Categories
Radio & Music Technology

How to fill a 5Gb iPod…

So here’s what the thing would be if I were really bored and absurdly anal on a Sunday early-evening… As an iPod early-adopter, I may be the last person left in the Western hemisphere with a mere 5Gb to fill. And I’ve ripped nearly 12Gb of my own albums onto my computer. So I clearly have a problem here – how do I choose what to put onto it? Of the 2822 songs currently on my Mac, only 535 songs have yet to be rated (star ratings out of five – rating songs is the new ‘alphabetising my record collection’), and these include several new albums I’ve bought recently along with some older stuff that I don’t really want to write off just yet, but at the same time can’t really say that I’ve demonstrated much inclination to listen to either…

There are several considerations – one being the highly frustrating way that the iTunes randomise function only randomises the first time you make a playlist – so you can’t have one list that changes dynamically. This is highly frustrating.

Quickly – some dumb figures: 5 star songs: 281, 4 star songs: 991, 3 star songs: 691, 2 star songs: 286, 1 star songs: 38. In principle the scoring mechanism that has evolved works like this – I would not be averse to listening to at any time any song that gets three to five stars. Two star songs are an irritation. One star songs are just irredeemably awful. One star songs are mostly from Moby’s “18” or represent the most self-indulgent of The Magnetic Fields’ oeuvre… Other considerations to bear in mind – classical music is interesting, but not always ideal for listening to on the bus or down the street, I add new music all the time and wish it to be represented on my iPod before I have had the chance to assign it a rating, humourous songs often become less humourous after a fairly short amount of time. There are many many other criteria in play here as well – so many in fact that ideally I would need someone to do some comprehensive analysis on the subject and return to me with a set of criteria that one could use as a basis for evolving appropriate smart playlists. Here’s my attempt so far…

  • 5-star songs:
    This smart playlist operates according to the following principles:
    My rating is 5-stars, Genre is not Educational or Classical, limited to 1210 Mb selected by random.
    Significantly, this represents essentially every single 5-star song I own except for a batch of scientific songs by Tom Glazer that may be empiracally wonderful but can become irritating, and classical pieces which are often simply too long to warrant inclusion.

  • 4-star songs:
    This smart playlist operates according to the following principles:
    My rating is 4-stars, Genre is not Educational, Classical or Humour, limited to 650 songs selected by random.
    This represents 2.69Gb of the space on my iPod, except that since several of my smart playlists randomly select some of the same tracks, there is no particular benefit in specifying exactly how much space it should occupy. This is a fairly savagely cut selection from the 991 songs that I’ve given a full four stars too. There simply isn’t space to do otherwise. Note the addition of humour to the banned genre… Five star comedy may remain entertaining… Four star may not…

  • 4-star songs (running):
    This smart playlist operates according to the following principles:
    My rating is 4-stars, Genre is not Educational, Classical or Humour, limited to 200 songs selected by date added.
    Now we have to add an element of movement into the organisation of the iPod. Since we can only have a selection of the four-star songs that are available, a choice has to be made as to which ones will be chosen. A purely random selection that isn’t dynamically different each time would relegate some songs to obscurity, and more importantly would remain uniformly distributed – there would be no sense of progress… By choosing the 200 most recently added 4-star songs, I make sure that my iPod changes and responds to my daily whim…

  • Most Played:
    This smart playlist operates according to the following principles:
    My rating is better than 2 stars, limited to 300 songs selected by most played.
    This list has precisely the opposite function of the one above – to be a stable chart of songs that I seem to not want to be without. The removal of anything with a rating under 3 stars makes sure I don’t accidentally get stuck with something I’ve come to hate.

  • Recent Additions:
    This smart playlist operates according to the following principles:
    My rating is not 1 stars / 2 stars / or 3 stars, limited to 60 songs selected by recently added.
    Essentially this list is simply to catch stuff that I’ve added but have not as yet come to a conclusion about. The removal of anything with one, two or three stars means that I don’t waste time listening to stuff that I have come to a conclusion about, while anything that has yet to be rated remains included.

  • Recently Played:
    This smart playlist operates according to the following principles:
    My rating is not 1 or 2 stars / or 3 stars, limited to 40 Mb, selected by last played.
    Because it’s possible that I could have a song in my head when I leave my flat, and for it not to be on my iPod, I instituted a list that kept track of what had been listened to in roughly the last forty minutes and made sure that was always with me. A tiny additional list that has been a complete life-saver many times…

This feels a bit Open University-esque. It’s an almost total waste of weblog space as well. Except that I am serious – if someone out there has done some research into these things I’m sure that a company like Apple would be interested… A default ‘my music collection’ playlist that intuited what kind of things you actually wanted to carry around with you (with whatever obscure algorithm it used) would probably be quite appealing to some people…

Categories
Technology

Domain names & the "Google Effect"

Dan Gilmour has recently argued that the “Google effect” – ie. the fact that Google and other search engines are now so good that they can locate extremely accurately what someone is searching for – will reduce the demand for new domain names.

“The most interesting from a domain-name point of view is this: With the rise of search tools that unerringly bring you to the page you want, the need for a highly specific domain name — one that a casual Web user would be able to guess — has practically disappeared.”

I should start off by saying that this statement is of course true. But it’s not the same as saying that the use of search engines like Google will radically decrease demand for domain names. Because what he’s concentrating on is an individual’s attempts to find for the first time a specific type of website or online service by guessing the name. There are many other types of searches, and many other reasons for a domain names existence. But I’m getting ahead of myself…

Here, in a nutshell, are my five reasons why the Google effect will not seriously compromise the take up of domain names:

Boring reason 1: Repeat traffic. If I am an e-commerce based retailer then it does me no good if people only come to my site for a very specific purpose (one product) and then depart never to return, unless of course that’s the only product I sell. Because relationships between shoppers and shopping sites are developed over time – the shopper has to trust the company to sell to them. And that means that they have to visit a number of times – each time with a slightly different agenda. Other sites also rely on repeat traffic – even personal site like weblogs operate on the principle that an individual will become involved in reading the content, and return regularly. This actually goes right down to the point where it can be recognised that to a large number of people – finding a site like a weblog through a search engine probably isn’t going to be useful to either party. The weblog is unlikely to answer their query like a pure information site would – and is designed for a completely different approach to reading. Memorable site names aid repeat traffic.

Boring Reason 2: Authority. Do you believe what you read on the internet? All of it? Without question? If so then you are incredibly foolish – because for every light-hearted spoof there is a hate-mongers site or a rampant charlatan. If I search for a page about cancer treatments, I want to go to a site that has a name that I trust – my first point of call wouldn’t be to a page on GeoCities. The (right) domain name is like a badge, which affiliates you with the institution that you purport to be from much successfully than merely saying that you’re from the FBI or MacDonalds. All sites which have a function that requires a relationship of trust between the parties or which needs to come from a trusted source is more likely to achieve that function with a domain name than without one. A domain name suggests you are who you say you are.

Boring Reason 3: Advertising and branding. Because repeat visitors and a relationship of trust are so important to most serious sites, advertising and branding has become tremendously important – from the top (Amazon) to the bottom plasticbag.org). Domain names make this process significantly easier.

Boring Reason 4: How people actually search. The previous three sections have described reasons that domain names remain important. This section goes the other direction, and suggests that while they are important, they have never really been important in the way that Gilmore describes. There are lots of types of searches that people make on the internet. Most of these searches do not fit into the two narrow categories that he elucidates. In fact most of them have never had any relationship to domain names. If you are attempting to ask a question by reference to the internet – you are unlikely to end up at the front page of a site. Once you get past people searching for things like “Hotmail” or “Amazon.com” you are left with searches based on current events, celebrity gossip and searches made to answer all the hundreds of thousands of questions that people want answered. Gilmour makes it sound as if searching had previously been synonymous with the guessing of domain names (albeit much more successful) when in fact guessing a domain has never been a particularly important path to most of the pages on the World Wide Web.

Boring reason 5: And anyway – doesn’t the domain name help the search engine? I’m not going to push this too heavily, as there are now many other factors that a search engine takes into account when it’s figuring out the relevance of it’s search results – but the presence of a domain name certainly doesn’t hurt

Let’s go back to the beginning and think about it logically. In fact the search approach that Gilmour describes has only really ever worked in very specific circumstances – like when you’re attempting to find a company by using it’s trading name or when you’re searching for generic domain names – like sex.com or bookshop.co.uk. Once you’re outside the remit of those particular limited circumstances, his argument loses much of its plausibility…

Categories
Technology

On ‘Smart Tagging’…

An interesting piece on Microsoft’s Smart Tagging Plans doesn’t say anything I haven’t heard before. The plan, it appears is for software that automatically inserts links into browsed web pages, links that Microsoft is the point of exchange for. Hence, your anti-gun site could be filled with links to promotional literature for the NRA. Or more prosaically, a web business that’s already reeling from an inability to make money will find itself crippled by drops in traffic. Is this the future of the web? Certainly it seems to be a step away from the madly anarchic non-hierarchical net of today…

Categories
Technology

The Complete Human Genome on CD…

I went to a meeting today with some people who work at Prospect magazine, which markets itself as “the magazine for the intellectually curious general reader”. I wasn’t familiar with the publication before the meeting.

I pick up a copy of the magazine and there’s a CD attached to the front, which I glance at briefly. And then I do a comedy double-take. The CD has written on it in large blue letters on a silver background:

“The Complete Human Genome On CD”

I stood there in some kind of dumbfounded silence for a few seconds, looking at this astonishing declaration with a mixture of feelings.

First came the awe at being in the presence of such an astonishing accomplishment – the same awe I would have felt the CD had contained every book ever written in English.

Then came a certain amount of fascination, and a kind of utopian wonder at the accomplishments of humanity.

And then came a bizarre feeling of betrayal and disgust that I was really not expecting. The same space used by magazines for distributing computer game demos, dodgy shareware programs that crash your computer, picture libraries of animals running or sitting or eating each other – this pure commercial space designed to flog both magazine and product – was now being used to contain an artefact with almost arcane power as if it were the toy in a Happy Meal.

Whether they know it or not, it seems like evidence of the increasing commercialisation of our lives – the ever present opportunity to repackage, rebrand or market anything and everything as a means to the tripartite commerical gods: “Sales”, “Sales” and “Sales”. Something important has been broken.

Categories
Technology

Thoughts on Encryption and Privacy…

  1. FACT ONE: There is increasing invasion of privacy by governments. There is no denying that the surveillance of the public is at an all time high in Europe and America at the moment. In London we find ourselves routinely watched everywhere we go by closed-circuit cameras everywhere we go. In fact in Central London, the Big Brother TV show is advertising itself by putting up stickers pointing towards cameras. They cover the city like a blanket. The same is true on the net. Projects like the intimated Echelon and the FBI’s Carnivore program are designed to search through all e-mails and track certain words like “KILL PRESIDENT CLINTON, DAMMIT”. The functionality of such machines however is prone to abuse. Because you have to scan every e-mail that goes through, without the slightest evidence that the person who writes the message has committed any kind of crime, everyone’s privacy is invaded.
  2. FACT TWO: Computer power and technology is improving the ability to track and collate information. It makes sense that the programs that discern whether or not an e-mail is from a terrorist will improve in time, but all this means is that individual types of people will be easier to track. And as computer power increases as well, we could be getting to the stage where a search placed could produce a list approximating every active gay or black person in a certain area of the world. Look out for words like Popstarz and Old Compton Street. All this really means of course is that if someone wants to find out about you in intricate detail they will be able to. Of course, there has to be some good reason to want to, doesn’t there? Doesn’t there?
  3. FACT THREE: Encryption is the best chance for private communication. If you want something private kept private, the best option is to encrypt it. However problems arise through this as well. While most e-mails are not (for example) PGP encrypted, the people who do decide to encrypt will be immediately noticeable through the same keyword searching process (in this case, PGP would probably do it). Suddenly all people who wish their correspondence kept private are marked as exactly the kind of people who probably shouldn’t want to.

All of which leads me to this conclusion: It’s only if we routinely encrypt our trivial e-mail that we have the slightest chance of maintaining the privacy of our lives. Enough encrypted trivial e-mail should swamp the stuff that needs to be completely private amongst a wash of personal correspondence. I urge you all now. Get PGP now and send me a message to tom@no-spam-please-barbelith.com.

Categories
Technology

Tom's thoughts on cryptography…

Tom’s thoughts on cryptography

  1. FACT ONE: There is increasing invasion of privacy by governments
    There is no denying that the surveillance of the public is at an all time high in Europe and America at the moment. In London we find ourselves routinely watched everywhere we go by closed-circuit cameras everywhere we go. In fact in Central London, the Big Brother TV show is advertising itself by putting up stickers pointing towards cameras. They cover the city like a blanket. The same is true on the net. Projects like the intimated Echelon and the FBI’s Carnivore program are designed to search through all e-mails and track certain words like “KILL PRESIDENT CLINTON, DAMMIT”. The functionality of such machines however is prone to abuse. Because you have to scan every e-mail that goes through, without the slightest evidence that the person who writes the message has committed any kind of crime, everyone’s privacy is invaded.

  2. FACT TWO: Computer power and technology is improving the ability to track and collate informatiom
    It makes sense that the programs that discern whether or not an e-mail is from a terrorist will improve in time, but all this means is that individual types of people will be easier to track. And as computer power increases as well, we could be getting to the stage where a search placed could produce a list approximating every active gay or black person in a certain area of the world. Look out for words like Popstarz and Old Compton Street. All this really means of course is that if someone wants to find out about you in intricate detail they will be able to. Of course, there has to be some good reason to want to, doesn’t there? Doesn’t there?

  3. FACT THREE: Encryption is the best chance for private communication.
    If you want something private kept private, the best option is to encrypt it. However problems arise through this as well. While most e-mails are not (for example) PGP encrypted, the people who do decide to encrypt will be immediately noticeable through the same keyword searching process (in this case, PGP would probably do it). Suddenly all people who wish their correspondence kept private are marked as exactly the kind of people who probably shouldn’t want to.

This leads me to this conclusion: It’s only if we routinely encrypt our trivial e-mail that we have the slightest chance of maintaining the privacy of our lives. Enough encrypted trivial e-mail should swamp the stuff that needs to be completely private amongst a wash of personal correspondence. I urge you all now. Get PGP now and send me a message to tom@no-spam-please-barbelith.com.

Categories
Technology

Internet Time (Part Two)

I am back at work, and thrilling it is too… I am really taken with the way that metajohn has incorporated the date information into his blog. I may be forced to steal valuable sexy ideas from him.

INTERNET TIME PART TWO:

So the response to my first conception of Internet time wasn’t particularly overwhelming. People seemed (how should I put this) … confused … by its purpose and/or functionality. Looking over it again they are probably correct.

So I have a new and better idea. Imagine this – you wish to organise a meeting with someone at 1pm your time. You decide on e-mail. You click on the new e-mail button, write what you want, and when it comes time to put in the time of the meeting you click on the clock icon on your toolbar. This icon triggers a “realtime” plugin which would bring up an image of a clock which you could click on, thereby selecting a time. This then writes a small piece of code into your e-mail which you would see as 1pm.

However this code at the OTHER end would be read by the plugin and automatically translated into the equivalent in their time-zone – thus your meeting partner would view it as 3pm (for example).

This plugin could then be added to any number of packages, automatically translating a particular timecode into what time it should be in the place where it was read. (Packages like word processors and web browsers are obvious examples – you can do something on the web like it with applets or javascript, but they go no further).

If you added to the functionality of the plugin by having it check the time against some form of authority (such as GMT) and automatically setting the computer’s internal clock accordingly, you could have a really handy nice little device that would completely invisibly generate a standard for Internet Time.

And to make it more entertaining, you could have any number of additional translation timescales in place – you could specify you wanted to see times sent to you in Swatch Internet Time (automatically translated with Swatch’s permission) or any number of wacked out systems. Think of it – on your home computer you could choose to work completely in StarDates (with the permission of Paramount of course). And if you wanted to extend it to dates, you could operate in a completely different calendar (although that might get a little too confusing). And it wouldn’t matter at all, because to the people who you communicated with would all think you were making all these appointments in whatever system they preferred as well!

Is there anyone out there interested in developing such a mini-app? I even have a snappy name for it: THERMIDOR, after the month of the French Revolutionary calendar in which I was born.

More ravings tomorrow.

Categories
Technology

Internet Time (Part One)

Mr Blair, the prime minister of the United Kingdom, has given his approval to the creation of GeT Greenwich Electronic Time. The site for this is incredibly boring, and I can’t help feeling rather misses the point. I mean of course it is handy to have a standard to measure time on the net, but Swatch‘s system seems to me to have more going for it (although it isn’t particularly intuitive). It seems to me that one of the functions of Internet time must be to make it very easy for people to agree to meet online without having to take into account regional time differences. Therefore the system must be intuitive enough for people to be able to understand, while different enough from the standard layout of time to not be confused with it. It’s important that people can say lets meet at XXXX and understand where that is in relation to their own time even when their own times are very different.

So here is a preliminary statement regarding BIT – Barbelith Internet Time. My suggestion is to keep with the 24 hour clock (because people have to be able to work out its relation to their own time when things like daylight saving and stuff kick in in the real world), keep it running from Greenwich (why the hell not, everything has done for years anyway) and only abstract its presentation. Something like 3113 for 1:13 pm GMT, minutes first, hours second. I know it seems ridiculous, but it should make time translation really easy while not allowing it to be confused with local time. Six figure numbers would have the seconds at the front – hence 003021 would mean 9.30pm exactly GMT. Someone who currently uses Eastern Standard Time would be able to say to someone two hours ahead of them to meet at 003018BIT which would be half past one EST and half past three for the other guy.

Has anyone got any other ideas?