N excellent and comprehensive review of the iPod nano comes from ars technica. The review is full of clean shots of the nano. To spice things up, the durability of the iPod is put into the test with cars running over, bashes into the ground, etc. I warmheatedly recommend this visually-rich review, which is not vandalistic in nature. The picture shown above is somewhat misleading.

Manufacturer: Apple (product page)
System requirements: Macintosh computer running Mac OS X 10.3.4 or later, USB port; Windows PC running Windows 2000 SP4 or Windows XP SP2, USB port
Price: US$199 (2GB), US$249

Also see: Sony’s ‘iPod’

The display blends with the player’s body

AVE you had a peek at the new Sony Walkman, which aims to compete with Apple’s iPod? Since Apple have introduced the ultra-thin iPod nano and will soon incorporate a mobile phone, Sony appear to have fallen behind already. Their 20 GB player does not look elegant to me. Have a look, however, at how its display blends with the actual body. The white captions are part of the display, which is almost unbelievable and probably unprecedented.

Even the Queen of England might fill out the form

Dear iPod users,

Be sure to read this earth-shaking item and fill the form if necessary.

A San Mateo County judge on Thursday approved the settlement of a class action suit that will offer relief to as many as 1.3 million iPod owners who may have been victim to poor or defective batteries.

• Claim Form

enjoy listening to music, but yet again who doesn’t? There is always catch though — a pitfall if you like: records and files get ‘used up’ after a period of time. Obtaining new music is a time-consuming process whether it involves walking, browsing or loading CD‘s.

Since peer-to-peer (P2P)) downloads are illegal, I have been downloading a lot music from small bands’ Web pages en masse. What I do not like, I can immediately dispose of. Using the command line, I am able to download music in the background without any manual intervention, simply by spidering the Internet.

I have come to realise that I do not know the names of songs that I listen to. I rarely bother to look at the filenames or tags because it is the same time-consuming ‘luxury’ mentioned above. In most cases, I fail to remember the names of bands too. To use an extreme example, a song which I must have listened to 50+ times in past few days has been completely obscure to me up until now. To repeatedly play it, I would typically use the “jump” function followed by the string “be” (for the word “belief” contained in the song’s title). This plays the song merely at the power of will, taking no more than a second to find and play it. It all leaves song titles and artists’ names in the dark unfortunately.

While on this issue of ‘music productivity’, XMMS accelerators, for instance, are used to control the music player without changing application focus. In other words, key combinations involving SHIFT, CTRL and/or ALT work universally and affect the player’s state at any time. You can change tracks, adjust volume etc. while in the midst of writing a sentence. The mechanism which enables this is described (also visually) towards the end of an old item about Music Log Files.

I have reached what I consider to be a Utopian situation. This is analogous to have a radio station with infinite supply of music. I have the ability to skip tracks, repeat tracks and organise the music supply at great ease. Moreover, the music supply can be handled purely by robots (cron jobs), as described in an old item on music mass-downloading.

Yahoo have recently introduced nice features such as news delivery as RSS feeds. They now offer an extensive audio search which encompasses on-line music stores as well as smaller sites and self-promoting bands. It appears to work well because it managed to pick up some long-forgotten files — the stuff I recorded in 2002 or thereabouts. It’s all quite embarrassing, but I am flattered to find my name among the results pages. Raves and insults in the box below….

Spreadsheet representation of music history (click to enlarge)

ome weeks ago I wrote about an XMMS extension that I had devised. This collection of steps which I described could make XMMS store simple log files for music (music ‘history’). That work has now been thoroughly extended to produce spreadsheets (comma-separated values (CSV) to be precise) rather than flat log files.

In brief, what is obtained using this trick are music history logs, one file for each day. Each file contains full, detailed history of the tracks listened to. Interesting information can be derived from these, e.g. most frequently played tracks, tracks quickly skipped (implying the possibility of deletion or removal from playlist). As already said in last entry:

A music log file results in even more redundant data to store. According to a rough calculations, I will see it growing by 1 megabyte every month or so. Compression, however, should make it only 10-20% of its original size.

You can find all the details on this work (installation instructions included) in my Linux Utilities page.

ow does one identify music which has potential of being liked? Music, unlike textbooks, does not contain text or keywords. Its tags are not always valuable either. An interesting paper from Trinity College Dublin describes a method by which music adapts to the preferences of listeners (PDF). However, can this be done purely based on prior data? Data that is provided in advance unlike in real-time? A List of records maybe? Playlists perhaps? We seem to be coming closer to realisation of this idea.

Image similarity measures are one focus point of my research; also sparks to mind is Google’s notion of ‘Similar Pages’. Why not apply similar principles to music? I now collect big daily dumps of music that I listen to (output to files using the following technique ). Bound to each entry is the time when a track started. From this, one can infer which tracks are being skipped. Alternatively, full, raw playlists can be of use and might, in fact, be more manageable as well. By exploiting a large collection of playlists, the nature of the genres can be better understood.

Given all of this data, it can potentially be used for collabortive playlist sharing, somewhat like del.icio.us (see previous reference to del.icio.us with a gentle introduction). Users can then discover other songs they might like based on other people’s playlists. The more data, the more accurate statistics will be. Getting large lumps of input (playlists) is effortless too. Just imagine yourself the scenario:

You can automatically find playlists most similar to yours and recognise the most-played tracks on that playlist. Social software has seen great success recently, so exchange of music preferences and recommendations is probably the way to proceed.