Getting Eclipse 3.4 working on Ubuntu

Lots of people might be planning to develop Sametime plug-ins for Hackday. While the Sametime plug-in development environment is really easy to use once you get it set-up it can be tricky to get your environment configured. I always tell people that they must read the instructions in the Sametime SDK very carefully and follow the instructions exactly.

One thing that is slightly annoying is the fact that the Lotus Expeditor Toolkit which is needed to configure your Sametime launch configuration can only be installed with Eclipse 3.4. This is annoying to users of any platform who may prefer to use a more recent version, but it is especially annoying to users of Ubuntu and other variants of Linux because there is a well known bug in this version which can stop Eclipse from launching.

Fortunately there is a simple work around. You just need to edit the eclipse.ini file (which will be ins the same folder where the eclipse executable is installed) and add this line

-Dorg.eclipse.swt.browser.XULRunnerPath=/usr/lib/xulrunner 

One you have made this change, eclipse will launch properly and there will be nothing to stop you from doing the coolest Hackday project ever :-)

Hackday Project results in Patent

Hackday 9 is taking place tomorrow and due to a training course, I won't be able to take part for the first time in several years. Therefore it was ironic that I just got notification of a patent that was just issued based upon a project I did back for Hackday 6.

The patent is entitled "System and method for client-based instant message monitoring for off-line users" and it deals with an innovative way that I devised to get around the fact that I was forced to implement my Hackday project on the client side due to the fact that I didn't have access to change the server. If you want to find out more, you can access the patent via Google Patents.

Good luck to everyone taking part this year. I hope you succeed in generating several more patents.

My Hackday mini-Project

Due to schedule conflicts, I won't able to take part in Hackday 9 which is due to happen on Friday. However, I felt it would be setting a bad precedent if I completely ignored the event. Hence I decided to tackle a mini-project the last Sunday at home.

I have been recently working with some local primary schools helping them to do projects relating to Smarter Energy for display at the BT Young Scientist Exhibition next January. In order to help introduce the concept I was decided to build a simple test rig with several different types of light bulbs that the students can use to measure the efficiency of the bulbs. This was not a technically very challenging job, but it did involve a few hours of hard work stripping wires and screwing in light fixtures. On the right you can see my completed handy-work.

I have also purchased a simple plug-in power meter so that the students can easily measure the power consumed as each of the lights is switched on. Of course the power consumed is only telling part of the story - I also needed to measure the light output of the various bulbs because as well as having a huge variation in power consumption the bulbs also produce radically different amounts of light.

Rather than purchase a dedicated light meter I decided to see if I could find an application for my phone which was capable of taking the measurements. After a bit of experimentation I settled on using the Light Meter Application from Borce Trajowski. As well as producing reliable measurements of light intensity, this free application also has a really attractive retro-style display. Of course it is only fair to mention that I also tried a different application called Light Meter as well as ones called beeCamLightMeter and  Lux Meter which didn't seem to work reliably at all (which just goes to show that you can't totally trust the application descriptions in the Android market).

I plan to use this test rig in some upcoming visits to schools by myself, but of course I am happy to lend it to anyone else who would like to use it for similar events.

Where should electric car charging points be located

eCar Charger Coolmine

Apparently electric cars are set to become much more common in Ireland in the near future.  However, before this can happen there will need to be a network of charging points for electric cars located at convenient locations throughout the country. The ESB is currently rolling out a series of such charging points.

Obviously such charging points are going to be located at Motorway service stations to facilitate motorists who are trying to undertake a long journey across the country. I was also interested to see that one has been installed in the "Park and Ride" facility attached to Coolmine rail station (near where I live). The idea must be that they expect that local commuters will use their eCars for the short trip to get to the rail station before taking the train.

This seems like a sensible idea, however the only problem is that the charging point is located in the area reserved for disabled drivers. Do the ESB expect that eCars will only be driven by disabled drivers or do they plan to give eCar drivers special permission to use the disabled parking spots?

Hackday is coming again

As I have mentioned on my blog previously, Hackday is a wonderful event which is held at IBM each year. The general idea is to see what cool "hacks" employees can build in a single day. These events are a wonderful way to encourage innovation and also build morale because employees get to work on a fun project of their own choosing. I have been involved in the Hackday event since shortly after it started and I entered every Hackday since Hackday 2. As a result I once was awared the title "The Oldest Hacker in Town" in recognition of my long term commitment to the contest.

This years event "Hackday 9" is scheduled to take place next Friday and the newly redesiged contest site is now up and running for employees to register their projects (unfortunately the site is not accessible outside IBM). Although there is a local event planned for the Dublin lab I don't think I will be able to take part myself this year due to the fact that I will be on a course. However, you can rest assured that I will definitely be thinking of all the IBM hackers and I will also be helping out with the judging process later.

Variable rate mortgages are fundamentally unjust

The vast majority of mortgages offered by Irish banks are variable rate mortgages, so most consumers have no option but to choose one. However, I have always struggled to see how this system could be considered fair. Consumers have to choose a lender based upon what rate they are currently charging, but since the rate will change several times over the life of the mortgage there is no guarantee that the lender they choose will be cheapest overall.

The way this system works is that the banks constantly alter the rate of interest that they charge mortgage holders based upon the rate that the bank have to pay on the inter-bank markets to borrow money (plus a profit margin of course). It might seem at first glance that this is fair, but surely the bank only need to source money on the inter-bank markets when they give out new mortgages rather than constantly through the life of a mortgage. If most mortgages are for a term of 20 years or more, this means that on average mortgage holders took out their mortgages about 10 years ago and hence the inter-bank rates charged 10 years ago are more relevant than the inter-bank rates now.

It could be argued that consumers are free to choose which bank they want to get a mortgage from, but since consumers need to pay considerable administrative fees to switch from one mortgage provider to another, they don't really have a choice to switch from one bank to another in the middle of the term of the loan. Some banks exploit this fact by offering new customers a special introductory offer for the first year, which means that they are using their existing customers (who have no choice) to subsidise potential new customers (who do have a choice). Surely this is grossly unfair and should not be allowed!

A few Irish banks offer fixed rate mortgages where the interest rate is fixed for a limited term (e.g. 3 or 5 years), but since most consumers want a mortgage over 20 years or more this is of limited value to consumers. I wonder why no Irish bank offers customers the option to lock the interest rate for the entire life of the mortgage - banks in other countries offer this. 

How much energy does it take to watch television

The Individual Appliance Monitor (IAM) I ordered from Current Costs arrived yesterday. This is a device that I can insert between the plug of an appliance and the wall socket that will then send data to my home electricity monitoring system about how much electricity is being consumed by the device(s) plugged into this socket. This will enable me to analyse the power consumption of individual devices more accurately. The first thing I decided to look at was the amount of energy being consumed by the television in my living room.

I have often heard people say that it is possible to save a significant amount of power by completely switching off your television rather than just leave it on stand-by where it can be switched back on again with the remote. I have always been sceptical of this information and the measurement from the IAM indicate that it is not drawing any power while in stand-by mode (or at least the power being consumed was too small to register and the IAM is capable or registering a single watt of power). Therefore, I conclude that it is not worth my while to unplug the TV from the wall every night.

When I turn on the TV, but leave the cable decoder box switched off the TV begins to consume about 100 watts of power (it is showing a blank screen). When I switch on the cable decoder box and begin watching something on the screen, the amount of power being consumed by the TV increases to about 200 watts. Therefore leaving my TV permanently switched on would cost about 20 euro per month, which would be a significant amount of money to waste.

I get television service from UPC and the package I have signed-up for includes the ability to automatically record all episodes of a program. Naturally for this feature to work I need to leave the decoder box constantly connected to power and in stand-by mode. When the box is completely in stand-by mode it consumes about 8 watts of power, but once it is either switched on or starts to record a program from my preferences its power consumption raises to about 20 watts. Therefore the TV recording service is effectively costing me between 1 and 2 euro per month in power in addition to the subscription costs from my cable provider. I think this is quite reasonable.

Circles, Friends and Followers

There has been a lot of interest in the recent launch of Google+ social networking service. One of the features that has gained most attention is the fact that  Google+ use the concept of circles to track the relationships between people as distinct from how Facebook uses Friends and Twitter uses Followers. However, many people are confused about how Google+ uses circles so I thought it might be helpful to write a brief explanation.

Facebook was originally based upon a simple model of symmetric friend links, this means that for if Tom is Jane's friend then Jane must also be Tom's friend. It is for this reason that when you add a new friend on Facebook the other person must confirm that they also wish to add you as a friend. This simple model of friendship works well for the friendships that we establish in school (and hence it is a good match for Facebook which originally was most used within a school setting). However, real world relationships are much more complex and hence mode advanced models of social relationships are sometimes needed.

Not all friendships are the same. For example, most people would have different criterion for what they would be comfortable sharing with their work colleagues and what they would like to share with their close friends and family members. For this reason, Facebook introduced a feature called 'lists' where you can divide your friends into different catergories. When you share something on Facebook you can choose whether you want to share it with all of your friends or else restrict it to only the members of a particular list. Since this feature was not originally in Facebook when it launched many users are not even aware of its existence and hence they tend to share everything with all of their friends (and this is the default anyway).

The other weakness in the Facebook friendship model is the fact that it is not necessary true that the people whose updates you would like to read are the same as the people whom you are happy to allow read what you post to the site (Facebook uses your friend list for these two different purposes). Systems like Twitter recognise this difference by implementing asymmetric links that they call 'followers'. By default all updates sent to Twitter can be read by anyone. If I add someone to my 'follow' list it simply means that I would like their updates to appear in my default stream and does not imply that there is any relationship between us. For example, I follow the Dalai Lama on Twitter because I am interested in reading the profound thoughts that His Holiness posts to twitter, but I have no reason to think that His Holiness is interested in reading the boring stuff that I post. For this reason the Dalai Lama has over 2.5 million followers on Twitter, although he doesn't follow anyone.

Google circles are asymmetric links (like Twitter) because you can add anyone you want to your circles and there is no need for them to approve this and/or add you to one of their circles. However, Google circles are not like Twitter follow lists because circles are used to control who can read your updates as well as whose updates are presented to you in your default stream. Everytime you post something to Google+ you are asked to specify which of your circle's members should be allowed to read the update (there is a special group called 'public' that you can share with if you don't care who reads this update).

By default Google+ shows you all of the updates you are allowed to see from any of the people in any of your groups, but you can also click on a circle name to only show updates from people who are in that circle. Therefore in order for you to see a particular update it will be necessary that you have added the person who posted the update to one of your circles and it will also be necessary for you to be a member of one of the circles that the person posting the update chose to share it with. Hence the logo shows intersecting circles.

My trip of a life time

I have been quite actively blogging in the past while, but for the next 3 weeks I will probably not be active on the internet very much because I am going to be on holiday in South Africa.

I am very excited because this will be my first trip south of the equator and I expect that it will be a very different country, especially during the Safari we have planned. In addition I will be meeting up with an old friend from school and college whom I had lost touch with until recently.

I am sure I will have lots of lovely pictures to share when I come back because my sister has given me the loan of her fancy DSLR camera to ensure we capture all of the adventures.

What is happening in my house when I am not there?

My ENVIr energy monitor system from Current Cost is constantly monitoring my electric usage and posting these statistics to the CurrentCost web site where I can view lots of interesting graphs.

I was recently away from the house for a few days, so it was interesting to see how much electricity was being used when there was nobody in the house. As you can see from the screenshot below, their is constant electricity usage rate of roughly 170 watts (probably a sign of the large number of electronic devices on stand-by).

I wondered what this background electricity usage is costing me. For simplicity sake I assumed 30 days in a month, so 170 watts would equate to 122.4 Kilowatt-hours (170x24X30/1000).  If I assume that my electricity rate is around €0.14 per Kilowatt-hour this means that I am using about €17 worth of electricity before I even switch anything on. This is not a huge amount of money, but it world be worth reducing if possible.

Every 2 hours or so the electricity jumps up by 100 watts or so for a period of about 10-15 minutes before dropping back again. I am not certain what causes that, but I guess it might be the fridge/freezer which needs to turn on its motor occasionally to keep the food cold.Again this might be worth investigating further, but I will have to wait until my individual appliance monitors arrive..