Building Ambisonia.com

1 year would give you $8 tipping credit

This is where I’m at. $14.95 will buy you 3 months subscription to Ambisonia.

For that, essentially, you get to download everything in a range of formats.

I havn’t yet decided if whether or not *.amb files should remain as free downloads or not.

What I’m looking for is the XXXXXXXX in the picture on the right. What is it? …. what’s that last little feature that will really add to people’s experience of the site? Its got to be

1. funky in some way. a ’super’ feature,
2. accessible to newcomers. The business model cant rely on features only meaningful to a tiny minority.

Essentially, if that XXXXXX is good, then I can make *.amb files free for downloads. That will be the clincher.

Maybe its an automatic-convolution service where you upload the data you want Ambisonia to convolve each tune with (for Binaural or room correction). I dont know.

Can you help?

Clearly, for $14.95 for 3 months, what you are buying is accessibilty. It really feels like there should be one more little thing involved… some must-have feature that will open up the can to excitement. The questions is what!

Its 2015. Ambisonics isn’t mainstream, but there is significant market awareness. Jane and Bob want some of the action, and so they decide to set themselves up with an Ambisonic listening rig.

They walk into an audio hardware store and announce their intentions to the sales rep.

Jane: “Hi, we’d like to get setup at home with a surround sound system that can play Ambisonics”

rep: “Ok, great, we have everything you need here, so what exactly are you after?”.

Bob: “Well, we’ve heard that Ambisonics is the best surround experience, and we love our music, so we’d like to get set up”.

rep: “Ok, so what kind of speaker set-up are you after?”

Jane: “Something so we can listen to Ambisonics”

rep: “Ok, Ambisonics can be played back over practically any speaker array.”

Bob: “What’s a speaker array?”

rep: “Oh, sorry, you can use as many speakers as you like. Generally, the more speakers the better.”

Bob: “Ah, I see, so what’s the ideal number of speakers so we get the most out of Ambisonics?”

rep: “Well, you could have a 5.1 style set-up, or you could go a bit extra and have a hexagonal array. If you’d like to experience height then you might consider a cube array. You could also go for a cube on its side type of array, then you can have 2 rows of 6 speakers, and generally any vaguely symmetrical array is valid.”

Jane: “Ahhh, Ok”

Jane: “so… what’s recommended?”

rep: “Well, it depends really. What kind of room you are in, and how much money you have, but above all, which AV receiver you are going to use… not all AV receivers decode to all those different speaker arrays. You see, that’s the great thing about Ambisonics, it can be played back to any speaker array”.

Bob: “That’s great that Ambisonics can be played back to any speaker array… but that doesn’t help me choose which array”.

etc. etc.

My point being that choice is often an inhibitor to technology uptake because it adds complexity and introduces the need to make decisions. Making a decision is work, and if you dont know much about what you are making a decision about, its hard work.

So it is my belief that Ambisonics needs to define a set of recommended speaker layouts. This will make it easier not only for consumers to make choices, but also for maufacturers to make decoders.

Above all, I believe that there should be 1 overriding recommendation. A recommendation that can be sold to consumers who have no means, no mechanisms, by which to make a decision about what kind of speaker array / AC receiver combo to choose. Maybe that recommendation should be a cube layout … just so that height is a factor.

Just found this excellent post on the concept of Tipping. The (opt-in) ‘patronage’ aspect of tipping is important. The challenge is always going to be how to aggregate small amounts of money to avoid drowning in credit card fees. PayPal is one (not very popular) option.

My thought would be that people would purchase ‘tipping credits’ like $10 or $20 amounts (like one does on Skype), hence only causing one credit card transaction (fee).

The issue is whether people would be willing to spend that kind of money on ‘future’ tips. I think its worth a go.

A while ago, Kevin Kelly proposed that all that any artist required for survival was 1000 true fans. He was essentially rebuked by people saying that 1000 true fans is actually a hell of a lot to achieve… and to have each fan spend $100 per year, is also a large somewhat unrealistic sum. More likely you’d need 5000 fans spending $20 - $30 per year… 5000 fans is a lot to achieve (without commercial exposure).

More appropriate to the web is Scoble’s estimation that all a web site needs is 100,000 passionate fans. The figures I’ve done on Ambisonia’s business model suggest that I need 150,000 monthly visits to achieve ‘break even’ point (i.e. the site sustains my wage, and hence has a full-time staff of 1). Perhaps I have been pessimistic with my figures (I dont think so)… or my business models are not right.

But most interesting of all is an other post I just discovered by Kevin Kelly. “Better than Free” is Kevin’s attempt to break down what is worth value in an essentially “free as in Google” world (that’s my phrase, not Kevin’s).

Its an excellent break down. He proposes 8 areas where value still exists in a digital world. I’m going to attempt to very quickly elaborate these 8 areas where they relate to Ambisonia.

Immediacy.

Pay $5/month and download the latest 20 tunes 1 week before everyone else? Not quite the right model, since very few tunes posted on Ambisonia relate to a highly anticipated ‘live’ performance.

Pay $5/month and access the ambisonic recording immediately (i.e. streaming then and there)? That would work.. its real value. But I want newcomers to have access to ambisonics as quickly as possible with as little barrier as possible, so not ideal for Ambisonia right now. That’s a feature I cant put behind a financial transaction.

Personnalisation

Pay $5/month and get all the Ambisonia tunes mixed down to binaural using your personalisd HRTFs. Absolutely. This would be a champion model. Excellent value, I’d imagine people would get very excited about it. Its a very positive model.

Unfortunately, I dont think there’d be enough people with their own personalised HRTFs to sustain the site. Maybe this could be a part of an other larger model.

Interpretation

This is the Red Hat model … get the software for free, but you pay for the manual. Perhaps Ambisonia could produce a super-easy to use Ambisonic player… and charge for it. Hmmm. $10 one off purchase for a super-easy to use Ambisonic Player. For 100,000 passionate users, if 1/10th bought the software that would generate maybe 1 year’s sustainability. Not really a long term prospect. Wouldn’t be a bad project though.

Authenticity

Get all the tunes uploaded to Ambisonia this month on a DVD with signatures of each uploader. Not the right angle. Decodes could perhaps be considered ‘authentic’. Get an authentic Ambisonia 5.1 decode of such and such. It would have to be a demonstrably better decode than anyone else’s. Unlikely (at least not without cash input).

Accessibility

This is the one I am likely to go for. You can stream the track directly from the browser (free accessibilty to pull in newcomers) … you can download the *.amb via bittorent (perhaps, not sure about this). For $5/month Ambisonia will give you access to AC3, AAC, DTS and stereo… in an iTunes compatible podcast.

Embodiment

Something physical. How about a Sound card that is configured to work out of the box with Ambisonia (or for ambisonics more generally)? Or the DVD idea again.

Patronage

This is key! As Kevin says “It is my belief that audiences WANT to pay creators“. I absolutely agree. And the “tip” system will allow people to tip the uploaders. I think it would also be great to report on patronage (at the patron’s agreement of course).

Findability

I guess this is a curatorial angle. Help people find the good content. Ambisonia already kind of does that via the ratings widget. Not sure how else I could work this angle.

These are just knee-jerk initial thought. Nothing ground breaking. But I’m going to mull over them in my head. Would love to have people’s suggestions.

Like an eagle hovering before the kill, I’ve been making circles above my target business model for Ambisonia for a while now.

Its not simple. There are so many minor business model tweaks that might have srious impacts. I need to speak to an expert … but who?

There are essentially 2 ‘transactions’ I’m aiming for. One is for Ambisonia (subscriptions), the second is for the content creators (tips). I ’sense’ that its too complicated… having 2 separate transactions. I need to simplify the model… but how ?

Perhaps credits for the tunes could come out of the subscription fee. For every $5/month subscription, $2 is credits to allocate to the tunes.  Or should it be $3 for a subscription … but then I encourage people to buy at least $2/month of tippings? I think that’s too complicated.

One other option is that downloads remain free, but anyone with their personal HRTFs can upload them and then get all of Ambisonia’s content rendered to binaural with their personal HRTFs. But how many people would go for this? Its a limited audience … and the site will start being very complicated.

This gentleman (a vc) says that the services have to remain as simple as possible. He also suggests that the services should be pluggable. Having pluggable services means that other people can produce UI’s and features that leverage the existing infrastructure. i’m not sure how that could be done on Ambisonia… perhaps an API for accessing recording’s metadata? … ideally, it would be an API that could allow others to create their own decodes of Ambisonia tunes … but these people would need heavy infrastructure to do that.

That gentleman also argues that services like Twitter actually empower themselves with super simple constraints. Twitter has a 140 character limit. A powerful constraint that allows it to adapt to other platforms very easily. How could this kind of thinking relate to Ambisonia? Perhaps limited time for surround sound tunes? 20 seconds? hmmm … too restrictive… but would make the use of an external API much more likely.

8tracks unlike Pandora (which I cant access because I’m not in the US … der … internet-not) suggests putting 8 tracks together to form a set. I dont know why, but psychologically, doing this works. I can ‘imagine’ choosing 8 tracks to make a great set … its more difficult to get excited about putting together an undefined set of tracks to make a great set. Perhaps it is the empowerment of being freed from too much choice. How can I do that on Ambisonia?

The business model has to be simple. How can I merge 2 transactions, subscriptions and tips into the 1 model.

I wish there was a methodology for developing and adjusting business models ‘on the fly’ … like in ‘agile’ software development, where you just start writing code but then ’stay’ loose and flitter back and forth until you find the right code architecture.

I’d like to start charging people $5/month to download Ambisonia’s content. Then I want to change that, a couple of months down the track, to something a bit different… then I want to change it again, and again, until I find a ‘business model’ that fits both the business and the audience.

My guess is that doing this might be commercial suicide. But I dont know.

Note to self: have a look at this site, youlicense.com. If I understand right, they act as an intermediary for licensing music, whether it be a commercial license or other.

If they had some kind of API that ambisonia could hook into, then it could be an opportunity for facilitating commercial sales. Commission (again if I understand right) is 9%.

The aim is to provide a fast website which can, if/when necessary, scale to serve hundreds or thousands of concurrent users.

Doing this is non-trivial. Scaling websites is something that I do not know much about, except that its a bit of a black art.

I’ve enlisted outside help… and the resultant architecture is shown in the image below.

The future Ambisonia site will sit on EC2, Amazons cloud computing service.

You’ll notice the diagram shows 2 Amazon EC2 virtual machine instances. Only one (the one on the left) will be deployed to start with. The point here is that I will be able to deploy as many concurrent virtual machines to cope with whatever traffic demand I need to cope with. The second virtual machine (on the right) can be deployed in multiples and will share the load in rendering pages and executing requests to the database.

In the diagram on the left, the majority of visits will be served by Varnish. Varnish is a cache server which (most of the time) wont even bother to go back to Zope/Plone to render a web page … it’ll just spit back the same page that was requested 5 seconds ago…. no requests to databases, no rendering of HTML etc. Super fast.

When a web page is requested that has never been requested before (or the user is logged in, or its time to get a new version of the page), then Pound (a load balancer) will decide which Plone instance (there could be as many of these as necessary) is the least busy … and get that instance to fulfill the request.

Here I am using a product called ZEO … ZEO essentially caches the Zope/Plone database (ZODB) on each Plone instance. This time it is not the web page rendering that is being cached, it is exclusively accesses to the database.The contents of each database will be cached in each Plone instance.

This architecture is already implemented and sitting on EC2. My next tasks are to implement the new Ambisonia skin, migrate all data to Amazon S3, implement the Premium accounts and other e-commerce services. The Uploader is already integrating with the EC2 instance, and uploading to S3.

Still a lot of work to do, but this is the _final_ architecture, no more looking back wishing I’d done it ‘properly’ … this is properly.

The planned business model for Ambisonia is essentially known as the ‘freemium’ model. It is one of the most successful business models practised on the Internet.

The basic services are free … and then a percentage of the audience, often around 0.6% of all active accounts pay for extra premium services.

This is how Flickr works, they charge $2/month (on a yearly basis) for unlimited disk space. Last.fm has premium accounts for 2.50euros/month (extra features). NicoVideo.jp charge $5/month (extra features). More business oriented software, like 37signals.com charge from around $30 - $150 per month.

I’m aiming for $5/month…. ok … but what do you get in the Ambisonia Premium accounts?

Well … that’s the tricky bit.

I’ve made the mistake, you could say, of offering everything for free on Ambisonia. Disk space is free, downloads are free (note: the cost of bandwidth is essentially absorbed by the generous torrent mirrors). RSS feeds are free, everything is free.

My initial non-business oriented idealistic impressions of the state of things sold me the notion that ’some way of making money would come up’. It hasn’t. It just doesn’t work that way. And not just for me but for thousands, if not hundreds of thousands of web sites out there. Making money from websites is difficult.

So I’ve been busting my gut trying to come up with the ‘extra’ services that I can offer that would be worth $5/month. I havn’t come up with any glaringly clear ideas. If you have an idea … please throw it at me.

At first I thought that it would be a strategic mistake to ‘remove’ existing functionality from the site … then try and sell it back. I’ve had to take a step backand have a look at how I would model the site had it never been released. There is a very simple model that (I believe) would work.

All files would be streamed for free, directly from the web browser in both 4ch AAC (for people with 5.1 on their computers) and 4ch AC3 (for people with optical connections to the AV receivers). For $5/month, you can download everything in all formats. The advantage of this model is that newcomers will always have full access to listening to the content. Those who dont want to subscribe, will always have full access to listening to content. So there is no real restriction or hindrance for the site to welcome newcomers to ambisonics.

But if you want the added benefit of having the content stored locally, or having access to the full B-Format files to decode to your own array, then you need to subscribe.

I’ve decided this is the model I’m running with. But I’m still very very open to other ideas.

Lastly… but very importantly, if people are going to be paying $5/month they will be expecting professional service levels on the website. The website has to be responsive, it has to be rarely (if ever) down, it has to be consistent in its communication and message. The website has to be able to support 150,000 visits per month! The design has to be slick etc. I have been working on all this for months (almost years) … and I am making headway. The next post will detail the upcoming Ambisonia enterprise grade ’scalable’ architecture.

I’ve been working on a Business Plan for a couple of months. I’m going for the warts-and-all Business Plan. Figures are projected for 3 years, I have an optimistic, a pessimistic and a target scenario.

I’ve developed extensive spreadsheets (I’ve got about 15 pages of spreadsheets) including cash flow projections, profit and loss projections, etc. And I tweak and tweak and tweak to try and reach a plausible level of realitity in my figures. There’s no point if I’m not being realistic.

I don’t want to give away all the details, but here’s a very basic description.

Executive Goal.

Firstly, earn revenue to sustain effort on Ambisonia

Secondly, allow contributors to earn revenue so they can sustain their efforts in creating ambisonic work.

Thirdly, make ambisonics as accessible as possible to new audiences.

Revenue streams.

I have slated 4 revenue streams. Listed here in order of deployment.

1. Premium Accounts (at $5 / month). The principle revenue stream. More on these in an other blog post.
2. Tips … sustenance for the contributors (70% of tips go directly to contributor)
3. Advertising. I dont want to depend on advertising revenue. It is potentially volatile and unpredictable (from the point of view of the ad-space retailer)
4. Affiliate sales. A bit of extra income.

Long term, direct sales of surround hardware is an option.

Required Traffic volume for break-even point.

Targeted projections model Ambisonia reaching its break-even point at around 150,000 visitors per month. By ‘break-even’ I mean that I will be able to afford to work full time on Ambisonia.

Targeted projections model that 150,000 visitors a month being achieved in July 2010 (that’s in about 24 months).

If my calculations are in the ‘realistic’ ball park, then once I’ve reached the 150,000 visitors / month mark, Ambisonia should be generating around $1900/month in revenue for contributors. I’d say there’s potentially an error factor of 2 in that figure.

These firgures refer to the ‘target’ projections. Optimistic projections model reaching the break-even point at just after 1 year. In my pessimistic projections, the break even point is not reached in 3 years.