Small addendum to point 1: Currently the "replicas" (i.e. the change units that need to be replicated between the peers) are managed in persistent queues and indeed unencrypted. This are temporary queues until transmission completes. The RDBMS itself is PW protected and the transport is - as already mentioned - DES/AES encrypted. We will likely make the local storage encryption an option in the future so that users can switch on/off based on preferences. Handling of large files will be much faster with no encryption.
Thanks. Good to see you are test-driving it and I can gladly provide more details:
It actually *should* be encrypted. If it is not, this might have slipped during the build. I will look into it. We ran tests with both options: using disk encryption from the RDBMS and applying our own encryption algorithm to the content. We also changed our embedded database some time ago, from QED to Derby (Apache project).
Traffic over the wire it is encrypted end-to-end using AES-256.
Everything is stored locally in the Derby RDBMS. All team members of a workspace store all the content (full redundancy). The only intermediary services are the following: "Rendezvous" is used for presence where every peer announces itself when online; "Relay" used to help route traffic if a direct TCP connection between two peers cannot be established. No content is stored on these servers. See http://www.jxta.org/ for more information on basic P2P libraries that we use.
Peers exchange an auto-generated UUID as their basic credentials. On application level we don't offer additional authentication yet. We are currently thinking about how to add the next element here. Maybe a "please you tell me your fingerprint" cross-check approach (like in Groove), but we want to explore other alternatives as well.
End of March we will release a new version with a "Conflict Bin" feature. This will "save" your data in situations of conflict.
Your other comments: The invitation process should improve with the Central User Directory - CUD (end of March release). The CUD will add off-line invitations management. And yes, the only context is the workspace at the moment. We will add more features "outside the workspace context", such as a unified contacts list, and much improved IM-behavior outside of workspaces.
How much is what? Price? Well, the basic version is and will stay free with actually little to no limitations. We don't like the concept of selling "Premium Software Licenses" (a Workplace is a Workplace!), but we plan to add premium subscription services, for optional user benefits such as a permanent peer capability for workspace management, backup, Web-access, etc.
Thanks again and we look forward to welcoming you to the Collanosphere soon!
Hi, I am the founder/CTO of Collanos, and we are offering something quite similar to Groove. We are a small start-up and don't have $150 Mio. to invest, but we do have talented developers and came already a long way. Our main differentiators are:
- Multi-platform: Runs on Win, Mac, and Linux's
- All P2P and built atop Sun's OSS JXTA.org libs
- Object-oriented, not Tool-oriented, i.e. you can structure your workspaces and mix and match any object type (e.g. group discussions are objects, not a another tool tab)
While we don't offer the functionality and richness of Groove (yet), we are going to ship a number of new releases in the next few weeks and months. We are following a community-oriented open approach and engage in co-innovation and discussions with our user base as much as possible. Please have a look at what we're doing and tell us what you like, what you don't, how you would spin things instead, etc. Thanks in advance!
Franco
Slashdot Mirror
Small addendum to point 1: Currently the "replicas" (i.e. the change units that need to be replicated between the peers) are managed in persistent queues and indeed unencrypted. This are temporary queues until transmission completes. The RDBMS itself is PW protected and the transport is - as already mentioned - DES/AES encrypted. We will likely make the local storage encryption an option in the future so that users can switch on/off based on preferences. Handling of large files will be much faster with no encryption.
You can read a bit more about the near-term road-map here: http://blog.collanos.com/index.php/2007/03/01/glim psing-into-the-future-of-collanos/
Your other comments: The invitation process should improve with the Central User Directory - CUD (end of March release). The CUD will add off-line invitations management. And yes, the only context is the workspace at the moment. We will add more features "outside the workspace context", such as a unified contacts list, and much improved IM-behavior outside of workspaces.
How much is what? Price? Well, the basic version is and will stay free with actually little to no limitations. We don't like the concept of selling "Premium Software Licenses" (a Workplace is a Workplace!), but we plan to add premium subscription services, for optional user benefits such as a permanent peer capability for workspace management, backup, Web-access, etc.
Thanks again and we look forward to welcoming you to the Collanosphere soon!
Hi, I am the founder/CTO of Collanos, and we are offering something quite similar to Groove. We are a small start-up and don't have $150 Mio. to invest, but we do have talented developers and came already a long way. Our main differentiators are: - Multi-platform: Runs on Win, Mac, and Linux's - All P2P and built atop Sun's OSS JXTA.org libs - Object-oriented, not Tool-oriented, i.e. you can structure your workspaces and mix and match any object type (e.g. group discussions are objects, not a another tool tab) While we don't offer the functionality and richness of Groove (yet), we are going to ship a number of new releases in the next few weeks and months. We are following a community-oriented open approach and engage in co-innovation and discussions with our user base as much as possible. Please have a look at what we're doing and tell us what you like, what you don't, how you would spin things instead, etc. Thanks in advance! Franco