It was the first winter, yet it didn’t offer anything more than spectacular display, and has been more of a nuisance. Of course there were few other factors that influenced me to choose california for winter break instead of spending the holidays in my new home – Cambridge. I flew to california on snowy sunday morning in United airlines from Logan international airport, Boston to Los Angels. Following a simple pattern I had adopted recently, I will tell the story in three parts – concerning three different people.

I, being someone who like to be organized and prepared in advance, purchased and collected all I needed for the trip a week advance, did laundry in four days advance, and packed everything three days in advance; indeed I had been wearing the same clothes at home for last three days. I checked-in online Saturday afternoon and made the payment for the checked-in luggage. Over a surprise call on Saturday evening, I was informed by a computer voice that my flight from Washington to San Diego (I had the connection at Washington) had been cancelled. I was shocked as I had been preparing and eagerly waiting for Sunday – the day of my flight’s departure. I made few attempts on buying a new ticket, rebooking, calling customer care, and even considering taking the flight to Washington (which was just delayed), camp there and look for the first flight to leave, yet at no avail. Fortunately there was another friend of mine who also had his Sunday morning flight to Chicago cancelled. We decided to go to the airport and check, as per previous plan, at the early morning, and here I am typing this blog from the flight from Boston to Los Angels.

United airlines, something rather picked by Bing Travel rather than my personal preference, may not be the best. I was really annoyed by their computer operated service which couldn’t recognize my first or last name – of course my machine learning background helps me understand the complexity of HMM models and their less perfectness. Yet, voice recognition is theoretically a solved problem, and there are alternate ways of confirming the customer – perhaps by asking the customer to enter eTicket number and date of birth, still following the protocol of two independent identification methods – one public and one private. The lady at the checkin counter appeared as if she really wanted to do something to help me get out of Boston, yet didn’t convince me in doing so. She tend to get offended when was politely asked to check for different connections – Texas, Arizona, LA, etc, different number of stops (I didn’t care how many stops it would take, rather I just wanted to get out of the fully snow carpeted Logan international), and different destinations in California. She was rather interested in finding a route that exactly matches my original itinerary (of course this means less effort and computer does the searching). Finally she was able to find me to book me for Wednesday three days from my expected departure date.  Having noticed two direct flights to LA being not cancelled, but rather only delayed, I offered to take a chance on standby which wasn’t neither suggested by her in the first place nor enthusiastically welcomed when I suggested, I took the chance, and United Airlines staff did their best to help the place take off. While other airports in New England and east coast, have ceased their operations owning to this massive snow storm, Logan was still operating, occasionally canceling few, and trying the help as much as plans get off; the airport staff deserves a big applause for this. The flight attendant at the boarding gate was, in stark contrast, extremely friendly, and yet managed to help as many as standby passengers board, and here I am realizing how United could still manage to hold few loyal customers and succeed in attracting new frequent flyers.

Omar Ramos, is a friend I came to know through Google Summer of Code 2009. Out of my applications to this competitive open source coding program, Joomla wasn’t my first choice, nonetheless, now I am glad that I chose it. Omar was my mentor in the project, though we started to get in touch in casual terms only after the end of the project. Perhaps, we shared the same profile, we both expected the same from other,  or rather we both wanted a brother – perhaps younger brother, yet we managed to make the acquaintance a friendship in no time; I met him for the fist time at JDC 09 at NYC, he came stayed with me for couple of days and checked out places in and around Boston. Also he invited me to his place, offered to pick me up from San Diego airport – which is about 2 hour drive from El Centro. And this latest mishap had made me seek an alternate destination – LA which is 2 hour drive from San Diego. Still he enthusiastically offered to drive upto LA which puts me in a greater debt than I anticipated, and later I learnt that he was even waiting at the airport for 2 hours. This quality, not very common in America  - or for that matter even in my native country – Sri Lanka, where family, relationships and social life are valued the most, is something makes the friendship invaluable – which must be cherished and preserved.

There is only a narrow line between liability and feasibility, and it is we who choose to view through either glass of paradigm and everything appears accordingly.

Related posts

• From 6.840 to 754 (November 13, 2009)">From 6.840 to 754 (6)
• Tactical GRE (March 7, 2009)">Tactical GRE (3)
• Should learn to bargain and treat everything as a business (July 29, 2009)">Should learn to bargain and treat everything as a business (1)
• Increasing the relevance along with the ontogeny of a community site (July 29, 2009)">Increasing the relevance along with the ontogeny of a community site (1)
• From small miracle to real miracle (October 13, 2009)">From small miracle to real miracle (15)

6.840 is a class – Theory of Computation, taught by Michael Sipser,  who wrote the popular book Introduction to Theory of Computation; well, being at MIT is like, getting the blessings from the God himself. Many theorems appear in the class are also attributed to other MIT alumni – Noam Chomsky of Chomsky notation, the reduced form of Contest Free Grammar (CFG), Rivest, Shamir and Adleman of RSA,  Sleyer of PSPACE-Completeness, etc. I have heard people mentioning,  how people go bewildered when, a decade back at University of Peradeniya, late Prof Thurairaja, used to start his lectures by writing Thurairajah Theorem on the board; I am witnessing it myself only now. I have attended lectures given by professors, senior lecturers, lecturers, probationary lecturers and even instructors in my alma mater, where slides were being skimmed through, jargon and illegible words were being scribbled on the board, and the voice that couldn’t be heard at even the front row being uttered. But what makes you stay at in this particular lecture is the flow and elegance of presentation – the art of teaching. I know someone who is taking this class and coming to lectures, albeit having taken all the required graduate classes in his senior year when he was an undergrad here, only to listen to Sipser and enjoy mathematics (too bad, I am from the other side of the planet).

6.840 comics

6.840 is offered under course 18 – applied mathematics and course 6 – electrical engineering and computer science. In course 6, area 2, which is Computer Science major, inorder to satisfy Technical Qualifying Examination (TQE), at least one class from Theory (there are three sections – AI, Theory and System) is required and could you guess what would be the first choice among non-thoery people? First let me start with the book Introduction to Theory of Computation. It is a graduate level (or senior year undergraduate) material, yet first chapter (chapter 0) is dedicated for basic mathematics – definition of probabilities, inequalities, etc. The text then delves deep into the subject often providing examples and analogies. Could I ever have had a droplet of this comparatively abstract material otherwise?

6.840 defines the essence of teaching – thorough understanding, vast knowledge around the subject, simplicity, empathy, assuming no previous knowledge and preparation. Having written a book, and being able to conduct the lectures without any notes or slides Sipser satisfies the first two conditions. He only uses the chalk board – except for one time, when he used an OHP to show the historical letter written to Von Neuman on what is known as the earliest discussion on Theory of Complexity; he invites questions and often pauses, makes sure we are following him and if not puts his strongest effort to transfer the concept.  Probably it is the preparation, that is highly regarded as the basic requirement of teaching in the United States  – perhaps because even a monkey could teach if it could memorize all, and an interesting occasion reflecting this was when he tried to explain Word Ladder game by offering the example – Lead to Gold and he was taken aback by the solution class offered in seconds, which allegedly took 15 minutes earlier when he tried during breakfast on prepping up for the lecture. Even simple – stupid examples are worked out beforehand; isn’t that better than appearing stupid in front of the class?

C2867 is the class I am going to teach for middle / high schoolers as part of Splash. MIT has been always keen on giving back to the community, and in this context, it conducts a series of short courses under a program called ESP. The teachers – volunteer undergrads and graduates students are required to fill the applications, provide course details and attend a mandatory teaching session; the session I attended was carried out by two girls – presumably undergrads in their junior years. What makes it worth mentioning is the way it was conducted and consideration for people’s (including school kids) time shown, especially when considering the fact they are still young undergrads.  She wanted to pass the message that teachers are encouraged to come 20 minutes before the class, and not to disappoint the students by not showing up or coming late. First she used to add “where you are supposed to come 20 minutes before the class” with funny body expression and tone, every time she uttered the word Bush room (the place that will be used as office room for Splash). “Suppose you will be little late, say you have a real emergency, for example you are running to Splash and a meteorite fall in front of you and you will have to go around it, take your phone and let us know that you are on the way and will be little late”, she repeated at least five times. Presumptuous faculties who have little concern for other people’s time have a lot to learn from these kids.

C2867 is one of the numerous classes offered in Splash 2009 for which well over 2000 students have registered. Splash is one of the examples, though not perfect, of student responsibility, commitment and pro-activeness. It is completely organized, publicized, and conducted by students – mostly undergrads. Unlike exhibitions or displays, this type of activities are of high return to the community compared to the spending in time and money. Although similar programs are conducted by many other universities around the world, the significance lies in the organization, effectiveness and gain; of course neither graduate students would volunteer to teach nor school kids would attend otherwise, considering the weight given for the productivity and time spent in the United States.

754 is the number of votes (of total 16000 ballots) Leland Cheung, Cambridge  councilor – elect, received in the recent city council elections. Joint MIT/Harvard MBA student Leland Cheung made history on Nov. 3 by becoming the first university student and the first Asian American to be elected to the Cambridge City Council.  Participation in political activities is highly disowned at my alma mater, and politics is considered taboo. However, in par with elsewhere in the United States, dialogue and discussion are hugely welcomed and encouraged at MIT.  Starting from Open House for every courses where professors engage in casual conversation with students, to “Two Dollar Dinner Tuesday” (Alumni Dinner Series) where Dean of Graduate studies, alumni and students involve in informal discussion over dinner, to task forces comprises of student groups, fraternities, faculties, and other officers to formulate or reevaluate policies, diversity, discussion and democracy prevail.

754 may not huge, but yet significant. As a MIT – Sloan School graduate student, Leland was expected to receive backing from student voters, yet Cheung was not, in fact, carried into office by waves of MIT and Harvard students; though he did well among the relatively few students who voted, Leland’s votes were spread uniformly across the city  (MIT news). What is remarkable, though not very uncommon in the United States,  is that he was able to convince the general public all around the city, and be a role model on the topics concerned, through transparent communication; again, when compared to my alma mater where undergrads are used to picket, wear arm-bands or even throw stones at each other to communicate their message, and officials who are used to pay attention only when the situation bothers their daily activities, the former is essentially nourishing.

I told you three stories on three different arts, and, I feel, some stories must end without conclusion, and let the reader connect the dots.

PS: If you find the connections, leave it in your comment.

Related posts

• From small miracle to real miracle (October 13, 2009)">From small miracle to real miracle (15)
• Tactical GRE (March 7, 2009)">Tactical GRE (3)
• Should learn to bargain and treat everything as a business (July 29, 2009)">Should learn to bargain and treat everything as a business (1)
• On a snowy sunday from Logan International (December 26, 2009)">On a snowy sunday from Logan International (0)
• Increasing the relevance along with the ontogeny of a community site (July 29, 2009)">Increasing the relevance along with the ontogeny of a community site (1)

University of Moratuwa, well, the majority of expected audiences are from Sri Lanka and lengthy description is extraneous; nevertheless “Lets bring technology and professionalism” is worth reading.

MIT, the leading technical university in the world, having been continuously ranked first in various fields of engineering and technology, contrasts with its Sri Lankan counterpart in every single area, albeit the fundamentals being the same. The institute is famed for its passion for numbers, pranks and standing out from the crowd. From naming the buildings with numbers to calling the courses by decimals to the design of buildings, it chooses to be different, unique and relevant. The conspicuous element out of all is the  design of stata center.

The building has been planned so as to foster interactions not just within a department but also between departments, under the assumption that “scientists are not just cogs in a research machine, but highly creative – and highly social – thinkers”. The center invites people to mix in every possible way. — (more)

The most noticeable is the focus on objectives and flexibility surrounding the non-essentials. For instance, MIT’s undergraduate curriculum involves the hardest coursework, to be referred as “drinking from a fire hose“, yet freshman year is not graded and sophomore year only records passes. The curriculum requires an extensive list of courses, physical education, and interestingly swim test.

You can get a perfect GPA and lead every extracurricular activity on campus, but if you don’t pass a swim test and complete 8 Physical Education (P.E.) “points,” then you won’t receive an MIT diploma. Harsh, eh? That’s right, MIT requires you to graduate with a strong mind and body, while ensuring that you will not drown if your get knocked into the ocean while taking measurements of toxic algae concentrations while you’re on a boat in the Pacific (for example…) — (MIT Admissions)

The graduate course, different in expectations, focuses more in research and discoveries. (MIT receives its capital by endowments, operational support through federal and other grants, and generates one of the largest research revenues in the United States.)  Although there is a system of course requirements you could literally petition for anything and choose to take any course as you wish.

TQE requires four, H level courses (advanced graduate courses). It is not uncommon to see significant amount of undergraduates taking H level courses in senior years. The method of teaching comprises of two, one-and-half-hour long weekly lectures and recitations conducted by teaching assistants (TA) for one hour. Coursework generally comprises of Problem Sets (pSets), mid-term examinations, projects and final examinations. Both professor and TAs hold office hours every week, and usually well attended. How hard is to drink from fire hose? Generally mid-term examinations and finals are open-book, after all the life provides you all you require and expect you to solve your problems on your own, and why not MIT’s grad course? Collaboration is usually encouraged in pSets, nonetheless they are individually attempted and solutions are discussed in groups. The pSets are generally tough, often newly created set of problems for which sometimes professor wouldn’t know the answer even, weekly or biweekly posted, superficially discussed in lecture hours, personally worked by the professor well before the deadline, and occasionally graded only for submission in which case it would have been discussed extensively during TA office hours.

There is no silly question! People here are trained to live independently and question everything they hear and see. I believe Buddhism also teaches the same, yet I was able to witness it in practice here. Any lecture wouldn’t go without at least ten questions asked, and initially I, as a student, used to get annoyed by such interruptions, while it is encouraged and well received by the professors. There is no silly question (purposefully repeated)! However simple a question is, it is not uncommon to be referred by “that was a great question, let me explain this way”. I just remembered, how I was told not to ask more in my second alma mater! I wasn’t the Oliver Twist, and it wasn’t the London orphanage feeding under poor law.

Further it is very common for a student to disagree with a professor, and sometimes professor, himself would volunteer and ask whether anyone disagreeing, and continue by telling where someone could have disagreed and why his statement remains valid. Such conducive environment is what makes impossible things possible, not when ideas are stifled and nipped in the bud.

The research work, demands intellectual findings and hard work, and generally hacking and smart engineering are discouraged. The importance is given to research in graduate courses. Number of publications, nominations and awards measure the success, recognition and respect are gained by hard work and significant achievement. I have to stress on publications. MIT is well known for entrepreneurial activities (100k is one of the most popular initiatives), the institute, city of Cambridge, and New England as a whole provide the best atmosphere for any such initiatives; many – both graduate and undergraduate students – go on starting their own ventures immediately after graduation, in the areas ranging from food trucks to advertisement agencies to high-techs; yet publication is what get you through your graduate career.

I was told to call my supervisor, John Guttag, by his first name from the first VOIP conversation I had with him. Well, it took me a while to stop conferring him the knighthood which I was trained to in both my almae matres. Oddly he does not fail to apologize when he gets late for a weekly scheduled meeting, which is quite opposite to what I used to be, waiting for hours for someone to show up at their office. The conversations are highly informal, often taking interest in personal matters. For instance, he would ask how I am settling in (indeed with a question mark), and I would ask about his house that is being rebuilt.

The carrot and stick theory is long gone! I was used to read that only in Organizational Management text books till now. No one asks you to do anything! No one is obliged to do anything, except for iPhone owners, well they are bound with AT&T for two years. I have seen, and been to competitions and events with as few people as four, and struggled to get into fully packed auditoriums. Organizers understand their duties and they do enough publicity, and provide adequate incentives – the main and most common are free food (lunch, snack, refreshments or dinner), gifts, raffle draws, etc. There hasn’t been a single event without free food and drink. Motivation theory is well understood and well practiced by everyone, starting from a hair dresser, to waitress, to organizers to professors. The exams always carry 2 points only for writing your name and registration number on the question paper; and they never berate you for not attending some boring meeting, instead would focus on making it more appealing.

Rules and laws are there to guide not to control. The best thing about MIT and especially the system in USA is, there is no red-tape. Anything and everything is petition-able, and adaptable. It is very common for people to switch their degrees, do both MD, and PhD (need not be in biology), and basically do things outside the rules. The system bends as far as it serves the objective and best interests of the individual and the institute, which contrasts a lot with where I came from – where the law bends to serve the best interests of the fittest of all, or remains inflexible. Here, often deals are made; compromises are very common, and result is an improvement over whatever status quo it was. For instance, as per PhD requirements, one is expected to take two minors – graduate level courses outside the research area – and foreign language courses are often not allowed, as they are often of beginners level and hardly offer any significant benefit. Had this red-tape been enforced as in Sri Lanka, would there be someone who did take beginners level course in Mandarin and went on to do Chinese literature, or someone who fell in love with a Chinese girl and mastered the language through his minor requirement. For someone who was prevented from doing CS courses in his undergraduate and faced with difficulties in applying for transcript this must be welcoming.

It might sound as litany of complaints, but again it requires a paradigm shift to view them as observations based on direct comparisons flavored with little satire. The life is not necessarily the happiest here. After all, people here work hard, and play hard. I have seen people only watching tech shows in TV and only listen to iTunes U in iPod. I sometimes wonder whether they live. The path is always decided by efficiency and often harnessed by the objectives; and every second is measured in terms of productivity. Unlike what you see in TV series and movies, they do go to work, and don’t wait to finish their cappuccino in the coffee house.

Related posts

• Tactical GRE (March 7, 2009)">Tactical GRE (3)
• From 6.840 to 754 (November 13, 2009)">From 6.840 to 754 (6)
• Should learn to bargain and treat everything as a business (July 29, 2009)">Should learn to bargain and treat everything as a business (1)
• Increasing the relevance along with the ontogeny of a community site (July 29, 2009)">Increasing the relevance along with the ontogeny of a community site (1)
• Talentday 2007 (August 19, 2007)">Talentday 2007 (0)

It has been quite long since Joomla! 1.6 Alpha came, and there have been a lot of changes made in Joomla! 1.6 since then. Further, it was also decided to adapt the architecture slightly to follow Joomla!.

The major changes since beta, excluding bug fixes are

1. Taxonomy TreeMaps are made standalone entities (equal to Trees and Leaves), with CRUD.
2. Utility functions are moved to JTaxonomy class which is the primary entry point for taxonomy extension.
3. Autocomplete feature is further brushed up to provide better user experience.
4. Taxonomy Cloud and Taxonomy Cumulus are provided.

The package provided below contains

1. Taxonomy Library : API classes
2. Taxonomy Component : To provide management forms in the backend and listings in the frontend
3. Content_MapTaxonomy Plugin : To store taxonomy-content mappings in the event of a post save.
4. Content_Taxonomy Plugin : To provide taxonomy links inside the article in frontend.
5. Taxonomy_Content Plugin : To generate output of content posts under leaf listing
6. Taxonomy Cumulus Module : To generate cumulus and cloud of taxonomy terms
7. Modified com_content file : To insert Taxonomy options into content edit form.

The suggested installation order is as follows

1. Copy the following directories into Joomla! installation
   • libraries
   • administrator (will replace a com_content file)
2. Install com_taxonomy.zip : Under taxonomy component, go to configurations to fine-tune
3. Install plg_content_maptaxonomy.zip
4. Install plg_content_taxonomy.zip : Go to settings to style up the output
5. Install plg_taxonomy_content.zip
6. Install mod_taxonomycumulus.zip : Add one or more instances (there are two modes available – cloud and cumulus, and you might want to add one from each, configure them from settings)

You might also want to take a look at the demo, before digging in

Visit : http://demo.theebgar.net/gsoc2009

1. There are two taxonomy trees created and mapped to com_content. One is used to represent the language of the post, and the other for free terms (tagging) with autocomplete. Both are not controlled (i.e. properties can be edited and member leaves  can be added and edited from com_taxonomy backend)
2. On the frontend
   1. Sidebar
      1. Mod_taxonomyCumulus is used in Cloud mode to represent Language Tree (<br /> delmiter is used with sorting by weight in descending direction)
      2. Mod_taxonomyCloud is used in Cumulus mode to represent tags (white background with no transparency is used)
   2. Content Posts
      1. Taxonomy links are presented before content with “,” delimiter, “<em><small>” prefix and “</em></small><br />” suffix.
      2. By clicking a taxonomy term in that link you could navigate to taxonomy listing page for that term
   3. Taxonomy Link from menu
      It will take you to the listing pages
3. On the backend

1. Send a request for access details (use the contact form here) .
2. By using component menu navigate to Taxonomy Management Pages
   1. Have a look at Taxonomy Configurations, and follow the help provided
   2. Add / Edit trees
   3. Add / Edit Tree maps with other components
   4. Add / Edit leaves to a tree
   5. Have a look at the tree hierarchy and leaf maps
3. Navigate to Article Managment and add an article or edit one.
   Under Taxonomy Panel from the sidepane, you could notice that the terms from Taxonomy Trees mapped are listed

Screenshots

Related posts

• Taxonomy Extension – Project Proposal (May 9, 2009)">Taxonomy Extension – Project Proposal (0)
• GSoC 2009 – Taxonomy Library API (June 19, 2009)">GSoC 2009 – Taxonomy Library API (0)
• GSoC 2009 – Taxonomy Architecture (May 11, 2009)">GSoC 2009 – Taxonomy Architecture (1)
• Lets bring technology and professionalism into FYP (July 15, 2009)">Lets bring technology and professionalism into FYP (5)
• GSOC 2007 (February 21, 2008)">GSOC 2007 (0)

The discussion about money is treated as taboo in many cultures not limited to Asian, and we are quite uncomfortable talking about money, especially when it is related to Health and Education – the professions we treat as holistic. We tend to avoid asking the fee beforehand in either cases, and expect good service at a reasonable payment.

I had to complete a form on medical test results and vaccination that is required by MIT before registration. Immunity for the following were required either by taking vaccinations or by proving the presence of antibodies through positive test results.

• Measles, Mumps and Rubella (MMR)
• Hepatitis B
• Tetanus/diphtheria booster
• Meningococcal
• Varicella (chicken pox)

I chose to go to a medical center, which I got to know through a friend of mine.  I liked the idea of going to a small place, as it would be less crowded and won’t reek with hospital smell. Also channeling is done by appointments which seemed convenient. The place is also in the suggested list of places for medical check-up for Canadian visa.

The physician, was amiable and nice. Although I had to wait for about 20 minutes on my second appointment, overall the experience was  pleasant. The physician was attentive and empathetic, and room was clean. I was given two injections (MMR vaccination and Menomune vaccination). Also blood was taken to test for antibodies for Hep B and Varicella. So far so good. I was shocked, only when I saw the bill for 10500 LKR for channelling and vaccination, and another 6000 LKR for blood tests. Interestingly, I had had chicken pox and hence, the test was not required which I attentively mentioned. To exacerbate, when she filled the form the other day she didn’t even bother to take out the test results, and she circled the option that I had already had chicken pox, which of course she could have done without ordering for the test results. Damn-it. I must have protested earlier.

I had to go two days later, so that she would consult the test results and complete the form (I required the signature). The happenings were quite fast, that before I could ask for the price or say no, the injection was ready, and another surprise was waiting when I learnt it was another 8500 LKR from my pocket. For my bad, I had already checked outside, and it is priced at 1500 LKR.

However lesson was learnt, signature on the form and list of vaccinations I had, have been obtained, and I am getting the next doze at a dispensary near by.

PS: Today I learnt Hep B vaccine costs only 50 USD in the US which would be about only 60% of what I had paid.

Related posts

• Tactical GRE (March 7, 2009)">Tactical GRE (3)
• Increasing the relevance along with the ontogeny of a community site (July 29, 2009)">Increasing the relevance along with the ontogeny of a community site (1)
• From small miracle to real miracle (October 13, 2009)">From small miracle to real miracle (15)
• From 6.840 to 754 (November 13, 2009)">From 6.840 to 754 (6)
• On a snowy sunday from Logan International (December 26, 2009)">On a snowy sunday from Logan International (0)

SL2College

I was requested to provide a small write up on how a forum (sl2college) helped me in applying for the universities in US.

I got to know to sl2college during a session at US Fulbright Commission, Sri Lanka, when it was just a yahoo group and I am great to see that evolved into what we have a CMS powered forum + website. But how about its usefulness and relevance?

While something might not be making a difference to a wider audience it could be bringing huge difference to single person. In that context, I see the objective of the sl2college to be, a forum to share the ideas, a market place to facilitate the meeting between providers and consumers, resource library that archives the answers for well known or already answered questions, and finally a place to get the questions answered by voluntary resource persons. How it is being used? I have seen activity in all aforementioned aspects, and for me, it has served as only a resource library thus far.

From my GSoC experience, in a community I see 4 types of participants. Active givers, who likes to answer questions and help people for some reasons, active takers, who suddenly appear with questions not necessarily appropriate, relevant and well formatted, contributors, who answers the questions directed to them and passive members, who do only reading possibly through feeds and mail-lists. I had been a passive member in my opinion.

I believe, for the betterment of the group, we should encourage the participation of the first and last type of people by easing up the process and, welcome the second and third for better utilization of the resources that get accumulated. Also the communication can be strengthened if the latter takes some responsibility before requesting the help.

To make it easy to contribute, we could bring the following 3 aspects into the sl2college site.

• Posting an article by e-mail (if it could support a blogging application like Windows Live Writer it would be excellent, which is supported by Drupal and Wordpress)
• Simply sharing a link (perhaps weblinks approach in Joomla or blogroll in Wordpress). It could be the author’s blog post like what you are reading now, or an article that seemed significant to be referred.
• Encouraging users to contribute in the form of articles and blogs that will be followed up by comments. How many feel comfortable with forums as a resource repository?

This will also ensure a follow up, as users will be interested in answering a question posted under their article than an arbitrary question.

Although it will alleviate the need for QnA platform like a forum, still users might  not find all of their questions answered even if they have done their homework. In which case, they could write a blog in question form, providing a small introduction about them, and the overview of the problem followed by specific questions. This will not only be beneficial to future readers, but also make an itch on the contributors to answer when they find that the users have done their homework. I find it quite essential, as it was annoying to see common, already answered questions like “how to attempt GRE?”, “how to apply?”, “what are the best universities for this area?”, etc being posted. A single Google search would have led to the answer instead (of course at the cost of 7g CO2).

Further, the sl2college must incorporate at least basic level of social networking features like, members directory, member contact forms, and possibly twitter and / or Facebook integration. Also it must be taken into consideration that people tend to spend more than 80% of their online time in only a handful of selected sites like gMail, Facebook, Google reader, etc and it is paramount to increase the presence of sl2college in those sites to improve the utitlization rather than ringing the bells at sl2college itself.

For some reference in asking questions ..

• How To Ask Questions The Smart Way
• How to Ask a Question Intelligently – wikiHow

And finally a quote from WikiHow

Never ask a question just for the sake of it, whether it be to bring attention to yourself or for appearingsmart. That is the worst possible motivation for asking a question.

Related posts

• Tactical GRE (March 7, 2009)">Tactical GRE (3)
• Should learn to bargain and treat everything as a business (July 29, 2009)">Should learn to bargain and treat everything as a business (1)
• From small miracle to real miracle (October 13, 2009)">From small miracle to real miracle (15)
• From 6.840 to 754 (November 13, 2009)">From 6.840 to 754 (6)
• On a snowy sunday from Logan International (December 26, 2009)">On a snowy sunday from Logan International (0)

I was getting angry easily, wanted to detach myself from others, lost the usual satire in speech, and strangely talking, although not pleasantly always, with my parents more. I also felt like doing nothing, enjoyed the sky and blue mooon, and was rather interested in choosing a new design for the blog and doing small modifications to suit my expectations whereas I could have done more classes, finished the projects or looked into other ways of making more money. I was feeling ravenous, and even after taking benzimidazole drug. Hardly it could have beeen a captivity of negativity nor depression.  I was searching for the answer for what is happening to me.

I happened to flip over my Onenote – note page on 11 minutes book and realized probably it is the strange feeling of codependency makes me so ..

Do you know who the loneliest person in the world is? The
executive with a successful career, earning an enormous
salary, trusted by those above and below him, with a family
to go on holiday with and children who he helps out with
their homework, but who is then approached by someone like me
and asked the following question: “How would you like to
change your job and earn twice as much?”
‘The executive, who has every reason to feel wanted and
happy, becomes the most miserable creature on the planet.
Why? Because he has no one to talk to. He is tempted to
accept my offer, but he can’t talk about it to his work
colleagues because they would do everything they could to
persuade him to stay. He can’t talk about it to his wife, who
has been his companion in his rise up the ladder of success
and understands a great deal about security, but nothing
about taking risks. He can’t talk to anyone about it and
there he is confronted by the biggest decision of his life.
Can you imagine how that man feels?’
No, that man wasn’t the loneliest person in the world.

- From 11 Minutes, by Paulo Coelho

Related posts

• No related posts.

I particularly identify the following areas to be improved.

Group formation

The list of names of group members (4 undergrads per group) was called in March, 2009 (if I remember correctly). The groups were mostly created among friends, colleagues (those who did the internship at the same place), and based on the popularity of members.

Although I am not confident about its practicality, I think, it should follow “idea-incubation stage”, so that members  could choose to join a group based on the idea and complementing merits of others who have also expressed their interest on a topic.

Idea incubation

Project proposals were called following the a session where we were introduced to the procedures. Approximately two weeks time was given. However proactively we had planned a fortnight ahead and came up with a list of 13 projects on our own and chose 5 top projects, that would meet the final year project’s complexity and requirements. Few to note were, a shopping guidance system with RFID or any others(we later learnt the possibility of various indoor transmission, localization and guidance systems) to assist buyers, movers and managers in a large shopping complex, a system to transmit mobile phone calls over land lines, when the call is made from inside the building and Zigbee based intelligent home automation, and X-bee/Zigbee based traffic light management + bill board management. We had mailed two staff whom we thought would be interested in the selected topics, and met one, but apparently they were more inclined towards their projects.

The identified difficulties are

1. Although about 2 weeks time is given to select a project, not all lecturers post their ideas immediately, and most of them add more later. This effectively reduces the time left to weigh the options.
2. Although it was emphasized that student ideas are preferred, it is yet to be seen in practice, and even a mild disinterest that appear from a staff will be hugely discouraging as undergrads tend to belief that staff will grade their own projects favorably.
3. There is no common forum to discuss the ideas, comment or follow up. No transparency at all.
4. When staff proposed projects are listed very abstractly, it could lead to two possible disasters: students choosing it based on the appealing title and get disappointed when they find it was not what they intended, and students leaving the topic because of the ambiguity. In such circumstance, it is wise to consult the staff who proposed it. However, few staff are quite busy with other works, and generally not available for discussion in person and / or takes longer than reasonable to respond (provided that students have less than a week to choose). In such situation, it is better to follow another project as the students will be left helpless if this behavior continues in future as well.

I believe staff should become easily approachable, available at least over the email or by appointment, and the communication should be more transparent, friendly and objective.

By looking at the statistics on the projects that made any significant changes  we could evaluate the efficacy of the system.

I see the main reasons for the projects being not pursued further as follows

1. Students are not passionate about the project : while it is unavoidable, the rate could be reduced by allowing more student proposed projects. It is those individual itches that are strongly pursued after, not someone else’s.
2. No immediate financial viability foreseen in the projects : most of the projects are complex in breadth but not in depth, and it requires larger infrastructure facilities to mass produce. Project ideas are not easily convertible to money.
3. Support from industry is inadequate : Although it could be a proactive nature for a student to acquire support from an outside establishment, it behooves academic staff to build relationships not necessarily limited to Sri Lanka only.

I suggest, that following carpenters rule of measure twice, cut once, minimum a month should be allocated for this process and it should be done over an online discussion platform with the support to topics subscription, follow ups, comments and ratings. Students should be encouraged propose their ideas through this medium and staff members should actively engage in the discussion. Also they could propose their ideas and let students discuss on it. By their participation in the discussion under a topic, students could identify their possible advisor and vice versa. By the deadline students could submit their proposals for more than one project and staff members also could pick more than one. With the availability of proper tools, this could be easily done, and conflicts could be resolved a over a meeting possibly online with students and staff involved.

Google groups, probably coupled with a messenger (google talk or Skype) and / or IRC (Freenode server is one of the popular) meets the above requirement. Also it is compatible with Melange (mentioned below) and hence a fully integrated solution can be easily put on.

Project proposal

While even state institutions in the United States are trying to minimize paper work by utilizing e-formats, they are still required extensively in the only technical university in Sri Lanka. Not only it is a waste of money, but also a pollution. What happens to those printed papers by end of the year?

I suggest that proposal submissions be done online. Also, what I had noticed  was both staff and students are interested in seeing more graphics, pictures and tables than actual content. It is quite understandable as it eases understanding, it also implies lesser original work from student, as a picture takes about 10 times of time to make compared to paragraph of content. Therefore, I suggest making the proposal through a WYSIWYG powered editor online where students could provide links and references to other resources easily instead of copy/pasting and thus filling pages. This not only improves the quality of content, but also eases the unnecessary work load related to formatting and editing. In addition, it enables the plagiarism testing, spelling and flow checking, etc.

The department is using Moodle for the purpose. Being a CMS developer, I am not appealed by Moodles design and usability.

Melange, a projects management framework backed by Google, (accessible at http://code.google.com/p/soc/wiki/MelangeIntro) could be used for this purpose.

Being a 3-time Google Summer of Code participant, I am hugely impressed by the professionalism of the coordinators, appropriateness of the tools involved, and passion showed by the participants (of course the 4500 USD stipend involved makes a huge difference) albeit the scaring total of more than 1000 participants, 1000 mentors, and more than 100 institutions.

Why can’t we do the same with 100 students and 10 staff? Perhaps it is time to bring more technology into HOW THIGNS ARE DONE.

Communication

Although we were fortunate to have a supervisor who wouldn’t mind us bugging over email nor call his mobile, few were waiting for their supervisors for hours. It is understandable, that supervisors will have other commitments that have more priority. However, no regard shown for others’ time portrays the arrogance of the staff. It is becoming common for students to involve in outside activities, including extracurricular, co-curricular, entrepreneurial and other professional activities, and for few university career brings a smaller benefit compared to others, let alone FYP. Especially the economic crisis and job-cuts made other activities more valuable, and continuous disregard for undergrads’ time will only make them get fed up and give up eventually.

Both could have a win-win situation by setting a prior appointment or by having a fixed time assigned to meet FYP groups. This is possible over an email, but not everyone is organized enough to keep their words and appointments. A micro-blogging framework with the ability to add groups (or channels), follow selected groups, persons and topics, pinging, personal messaging, and status messages will alleviate the problem as it enables the staff to contact students easily, and quickly, while students could also respond instantly. Further students could easily check the availability of the lecturer from the status message and/or by pinging them.

In the second part of this post, the following areas will be discussed.

• Logistics and Resource Management
• Funding
• Procurement
• Project Evaluation

Related posts

• GSOC 2007 (February 21, 2008)">GSOC 2007 (0)
• Taxonomy Extension RC1 (August 3, 2009)">Taxonomy Extension RC1 (21)
• Taxonomy Extension – Project Proposal (May 9, 2009)">Taxonomy Extension – Project Proposal (0)
• Talentday 2007 (August 19, 2007)">Talentday 2007 (0)
• GSoC 2009 – Taxonomy Library API (June 19, 2009)">GSoC 2009 – Taxonomy Library API (0)

I had some discussion with the mentors as where to place the taxonomy library folder, and finally it was decided to go under libraries/joomla/taxonomy, although it could be moved later to to libraries/taxonomy if it is deemed to be more appropriate.

The latest code can be checked here :�http://labs.joomla.org/svn/labs/playground/people/garthee/trunk/

The demo can be tried at (need not parallel the version in SVN) : http://gsoc.theebgar.net/administrator/index.php?option=com_taxonomy

JTaxonomy

At the root JTaxonomy class which extends JClass, exposes the full set of features. It is used in a singleton form and must be invoked as

jimport(’joomla.taxonomy.taxonomy’);
$taxonomy =&JTaxonomy::getInstance();

This class provides the following set of methods

1. Tree related

&getTreeList($extension = null)
&getTrees($extension = null)
&getTree($id = 0, $options = null)

First returns an array of tree names keyed by tree IDs, second returns an array tree objects keyed by tree IDs, and the last gives tree object for the given tree ID. They are eventually routed through JTaxonomyTree which provides the implmentation for trees.

2. Leaf related

&getLeafList($options = null)
&getLeaves($options = null)
&getLeaf($id = 0, $options = null)
&getLeafHierarchy($tid, $options = null)

The last function serves a important function implemented in a helper class. It builds a complete leaf hierarchy under a given tree.

&getMap()

JTaxonomyTree

JTaxonomyTree extends JClass, and mainly handles the CRUD operations which are routed through its JTable subclass JTabeTaxTree class.

JTaxonomyTree Class

JTableTaxTree Class

JTaxonomyLeaf

JTaxonomyLeaf class

JTableTaxLeaf class

JTaxonomyMap

This is class handles the object – leaf mapping, the fundamental operation of taxonomy frame, through standard methods. The implementation is expected to change with the requirements imposed, which will be actually finalized along with the implementation of com_contentPlus (a content component with taxonomy support). �This class together with JHTMLTaxonomy is expected to provide a complete form support to the integration with any components.

JTaxonomyMap Class

JHTMLTaxonomy JHTML subclass

Summary

Although the feature sets of Taxonomy Library is currently completely determined, it might evolve with the requirements of components and plugins as demanded. Hence it is provided here for reference and feedback only. In the next blog I will discuss the dilemma we had in compromising between performance, standard pattern and generality.

Related posts

• Taxonomy Extension RC1 (August 3, 2009)">Taxonomy Extension RC1 (21)
• Taxonomy Extension – Project Proposal (May 9, 2009)">Taxonomy Extension – Project Proposal (0)
• GSoC 2009 – Taxonomy Architecture (May 11, 2009)">GSoC 2009 – Taxonomy Architecture (1)
• Lets bring technology and professionalism into FYP (July 15, 2009)">Lets bring technology and professionalism into FYP (5)
• GSOC 2007 (February 21, 2008)">GSOC 2007 (0)

The primary objective of classification frameworks is to decouple categorization from object creation and management and provide a human friendly, visually recognizable, flexible alternative to search indices of objects.

Starting from directory architecture used by virtually any operating system, methods of classification have in been in use for long time, under the principle of decoupling classification from creation, management, storage and identification of objects.

Taxonomy classification, and its currently popular variation – tagging are widely used in many web frameworks and desktops. Gmail started it with replicating a folder structure, with the concept of placing an object in virtually into more than one bucket perhaps following soft linking that existed in file systems for decades.

This has been given a new look when web services like Gmail and Flickr adopted taxonomy classification for a very specific purpose of classifying only few selected objects. For example, we could choose to apply the classification or live without it. Later similar classification concept has been adopted to many desktop applications such as iTunes, Explorer in Microsoft Vista OS, etc and web frameworks.

Apart from classification framework’s primary objective of decoupling the categorization logic from content management, we will focus on the following indispensable entities in a web framework. This must also take the frequency of operations into account in determining the crucial factors. Implementation specific details are discussed under design considerations.

Database is normalized to 3NF to ensure extensibility while being scalable. Frequently used queries like relationships, object mapping and leaf membership are designed to be scalable.

Reliability is achieved through both hooks on different states of an object and vice versa and possibly through cron tasks.

Vocabularies

Leaf : The atomic unit of taxonomy labeling, that will be associated with the items to categorize them. Here it is referred to as terms and tags interchangeably.�Leaves might be called with similar names and we name them alias

Alias: For example, university and college may mean the same thing and it is redundant and misleading to have two different labeling for a single term. Hence university can be aliased with college so that there will be only one term that is university is used and whenever user specifies the term college, it will be interpreted as university.

Tree: This is an umbrella unit for leaves that act as a bucket and define the properties, rules and interaction leaves underneath. It will have the following properties

• Hierarchy : The type of hierarchy involved (refer structure below)

• Controlled: Whether leaves can be created outside the admin forms (com_taxonomy)
• Relations: what type of relations are allowed. Only parent-child or peer-to-peer as well.

Tree mapping : Trees are useless unless it is mapped to an object manager, such as a component. A particular mapping involves a tree and an extension. Further it defines the following rules

• Required: Whether each object must be associated with a leaf from this tree under this association
• Multiple: Whether multiple leaves can be associated with a single object
• Weight: The factor that determines the priority for a tree when an extension is calling for all associated trees or all associated terms for an object under that extension. Heavier weights sink.

Leaf mapping: Guided by the tree mapping rules and the properties of trees, a leaf is associated to an object coming under the extension given in the tree mapping. This is also known as labeling or tagging.

Although a unified tree assimilating the structure of domain names will be adequate we consider a forest of trees for the reasons explained earlier. The top level member will be a tree, whose attributes define the usage of the tree and its structure. Leaves are members of a particular tree, albeit an unrestrained membership is considered for many reasons explained under design considerations a leaf will be confined to only one tree, but can be used by multiple extensions, still abiding by the concept of define once, use everywhere.

There are several kind of structures that can be built under a taxonomy framework. However there are three fundamental structures on top of whom a complex design can be built on and we call them hierarchies.

Hierarchies

1. Free terms

This is also known as flat structure, tagging, floating terms, etc. The tree built with such a structure will be forwarding a quite straightforward simple taxonomy system to its end users.

Flat hierarchy

2. Single Hierarchy

This is commonly understood as a taxonomy tree, where a leaf is either the root leaf or a child of another leaf. Therefore, there can be only one parent, but many children resembling top-down hierarchy.

Single Hierarchy

3. Multiple Hierarchies

The requirement flexibility may bring some rarely used features. Multiple hierarchy is such a feature, nonetheless still essential for completeness. It allows multiple parents , so that branch off is possible upwards and downwards

Multiple Hierarchy

The intention of separating the logic of structure from its building block is to maintain the maximum flexibility that is being able to virtually achieve any level of complexity in building a taxonomy system, while ensuring usability that users are not lead into wilderness. Under this design, structures will be built on the aforementioned blocks seamlessly .

Popular structures

1. Free tagging – similar to the tagging feature available in Wordpress, Gmail, or Flickr. This is of flat hierarchy, multiple, uncontrolled tree, and possibly not required.
2. Categories – Similar to categories used in Joomla! Or in wordpress. Hierarchical trees with single hierarchy. Mostly controlled, and possibly required and single. Which means user must opt for one and only one leaf per object
3. User groups – Similar to Organic Groups used in Drupal. Hierarchical trees with single hierarchy, controlled, single, and required.
4. Book – Hierarchical trees with single hierarchy, not controlled, required and single.
5. Channels – Multiple hierarchies, not controlled, multiple and possibly required.

Above examples emphasize the coordination required between the creation of a tree and mapping it with an extension. Although user or an extension could do that by directly entering in the table or using the backend forms and build a model, well thought planning is indeed needed.

Database Structure

Both the ability to handle large number of records and extensibility are considered in the design and hence the schema is normalized to 3NF. �The following tables build the taxonomy framework

• tree : It builds the forest of trees that stores the complete information necessary to build a complete tree. All the attributes are editable through management component, however, a quick build will be made available for common types of trees.
• tree_map : It stores the correspondence between a tree and another extension, for example, content component. The tree can be completely linked to an extension. such a normalization would enable reuse of a single tree for multiple applications. Weight determines the priority a particular linkage should take over a similar linkage.
• leaf : Contains the atomic information about the term, the tree it belongs to, etc. Also weight is added to ensure, the preference order when multiple leaves line up for a particular request.
• leaf_map : It is used to match the terms with an item, for example a content post
• leaf_hierachy : It contains the relationship between two leaves. The type of relationships currently supported are parent and peer, which will translate to parent-child relationship that makes up a hierarchical tree and peer-to-peer relationship that makes up a cluster tree respectively.
• leaf_alias : It contains the list of alias for a leaf. It is essential when taxonomy framework is used for components that interact with human directly where responsibility of remembering the right term cannot be enforced. Under such circumstances the ability to handle aliases, for example, words like university, universities, etc could all mean the same term – college.

Design Considerations

Hierarchy / free terms :

There are two popular taxonomy structures. Hierarchical tree (mostly single hierarchy) and free tagging (flat hierarchy). Although it is quite sufficient to content management tasks, complex hierarchies like albums with free tagging, user groups, user roles (for privilege granting purposes), etc will require variety of features. This is achieved by isolating structure from its building blocks – hierarchies, extension mapping (and properties) and tree properties.

Concentrated / Distributed

Sometimes it is necessary to focus on 90% used features while giving up the total flexibility in order to boost performance.

In design there is no need for a tree, as implemented in domain names hierarchy where everything starts at 0th level – root domain – the dot, and expanded by 1st and 2nd level domains, all are equal in representation. Similarly only leaves could have formed the taxonomy framework with the ability to map the extension to particular leaf which will govern all the children.

However it is chosen to implement multiple trees, sacrificing such flexibility for the gains in performance and more importantly support multiple hierarchies or more precisely multiple parents. It allows greater extensibility in breadth and depth.

Define here, use everywhere

Following the popular paradigm “write once, use many” taxonomy framework is expected to be the unified solution for all classification requirements. Thus must be able link with any implementation seamlessly. This is achieved by mapping a tree with a particular extension, thus make all the terms underneath available at the extension’s disposal.

Related posts

• Taxonomy Extension RC1 (August 3, 2009)">Taxonomy Extension RC1 (21)
• Taxonomy Extension – Project Proposal (May 9, 2009)">Taxonomy Extension – Project Proposal (0)
• GSoC 2009 – Taxonomy Library API (June 19, 2009)">GSoC 2009 – Taxonomy Library API (0)
• Lets bring technology and professionalism into FYP (July 15, 2009)">Lets bring technology and professionalism into FYP (5)
• GSOC 2007 (February 21, 2008)">GSOC 2007 (0)

Joomla! is a powerful Content Management System (CMS) and a popular publishing framework known for its aesthetics and Web Application Development. However it is limited by the simple section/category/article hierarchy for organizing content.

The project intends to address a greatly missing feature of Joomla! that is a well structured Taxonomy Component (API, frontend and extensions / modules) to provide proper organization of content through Taxonomies and Tagging. Almost all content needs organization for easy access and a well-designed Taxonomy Component increases the accessibility and usability of the site dramatically.

Personal Details

Name: Gartheeban Ganeshapillai
Email: garthee@theebgar.net
GTalk: garthee@gmail.com
Skype ID: garthee

Project Proposal

1. Introduction

A robust Taxonomy Component will eventually supersede similar but inferior implementations such as keywords, groups and other classifications, and eventually replace them with greater functionality. Other foreseeable areas of applications are

1. Replacement of section / category with Taxonomy
2. Use of Taxonomy for any such classification required by creating new Taxonomy trees for different applications
3. Powerful search functionality, also known as faceted search
4. Dynamic client side filtering, an innovative feature provided by Exhibit [1]

In bird’s view, a Taxonomy Component includes the following components

1. API (underlying layer for frontend modules and extensions, and backend management forms)
2. Frontend modules
   a. Sidebars – for related content based on Terms attached, Tagcloud, Cumulus [2], etc
   b. Form integration for content creation (through extensions)
3. Backend Management forms
4. Support for other 3rd party extensions (such as exhibit to provide dynamic faceted presentation, Opencalais for auto tagging, etc)

Also in design the taxonomy system include

1. Hierarchically organized terms also known as taxonomy tree
2. Flexible free terms also known as tagging

I propose the development of a Taxonomy Component that is scalable and robust so that it can be integrated into the core and extensible to cater future needs.

2. Based on the proposed idea:

Taxonomy Extension :“Create a 1.6 Taxonomy extension with the goal that this work could be integrated into the core Joomla CMS for future releases. The Taxonomy Extension will allow the ability to organise content for classification, improving on the current Section and Categories classification….” [3]

Expected Mentor : Allan Walker

3. Benefits to Joomla!:

A site’s popularity depends on not only its content but also the accessibility and usability. People tend to spend very little time looking for an item in a site, even if they are confident about the existence of the item they are looking for. A fine organization of content, will enable users (when searching) to obtain the most relevant information they are looking for by

1. allowing content creators to organize content
2. allowing admins to guide content creators in organizing content by defining protocols such as taxonomy trees, related items, etc
3. allowing users to access content easily by providing interfaces that expose the classification system
4. allowing the extensibility of the system
   • in creating terms, so that possibly other plugins or 3rd party tools can assist (this is in view with allowing features like auto tagging in future)
   • in presenting the UI to the users, so that users can choose different formats to expose their classification system (tag clouds, taxonomy trees, cumulus tag clouds, exhibit integrations)
   • in managing the classification system, so that admin controls can be enhanced to address future management needs.

4. Project Details:

Project consists of the following phases.

1. API

This section deals with the database and provides methods to create taxonomy trees, managing them, etc. The intended structure allows greater flexibility and organization such that the following functions are made inherently available

1. Multiple select
2. Free tagging
3. Hierarchical representation
4. Relationship between terms (such as similar term)
5. Separate trees for disparate classification

This can better be explained through the following table structures, where 3NF is maintained (jos_taxonomy_relationship can be further normalized)

• jos_taxonomy_tree : |id|name|description| other fields … |
• jos_taxonomy_leave : |id|tree_id |name|description| other fields … |
• jos_taxonomy_relationship : |id1| id2| type_of_relationship (can be parent, similar, etc)|
• jos_taxonomy_mapping : |term_id|content_id| other fields .. |

Further considering the scalability in mind, for appropriate functions, the generated results from a complex query such as a whole taxonomy tree, will be kept in static variables so that repeated calls are answered without affecting the performance. Such measures will be essential, when Taxonomy is integrated to core and provided as an answer to virtually every classification requirements.

2. Front end modules

The first objective of this is to expose the taxonomy classification to users. For example, we could provide tag clouds (or fancy forms of it such as Cumulus), taxonomy browsing (as an alternative to menus) and other features. I would like to restrict it for Tag clouds, Related Content, and Taxonomy browsing (optional) for the scope of GSoC. However, with a flexible, extensible API, it would not be hard to extend as we see in future.

The second focus of this area to let the content creators create or apply taxonomy terms to the content (the mapping is performed at this stage). This can be extended to provide
• Autocomplete of terms as you type
• Suggestions of terms (based on the content body)
• Auto tagging (using third party tools)

I intend to provide autocomplete feature under the scope of GSoC. However, I would love to see Opencalais integration in near future too, as it will be hugely helpful in a community site where all users cannot be burdened / entrusted to submit corret terms, if they submit at all and auto tagging is essential.

3. Backend Management forms

In addition to general settings, Taxonomy Component will require many administrative configurations such as creating a new Taxonomy tree, determining who can add new terms while creating content, if it can be added all, etc. Further administrators might also want to perform actions affection masses such as tagging content in bulk, renaming or remapping tags, etc.

I would like to cover as much as possible while certainly completely general management tasks within GSoC. This task is to be clearly defined upon the discussion with the mentor.

4. Integration with 3rd party tools to enhance Taxonomy Component

First is the integration with Exhibit, a dynamic faceted browsing tool [4] from Haystack group, MIT [5] that uses many factors including taxonomy to filter and present content in real time. In addition, we could use Cumulus for tagcloud and Opencalais for autotagging.

While emphasizing the fact that a well-structured design can be extended easily to accommodate many novel features and tools around Taxonomy system, I would like to focus mainly on the well-designed backend within the scope of GSoC.

5. Background and foundation:

I understand the development of this particular module requires fluency in fouor basic areas that are Joomla!, PHP/MYSQL, Taxonomy systems and Tools around Taxonomy system. I am well experienced with the latter three and I am confident mentors and community will complement my lack of experience in Joomla!. If the proposal is accepted, I intend to use the community bonding time to learn the design principles of Joomla! (mainly from books [6]) and to finalize the design framework with the help of the mentors and community.

This proposal is highly influenced by the experience I have gained in the development of modules around Drupal’s taxonomy framework and my work with Exhibit tool to facilitate “Faceted Search” and “Faceted Browsing”. I invite you to look at the implementations of them in my sites (few are customized by me to suit my requirements, however written for a different platform)

1. Faceted Search [4] : http://theebgar.net/all/results
2. Exhibit in action (with timeline) [5] : http://theebgar.com
3. In addition, I have worked on a similar project enabling dynamic viewing of filtering which can be found at
4. http://old.theebgar.net/history – where with extensive Ajax use real time filtering of content is provided regardless the amount of content available in the site.

6. Risks

The main risk I could see is any inherent limitations in achieving the aforementioned objective due to the structure of the Joomla! that I am highly confident that the community members and mentors will be able foresee more easily than I do.

Considering the tight deadlines and conditions involved with the client projects I have worked with in the past and the amount of free time I going to have between graduation and enrollment at the Grad school I hardly see a possible conflict in time management.

Roadmap

This Component will address the aforementioned needs in its own, independently and modularly as explained above.

1. Deliverables:

1. API
2. Frontend modules for sidebar and forms
3. Backend Management Forms
4. 3rd party integration – Exhibit, Cumulus

and OpenCalais (if time permits)

2. Project Schedule :

I have already laid the foundation of the project and expecting to start working immediately if the proposal is accepted

• First week of May – Learn the internals of Joomla!, how the system works, coding conventions, and design patterns
• Mid of May – Complete the design of framework and API, after discussing with mentors
• End of May – Complete the basic development of the API
• Mid of June – Complete the testing and review on the API and Management Forms
• End of June – Complete the development Content creation form integrations
– Mid term evaluations –
• Mid of July – Complete the development of modules for sidebars
• Mid of July – Integration with Exhibit, Cumulus and Opencalais
– Final evaluation — Bug fixing and Documentation

Bio

Open Source Development Experience

I have been in programming since my childhood, and in recent years I am trying to involve in major projects like Drupal, Audacity, Wordpress and Joomla!

I took part in GSOC 2007 with Drupal, for ULINK project [7] and developed ULINK module to generalize filtering and UAUTO to auto complete links through popup with suggestions. The latest releases with demo can be found at [8a] and I released a restructured version [8b] for Drupal 6 in February 2008.

I am a freelancer in open source web development and an Electronic and Telecommunication Engineer, and worked with several client projects [9] in this context.

Work/Internship Experience

I did my internship at Motorola [10] from 2007 September to 2008 April, mainly working on power management units, where I focused on writing applications and components of drivers to automate testing. During my latter part of the internship, I was researching the potential of improvising high contrast cameras from 2D barcode scanners for authenticating users based on palm print recognition. Further, I did fair amount of research on my own and published a paper post internship on using Principle Component Analysis (PCA) to classify multi-dimensional objects in unsupervised manner.

Since my participation in GSOC 2007, I have been involved with many client projects on web development and have worked with Figment SRL, Italy and Research Applications and Financial Tracking (RAFT) INC, USA [11]. The projects I have handled and modules developed are listed at [9].

Academic Experience

I am currently in the final months of my first degree in Electronic and Telecommunication Engineering [12a] at University of Moratuwa [12b] where I am currently ranked first in the faculty with a GPA of 4.15 in 4.2 scale.

I have been admitted to the PhD program at Massachusetts Institute of Technology (MIT) [13a] and will start working at the popular CSAIL [13b] (Where Exhibit and other popular Web2.0 tools were developed) from September 2009. My interest lies in the area of information retrieval and information management and in this regard, this particular project (taxonomy system) hugely attracts me. I will be graduating in 2009 April, and looking forward to spend the gap-time in Open Source Development.

Motivation

I have a strong background with PHP/MYSQL and I have been doing web development (since 2005) and with Drupal (since GSOC 2007) for long time. However, as I have become fully conversant with Drupal I am looking forward to work with similar publishing platforms to expand my horizons, and in this regard, I chose to work with Joomla!

Further, I understand this is a great opportunity for me to start a new road that is with Joomla! to put the knowledge and design skills I gave gained over the past few years, in practice.

References

[1] http://simile.mit.edu/wiki/Exhibit
[2] http://wordpress.org/extend/plugins/wp-cumulus/
[3] http://docs.joomla.org/Summer_of_Code_2009_Project_Ideas
[4] http://en.wikipedia.org/wiki/Faceted_browser
[5] http://groups.csail.mit.edu/haystack
[6] http://www.amazon.com/s/ref=nb_ss_gw?url=search-alias%3Daps&field-keywords=joomla&x=0&y=0
[7] http://drupal.org/project/ulink
[8a]http://project.theebgar.net/drupal-modules/ulink/ulink-52
[8b] http://project.theebgar.net/drupal-modules/ulink/ulink-61
[9] http://project.theebgar.net
[10] http://theebgar.net/all/results/taxonomy:18
[11] http://openbioraft.com
[12a] http://ent.mrt.ac.lk
[13b] http://mrt.ac.lk
[13a] http://mit.edu
[13b] http://www.csail.mit.edu/

Related posts

• Taxonomy Extension RC1 (August 3, 2009)">Taxonomy Extension RC1 (21)
• GSoC 2009 – Taxonomy Library API (June 19, 2009)">GSoC 2009 – Taxonomy Library API (0)
• GSoC 2009 – Taxonomy Architecture (May 11, 2009)">GSoC 2009 – Taxonomy Architecture (1)
• Lets bring technology and professionalism into FYP (July 15, 2009)">Lets bring technology and professionalism into FYP (5)
• GSOC 2007 (February 21, 2008)">GSOC 2007 (0)

Graduate Record Examination abbreviated to GRE is a standardized test that is required by �many grad schools in USA, Singapore, Canada, etc. Let’s put aside the debates on its accuracy, relevance, and resilience, �and keep in mind the final objective – getting into a renowned graduate program.

However, I would like to mention few resentments shared by many GRE�takers. First of all, it is commercially driven, and hence by all means expected to maximise profits. They are more interested in seeing you taking GRE next time in vain, rather than letting you enter the Grad school, where their accuracy and abidance by ethical codes may be called into question. I learn that many second time takers score fairly better than first timers, perhaps not always due to the experience they gained nor extra work they put. Secondly, the scores are now sent electronically and therefore does not cost a peny more in sending it to another instituition. However, they require you to pay 20USD, and also send you a copy of the results sheet everytime you ask them to send a score to an instituition. Thirdly, albeit the introduction of computerized exams they are still in the 21st centuary, and results are available only through snail-mail (20 USD�for each extra request and takes ages to arrive) and through an automated phone call (also charged). A web based implementation, at least a rudimentary setup like what is used in few British Councils (BC in SL has this) for IELTS would be highly beneficial.

Two types of tests are available in two modes.

General Test is required by almost all grad schools (AFAIK�except MIT and Purdue) in USA. Subject tests are recommended espcially if you are applying to a different field from your undegraduate area and thus do not have enough academic records to support your ability.

In Sri Lanka, at the time of writing only paper based tests are available and they are highly recommended �mainly because

• it lets you answer sanely
• it lets you re-check / re-do / correct your answers
• it lets you skip questions and answer them later
• but it is quite inconvenient for analytical writing especially for those who are used to press backspace more than enter key.

Therefore, if it is available, I suggest that you take paper based test.

Plan IS�A�MUST

In one aspect where I must agree that GRE�is preparing one for grad school indeed, is where it necessitates independent work and proactiveness. A plan is indispensable and you would have come with the following when you are ready.

• Why Take?
• Should I�take?
• When to take?
• How to Prepare?
• Long term plan
• Short term plan

After short listing the potential countries to pursue the graduate studies in, make sure that it is very essential for you take GRE. For instance, �for those who are planning to do in Australia and Canada, GRE�is not advisable provided the expenses, pain and time invested. However if your first and/or only choice is USA it is better that you do.

In any case, unless it is required, taking GRE�does not necessarily need to boost your application, and I suggest you spend the valuble time in a research work. We must understand that what the Universities are expecting is not skills in performing exams which few of us mastered and that is not commensurate with the knowledge. They use GRE only as a mean to evaluate you in a common scale, when other convincing entities are absent. Therefore sound research work, recognized publications, international awards and grants and a recommendation from a renowned personal from academia are more compelling and recommended. In�this regard, I chose not to do any subject tests.

For us, non-native speakers, it will require some extra effort in tackling verbal section, and requires longer preparation.��I suggest you start at least an year earlier to prepare, and as paper tests are conducted in October and February / March, it should not be later than January for October exam. I�started it in December, if I remember correctly.

Long term plans

The objective of this to make sure that a month before exam you remember all the words, meanings and similar words by heart.�It is said that, one is expected to learn or come across around 10000 new words in their undergraduate career and therefore can be tested for any of those.�Barrons extensive list comes with 3000+ words and big book with little more. I suggest that you first memorize the Barrons extensive list, and make sure that you master Barrons most frequent words�(about 300) so that you know A-Z on those 300 words. This can be supplemented by Bigbook, Kaplans list, etc.

I found it effective to memorize a list in the morning, and listen to the audio for 2 or 3 times in that day (you can use the commuting time if you have an iPod) followed by a recap that day night. This way it needs only 50 days to complete the extensive list, however I am quite sure when you receach C you would have forgotten A. Therefore it requires perseverance and repeated approach.

However the last month requires a totally different approach that I will detail below

Short term plans

It is essential that you stick with your long term plan and follow it daily. For instance, an outing or a project load might discourage or make it impossible for you to stick with your plans. However, what is more important is that you don’t give up. I�have seen many people start with a great plan one day to waste time next day, to repent the day after and to give up the following day. Therefore missing your schedule one day should not lead to total abandonment. In this regard it is not foolish to make new plans every week or even every day and I call them short term plans.

Month before the test

- TBD

Just before the test

You will have to mark (shade) the choices with a 2B pencil (I opted 4B). So take 2 pencils, both side sharpened (so that you would have decreased the frequency of sharpening by 4 times) and a good eraser. The advantage of using 4B is that minimal pressure is required to leave a thick layer of lead, and the disadvantage would be that you will have to sharpen more often than with 2B.

Take at least 2 sharpeners, 2 erasers, 2/3 – 2B/4Bpencils, water bottle, sweater (even in a tropical country like SL as they keep AC at maximum level – perhaps they wanted to standardize the environmental conditions too – the it was really chilling and I wished I had taken a jacket) and 2 recognized photographic IDs (Passport and NIC).

Exam Hacks

Verbal section (2 papers – 38 questions in each – 30 minutes – each with 2 reading comprehension sections)

• It should be the toughest for a non-native speaker, and the presence of two reading comprehension sections exacerbates this. I suggest that you attempt antonym, analogy and sentence completion parts first and try to finish them within 15 minutes. Try to spend 3 minutes in reading a paragraph and 3 minutes for answering. For the larger paragraph it might take longer and in my case it took 10 minutes, however I was quick enough in other sections.
• This requires you to have words and meanings on fingertips, and know the roots, prefixes and suffixes, so that an unfamiliar word’s meaning can be guessed.

Math section (2 papers – 30 questions in each – 30 minutes – look for tricky questions)

• It must the be easiest for Sri Lankan exam takers as it is just above the standard of a typical GCE�O/L maths paper. Be careful in attending the questions that look very obvious or simple as there might be a trick behind. You could easily score 800 / 800.

Analytical Writing (2 sections – 30 and 45 minutes – prepare your structure, style and use of words earlier)

• I prepared my structure following Barrons and I suggest that you do not. A hackneyed structure and style of expression will badly affect your score and I�strongnly advise that you come up with your own style first, then compare with Barrons, Kaplan and other sources, and refine your’s. It is very imperative that you go to the test with all these prepared beforehand as you would not have enough time to even think through the topic.
• Suggested time spending for issue presentation section (45 mins)
  • 5 minutes – brainstorming – come up with points for and against
  • 3 minutes – filtering points and structuring
  • 5 minutes – write the part of the �introduction – following the style and format you had already planned
  • 20 minutes – write the body – again according to your plan
  • 5 minutes – write the conclusion – do not forget to prepare a generic style for this too
  • 2 minutes – complete the introduction – you would have come up with extra lines that need to be appended to your introduction while writing the body
  • 5 minutes – check for errors in spelling, flow and grammar
• Suggested time spending for argument analysis section (30 mins)
  • 5 minutes – brainstorming – according to the plan come up with points – do not waste time in thinking on extraneous points
  • 2 minutes – filtering points and structuring
  • 5 minutes – write the part of the �introduction – following the style and format you had already planned
  • 10 minutes – write the body – again according to your plan
  • 5 minutes – write the conclusion – do not forget to prepare a generic style for this too

3�minutes – check for errors in spelling, flow and grammar

After the test

You must make sure the test scores are sent to the institutes on time, and therefore it is essential that you submit the request to send scores as soon as you get results and find the them satisfactory.

Also it is said that the scores will be out in 6 weeks from the day of the test, it might take more than 2 months for the mail to arrive. Therefore if you are near the deadline of the application you had better call them and check the scores (which will be available in exactly 6 weeks from the test date – which was Nov 29th in my case).

If you are planning to do IELTS or TOEFL, I suggest that you take IELTS and with the vocabulary and writing skills you would have mastered by then will definitely get you a good band in IELTS.

Resources

• Barrons Extensive word list (attached) (also part of Barrons GRE book)
• Kaplan lists on grouped words (attached)
• Sample Questions (attached)
• Flash cards�(attached)
• Audio recordings on wordlists�(available in eSnips)
• However I have a volume boosted, edited and organized set of recordings, and if you are from SL I might be able to provide you (of course over a dinner that you will have to host:) )
• I have a collection of analytical writing typical answer book (in PDF) which is quite big for me to attach here (first 10 pages attached)

Links

• http://gre.org
• http://en.wikipedia.org/wiki/Graduate_Record_Examination
• http://www.esnips.com/web/sujith-GRE

Photos :�2008 – Events – On GRE day

Related posts

• From small miracle to real miracle (October 13, 2009)">From small miracle to real miracle (15)
• Should learn to bargain and treat everything as a business (July 29, 2009)">Should learn to bargain and treat everything as a business (1)
• Increasing the relevance along with the ontogeny of a community site (July 29, 2009)">Increasing the relevance along with the ontogeny of a community site (1)
• From 6.840 to 754 (November 13, 2009)">From 6.840 to 754 (6)
• On a snowy sunday from Logan International (December 26, 2009)">On a snowy sunday from Logan International (0)

Related posts

• No related posts.

In CETK, as in any other unit tests, we prepare test cases and write functions to utilize the existing function calls and APIs, often repeatedly, and it is uncommon to go to the abusing extent to verify the robustness of the code.

The second task I handled was related random numbers. Along with its next mobile platform architecture MPA 2.0, Motorola is going to introduce battery authentication, where unless the attached battery (also known as gifted batteries) responds correctly it won�t be allowed to power the device. This can prevent the use of non-Motorola authorized batteries in the devices. It is going to be managed through the existing power management unit � power micro � by challenge response sequence.

The challenge will be encrypted through CHAP 1 algorithm implemented in a Texas Instrument chip which also contains the private key. The battery responses to the challenge with a response produced with the public key provided by Motorola. However to increase the security, random number will be used to produce the challenges.

As a physical parameter needs to be used to make a bit sequence real random, we used a pseudo random number generating function built with several shifter s and XOR operands together with a reseed from physical parameters such as Voltage, Current, Charge and Temperature. The next step was the test for randomness, which was implemented with a histogram generator. Initially a string of 64 bit length was used, and an unsigned long integer was sufficient for calculation. As around million samples were used to generate a histogram, to speed up the calculation and reduce the memory consumption, two arrays were used where one for bit string and the other for frequency. Upon the arrival of a bit string, if it is already found in the first array, the index is used as key to the next array and value is incremented. If it is not found, it is added to the array and the new index is also added to the next array and value is set to one.

Secondly I wrote the random number generator in code as per the algorithm implemented in hardware, and produced hundred thousand bit strings and checked its histogram through the code written previously where all were of frequency 1 which showed an excellent randomness. However, as it was decided to use 160 or higher bit length, we had to change the program. I wrote a variable bit length class and implemented operator overloaded functions to allow the class to be used as an integer.

As we reached the end of the training, we started to write our reports for which we started to collect details about the company, history, finance, structure and performance. Although many details were obtained from Motorola public web site, Motorola Corporate site, annual reports and various other publications, the evolution of the Sri Lankan branch was only learnt from a senior colleague and worth sharing.

It was founded in 1999, by Llavan Fernando as MobiNetix PVT ltd, under the aegis of US based Receiptincity.com, a dot com company to offer interactive point-of-sale transaction systems and electronic signature capture devices. Later in 2001, it underwent a name change to @pos.com. The company was then handling both hardware and software aspects of Point of Sales (POS) terminals, where Sri Lankan branch was concentrating on Software field. In 2001, amidst the dot-com bubble, the company�s profit soared and in order to address the scalability it was split into two; Crossvue.com to handle software area and @pos.com for hardware, where Sri Lankan branch naturally came under Crossvue.com. However the recession followed the dot-com bubbled made the management of two different companies difficult and Crossvue.com was absorbed into @pos.com. However, Sri Lankan branch remained as Crossvue PVT ltd. The enormous growth and stunning revenue of the company attracted Symbol INC, a US based technical leader in similar industry, which made an offer in 2004 and acquired @pos.com (including Crossvue PVT ltd in Sri Lanka) for 3.5 million USD. Motorola, Inc offered the acquisition in the beginning of 2006 and completed its acquisition of Symbol Technologies, Inc. for the equity value of approximately $3.9 billion. Sri Lankan branch officially changed its name from Symbol Technologies PVT ltd to Motorola Technologies PVT ltd in February 2008.

Related posts

• Third month at Motorola (February 24, 2008)">Third month at Motorola (0)
• Second month at Motorola (February 20, 2008)">Second month at Motorola (0)
• Fourth month at internship (February 26, 2008)">Fourth month at internship (0)
• First month at Motorola (November 20, 2007)">First month at Motorola (0)

Finger print recognition, the most prominent details extraction method, uses a simple but elegant algorithm. Once the image is acquired, borders are trimmed to get the core picture. Then histogram equalization is performed to increase the perceptional information. Then image is enhanced by taking FFT on a 32 x 32 region, multiplying by the determinant of the resulting matrix and taking Fourier inverse of it. Thus max frequency is enhanced and features are made consistent and any noises are suppressed. Then image is segmented for Region of Interest (ROI), thus proper orientation of the image is found and it is made erect.� Using SOBEL filters gradient along X (cosine) and Y (sine) direction are calculated and gradient of the image (tan) is obtained thereby. Finally image is binaraized along the right direction and ridges and joints are searched with proper filters.The initial part of finger print recognition is adapted for our use, but unlike the finger print, palm print contains multiple locations of interest, lines of various thicknesses and as the image is not taken closer to the object added noises through shining and shades exacerbated the situation.Secondly image details reduction was used and my work primarily consisted of research and development on adoption of PCA for palm print recognition.

Principle Component Analysis (PCA) is theoretically the optimum transform for a given data in least square terms and fundamental of Face recognition.� In pattern recognition it is primarily used to reduce the dimensions in a data set y retaining the characteristics that are prominent. Using Eigenfaces, which in turn used PCA we tried to find an algorithm.

The suggested method we followed was:

1. Acquire the training set of images and calculate the eigenfaces that define face space.
2. To get the match, project the new image onto each of the eigenfaces created before.
3. Determine if the image is a face at all, by checking whether difference from the training set is within a threshold.
4. If it is a face, enforce greater threshold to confirm that it belongs to the same face.

We, before testing its suitability for palm print recognition, tried on face samples where it showed a 0-9000 difference on 5 images from the same face (out of 20 used for training set) and� > 14000 for a different face (controlling set) which means for a threshold of 10000 it is likely to have 25% success rate in identifying the face and 100% success rate in rejecting the wrong one. However, when it was adapted for palm prints results were futile. In addition, I helped a colleague in extracting bit data from a bitmap (bmp) file and outputting it in an array defined in a header file so that any program could use it instantly and effortlessly.� Although initially it was written in Matlab, due to speed limitation, size and portability, it was later re written in C. As the picture could be presented in binary (black / white � 1 bit ) image or intensity (8bit or 16 bit gray scale) image, chosen to use Matlab to write the program as testing and debugging will be a lot easier. Following the BMP standard, file size, data size and data start, height (in pixels), width (in pixel), depth (bits) and compression were read from the header of the file. An odd behavior was noted. A file may contain extra bits at the end of each line, which is not included in width information. Hence to find offsets it was necessary to use (size � start) / height to find the actual length of the information of a row.� Another peculiar feature of BMP file is when height is specified in positive figure (which is the most common) the last row of the image is written first and the first row at the end. i.e. an inverted mirror image is stored. Hence to print the first row, it had to be read from (start + ( height -1) row_length ).

Further, in order to support both the endiness, a bit reversal operation needed, which was implemented in a macro, instead of a �for� loop or function to save the additional over head that would have occurred otherwise.

Related posts

• Fourth month at internship (February 26, 2008)">Fourth month at internship (0)