Difference between revisions of "GSoC Ideas 2019"

From Audacity Wiki
Jump to: navigation, search
(notes.)
Line 179: Line 179:
 
* 'true/false' synonyms.
 
* 'true/false' synonyms.
 
** Many functions have bool parameters.  A function call like DoUpdates( mMessage, true, false, false ); is not self documenting.  Fix it.
 
** Many functions have bool parameters.  A function call like DoUpdates( mMessage, true, false, false ); is not self documenting.  Fix it.
 +
* coding standards
 +
** Apply Google's C++ coding standards to the naming of variables in the main code.
  
  

Revision as of 12:32, 15 February 2018

Google Summer of Code (GSoC) is Google's program for promoting Open Source Software development. Audacity was a mentoring organization for five students for Google Summer of Code 2008, and mentored two students in 2009. This page contains our ideas for suitable projects for 2019.
For information about our future plans and about Audacity software development, please join our developers mailing list
 
Related article(s):


Introduction

The Audacity audio editor is downloaded around 18 million times each year. It's a large, mature and clearly popular open source project. In 2008 and 2009 we participated in Google's Summer of Code, with 5 students in the first year and 2 in the second. If we participate in 2019 we will likely try to mentor two students, so that we can give them enough attention that they have both challenges and every chance of success.


In many ways Audacity has 'grown up' between 2008 and now. We've become more quality conscious. We've become more aware of both the need to and how to bring on new developers. Because of this we've changed what we ask for in project proposals. Your project proposal *must* be presented as a collection of smaller features each of which is useful in itself. It's no good proposing one single big change which can't be trialed by end users during the GSoC period. A mix of smaller related features - along with related bug fixes is good. If your proposal is different to that, if it can't be broken up into smaller useful pieces, if it does not include any bug fixing, or if the pieces are not related and working as a whole, you'll have more trouble convincing us that it will be successful.
We *expect* incremental development, with near-daily check ins to GitHub of work in progress, and work that is ready enough that your mentor can easily try it out. In return, mentors will be available to guide and give feedback. We work by email, not IRC nor video chat. It's what works for us.

Idea Seeds

In past years we provided pretty well fleshed out proposals as starting points.
  • We expected applicants to provide more.
  • We found that some students just echoed our proposals back to us, making it hard for us to tell if the student had any knowledge or interest.

This year we are providing seed project ideas. They ARE fairly well worked out by us, BUT to apply with one of these ideas you will need to do some work too. That's the 'show us you can...', though you are welcome to find other ways to show us that you can do the project.

  • To grow the seed you will need to hunt through our code and read Proposals on this wiki.
  • We have a developer section on this wiki with information to help you find your way around.
  • You will need to compile Audacity, and show that you can code.


Personality Matrix

Low Risk Big Payoff Notes
Core Refiner Miner Big Payoffs tend to come with higher risks that the project not succeed
Fun Explorer Magician Core work tends to be harder and on older code, and less fun than work on the new edges.


Seed Project Ideas

Name Personality Type Risk / Reward Core / Peripheral Speciality? Idea Desired Outcome Show us you can...
Nouns and Verbs Refiner Low-Risk Abstraction Architecture Audacity classes often mix data (nouns) and functions (verbs).
  • Propose and implement specific refactoring that enhance the distinction.
Code that can be better combined combinatorially
  • Show us which classes would benefit most from this and say why.
Redux Miner Big-Payoff Core Maths Find ways to code Audacity at a higher more mathematical level - that are 100% compatible with our existing code. More Audacity for less code.
  • Show us where we have repetitive code that could be reduxed.
Build Systems Explorer Low-Risk Fun SCONS or CMake Fix and document our build system Less time wasted on build-system gotchas.
  • Explain to us what would change with the new system.
RTL Refiner Low-Risk Core An RTL Language Get Audacity working well with a Right-To-Left language. Support for Hebrew or Arabic
  • Show us which classes would benefit most from this and say why.
WIT Explorer Low-Risk Fun Integration Enhance the (MIT Licensed) WIT system that we use in-house. Better tools for developers and support, using Javascript.
  • Spell out how the enhanced system would save us more time or enable us to do more with the same effort.
Hard Test Automation Magician Big-Payoff Fun Test Identify the hardest things to test automatically, and propose and implement solutions. Increased confidence / perfection in Audacity.
  • Show us something you could do in this line.
Bear Traps Refiner Low-Risk Core User Interaction Fix our various (User) Stuck-in-a-mode Issues. Simpler less sticky GUI
  • Organise the bear-traps showing your preferred option for fixing each.
  • Fix one to show you can.
BlockFiles Miner Big-Payoff Core SQL Offer SQLITE as a BlockFile alternative. No more fragmented projects.
  • Show where it would slot in to existing code.
Wavenet Magician Big-Payoff Fun Machine Learning Offer GUI enhancements for working with machine learning
  • This is for researchers more than for end users.
  • The features DO need to have relevance to ordinary users too.
Text-to-Speech and Speech-To-Text GUI support. Present a plan that shows:
  • What GUI features would be provided, and why.
  • How GUI would be developed and tested.
Error Analysis Refiner Low-Risk Core Cause Analysis Find patterns in the kinds of bugs we report. Implement measures to fix categories of errors, and avoid them in future. Issue avoidance.
  • Do a small part of it already.
Headroom Explorer Low-Risk Fun Metering Rather than detecting failure when it happens, design and implement measures that monitor and report on headroom. Earlier warning of issues.
  • Set out a plan of what you would measure, and where in the code you would measure it.


Notes on micro tasks

Small tasks that give familiarity with code

  • 'true/false' synonyms.
    • Many functions have bool parameters. A function call like DoUpdates( mMessage, true, false, false ); is not self documenting. Fix it.
  • coding standards
    • Apply Google's C++ coding standards to the naming of variables in the main code.


Past GSoC Ideas

These will all move to More GSoC Ideas

Subheadings for each idea:

Possible Mentors These are, in order, the most likely or best mentors for this project idea.
Skills For nearly all project ideas, wxWidgets and C++ programming are essential. Some projects need additional skills too.
Difficulty 'Easy', 'Moderate' or 'Hard'.

Hard tasks can be made easier by solving a simpler problem. That's a decision that needs to be made early on. 'Easy' problems are lower risk. They are better suited to being done in separate smaller pieces if the going gets tough. So take the 'difficulty' grading with a grain of salt. It's only a guideline of how hard we think the problem is.

Early Spinoffs We regard it as vital that projects have early spinoffs that can be completed well within the time. These early spinoffs help to ensure that the code is useful to us. We don't want to end up with 'almost complete' code that we cannot use!


Your Proposal

Possible Mentors:

  • TBD

Description: We've found that the best proposals come from students who are passionate about some feature that they want to make happen in Audacity. We're very positive about suggestions. Bear in mind that if it is a big new feature it will almost certainly need to be in a plug-in. You'll need to have something that is usable by the mid term. To be sure of that you might be best off scaling a big proposal down a little and adding to it smaller pieces such as bug fixes and refinements to existing features which you are sure of having done by the mid-term.

Skills:

  • wxWidgets and C++

Difficulty:

  • Easy, Moderate or Hard depending on your proposal.

Early spinoffs from this work:

  • To be specified in the proposal.


Oscilloscope Mode

Possible Mentors:

Description: Improvements to scrolling and zooming behaviour for close up work. In particular in Oscilloscope mode the 'play' or 'record' line will stay still and the wave form scroll past. At high zooms this will only look good if the section to show with each refresh is chosen carefully, picking up in some easy to compute way on some regularity in the audio sounds. Some performance improvements in drawing the graphs would help make this mode possible on lower powered machines.

Skills:

  • wxWidgets and C++
  • Understanding of how oscilloscopes 'trigger'.

Difficulty:

  • Easy to Moderate.

Early spinoffs from this work:

  • This would no way be enough on its own for a complete proposal. A viable proposal would have multiple parts, and some features would need to be usable by end users by the mid term.


Audio Diff GUI Plug-In

Possible Mentors:

Description: Ability to compare and align two sound sequences just as one compares text using diff would be a powerful new feature in Audacity. It would greatly facilitate the combining of sounds from multiple 'takes' of the same track. It would also be of use to people looking to identify particular known sounds, e.g. repeated themes in birdsong, in very long recordings.

We proposed implementing Audio Diff including a GUI for it as a project idea for GSoC 2008. It was the most challenging of the ideas offered then. This year the focus is on the GUI side of audio diff. An existing audio diff algorithm could be ported or a very very simple one used as a demonstrator. The key extensions are in the interface so that diff becomes a useful tool. The interface will probably need to extend what we already do with multiple clips on one track. It is likely there will be a 'dotplot' display mode and an 'alignment' display mode.

Skills:

  • wxWidgets and C++
  • Familiarity with some diff algorithm

Difficulty:

  • Moderate to Hard.

Early spinoffs from this work:

  • 'alignment mode' complete and usable by end users at the mid-term.


Vowel Quadrilateral Plug-In

Possible Mentors:

Description: The vowel quadrilateral is one step in making Audacity more useful as a tool for foreign language learners. Long term we want to make Audacity part of a language learning eco-system. The vowel-quadrilateral project would also act as a driver for creating a real plug-in architecture for Audacity. Part of the project would entail converting some existing features, such as the karaoke display and cleanspeech into plug-ins. The project could be expanded to involve some collaboration with RockBox to work on mark-up of audio for language learners.

Skills:

  • wxWidgets and C++
  • Must develop on both Windows and Linux.

Difficulty:

  • Moderate.

Early spinoffs from this work:

  • Karaoke and cleenspeech mode as plug-ins. Preferences panel supporting plug-in preferences.

Custom Developer Tools

Possible Mentors:

Description: The key ingredient in this proposal is software that make Audacity developers more effective. This is seen as being of strategic importance to the future of Audacity. Some possibilities include:

  • Collaboration with wxFormBuilder to make a theming tool for easy changing of the skin of Audacity.
  • Collaboration with Melange team to improve Melange as a developer support tool based on the very specific needs of Audacity.
  • Enhanced built-in screenshot tools for documentation.
  • Enhancements to Doxygen output modes so that we can present more relevant and more dynamic information about Audacity architecture to potential developers.

Skills:

  • wxWidgets and C++. Possibly GAE and python too, or PHP depending on project details.
  • Prior experience on project(s) of similar size to Audacity.

Difficulty:

  • Hard.

Early spinoffs from this work:

  • One new tool that is already useful at the half way stage.

Hierarchical Track Panel

Possible Mentors:

Description: An experimental start has been made on this, but it needs to be taken a lot further. The key feature is to have a tree-like structure for a project with sub-mixes within the tree. The developer will need to break the TrackPanel class into smaller classes to get this level of flexibility.

Skills:

  • wxWidgets and C++

Difficulty:

  • Moderate to Hard.

Early spinoffs from this work:

  • To be specified in the proposal.


More Ideas

See Release Checklist for an understanding of what we need to do...