Thursday, December 22, 2011

Going off-line for the Christmas break

I am leaving tomorrow morning for a weeks holiday. Since the roaming charges on my phone are likely to be exorbitant and I won't be bringing my laptop with me I will be totally off-line for the week (definitely a change for me).

I think it is appropriate that the last thing I post to the blog should be my favourite Christmas tune which is Frank Kelly's adaptation of the classic song "The Twelve Days of Christmas". It is impossible not to laugh when hearing this song.

Tuesday, December 20, 2011

It takes power to stay connected

The big advantage of having broadband Internet access (instead of the ISDN access which I had before) is not the speed of the link, but the fact that my home is constantly connected to the Internet. This enables many applications, but it also requires that I keep the networking equipment in my home constantly powered on. I decided to measure the power consumption of the devices in my home that are permanently network connected to see how much power this is consuming.

I get network services from UPC. They supply me with a cable modem that consumes 6.1 watts and a wifi access points that consumes 6 watts or power, so this means that I need 12.1 watts to keep my broadband link active.

However, I also have some equipment at home which is left permanently switched on to use this network connectivity. I have a current cost electricity monitor that tracks my electricity usage at home and a bridge device that uploads the data to a tracking web site. In total this requires only 2.8 watts (it is good that the device is very power efficient).

I also have a Tonido plug  which is a small very small Linux server that can do a number of things. The most important thing from my point of view is that it uploads weather data from the weather station in my back garden to my personal weather web site and it also runs some software to help with my project to determine the accuracy of weather forecasts.  It consumes 5.1 watts.

This leads  to a constant power consumption of 20 watts, which translates to 14.4 Kilowatt-Hours  per month which would cost about €2. I think that €24 per year is a small price to pay for the convenience of being constantly connected to the digital world.

Saturday, December 17, 2011

Getting ready for BTYSTE 2012


The highlight of the holiday season for me is the annual BT Young Scientist exhibition. Irish people will all be very familiar with this long running event which happens at the start of each year, but foreigners are sometimes surprised to hear that a science competition for school children generates such interest and excitement that event receives over 40 thousand visitors and the announcement of the winner is normally the lead item on that night's national TV news.

For the last few years I have had the privilege of helping run the IBM stand at this exhibition. It is always invigorating to meet young people with such obvious passion for learning about science. I always go away from the event with renewed confidence that Ireland will have a bright future with youngsters like this joining our workforce in the future.

IBM will have a significantly larger stand this year and very exciting plans for what we will be demonstrating. The dedicated  blog has just re-opened for business and will carry all of the details.

My personal role on the stand will be in helping demonstrate the capabilities of IBM's Watson technology  which recently made headlines by winning the Jeapordy TV game show against the best human competitors. The video explains a little why IBM believes that the technology developed for a simple game can be adapted to solve real world challenges.

Saturday, December 10, 2011

Energy levels and my morning coffee

I am a big coffee fan. In fact I find it almost impossible to complete my cycle to work each morning unless I have topped up my caffeine levels with at least two mugs of the drug. While the coffee tops up my personal energy levels, I need to use some electricity to make the coffee. I decided it might be interesting to use my new energy monitor to measure exactly how much electricity do I consume to keep me fully caffeinated in the mornings.

I actually have three choices of how to make my morning brew (listed in the order of how much personal energy is required from me):
  1. I could use my old fashioned kettle to boil some water and then make instant coffee. This has the advantage of being really easy to prepare.
  2. I have a traditional filter coffee maker which makes reasonably pleasant coffee. The preparation takes a little more energy from me, but the taste is better.
  3. I recently got a fancy cappuccino making machine. Although there is quite a bit of work involved in using this machine, the result is a lovely tasty cup of coffee and so as a result this is my favoured choice most mornings.
The kettle does not consume any energy when plugged in, but not actually boiling water (I didn't expect that it would, but I checked to be sure). When I turned it on it consumed 2795 watts and took about 1.5 minutes to boil a half full kettle of cold water. This means that if I boiled the kettle once per day it would cost me about €3.50 per year in electricity (in reality I boil the kettle much more often than once per day).

If I use the filter coffee maker it consumes 928 watts of electricity and takes about 6 minutes to brew a medium sized jug of coffee which fills 3 mugs which is more than enough to kick-start my morning. If I used this device to brew my morning coffee every day for a year the electric bill would be just over 6 euro which is quite reasonable when you consider how much I would have to pay in a coffee shop.

However, I was surprised to see that the filter coffee maker consumed 2.2 watts when plugged in and not in use. I guess this must be because it has a clock built in and it is capable of being set up to automatically brew a jug coffee at a predetermined time so that I could have freshly brewed coffee available as soon as I step out of bed. Although this is a really appealing feature when the salesperson explains it in the shop, the reality is that I never use it, mainly because the mornings when I really need an instant cup of freshly brewed coffee are the mornings after the nights when I am least likely to set up the coffee maker correctly. Simply leaving the coffee maker was costing me about €2.70 per year in electricity - not a fortune, but I thought I might as well save that money so in future I will leave it unplugged when not in use.

Next I decided to analyse the electricity usage of the cappuccino maker. This is a more complex device and in fact the reason why I keep the filter coffee maker is because sometimes my personal energy levels are too low to make a cappuccino.

My first surprise was to see that the cappuccino maker consumes 1.6 watts when plugged in and not in use. It is hard to understand why this should be the case, unlike the filter coffee maker with its clock function there is no good explanation for why it should consume stand-by power because the cappuccino maker has no LEDs or other indication that it might be consuming electricity when not in use. I resolved that in future I  also leave it disconnected from the supply saving me a further €2 per year (every euro counts these days).

Next to look at the power consumption when actually in use.  The device has three different functions, each with their own power consumption profiles:
  1. The milk heater consumes about 466 watts when turned on, but the energy draw is not constant. The heating element turns it self on and off while the milk is being heated (presumably in response to a thermostat).
  2. The espresso brewer consumes 828 watts when turned on and it will turn itself off automatically once it senses that the coffee has been brewed (a little green LED comes on to alert me that it has done its work).
  3. The whisk is used to give a froth to the heated milk. It is recommended that I only need to use this for 30 seconds and the power consumed varies between 5 and 8 watts depending upon how deeply the whisk is immersed in the jug of milk. Actually I was surprised at how little power this was consuming because it makes such a loud racket that I would never accidentally leave it switched on - but the device consumes about 50% of the same electricity by simply being plugged in.
Overall the total power consumption to make a large cappuccino is 0.09 Kilowatt-hours. If I made an average of one cappuccino per day (which I do), the electric bill would be under €5 per year which is quite good value when you consider that there are some coffee shops in Dublin which would charge this for a single mug. However, I was surprised to find that I was spending an extra 50% on power simply by leaving it plugged in while idle.

Thursday, December 8, 2011

A lazy person's guide to staying active on several social networks at once

I am a keen adopter of new technologies and as a result I sign up for account on most new social networking related sites. People sometimes as me how I manage to find the time to keep all of these accounts active. However, the truth is I can't keep active on all of these different sites, but with clever management of the feeds between the different services it is possible to give the appearance of being active on all of them without actually logging into most of the sites in question regularly.

For, example the following diagram outlines some of the automated information flows that I have set up between the various social networks on which I am active:

  • While Facebook is most popular among young students, LinkedIn is a social network more popular with older professionals. There is a large overlap between the people registered on both sites, but if you want to maintain links with all of your friends, you really need an account on both services. Since people seem to spend more time keeping their status updated on Facebook or Twitter than LinkedIn, LinkedIn has a feature that you can enable to automatically import status updates from the other platforms. This has the benefit of making it seem like you are paying more attention to your LinkedIn profile than you really are, but you need to be careful because the social norms on LinkedIn are radically different from Facebook and what is seen as a really cool status update on Facebook, might seem highly inappropriate on LinkedIn.
  • I use Foursquare for sharing with my friends details of where in the world I happen to be located at any one time. I like Foursquare because it is most fun to use, but since only a sub-set of my friends are active users of Foursquare I have configured it to automatically share my activity on Facebook and Twitter. Although some of the non-users of FourSquare are puzzled to read the announcements that I have become mayor of some arbitrary place, I get more comments and feedback through these other platforms than I get on FourSquare itself.
  • I track my exercise using the MyTracks application on my Android phone. This is a really handy application which uses the GPS in my phone to keep track of how fast and far I have been running or cycling. The data collected by this application is truly amazing, but it is not really a very social application. So when I want to share my training data with friends I like to use the DailyMile site. This site can automatically share information with both Facebook and Twitter. Luckily a colleague from IBM has developed a really handy Android application called Tracks2Miles which automatically transfers data from MyTracks on to DailyMile (which in turn shares it on Facebook and Twitter - which in turn feeds LinkedIn).
  • Bit.ly is a popular URL shortening service. If you create an account on this service and associate your Twitter and Facebook credentials with your bit.ly account, then you can automatically share a status update to both services at the same time as shortening a long URL.
  • If you wish to interact with social networks inside IBM, security restrictions mean that web based tools can't help. However, BlueTwit sidebar is a Firefox extension that is available inside IBM. It is useful because it allows you to conveniently read and write status updates from several different social networks both inside and outside the IBM firewall from a single UI..It is available as a sidebar whenever you have Firefox open (which is most of the day for a lot of people).
  • In a similar vein, WildFire is an open source extension available for Lotus Notes that allows me to read and write social network updates to a large number of different social networks from a single UI. It is really handy for anyone who has the Lotus Notes client running all day long (most IBM employees fall into his category).
While these automated tools are really handy and save me a lot of effort, they do have one significant drawback. Although they give the impression that I am active on several sites, and alert human can easily detect the fact that I am not really engaged. For example, there is no way for an automated tool to read what someone else has written and leave them a thoughtful reply. Most social software experts advise that you should spend at least twice as much time reading as you do writing (in real-life nobody likes people who turn up at a party and talk incessantly without listening - this is what these tools make you look like online). In fact some people feel so strongly against automated tools like the ones I described here that they refer to them as "anti-social software"

This problem can be partially overcome by setting up email notifications (most social network platforms can be set up to periodically email you a summary of recent activity that might be of interest so that you can respond to some).

A service that I don't use much any more is ping.fm - but if you really want to automate your status updates this is the tool for you. It is capable of connecting a huge variety of social networking sites and it is even extensible so that if support for your favourite network is not yet enabled, you can add it yourself. A really neat feature of this platform is what they call vocabulary expansion, this allows you to put special tags into your status updates and have the system intelligently replace these tags with something different when propagating your status update each of the target platforms. However, this feature never really caught on since most social networkers are more interested in vocabulary contractions than expansion.

Tuesday, December 6, 2011

Is working from home a good idea?

One of the benefits of modern network technology is supposed to be the fact that "location doesn't matter" and "you can work from anywhere". While it is true that remote working is now very feasible for many occupations, I think that where you choose to do your work has a very significant effect upon your productivity.

I was recently at a conference and I noticed that during the lunch break many delegates chose to use the free time to catch up on the work they were missing by reading emails, checking voicemail, returning calls etc.. They were all a long way from their normal place of work so it is good that they could get some work done. However, they did not simply choose any location in which to do their work, instead there was a frantic search for suitable locations i.e. a quiet alcove where they had some peace and quiet as well as a place to sit and maybe even a place to rest their laptop.

What this means is that where you are located on a global scale doesn't matter. For example, you can easily do most jobs from New York City, but it would not be a good idea to base yourself in the middle of Times Square if your job requires some peace and quiet. Likewise you could probably do most jobs from a location in a remote wilderness location in African so long as you had power and in Internet connection, but you would probably need to move to a shady location ensure there was not too much glare on the screen.

IBM is quite liberal in terms of allowing employees to work from home if they want. In general the consensus seems to be that senior people can work productively even when remote from their colleagues, but junior employees benefit significantly from working in a team where they can learn from more experienced engineers. In some IBM labs in the USA, there are so many people working from home that people are beginning to complain that there is little point in being in the office since there is nobody else there to interact with, and the company has launched a "back to the lab" initiative to counteract the problem (which I guess is similar to the problem of city centres becoming empty shells when all businesses move to malls in the suburbs).

The factors influencing your decision about whether to work from home or not would include:
  • How far your home is from your normal workplace? I am luck enough to need only 20 minutes to cycle to work each morning, but many people live over 100km from work so they naturally don't want to make that journey if they can avoid it.
  • What is your home environment like? Some people are lucky enough to have a well furnished office space at home, but others might live in cramped accommodation shared with other people and hence working from home might not be feasible for them.
I am the only person from the team I am currently working with who is based in Ireland, so all of our team meetings are virtual meetings. However, I still find it useful to go into the office most days, becuase I know if I spent too long at home I would begin to suffer from severe cabin fever. It is great that companies allow people to work from home, but this does not mean that everyone could/should work from home on a regular basis.

Sunday, December 4, 2011

Does the clock on my microwave consume more power than the heating element??

According to this article in the Economist "a typical microwave oven consumes more electricity powering its digital clock than it does heating food". This seems very surprising to me so I decided to test if it is really true by using the individual appliance monitor that I bought from Current Cost Ltd. to monitor the electrical power consumed by my own microwave oven at home.

When not cooking food, my microwave consumes 2 watts of power to power the LED clock. This is actually higher than the 1 watt estimate used in the article and would represent a daily usage of about 0.05 Kilowatt-hours. When actively cooking food it uses about 430 watts, I was surprised how low this was since the microwave was sold as an 800 watt unit and seems to be capable of heating food as fast as expected by receipts written for an 800 watt model.

The total daily power consumed by the microwave naturally depends upon how much it is used. However, my figures were that the normal usage amounted to about 0.15-0.20 Kilowatt-hours per day. This means that for my microwave, the allegation in the article is not true. However, it is surprising to me that the LED clock is consuming 25-30% of the power of the heating element. Because of this information, I would like to plug out my microwave while not in use. Unfortunately the socket where the microwave gets plugged in is buried behind a press and it is not very easy to get access to so I will have to do some re-wiring before this is feasible.

Friday, December 2, 2011

Phone chargers are not all the same


When I initially tested the electricity consumption of my phone charger when plugged in but not actively charging, I was pleasantly surprised to see that no electricity was consumed. Since I was running that test with the current cost meter which could only measure to an accuracy of 1 watt, I decided to repeat the test with my new more accurate meter and also I compared three different phone chargers. The three chargers I tested are shown in the picture on the right.
  • On the far right is the charger which came with my phone (a Samsung Galaxy II).
  • In the middle is a charger that came with my first ever Android phone. I bought the phone from a clone maker in China via eBay because Android phones were not offically available in Ireland. The phone was truly dreadful, but the charger was really cute looking so I hung on to it.
  • On the left is a charger which was part of a multi-device charger set I bought in Lidl.
The results of my testing were that the chargers on the left and right truly consumed no power when not actively charging the phone. At least they consumed less than 0.05 watts since the measuring device reported the power consumption rounded to 0.1 watts. Unfortunately the charger in the middle (the cute looking one) consumed about 0.5 watts when plugged in but not charging. This is not going to cost much money (roughly 5 cent per month if left plugged in permanently) but since I have better charges available, perhaps I should consign this charger to its most appropriate use as an ornament.

All of the tests were done with the same phone. Obviously it has logic to stop drawing power once the battery is fully charged, I can't be sure that other phones would have the same feature. Therefore as the saying goes "your mileage may vary".