It’s often said that you learn more from failure than from success, and for sure, this project has taught me a lot. Most fundamentally, it has taught me several important things about myself that perhaps I knew but was trying hard to ignore:

• I prefer to pick a hard problem and spend all day thinking about it than to pick a easier problem and actually solve it. Instead, I should try harder to do something productive without worrying too much about the details or in which order I should complete things.
• I need to work with other people who are less imaginative but more productive. Not just so that I can sit in my chair and bark orders, but that that my team-mates can periodically encourage me to stop dreaming and do something useful for a change.
• Apart from occasional toilet and meal breaks, I really can surf the internet for an infinite period of time. Having regular contact with other humans, especially those that ask you “what have you done today?” helps a lot, but ultimately I need to recognise my knowledge addiction and keep it under control.

With that in mind, it’s not surprising that as my team fell apart, so did the project. Although I succeeded in finding some great people with the right skills and a genuine interest in the project, they all had successful businesses of their own as well as full-time study. I knew that it would be a problem from the beginning, but I had hoped that I might be able to persuade them to give me just enough time to make it work. As it turns out, I was being over-optimistic. In retrospect, I should have tried much harder to find team members with time as well as ability, instead of trying to “go it alone” with inevitable consequences.

It’s not all doom and gloom, though – many things have gone surprisingly well:

• Almost everyone I’ve talked to about my ideas have been very helpful, supportive and sometimes even constructively critical
• Being given funding from my University to “be entrepreneurial” was a pleasant surprise!
• Moving to Poland. Not only has it been a lot of fun living here, it’s given me a lot of confidence that I can live a semi-nomadic lifestyle without problems. I mean, if I can live in a country where even “hi” (cześć) is unpronounceable, the deputy education minister thinks that “the theory of evolution is a lie” and government officials are worried about school teachers and even the Teletubbies promoting homosexuality, I can live anywhere

Thanks to everyone who gave me their support and good luck to the IndabaMusic crew – it’d be great to see them grow rapidly and vindicate my ideas! I feel I’ve learnt a lot and am much better equipped to start my next businesses. Watch this space.

That’s not to suggest that nothing’s been happening in Flukebox land. A little while back, Last.fm‘s added two of our key features, downloads and concert listings, to their service. Then one of most enthusiastic and knowledgeable supporters, Andrew Dubber, wrote a blog post that wished our service a “swift and inexpensive failure” – with the smallest of get-out clauses in case I started crying. These, and a few developments from other competitors, made me increasingly worried that perhaps we risked being a small fish in an overcrowded pond. A quick look at some of the biggest Internet success stories in recent years (Amazon, eBay, Skype, etc.) shows that not only did they do something well, they also got established before anyone else had come up with something worth using. So, after a week of feeling decidedly moody and angrily throwing lots of ideas in the bin, I’ve emerged with a new plan.

Flukebox was always a reaction to two technologically-driven trends:

• the reducing cost of digital recording technologies are encouraging more people to record high-quality music
• the Internet is making it practically free to distribute music and is changing patterns and methods of consumption

Just like everybody else, we were focusing mainly on the latter trend and just accepting that the first would provide the raw material for our service. But why? Many musicians make music for the fun of it and don’t really care if anybody hears it or not. A bit like the way I write this blog as a way of getting my thoughts straight and keeping a diary, and am pleasantly surprised when I find somebody has actually read it.

A quick and informal survey reveals that people are mostly just using email and discussion forums to collaborate with other musicians online – if at all. Surely we can do better than that? Making music is different in several important ways – so why do we basically only have tools for doing it on a single computer? Why is there no proper online community for musicians?
Why indeed.

• Skype – I know plenty of people already use Skype to call their long-lost relatives on the other side of the world, but it’s actually pretty good for business, too. Although there are occasional issues with call quality, it’s very useful to be able to send files and web addresses to each other during the conversation. And setting up a conference call is as easy as clicking a few buttons – much easier than trying to figure out how to do that on a normal telephone.
• Google Docs - you might not see the point in struggling with the sometimes slightly awkward interface when you’ve got a perfectly good office suite on your desktop. The real reason to persevere is that Google Docs makes it really easy to work on documents together – even at exactly the same time. For instance, we can keep notes or jot ideas whilst in a Skype conference, and everyone sees what we’ve written. So no arguments about who said they’d do what!
• Gliffy – sadly Gliffy‘s updating isn’t so fast, so things can get a bit confusing when more than one person is working on a document at the same time. But it’s still a great way to generate diagrams together.

Of course, none of these technologies cope with the fact that it’s very easy for people to go incommunicado – after all, you won’t bump into them in the corridor. I think remote-controlled electric-shock collars are the best solution…

For technical reasons, the written word has been widespread on the Internet for much longer than any other form of media. Consequently, it would perhaps be fair to assume that is more mature – and perhaps provides some indicator of what is to come with images, sound and video.

The “music like water” makes the same mistake as looking at the Internet and seeing it as a cheap way to distribute newspapers. They’re not wrong – but it hardly describes the trend towards widespread publishing through blogs, forums, personal websites, social networking sites and so on.

I’d argue that we’re starting to see the same thing with music – and in a limited way, YouTube is showing how it can happen with video too. Those predicting that the internet will just make music so cheap that it’s almost free are missing the point and ignoring the real revolution that’s just starting to happen. Again.

This was also posted to the New Music Strategies forum.

I registered the company at @UK PLC whose basic package, at £23.49 all-in, is actually cheaper than you can do it yourself, and much quicker – electronic filing can only be done using special software or via an agent, and paper forms take around 8 days to be processed. It seems a bit crazy that I can do something as important as incorporating a company without ever signing anything, but I guess signatures are fairly meaningless anyway. Hopefully we’ll be able to conduct the more mundane government interactions over the Internet soon.

Once the registration has gone through, I’ll be able to open a bank account. Abbey offer free business banking (forever), so long as you don’t pay in more than 100 cheques or £3000 of cash or make more than 100 withdrawals per month. I that’s unlikely to ever be a problem for us. Their charges for exceeding those limits aren’t exorbitant, so it would seem like a good deal for most small businesses.

Once I have a bank account, I can apply for VAT registration, which is just a matter of filling in the right form, which looks fairly straightforward. So that was all reasonably easy, now I just have accounting, tax and employment law to get my head around. Joy.

I’m currently reading “Crossing the chasm: marketing and selling technology products to mainstream customers” by Geoffrey Moore. It’s perhaps jumping the gun slightly, as we don’t have any customers yet, let alone mainstream ones, but that’s where my aspirations lie so I figure it’s as well to keep an eye on the horizon, particularly as one of our closest competitors, Last.fm, seems to be stuck in the geek-phase and has yet to acheive mainstream exposure.

Perhaps this is why: it’s a pain to explain. “Well, it’s like this plugin for your media player that logs everything you play and then you can display that on your website and it will recommend artists you might like, and people who like the same things as you and then you can chat to them and listen to the music they like and…”. You get the picture.

Trouble is, Last.fm is probably quite simple compared to our offering. Essentially, it just recommends bands. The challenge that “Crossing the chasm” presents is to state why the idea is better than what people use already, and why we’re better than our competitors – that is, comparing, not just explaining. Oh, and you’ve got two short sentences to do it in.

This is where I will need some help. What services do you use to find music and gigs? What good about them? What’s bad about them? How do you think Flukebox might be better? Yes, I’ve heard it all before a hundred times, but I want to hear it again. Hit me with it.

About a year ago, I came up with some ideas about how various technologies could be combined to support independent artists, in a way that’s never been seen before. Essentially, the combination of:

• peer-to-peer networking, for near zero-cost distribution, made legal by Creative Commons-licensed music
• collaborative filtering-based recommendations, to sift the wheat from the chaff (regardless of the quantity of chaff!)
• online concert listings and e-ticketing for small-venue music events

None of which are particularly novel, but the combination of all three into a single easy desktop music-player application brings significant advantages:

• tracks can be downloaded automatically, based on recommendations
• recommendations can be used for “intelligent shuffle” playback
• music events can be recommended “on the desktop”
• e-ticket booking fees can provide revenue to keep the rest free

I hold the opinion that many internet ventures can, and therefore should, be done on a shoe-string. In keeping with that, I’m moving to Poland, where living is significantly cheaper, thanks in no small part to a beer costing less than €1! I’ll slave away on the code, enlisting the help of friends when I can and experts when I must.

On a more personal level, I see this as a chance to prove whether my dreams are feasible: whether I can survive in a foreign country where I struggle with even the alphabet, whether I can ‘work’ and travel at the same time, and whether I can manage a major project within the toughest of constraints. It might work, it might not, but one thing is for sure: it’s gonna be an interesting ride.

You know the best way to describe music? Don’t. Just give them a copy and let them make their own mind up. This is becoming steadily easier, thanks to advancing technology. Gone are the days of making a tape; just burn them a CD, email them an Mp3 or even just tell them the title and let them find it on a filesharing network. Okay, so there’s the small issue of copyright law, but that seems quite a distant problem compared to that of being able to share your emotion.

Despite the growing ease of sharing music, it’s still quite hard to find which friends share your musical taste, because everyone is so different. Effective music recommendation involves encountering someone who enjoys at least some subset of your musical tastes, remembering what they like, and then remembering to tell them whenever you encounter something they might like. It’s quite a difficult cognitive challenge, and unless you do it regularly, you’re unlikely to do it very effectively.

Online services that assist this process are great, but there has to be something to do with that information in order to make it useful. Like, for example, being able to hear tracks that these people like, but you haven’t heard. Guess what? That’s called collaborative filtering, and that’s exactly how most recommendation services services work, like Last.fm‘s Recommendation Radio.

Online “social networking” is the flavour of the day. Email, IM, blogs and social networking tools (like the ubiquitous MySpace) are becoming very popular, because they support and extend existing social networks. But the social network for music is ineffectual due to the communication barriers outlined above, and as such, a social networking tool for music deserves to be as sucessful as an architecural dance society.