Chris has been researching the area where essentially you have your own cloud on your laptop (replicated to your desktop and your phone) that stores all your data. This data can be accessible by whoever you choose.
If you have data wherever you want then you don’t need a web interface.
Specific software is available to create your own cloud. If you have the apps to access this then you still own the original data.
The data portability group is trying to do this.
The average person doesn’t care about their freedom and using open source software a the existing commercial software is easier.
Is there any commercial incentive to create this kind of software? In the long term, are business models based on getting people to your site going to last? These are bubble businesses. The innovation is needed in coming with ways of doing this commercially without relying on people coming to your site.
The data portability group got Facebook, Microsoft etc on board in order to find out sustainable models beyond advertising.
If you don’t own your data, what happens if Facebook goes under? Epsom users had 26 days to download their photos before they were lost forever.
There should be a place to find all videos so you don’t have to go to YouTube then go somewhere else if you can’t find the video you want.
You don’t lose any functionality by choosing to host your blog on Google or to create your own blog. It can still be found by search engines.
Getting basic data out of a database can be very time consuming.
There is a project called VRM (vendor relationship management) which is about licensing content. eg you can choose the terms in which you upload your content and then they bid for your business.
There are also questions around discoverability.
Related posts:
- Ryan McDonough: The future of netbook operating systems A discussion about the way we use netbooks, and the...
- Chris Dymond: What can we do with 100 MB/s broadband? Initiatives are underway to bring broadband connectivity to many more...
- Emma Persky: Social Media Consultancy – the trade that dare not speak its name Looking at Social Media Consultancy, the people that give it...
- Patrick Vyse: The Economic Tech Divide An open discussion on how economic giants segregate nations or...
Related posts brought to you by Yet Another Related Posts Plugin.
My notes from the same session:
There is a movement towards giving away software but locking in users’ data. Lots of people upload data to Facebook but then find they can’t leave without losing access to all their photos etc. If I can run my own email and talk to everyone else why can’t I run my own facebook and talk to everyone else?
You can have your own cloud – replicate your data between phone, desktop, laptop but data remains your own. If you have your data wherever you want you don’t need a web interface – can use desktop apps. Imap vs web mail. Blogs: can host your own or on blogger, but either way, people can interact with your data just as easily.
Example technologies:
CouchDB – drop-in replacement for MySQL with a layer on top that distributes the database for you. If one server goes down it will route arond the fault. Can be used to run your own cloud.
Midgard 2 – Like having a filesystem as a database.
The Data Portability Group are trying to do this. Various codes of conduct and ethical codes are being drawn up. However, although we can recommend software that does share data in this fashion most of it is currently linux based – not available to the majority of computer users yet.
Standards are required for interoperability. The problem is that most people don’t care about their freedom, they just want to use what their friends are using. Same problem as with Open Office – incompatible with proprietary formats. Format lock-in by vendors like Microsoft.
How do you switch to something else without dumping everything you’ve been putting into e.g. facebook?
Where’s the commercial incentive? No incentive for facebook to share the data it’s collected with other services. Are business models that require users to visit your site sustainable? Creates “bubble business models”. There is a need to divorce the infrastructure layer from the monetisation. Finding sustainable business models beyond advertising requires collaboration and open standards. Compete on quality of clients?! If people were willing to share more data with an open system this would be an incentive for systems to share that data?
Google wave: open protocol that anyone can implement. But the interface is horrible?! Not the point: can write any application to access protocol.
Need to say to people: don’t use this service ‘cos it locks in your data and you have to pay to release it. Most people won’t care but will be stung later.
Users do move around: open ID allows people to log on to various places with the same ID.
If you don’t have your data, what happens if Facebook goes under? Example: Epson photo service: if you bought an epson camera and used their default service, it went under, gave you 26 days to download photos (manual process) and then it went. Magnolia database fail. Various music and video DRM systems.
Status updates has value but not long term. Facebook sharing != photo archiving. Would users care if they lost all their data from a social network?
All web services are basically data storage services. Youtube should be able to find videos from wherever – even local – just through standard apis, messaging etc. However if you separate the database from the front end – can use any website or desktop app I like to see my facebook statuses – there’s no incentive to do the storage: nobody is necessarily using your front end so why incur the costs of storage.
If I keep a local copy of my data then at the moment people can’t socially interact with it (comments, calendar invitations etc).
New opera technology: web server built into the browser: exposes web applications that communicate over the web. But sends everything through opera.com! This may evolve over time to be more open.
RDF can be used to relate resources on the net but it’s difficult to use.
Microformats are simpler.
Vendor Relationship Management – VRM. Licensing your own content and uploading to flickr – flickr eats it. VRM service searches for hosting services that will offer publishing under a particular license. Bids and offers. E.g. Maybe Tesco might offer free storage for life but you have to give them so much data or sign up for clubcard.
Building an infrastructure on local hosting in the UK is difficult as net access is mainly ADSL – upload speeds << download speeds. Bittorrent works OK for this. Gittorrent: applies bittorrent to git. No need for a standard URL – just a tracker. Peers can disconnect and the files persist online.
Problem of discovery: concept of federation is a lot to take in. How to tell which version is which?