Rough roadmap. In no way is this binding, and there's probably important
stuff that I've forgot or that we'll discover we need in future:

Alpha 1: (target date 27 March; focus on basics and client
functionality; * = done already; ? = partially done)
- *Basic routing. May be susceptible to routing churn.
- *Encrypted connections, but susceptible to MITM; no node PKs.
- *99.5% Full FCP.
- *Persistent requests, but full state not saved; restart from scratch
  on startup.
- *Basic Fproxy. (no security).
- *Plugins, including basic search plugin.
- Daemonizable. (meaning can run in the background; turn off text mode
  interface by default, need to be able to do everything via fproxy or
  fcp).
- USKs. (Updatable Subspace Keys).
- ?Third party clients: Frost, FUQID. (FIW if we get a volunteer!).
- *Testnet mode ONLY.

Alpha 2: (target date ?June?; focus mostly on security)
- Nodes have PKs; safe encrypted connection setup; metadata saved in
  addition to the node reference.
- RSKs. (Revocable Subspace Keys).
- Official project freesite using RSKs.
- Basic request security.
- ARKs.
- Save full state on persistent requests. This makes multi-gigabyte
  uploads practicable even with sub-daily updates.
- Fproxy security (filtering) and i18n.
- Testnet mode OPTIONAL, as we provide some meaningful security in
  non-testnet mode by this point.
- Better searching?
- Darknet invites.
- Network size estimation.
- Research connection churn. Possible solutions: indirect lookup, bloom
  filters...

Alpha 3: (target date ?September?)
- Solve the connection churn issue.
- Translation infrastructure in place.
- Public appeal for translators.

Alpha 4: (target date ?January?)
- Opennet support.
- Debugging!
- Aim to have translations to as many languages as possible but
  particularly French, Spanish, German, Russian, Mandarin, Farsi, Arabic.

RC1: (target date ?March?)
RC2: (target date ?April?)
...

0.8: (target date 2008)
- Transport plugins, including connection-oriented transports.
- Passive requests.
- Pub/sub over passive requests.
- 1:1 messages/streams.
- Look into better fragmented darknet support.
- Premix routing.
(any of this might be deferred to 0.9)

0.9: (target date 2010)
- Debug debug debug!
- Make sure is scalable, easy to use and performs well.
(this is a general issue which needs to be addressed by every release)
- Possibly untrusted plugins.

1.0: (target date 2012)
- Debug debug debug!
- Make slashdot-proof! (When we release 1.0 we will likely have press
  coverage from CNN let alone slashdot!)

:)

Now back to work... :)
-- 
Matthew J Toseland - toad at amphibian.dyndns.org
Freenet Project Official Codemonkey - http://freenetproject.org/
ICTHUS - Nothing is impossible. Our Boss says so.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: 
<https://emu.freenetproject.org/pipermail/tech/attachments/20060315/4a68c90f/attachment.pgp>

Reply via email to