Development Team/Project Ideas

From Sugar Labs
Jump to navigation Jump to search

This is a list of project ideas. For now the focus is on giving GSoC students an idea of the kind of thing we're looking for. It also includes dumping ground for less-explicitly-explained crazy ideas at the end.

Priorities: see the ongoing discussion of priorities on our mailing list. To summarize that thread: the highest priority is for you to have a successful GSoC, which should bias you towards easier projects such as #Stand-alone activities. But if we had to rank this list in terms of priorities for Sugar, the #AJAX Sugar aka Karma and #Versioned Datastore ideas are probably our highest priorities, followed by the rest of the #Toolkits / Frameworks for developers. These are generally hard, though, so we would need you to give us evidence that you are capable to accept you for these ideas.

  • There is a similar project ideas page on the OLPC wiki (and a related category). OLPC-specific projects, including hardware- and scoholserver-related projects, should go there. Update: since OLPC did not get into GSoC this year, we would consider mentoring school-server related projects. Feel free to add below relevant projects from that list - perhaps 80% 95% of these could be appropriate Sugarlabs applications.

See something that interests you? To get in, you will need to design your project and find a mentor. On irc (#sugar on freenode) or the sugar-devel mailing list, you can briefly introduce yourself, state your interest, and ask relevant and specific questions about the state of the existing code. You should also do your own research for other open-source code that might help you. Get a development environment installed. We definitely appreciate it if you can show your readiness to help out - either joining bug squad and helping to triage, or actually helping fix some small bug.

When you're ready, figure out a very basic, preliminary design (where does your new UI fit in? what about the code - what talks to what?) and post that to the mailing list, asking if anybody would be willing to mentor you. You will almost certainly get some response, but you may not find a prospective mentor right away. That's OK. If your idea is really not a good fit for us, we will let you know; otherwise, remember that several mentors are holding back for now to see which projects shows the most feasibility, clarity, and creativity in their design ideas. Persistence should pay off.

Want to apply with an idea not on this list? That's fine. Read the thread linked above about priorities - and remember, our highest priority is for you to have a successful GSoC and hopefully continue to contribute afterwards. Do the steps above, paying particular attention to getting some community members' opinions on whether your idea would be valuable. The "iaep" mailing list may be the best venue for this. We will not all agree with each other about how important your idea is - that's normal - but our feedback will almost certainly help you refine your idea.

Template for well-explained ideas

If a project that interests you lacks a "potential mentor" here, or you want to know more about the current status of the related code, we would be happy to help guide you on irc (#sugar on freenode) or mailing lists (technical questions to sugar-devel@lists.laptop.org, educational/general ones to iaep@lists.laptop.org).


==== Project title ====
A quick explanation of the project idea
* Priority for Sugar: Low/Medium/High
* Difficulty (as a GSoC project): Easy/Medium/Advanced
* Skills needed: Experience with WikiCode and copy-paste.


Core Software

Versioned Datastore

  • To add Version support for Journal / DataStore: Start with (old) Olpcfs and (newer; less-documented; based on an RCS backend and a relatively small amount of fuse magic) olpcfs2. Get Sugar to mount OLPCFS2, a working virtual versioned filesystem, and keep its datastore there. Get datastore to create a new version for each save (automatic or manual). Modify journal UI to use these versions, fork from old versions, etc. Keep with the same name / tags, create a branch if metadata was changed. Allow the user to access "older" versions (Keeping and "old" version will create a branch) and view ancestry (tree of branches).

We would not expect a GSoC project to be necessarily ready to check into our trunk. For instance, you could avoid facing the issue of automated pruning of old versions for disk space, or not have a converter for existing datastores. However, it should work as a proof-of-concept with a variety of activities.

  • Priority for Sugar: High
  • Coolness factor:++
  • Difficulty: Hard
  • Skills needed: primarily Python UI (pygtk); also FUSE/file systems (this part is mostly done); and Packaging and building.

Implement existing UI design proposal

Look at Design_Team/Designs and Design_Team/Vision/Proposals, choose a proposal in one of those, and implement it. Obviously, you need to investigate existing work on these by testing/playing with a current jhbuild and by talking on #sugar.

  • Priority for Sugar: Medium
  • Difficulty: easy-medium
  • Skills needed: Python, PyGTK

Registry for people

Extend the interaction model to include real people beyond the user–laptop couple. This would extend the virtual network to include some very significant entities, such as family members, who may not have a physical computing device. See the suggestion submitted by User:Skua. The olpc:Record Activity could be used as a fun, instance-of-person creator and embellisher, by capturing an image or video of the person, and linking it to a new registry.

  • Integration of a person object into the Sugar architecture
  • Extension of the User/Group model to realistically capture the actual Person entity
  • Modification of at least one activity (Record ?) to support the new API
  • Extension of School Server registration model
  • Extension of an Internet person model to support a Person entity (e.g., to support a missing-person registry in the Sahana Disaster Management System)

Lots of extensions are possible, following a good, fundamental design

  • Difficulty: Medium - Hard (depending on scope chosen)
  • Skills needed: Data modeling, Core system programming
  • Potential mentor: User:FGrose for data modeling, collaborators needed for Core systems and Internet architectures

Speech Synthesis for Sugar

Integrate speech synthesis with all activities, not just Speak, and provide for karaoke coloring. See Mokurai's article on adapting Same Language Subtitling for literacy to the XO.

Desirable Features for Sugar Speech Synthesis Plugin:

  • Provide Sugar Speech Synthesis Configuration Management Tool
  • Karaoke Style Coloring in Sugar Environment
  • UI for configuration Control
  • Accent gets set on the basis of locale
  • Priority for Sugar: Medium
  • Difficulty (as a GSoC project): Advanced
  • Skills needed: Experience with GTK, decent Python ability, ability to integrate with existing code.

Print Support

Print support in Sugar would be useful in many scenarios. The ideal project deliverables would include

  • Integration of a printing infrastructure (CUPS ??) into the XO-1 software images
  • Modification of Sugar Control Panel to set up the printer (add/select default printer?)
  • Modification of at least one activity (Write ?) to support printing
  • Making a printing activity, that follows sugar GUI guidelines?
  • Extra credit: integrating a server, including permissions and quota management, into the XS image.
  • Priority for Sugar: High
  • Difficulty (as a GSoC project): Medium-high
  • Skills needed: Python programming, API design, some communications

Sugar Toolbar submenu support

The Sugar Human Interface Guidelines have a toolbar design that includes submenus (See Toolbar designs). The project would be to extend the existing Toolbar widget to include this new feature and then to work with a few Activity developer to incorporate the new design into their Activities. Possible candidate activities include Paint and Turtle Art.

  • Priority for Sugar: Medium
  • Difficulty (as a GSoC project): Easy-Medium
  • Skills needed: intermediate GTK and python skillz

Toolkits / Frameworks for developers

AJAX Sugar aka Karma

Create a proof-of-concept learning multimedia Sugar activity using javascript and html5 (for instance, Titanium-made apps). This activity should have animation, audio, persistent storage of user progress, and at least basic integration with the Sugar environment. For more information see the Karma wiki page.

JavaScript/Sugar Communication through the following strategies: PyXPCom, hulahop, xulrunner or Titanium. see also the mailing list discussion. The preferred solution should use a minimalist rendering engine like Webkit and a toolset for local file access like Google Gears. Such a solution would have much less overhead than embedding all of mozilla. For the moment we are focusing on Titanium.

  • Ideally, develop a demo activity which could be used as a template for sugarizing AJAX activities. The GSoC participant doesn't have to create her own activity but could simply recreate an existing activity such as one of OLE Nepal's flash activities.
  • This demo should have the following features:
    • Simple interactive animation and audio using html5 tags like <canvas> and <audio>
    • An assessment section that stores results of student's progress and gives them suggestions on improvement. Assessment info should be persistent.
    • Has embedded pdf for lesson plan that can be viewed w/in the activity
    • Integrates with datastore
    • Navigation and Help elements, ideally reusing widgets from popular javascript libraries like Jquery, Prototype, Mootools
    • Some element of collaboration using telepathy (This could be really hard, depending on the state of javascript bindings to dbus)

There is some discussion of the prior work on this question at User:Wade/Web_Activity_Spec.

  • Priority for Sugar: Very High ("never bet against the browser")
  • Difficulty (as a GSoC project): medium/hard Note: integrating w/ the datastore likely won't be too hard but utilizing Sugar's collaboration features could be very hard
  • Skills needed: Javascript/Python integration (PyXPCom, hulahop), CSS, knowlege of SQL
  • potential mentors: Wade Brainerd (wadetb at gmail dot com), Bryan Berry (bryan at olenepal dot org) can serve as project manager, define requirements and project deliverables

SWF Sugar

  • Integrate SWF (Flash/Gnash) applications into Sugar.
  • Ideally, develop a demo activity which could be used as a template for sugarizing Flash/Gnash activities.
  • Priority for Sugar: Very High ("never bet against the browser")
  • Difficulty (as a GSoC project): hard
  • Skills needed: SWF/Python integration

Improve Develop activity

There are several improvements that would make the Develop activity a more attractive IDE. Any ONE of these would be a good GSoC project.

  • Make a WYSIWIG GUI editor, like Glade. Note that GTK natively supports loading Glade-format interface definitions, although there would be some work involved making the Sugar interface elements available through this method.
  • Integrate Sugarbot and auto-testing facilities.
  • Integrate a debugger, based on pdb or other.
  • Priority for Sugar: Medium-High
  • Difficulty: Medium - Hard
  • Skills needed: Good python skills.
  • Potential mentor; Jameson Quinn (firstname dot lastname at gmail dot com)

"Translate Activity" activity

We will never finish localizing all our activities and base software for all our deployments - especially for places with high linguistic diversity like Afghanistan, Peru, Guatemala. So it would be great if there were an easy, discoverable way to translate any string on your machine; have the translation appear on your own machine immediately; and, assuming the activity has a link to a Pootle project, upload that translation to a Pootle server later. (For real-world use, these uploads would probably have to be cached at the school server level, but that is more complexity than we'd expect from a GSoC project.) Virtaal might be a good starting point for the UI.

  • Priority for Sugar: Medium-High
  • Difficulty: Medium to Hard
  • Skills needed: at least some experience localizing, to know what's involved; ability to do minor hacks on gettext in C and Python; work with localization formats (.po, etc.); Python for activity UI; some simple communications, to upload proposed translations to pootle.
  • potential mentor: Sayamindu Dasgupta (sayamindu at gmail)

SugarGames Pygame wrapper

The OLPCGames wrapper allows Pygame to run inside of GTK in Sugar, making Pygame based Activities possible. However, the wrapper is geared specifically for use on the XO-1, and does not necessarily reflect the modern reality of running Sugar on diverse hardware. It also does not allow for the use of other GTK Widgets in the Activity, restricting developers from adding UI features like pop up query boxes. This project would involve porting OLPCGames or writing a Pygame wrapper from scratch that would allow Pygame to run in GTK while still enabling the use of GTK UI elements. It would also involve making GTK events and other relevant parts of the Sugar API available to Pygame Activities.

  • Priority for Sugar: Medium-High (Activity development is limited between the graphics limitations of PyGTK and the UI limitations of Pygame)
  • Difficulty (as a GSoC project): Medium-Advanced
  • Skills needed: Requires a fair amount of knowledge of both Pygame and PyGTK.
  • Potential Mentor: Nirav Patel (nrpatel at gmail, nrp on freenode)

Generic linux application wrapper

Improve the ability to launch legacy Linux applications (or Windows applications in WINE) from inside Sugar. The issues are our window manager, our datastore/file system, and our security model (Rainbow). Basically, you'd have to fake all of these from one side or the other so that an average legacy app could live with Sugar.

This is a priority for sugar, and there is significant prior art and design thought. The first part of your job would be to search this wiki, the laptop.org wiki, and old mailing list archives for OLPC and Sugarlabs, and talk on IRC, to settle on a good design. Because of this, we'd consider accepting a strong application even if it left some design questions open; but it would have to at least suggest some basic shortcuts you'd take to make the task feasible, and argue why these shortcuts would leave acceptable usability.

  • Priority for Sugar: High-Very High
  • Difficulty (as a GSoC project): Advanced
  • Skills needed: Creative ability to use and quickly learn various tools, ability to focus on critical path (ie, build something quick-and-dirty that accomplishes the main goals rather than getting distracted by ideal solutions), experience with window manager and windowing systems (Gnome, KDE, etc.).

Stand-alone activities

Improved Read activity

Use Gecko to implement a reader for epub format ebooks. This is superior to PDF because such books can be reflowed to better fit the screen and user preferences. Also, (although it would break the standard) it would make it very simple to include AJAX-style active features to books.

Extra credit if you support textual and graphical annotation. Deployments have also asked for a page-turn animation. See also ml.

  • Priority for Sugar: High
  • Difficulty (as a GSoC project): Medium (w/o annotation); very hard (w/annotation)
  • Skills needed: Strong Javascript/DOM skills, some interlanguage integration (Python/Javascript), ability to adapt Read activity's communications code (Python).
  • Potential mentor: Sayamindu Dasgupta (sayamindu at gmail) (already has some code to start with)

Listen Spell activity

This activity has been last year GSoC project. Its code can be found here. Extending it activity wrt to following points

  • Supporting Speech Synthesis model of sugar (speech synthesis model of sugar is not designed yet and is again a gsoc project this year. This task would involve either using speech-dispatcher or GST-Plugin as of now and later adding the support for sugar model)
  • Multi player game over mesh network
  • User Defined word list. (Adding support to add new words to dictionary)
  • Test Mode: A teacher can feed the pre-defined word list on the network and activity is being used to conduct test/exam
  • Speaking sentences to make student learn grammar (this as a higher level)
  • Priority for Sugar: Medium
  • Difficulty (as a GSoC project): Medium
  • Skills needed: Python, GTK, Understanding of sugar mesh network
  • Potential mentor: Assim Deodia (assim.deodia at gmail dot com)

KDEEdu

Sugarize any KDEEdu activity, especially the ones which have no corresponding Sugar activity. This probably means recoding the C to use GTK instead of QT and to use Sugar conventions. It is doubtful that this process could be automated, so you'd probably just do one activity.

  • Priority for Sugar: High
  • Difficulty (as a GSoC project): easy-hard
  • Skills needed: C/C++, GTK.

Educational Toolkit

Either based on the existing educational toolkit, or starting from scratch, enable XO use in classroom scenarios. Such scenarios could include

  • Teacher shows slides, reproduced on child's screens
  • Teacher asks questions - either pre-prepared or on-the-fly
  • Students give answers via collaboration
  • Teacher or student chooses - explicitly or randomly - an answer for further discussion
  • Students split in groups and go from their individual answers to a collaborative answer
  • Teacher can review all answers later
  • Teacher gives individual or group feedback (offline) which will be shared with appropriate students when they come online
  • Teacher checks what's on a child's screen - (experience on other platforms shows this "look over shoulder" ability reduces goofing off even though it is rarely used.)

The low-hanging fruit on Educational Toolkit is the following:

  • Enable collaboration scenarios
  • Work on the GUI to provide support for multiple types of questions.
  • Add API to make it easy to add new question types.
  • Priority for Sugar: High
  • Difficulty (as a GSoC project): medium-hard
  • Skills needed: intermediate ability with Python and communications

Improved Imageviewer

Implement missing bits in Imageviewer, some of which are

  • Sharing support
  • Basic image effects support (grayscale, sepia effects, colorize, etc)
  • Exif support

There are more things that can be implemented, but the above are the basic minimum one should try to implement.

  • Priority for Sugar: High
  • Difficulty (as a GSoC project): easy-medium
  • Skills needed: Python, GTK. the Sugar collaboration framework


EduKT

Simple content - history creation tool that needs:

  • Re-design drag and drop features
  • Add audio support
  • Simplify the interface for little people
  • Better the saving projects process.

A lot of new features that simplifies to any level the building of interactive contents...

  • Priority for Sugar: Medium
  • Difficulty (as a GSoC project): medium
  • Skills needed: Python, Glade, GTK. the Sugar collaboration framework


Etc., Etc.

It should not be hard at all to imagine educational activities or games which would be useful for primary or secondary school education. Let your imagination run wild!

  • Priority for Sugar: Medium
  • Difficulty: Medium
  • Skills needed: Python, GTK, Sugar collaboration framework

Brainstorm / unexplained ideas

Sugar adaption for the Nasa

One of the 91 indigenous cultures that still exist in Colombia is the Paez people (aka Nasa). They have their own traditions, customs, world view, mother tongue (Nasa Yuwe), i.e. their own culture. It could be possible to take cultural elements into the Sugar Interface, not only language, to provide Nasa children a suitable and familiar interface. Santiago 18:01, 8 March 2008 (EST)

Core Software

  • Accessibility Support: Sugar currently doesn't have anything available for the visually impaired.
  • Improve automatic testing across the system. This would improve our check-in and build process immensely. Very high priority which nobody is addressing head-on.

Homework turn-in

  • "Homework turn-in" support: Certain metadata on a file causes new versions to be pushed out over the net (via SMTP, rss, or other; note that Moodle already has support for routing from special email addresses to a "location"). No new UI in Sugar, and a trivial amount of changes to Moodle.

Research projects: unpolished code

  • There is also Journal, reloaded, another research project with real code behind it that is promising but languishing. In this case, the idea is to make the journal "tagging" view transparently compatible with a traditional hierarchical directory structure.
  • bemasc's groupthink, expanded: The idea is to have a data structure which keeps itself in sync across many laptops "behind the scenes", thus providing drop-in collaboration as long as the structure in question provides the needed functionality. The problem is that the existing code is unpolished, and only supports some pretty limited data structures. I have some ideas of how groupthink could be more general. Homunq 00:43, 11 March 2009 (UTC)

Java

Package and integrate the IcedTea open source bootstrap of OpenJDK Java with browser plugin for the XO. Deliverables would include:

  • Binary, source and rpm dependencies for icedtea and icedtea browser plugin
  • Java enabled os image
  • Integration of packages into autobuild branch

(This is just to get Java into the build. Creating an application framework would come later.)

Graphical toolkit

Important work left to do:

  • Give focus feedback by showing a rounded rectangle in gtk buttons and HippoCanvas icons.
  • Implement keyboard navigation in HippoCanvas.
  • Implement accessibility hooks in HippoCanvas.
  • Improve keyboard shortcuts - make them easier to create and implement a UI to make them more discoverable, such as transparent letters which appear when you hold <ctrl>

Frameworks/Toolkits

Mono/.NET

The use of Mono could really enhance the number of Sugar developers due to the huge existing .NET community. Thanks to Torello Querci, developing a Sugar activity in Mono is already possible using the Mono/Sugar bindings Sugar.dll (more on Mono on Sugar here).

The idea for this GSOC project is to greatly enhance this binding:

  • Better integration with the Sugar look & feel and HippoCanvas,
  • Binding to telepathy API,
  • WinForm compatibility,
  • MonoDevelop integration.

More on this idea:

  • Priority for Sugar: Low
  • Difficulty (as a GSoC project): Medium-Advanced
  • Skills needed: C# programming, Linux programming
  • Potential mentor: Lionel Laské and/or Torello Querci


Activities

VideoChat activity

telepathy-python has support for audio and video streaming and has recently gained support for using gstreamer, which means that we could easily do efficient videoconferencing using fully open source codecs.

So a really nice project would be to do a proper Sugar activity for video conferencing.

Language Trainer

A language trainer with text to speech support would be very nice. Something that could start with letters and then teach words.

Thesaurus

Working together with openthesaurus -- someone could create a thesaurus for kids to learn different words (synonyms and antonyms)

Logo Activity

Logo is a computer programming language used for functional programming. It is an adaptation and dialect of the Lisp language; some have called it Lisp without the parentheses. Today, it is known mainly for its turtle graphics, but it also has significant facilities for handling lists, files, I/O, and recursion.
Logo was created for educational use, more so for constructivist teaching, by Daniel G. Bobrow, Wally Feurzeig and Seymour Papert. It can be used to teach most computer science concepts, as UC Berkeley Lecturer Brian Harvey does in his Computer Science Logo Style trilogy. — Wikipedia article on the Logo programming language

There is a "Sugarized" Logo—UCB Logo—but it does not record data into the Journal or use the standard Sugar toolbar.

There are two possible approaches we could take: (1) digging deeper into UCB Logo and (2) working with another Logo, possibly PyLogo.

  • Priority: high as Logo is an important tool engaging children in programming
  • Difficultly: moderate to high, depending upon the approach chosen
    • Integrating Pylogo would be relatively easy, but it is a very limited implementation of Logo that would need enhancing
  • Experience: some Python and C if the UCB Logo approach is taken

Other ideas for improving Sugar Activities

Broad project ideas

Activities Site (addons)

  • The activities http://activities.sugarlabs.org, is in need of a serious sugarization, a GSOC project could be giving some love to the dressing and coding of the underlaying activities site (based on mozilla's addons).

Packaging for specific distros

Debian

  • Help in maintaining and packaging sugar and activities in debian.
  • Including/adapting debian-edu .debs to sugar

Hello there, I am quite interested in Debian and want to help with this and all other projects. Please contact me (bjoern AT xruby DOT net) if I can be of assistance to the XO project or other things. I will start my PhD studies in April and have previously studied Computer Science. I am highly interested in helping where I can and want to bring the necessary technology to kids around the world.

from olpcwiki 2008

Preeti's list

Hi, I am Preeti, from New Delhi. I would like to get myself involved in this very interesting aspect of the OLPC software development. I have jotted some of my views on the same at:

http://wiki.laptop.org/go/User_talk:59.178.99.172