6 ms·
In 2005, I worked at a startup that attempted to solve the problem of privacy and security for personal information (photos, home videos, music, personal health
by vinay_ys 8y ago
In 2005, I worked at a startup that attempted to solve the problem of privacy and security for personal information (photos, home videos, music, personal health/finance documents, contacts etc) while also providing ability to share and collaborate.
The solution involved running a mesh network with nodes on user's laptop or desktop and a corresponding node in the cloud. These nodes would index local data and provide replication of metadata across nodes and backup of actual data to cloud node.
A locally running web app acted as replacement for 'windows explorer'. It allowed the user to access all their files and folders across all their nodes, access them (open document, play music/videos, see contacts etc), create smart collections and share these files, folders or collections with other users in a secure authenticated and private manner.
User got an identity - which comprised of a dedicated domain (or subdomain) and a PKI certificate tied to that domain. Each node had it's own private key and their public keys were tied together by the identity certificate.
All communication between nodes (of same user or across users) where authenticated and encrypted using these identity/node keys and certificates. No central node existed in the system that could spy on these activities. The architecture separated the network discovery cloud nodes from your data cloud nodes and architecture allowed for your data cloud nodes to be hosted separately anywhere (say, in your own cloud instances).
This is the only system I have seen that utilized zero knowledge protocols and made it accessible to common people to manage their data and share with others as well.
But unfortunately, as a business it never took off. It got acquired by emc and merged with mozy (good old data backup company) and then this product died a silent death in 2010.
Maybe it was timing, maybe after snowden, if this product had launched it would have done well.
But now, I think a more urgent and a relatively less complex problem to solve is one of distributed communication. In this era of always connected powerful devices (mobile phones, home gateways), why don't we all have our own personal email/chat servers that nobody else can spy on? Why does email and chat have to get relayed via big aggregators who mine so much data as well as metadata?
Not only do they violate privacy, they succumb to security breaches and cause serious damages.
I feel the stage is set for this disruption: crypto protocols, always-on cheap connectivity, compute power at the edge, and sensitivity to privacy/security in general population – all of these ingredients are appropriately set right now for this to happen.
- brabel 8y agoSounds like https://keybase.io/ https://keybase.io/
- lioeters 8y agoFascinating to read about your experience with the startup, and how the service was architected. > unfortunately, as a business it never took off Sounds like the timing was too early in 2005. I believe these days we're all so tired of the privacy and security situation, that the world is ready for something like this. > [it] utilized zero knowledge protocols and made it accessible to common people to manage their data and share with others This describes exactly what the re-decentralized web needs. Seeing the many attempts over recent years, it looks like there are significant technical, financial/business and social challenges - but I totally agree with your conclusion, that "the stage is set for this disruption". It also feels like the tide is rising, that the solution is being worked on from numerous fronts and eventually a more evolved system will be adopted by the public.