August'24: Kamaelia is in maintenance mode and will recieve periodic updates, about twice a year,
primarily targeted around Python 3 and ecosystem compatibility.
PRs are always welcome. Latest Release: 1.14.32 (2024/3/24)
Projects
This page details a bunch of projects, either in process, or which
could be done, or have been done.
Google Summer of Code 2008 Projects (don't follow the PTP format at
present)
Project Task Pages
Template
PTP/Guide
Unfortunately, this page is not as up to date as it could be, and will
be replaced with something relatively lightweight when practical. In the
meantime it is useful to give a flavour of projects Kamaelia has been
useful and used for!
Existing PTPS
PTPs to be written:
Background on Project Task Pages
Project Task Pages (PTPs) take a standard format, and recognise that
feature dev, bug fix, and any other task have commonalities. PTP's aim
to capture this information to make it easier to see what's going on,
and assign priorities. PTP's as a result have a template
format, which is based on the way Kamaelia itself is designed.
A project task page may cover a project as small as a single feature of
a single component of a single module, or something as large as an
entire application comprisig of several systems. The key aspect here is
that each of those subsystems would also have project task pages, as and
when they become necessary.
As a result Kamaelia development and management of components should
naturally reflect the component system itself.
PTPs to be fleshed out
Each project here will gain its own page as soon as practical - this is
a capture of current state. (This does mean also that there SHOULD be a
PTP for ch of the tools listed above)
Kamaelia
Macro - Timeshifting tools
- DVB Tools - Tools
for timeshifting in a storage efficient, automated manner
- Database Tools for working with databases in Kamaelia (at present,
very unclear what's sensible)
Mobile
Reframer
Multicast
RTP MPEG TS remultiplexer
Video
Cut Detector
Component Integration Tasks - Small ad hoc tasks of
documenting, reviewing and merging components into the mainline
codebase
/Sketches/MH/pixFormatConversion/VideoSurface
- pixel format conversion (yuv<->rgb)
- Pygame surface based video display
/Sketches/MH/dirac/*
/Sketches/MH/Video/VideoSurface.py
- yuv4mpeg file parsing and generation
- Cropping and scaling of video frames
/Sketches/MH/audio/WAV.py
/Sketches/MH/audio/ToWAV.py
- wav file parsing and generation
/Sketches/MH/MobileReframer/...
- rewritten UnixProcess component (possibly needs a rename?)
- Pipeline and Graphline that allows size limits to be set on
inboxes
- SAX stream based parser
- Sequencer for running components in sequence
- File reader that rate limits by the destination inbox becoming
full
/Code/Python/Kamaelia/Tools/Whiteboard/Whiteboard/...
- Some components are potential candidates for the mainline
codebase
/Sketches/MH/RTP/
- Selector (made more responsive to requests)
- Multicast_transceiver (reduced CPU usage by using the
Selector)
- ConnectedSocketAdaptor (minor mod to support multicast
transceiver)
- RTPFramer and RTPDeframer
Challenges Related - It's worth noting that as Kamaelia matures in
scope, it's capabilities are beginning to branch out well beyond these
challenges
Lunch - The fact the world is becoming massively parallel and we need
better tools to deal with this.
- Axon - Our toolset for
making concurrency easier to work with
Deliver - Tools and mechanisms for distributing and synchronising
effectively a huge database of content
- Macro - tools for
timeshifting, but also tools for delivering content via the
broadcast chain.
Share - Creation of tools for sharing that people can trust
- Tools to allow the audience to trust the content they recieve
- Tools to allow content creators to trust that their ability exploit
their copyright will be kept intact
P2P & Multicast joining
Thins that aren't appropriate as project task pages here:
- Moore's Law Growing both sides of an equation doesn't help - The Dirac project is largely
aimed at dealing with these, and is outside the scope of Kamaelia
- Open - This relates to the fact that open systems need to be able to
scale. This is inherently Kamaelia's core goal regarding online
delivery.
Projects are to be labelled with tags to help classify and organise
them. We've chosen this over having a predefined url based hierarchical
structure as it gives greater flexibility and acknowledges the fact that
projects will change in status and scope over time.
Possible tags to us include:
- Experimental (for stuff in /Sketches)
- Tool (typically the kind of thing that ends up in
/Kamaelia/Tools)
- Example (pieces of code that exist to demonstrate
functionality)*
- Mainline (work affecting the main /Code codebase)
These are relatively substantial systems in their own right. Their
development, and components driving their development are ongoing.
Summer of Code 2006
Project Ideas
Each of the ideas here have been fleshed out to some degree to or
another and are considered useful to differing degrees. If you're
looking for a project worth doing, these are worth looking at for ideas.
Aside from anything else if we've added them here there's a good chance
that they've had some serious thought behind them already, bootstrapping
your idea!
The following list contains the aggregated projects grouped under like
for like ideas. The order of projects here has no
significance
One of the most interesting aspects of this list of projects is that
many of them were proposed by undergraduate students, and all thought
they could achieve these projects over a 3 month coding period, most of
them after having done the MiniAxon
tutorial.
Completed Summer
of Code 2006 Projects
In some cases these were exploratory, some were concrete, some were
half way. All were useful for different reasons.
