wiki:org/projects

Version 23 (modified by nickm, 9 years ago) (diff)

--

Tor Projects Page

The Tor Project develops a number of software programs including the core Tor software and supporting applications. The Tor Project as an organization also maintains its server infrastructure, advocates anonymity online, and raises funds to pay its staff. This page summarizes the current products and projects we are working on.

In the following, we are using these terms:

  • Category: A group of related produts, like "Core software," or "Network analysis."
  • Product: A product is an ongoing artifact like a software program or a website that usually has no end or due date. We subsume ongoing activities like fundraising under products, too. Nearly every program is a product, but some products are not programs.
  • Project: A project is an endeavor to improving one or more products. Projects have a success criteria and in most cases an end date. Some projects are external deliverables that we have promised to one or more sponsors; other projects are internal. Some projects are activities we're currently working on; others may be in a blue-sky or planning stage. The scope of a project is usually on the order of days to months.
  • Milestone: A milestone is a defined sub-goal of a project that usually comes with its own due date.
  • Task: A task is an individual work item as part of a milestone or project. Tasks usually respond to a Trac ticket. The scope of a task is usually on the order of hours to days.

For example, "Core Software" is a category; "Tor" is a product, "Make Tor support all DNS types" is a project, and "Write a design proposal for transmitting arbitrary DNS queries over Tor" is a task.

See projects/HowWeDoProjectManagement? for info on how we think our project-management system is working and how it should work.

List of products and projects

We are working on these projects (third level), grouped by products (second level), grouped by categories (first level):

  • Core Software
    • Tor
      • Increase Tor's suitability as a circumvention tool based on designs stated in "Design of a blocking-resistant anonymity system" (sponsor A, deliverables 1 and 2)
      • Facilitate reconfiguration of clients to bridges (sponsor A, deliverable 3)
      • Hide Tor's network fingerprint even more (sponsor A, deliverable 5)
      • Enhance Tor's cell-based and directory protocol to improve performance in low-bandwidth/high-latency/high-packet-loss networks (sponsor A, deliverable 6)
      • Make Tor scale to 2 million concurrent users by splitting the network into multiple segments, switching to datagram-based protocols, and improving load balancing within the network (sponsor A, deliverables 7 and 13)
      • Improve usabilities, performance, and reliability of bridges (sponsor A, deliverable 12)
      • Research the option of providing incentives to run Tor relays (sponsor A, deliverable 14)
      • Split AES and SSL across multiple cores.
      • Instrument Tor's internals more thoroughly to identify bottlenecks.
      • Finish integrating a microdescriptor implementation.
      • Release candidate for Tor 0.2.2.x.
      • Perhaps, make all bridges also able to proxy connections to the Tor download site.
      • Add support for advertising directory mirrors that have no DirPort, but only support BEGIN_DIR.
      • Make Tor able to relay more open-formed DNS requests to the exit node's DNS; expand DNSPort to handle TCP dns requests as needed.
      • Add support for bufferevent-based data transports.
    • Vidalia
    • arm (Tasks & Milestones)
    • TorCtl
    • TorButton
    • Orbot
    • JTor
    • BridgeDB
  • Software distribution
    • Thandy
    • Debian packages/RPMs
    • Tor Browser Bundle
      • Research and develop Tor Browser Bundle as single end-user package, unless it turns out LiveCD is the better approach (sponsor A, deliverable 11.a)
      • Reduce footprint of Tor Browser Bundle usage (sponsor A, deliverable 16)
      • See also projects/TorBrowserBundle?
    • LiveCD
      • Research and develop LiveCD as single end-user package, unless it turns out Tor Browser Bundle is the better approach (sponsor A, deliverable 11.b)
    • GetTor
      • Develop more reliable download mechanism for Tor Browser Bundle by splitting downloads, implementing a lightweight download manager, or reducing the software bundle file size (sponsor A, deliverable 15)
      • See also projects/EmailAutoResponder?
    • Package archive
  • Network analysis
  • Infrastructure
    • Bug tracker and wiki
    • Blog
    • Translation portal
      • Provide translation portal for non-technical users (sponsor A, deliverable 17)
    • Nagios
    • Website & mirrors
    • Buildbot
    • Git & Gitweb
      • Get SVN projects and code that is not under version control in Git
    • SVN
    • Backups
    • Misc Infrastructure
  • Corporate

Automatically generated list of wiki pages in the projects/ directory: