Zander (former President of Sun & Ex-CEO of Motorola) interviewed Larry Ellison where he asked if Oracle would spin off MySQL. Larry answered with an emphatic No.

I thought the one big attraction of the Sun acquisition was MySQL and Java. We can debate whether the hardware business of SUN makes any sense or not. I also know that prior to the MySQL acquisition by Sun, Oracle attempted to buy MySQL. I do not agree with the statement that “MySQL does not compete with Oracle, but its main competition is DB2 and SQL Server”. That can not be true, as several Internet-age companies such as Google and Amazon selected MySQL over Oracle for specific applications. So, yes, MySQL is an attractive alternative to Oracle for some applications. Now that Oracle will be the potential owner of MySQL, it will continue to offer it as the open source offering in database. It can position both Oracle and MySQL for different types of applications – Oracle for heavy lifting with high scalability (with its RAC) and complex functionality. It may even show a migration path for those MySQL users to Oracle deployment when needed.

While speaking with a company in the Identity Protection space, I learnt that the entire system is built on the LAMPJ stack (Linux, Apache, MySQL, Perl/PHP, Java). The reason was simple – stay with open source and lower the cost, as long as MySQL is an “adequate” solution.

MySQL’s true competition is with other open source database products such as Ingres and Postgress, but it does have a large customer base.  The interesting thing to observe is how much is the overlap between Oracle’s own closed-source database vs MySQL.  Also unknown is,  how much R&D investment Oracle will inject into enhancing MySQL? Will Oracle’s sales force be motivated to push MySQL in the market for additional revenue. The “fee vs free” aspects of MySQL has been a challenge from the beginning. Will customers trust Oracle as the pusher of open source products? Time will tell.

Pair Programming

I have never heard of this term until I read this article in today’s NY times. The idea is to have two people doing any programming task. One is a driver and the other is a navigator. As the driver writes the code, the navigator keeps checking the code for bugs or any other discrepancy. They sit side by side physically all the time. The pairing gets changed all the time depending on the project needs. Sometimes a junior programmer can feel intimidated pairing up with a senior member. Then the trick is to make the junior member play the driver role. The article talks about other issues that may arise, such as disagreement between the pairs or personality clash.

The claim, as we can predict is better quality via early error checking. As they say, two brains are better than one. The other model is solo programming as the vast majority do, but break it up into 25 minutes chunks with a 5 minute break for checking email, do social networking, etc. This style apparently has a name – Pomodoro, which means tomato in Italian. When someone does not take the break, someone else shouts – respect the tomato.

Various styles of programming have been advocated over the years. Way back over 17 years ago, I was spearheading a concept at IBM to have a specially designated design room for reviews. The code gets projected on the wall/screen and a number of folks can stare at it and do the walk-through. That experiment yielded many early detection of errors and bugs. Management thought it was a very novel idea for bettering code quality. Remember, this was pre-internet, broadband and before we had Webex like collaborative tools.

I wonder how wide spread is this “pair programming” concept.

Europe, wake up to technology!

I just did a ten day trip to Europe and visited three great cities – Berlin, Amsterdam, and Rome. But let me tell you one thing. Europe seems to be in the past century when it comes to technology. One can clearly see where the priorities are – its history, have-a-good-time and don’t work hard, who cares about the customer?, and pay through your nose for everything.

I certainly enjoyed the architecture and history. But it’s almost impossible to get to the Internet. The hotels that let you do so, charge a lot. Try to use your iPhone from there and they do charge a hefty amount for roaming. Every time, I looked up my email, I was reminded that data rates are excessive while roaming internationally. It seems you hit technology roadblocks everywhere. Europe certainly does not want you to communicate.

While Berlin and Amsterdam were tolerable, Rome was another story. After paying $300 a night for the hotel, there was no Internet availability in the rooms. It reminded me of Bangalore where I paid $60 a night and had free Internet plus an elaborate breakfast. It tells you where the technology progress is happening. And if the hotel runs out of hot water, the receptionist shrugs his shoulder and smilingly answers, “What can I do?” That speaks a lot about customer service.

Then you travel to Asia like Korea, Japan, Singapore, etc. Broadband is everywhere and free internet access is the norm. While I love traveling in Europe, I was frustrated for lack of easy access to Internet. It is so good to be back in Silicon Valley. You don’t realize how good it is until you travel elsewhere.

Innovation-on-demand

This is an interesting concept that combines customized hardware and software for specific needs of clients. And guess who is pioneering this? Not IBM. It’s Accenture. This concept brings together off-the-shelf hardware with open-source Linux and other software components as needed. A very modular approach.

A quote from an article – “According to BusinessWeek, Accenture is now custom-building hardware for clients in concert with a New York start-up called Bug Labs. Bug’s devices are essentially a collection of components. They start with the open-source Linux operating system and add features as needed, including perhaps a GPS device or motion detector. To me, that approach is reminiscent of modern software design, in which objects — i.e., chunks of related code — are assembled into a greater, customized whole.”

“Software is the key here as well. Accenture plugs each new device into AMOS, an intelligent network that processes raw data to produce meaningful outcomes for clients. That’s similar to what Big Blue is trying to do with its varying technology specialties. Consider “Green Sigma,” an IBM consulting practice that depends on networked sensors and data analysis software to help companies and governments reduce carbon emissions and conserve water.”

“I’ll grant that’s an imperfect comparison. In both instances, IBM and Accenture are creating technology. The difference is that Accenture is selling Bug’s boxes and charging clients transaction fees for their use of AMOS. That’s much more like salesforce.com’s (NYSE: CRM) model than your typical consulting gig.”

An interesting development. Accenture is changing the playing field and IBM will have to respond to this new trend. The question is what about HP (EDS) and Oracle?

The NoSQL Movement

Now there is a new community called NoSQL and they   (150 enthusiastic developers of the cloud era)   met for the first time  last June in San Francisco.

“Relational databases give you too much. They force you to twist your object data to fit a RDBMS [relational database management system],” said Jon Travis, principal engineer at Java toolmaker SpringSource.  NoSQL-based alternatives “just give you what you need,” Travis said.

So get used to names like Dynomite,  CassandraDB, Voldemort, CouchDB, BigTable, MongoDB, Hypertable, SimpleDB, etc. These are all members of this club which was started by Google’s BigTable and its clones such as Hypertable (ZVent). Amazon is doing SimpleDB on Dynamo (not called a database, but a highly available key-value data store). Facebook is doing CassandraDB. Apache CouchDB is a free, open source, document-oriented database written in Erlang programming language. MongoDB is a collection of JSON documents ( no rows or columns), an open source document-oriented DB written in C++ programming language.

So what are the common factors pushing this movement?

• They can blow through enormous volumes of data. For example, Google’s BigTable with its sister technology MapReduce processes as much as 20 petabytes of data per day. We have not seen this volume in RDBMSs.
• They run on clusters of cheap PC Servers. Google has said that one of BigTable’s bigger clusters manages as much as 6 petabytes of data across thousands of servers. Oracle’s RAC (Rapid Application Cluster) can get there but at a much higher cost.
• They beat performance bottlenecks.  The phrase used here is “eventually consistent”, trading off consistency to maximize availability and scalability.
• While conceding that relational databases offer an unparalleled feature set and a rock-solid reputation for data integrity, NoSQL proponents say this can be too much for their needs. Hence the mantra is “no overkill”.

I think this is exciting development. When people like me worked on early days of RDBMS, we could not imagine the kind of scalability and data volumes being talked about now. It’s only natural that new approaches must be innovated to handle the demands of the Internet era. Eric Brewer of UC Berkeley floated the idea of this in his research work at least 6-7 years ago.

Although NoSQL movement is not a threat to mainstream database community yet, this may change in next 3-4 years time.