A gem of a track!

Was over on Soundcloud today looking for some new music to listen to. I sometimes hop over to Soundcloud to listen to remixes and things that aren’t available on Spotify. Came across this track, this it is possibly the most interesting track I’ve heard all year!

Summer Draws To A Close

3 items of big news (they always come in 3’s!):

I have secured a graduate job at Jaguar Land Rover after another awesome summer of undergraduate placement – really enjoyed myself and looking forward to coming back! :D For anyone else who might be interested in undergraduate placements (summer and year long) or graduate scheme, check out this page:

http://www.jaguarlandrovercareers.com/jlr-roles/graduate/

– My first ever research paper, “Optimisation of the material properties of indium tin oxide layers for use in organic photovoltaics”, written as a summary of my 3rd year dissertation (along with some help from J. Kettle and N. Bristow) has been published in the Journal of Applied Physics. You can see it (for free) here:

http://scitation.aip.org/content/aip/journal/jap/116/10/10.1063/1.4895552

– Last weekend I enjoyed a wonderful trip to Spotify HQ in Stockholm to meet the team (again) and meet some of the big names in the company. Was a superb long weekend, wish it could have been longer. A few of my favourite snaps are below! :D

Rock Star Team!

WP_20140830_11_30_42_Pro

 

Heading back to Bangor tomorrow, really looking forward to getting stuck into my final year!

 

Killing time on the boat

Just got around to uploading this (since I couldn’t get a reliable connection on the boat to do it then!)….

 

Well, since the Internet on the boat is being incredibly slow (guess that is what happens when 1000+ people try and share a 1Mbit sat. connection) I guess it is good time to blog – once again my absence from my own blog is shocking as it constantly falls to the bottom of my ever growing daily to-do list.

 

First thing I have to mention is Chip and PIN. Sitting on the boat with a table to myself (so I can type in privacy) at the end of the counter to one of the on board coffee shops, with a cup of rubbish machine made coffee and a great magazine (E&T although can’t remember which month and someone put the cover photo over the issue details – excellent design…) watching the poor single staffed barista struggle to cope with a huge queue of American tourists. The issue is, none of them have cash (and this boat takes Euros and Sterling, so not sure where the hell they have been not to have any of both currencies used on the island of Ireland), not an issue normally as us in the civilised world have the amazing (and secure) invention known as Chip and PIN, the Amercian’s aren’t quite so up to date. So bemused to ever seeing a card that doesn’t have a chip, the barista couldn’t even find a pen for the American’s to sign the receipts to had to run off to find one. You would think, one of the worlds richest nations, in the spotlight lately (and in the past) for high rates of credit card fraud and payment system hacking would follow the rest of the developed world. With mainstream banks in the UK and Europe already moving rapidly in the roll out of contactless and mobile payments, it does make me wonder how far down the line those technologies lie for USA consumers.

 

The other thing I have noticed courtesy of the American’s is their desire for (and our lack of) iced drinks. I’ve heard several requests for iced tea and coffee in my time sitting here, and I sampled both several times on my latest trip to the USA and I really like them but it is hard to find here. Unfortunately for them, they aren’t sold on the boat because obviously you can’t easily get it from a machine.

 

In the interest of furthering my research into how to get treated better as a non-business class customer (on this boat “Premium”) which started out with my travels on British Airways I have come up with a handy tip discovered by accident. Having two phones sitting on your table appears to portray the image you are a “business” person (although got to be said as I sit here in my jeans and Vans shoes I don’t look the part) does change the behaviour of staff and those around you.

 

Tested out in two lounges on this boat (on different decks, so completely different staff), in the first lounge having one phone out means you are treated just like everyone else, staff seemingly ignore your existence unless you go to the counter or flag them over and people sit at the tables all around you. In the second lounge, with two phones (and a laptop, which I am typing this on in both lounges) you find that people see a different image. Staff are more willing to make you feel welcome, come over to the table more often to clean up or have a chat and people tend to not want to sit at the table beside you. In actual fact my second phone, which is a real phone, has no SIM card in it and is purely being used as a music feed for my Bluetooth headphones. Seemingly I feel (although haven’t tested) if you look busy (like me now typing this blog post) people are also more likely to stay at a difference as to not disturb you.

 

As a young person, I very much enjoy travelling, but as an Engineer I can’t help but feel we are going backwards rather than forwards. Forgetting about big changes in travel, the end of Concorde springs to mind, travelling (at least by methods I use most often) takes longer than in my childhood. The boats between Dublin and Dun Laoghaire to Holyhead are ones that I have been using almost all of my life to go between Ireland and the mainland but since I started doing those journey’s as a child it now takes almost 2 hours longer than it used to. Seemingly blamed on the rising cost of oil, this boat now takes almost 4 hours (by the time you count loading and unloading) to do 100 miles. Even the “fast” craft now takes 3 hours instead of under 2. Surely there has to be a better way. My 4AM blast down the motorway from Belfast to Dublin reminded me of what great benefit Engineering projects can have, the cross-border motorway project which finally connected the two capitol cities with an end to end motorway slashed journey times compared to during the 90’s, but I fear travel outside of cities, were most of the attention is now with regards to emissions and traffic, lacks significant innovation.

An interesting by-product of my latest project.

Since I already have some data retrieved from the Spotify Metadata API relating to my playlists, I thought I would have a play about with it. I’ve mainly been collecting release years from the albums of the tracks that appear on a given playlist. So it’s interesting to plot the years as a percentage of tracks in a playlist.

For instance, take my New Music Playlist:

Running this data we get:

New Music by Year

As you can see, a large percentage of new music I am discovering (which is what that playlist is used for) comes from mostly things released over the past few years.

So what happens if we say compare this to a playlist of a particular decade, lets go for Spotify’s 80’s All Gold: Sophisticated Pop playlist:

Running those tracks we get:

80s Playlist by Year

Graph not what you where expecting? Me either! The issue? Oh yes it is known. It has been an ongoing debate on Spotify for years, it is the original release date debate.

 

And this in itself presents an issue, if all of the tracks in the above playlist (and I haven’t checked) where actually released in the 1980’s, then this “noise” in the results is going to cause issues when attempting to train a classifier to try and estimate someone’s age. However that being said, if everyone see’s the same amount of noise, then that becomes easier but I suspect from being on the community for so long some people go hunting for the original recordings rather than re-releases.

For example, lets take this user made Ultimate 90s Playlist:

What do you notice compared to the Spotify 80’s playlist above?

90s Playlist by Year

 

UPDATE: Shortly after writing this, Spotify have announced they are making changes to release dates and have already started the transition!

New Project – Using Pattern Recognition is it possible to accurately predict someone’s age by the music they listen to on Spotify?

I’ve missed doing some pattern recognition programming this semester after doing so much last year, so I thought I would give it a go myself. The question I have posed (to myself, blogging in case anyone else is interested) – Is it possible to predict someone’s age based on the music they listen to? Or more accurately in this case have in their favourite playlist? I plan to find out!

The plan:

– Write Matlab code to extract release date of albums from a Spotify track URI which can easily be copied in bulk from a playlist. Done

– Collect a testing data set from the Spotify community. Main issue might be getting enough diversity of ages to submit data for me to use, but I will hopefully be able to bug my friends over in Spotify to lend a hand.

– Select a suitable classifier and train on the test data which will be based on the release years of albums people have on their playlists.

– See if my trained classifier can classy ages of unseen people’s data correctly and how accurately.

Will keep you updated!

Exam Results are In…

4 exams, averaged 92% (pending exam board finalization) – not too shabby! Let’s hope I can keep that up for the next year and a half! Also moved one step closer to completing Bangor Employability Award, 28 more points to go – should be able to polish that off by the end of the year with some workshops!

Everything has been pretty standard lately, some truly mad weather yesterday but gave everyone a good excuse to have a lazy day and in fact I got quite a lot of Spotify Community and University work done, accompanied by a ridiculous amount of tea so it was a pleasant day!

This weeks top new music discovery comes from my friend Jack, excellent band thanks for the recommendation!:

In other news, small number of website tweaks! Added my Spotify Follow button over there (—->) so be sure to Follow Me! Also a cool article about my on the Spotify Blog about my amazing trip to Stockholm during summer to meet the team face to face!

Dissertation – What its all about!

I mentioned previously the title of my dissertation this year, thought I would take the time to write a quick summary of what it is all about here.

Organic Photovoltaic Solar Cells (abbreviated to OPVs) are solar cells where the active layer (the layer that turns light energy into electricity) is made of an organic material, rather than Silicon. OPVs currently suffer from having much lower efficiencies than traditional silicon solar cells, so my project was set up to attempt to optimise OPV efficiency.

All solar cells have a transparent electrode layer which must be transparent to light from the sun but also highly conductive – two characteristics which are normally mutually exclusive in common materials. However, there is a set of materials known as Transparent Conducting Oxides (TCOs) exhibit both of the required properties. A commonly used TCO in OPVs is Indium Tin Oxide (ITO) which is the main focus of my project.

This research simulates an OPV solar cell using a Thin Film Multiple Reflection model, and by varying all of the parameters of the ITO layer we can calculate and plot the efficiency of the cell. The simulation varies all of the main parameters of the ITO layer giving millions of possible combinations (which is made possible thanks to HPC Wales donating some Super Computer time to us) in order to find the optimum parameters for the investigated layer.

It is interesting stuff and it’s an area of electronics that has been investigated previously by others, but as far as I am aware the overall optimum model has never been calculated before.

I am currently midway through building the final simulation, but already it is possible to see a large range of device efficiencies emerging and this alone may lead to small improvements in OPV solar cell efficiency in the future.

Image-Eff