We already have a large number of activities in GCompris (about 120) and we are always looking at adding more. An important aspect is to keep the project focused on its core value: educating young children.
Each activity requires a good thinking of the educational aspect including a progressive difficulty, it must be easy to understand what we expect from the children by providing instruction and in game guidance.
If we consider the activity list, currently our biggest weaknesses are in the area of reading/writing and language.
In GCompris all the activities follow the same core values:
We started to rewrite GCompris in the Qt Quick technology. This will allow us to bring our work to tablet users. Here is a quick startup guide to start developping in the new environment. We maintain a page with the status of the port.
This is currently the most important task.
Wims is Web exercice server and HTML player. It contains a lot of mathematical exercices defined in a format named OEF. We can imagine a GCompris activity that can play some Wims exercices.
A very good resource of in class activities introducing computer science
For young kids, we could create an activity where they must classify object from larger to smaller, and so on. The size classification could be based on the real-life sizes. For example, a car is larger than a light bulb, but the actual images could be equal size.
We could also have classification based on shape, e.g. round apples and wheels, and rectangular books and doors.
Could teach also "in front of" and "behind" concept.
A symmetrical drawing activity, with various forms of symmetries. See this example.
Teach chromatic numbers to children. What is chromatic number: http://jdh.hamkins.org/math-for-seven-year-olds-graph-coloring-chromatic-numbers-eulerian-paths/
So, maybe a game with a drawn mesh and circles at the intersection of edges. No two circles connected by an edge can be of the same colour. We can start of by taking the largest chromatic number possible for any graph, and then slowly increase the difficulty by reducing the chromatic number.
We already have one in this area based on the puzzle activity engine. But is is not flexible enough and it would be better to have an activity dedicated to this concept.
In that activity, we could add :
- filling incomplete tables. The filled stop may be done by selecting images or by entering text or number. For example, it may also be used to try the children on multiplication table. - choosing the right heading for rows and columns of an already filled-in table
Of course we have to make it much simpler and stick to point, line, segment, perpendicular, parallel and circle.
This can be used: - as a free geometrical drawing activity - as a directed one, the children is given a set of instructions like draw a 3 cm diameter circle at position (3,3).
There is a possibility to save and load a drawing.
Like the drawing, animation, wordprocessor, we could provide a graph tool to let kids create simple graph for their experiences. They could load/save them. We could see it as a simplified spreadsheet. Here is an online example.
An activity to explain the different base. It present the concept, let the children play around with a virtual base counter and then ask questions to transform numbers from one base to another.
This is a Sudoku like activity where the goal of each puzzle is to fill a grid with digits –– 1 through 4 for a 4×4 grid, 1 through 5 for a 5×5, etc. –– so that no digit appears more than once in any row or any column (a Latin square). Grids range in size from 3×3 to 9×9. Additionally, KenKen grids are divided into heavily outlined groups of cells –– often called “cages” –– and the numbers in the cells of each cage must produce a certain “target” number when combined using a specified mathematical operation (either addition, subtraction, multiplication or division)
There is also a color based version.
Should be a good idea to find a way to generate some games with different level of difficulty.
We need something to introduce algebra to children. We can have a look at DragonBox which has a good educational approach.
We could have an implementation of this game. Addition exercises can be solved using various strategies. The Number Catcher takes its player, step by step, from basic calculation up to adult strategies.
Help students develop an ear for pitch by helping them notice differences between pitches (high vs. low, etc.) This is a very difficult activity for kids (and adults!) to master, so this activity will have to be very forgiving.
A beginner guide to the basics of music theory. Topics might include music notation, scales, chords, circle of fifths, etc.
We could implement a manipulation game aimed to represent numbers with hundreds, tens and unity representations. A good start to understand the concept is to look at the following app
Here some ideas to steal. TicTacToe and others variants of same game (QuinzeVainc, Sentier, Mots and Lettres].
Ultimate Tic Tac Toe is also a fun game, for older children - http://mathwithbaddrawings.com/2013/06/16/ultimate-tic-tac-toe/
Having a dominoes activity. It would be very nice, and maybe not too difficult to write (I don't speak for myself...) Easy levels could be with colors only, then images, medium level, images and dots, high level, only dots.
Nine Men's Morris is an abstract strategy board game for two players that emerged from the Roman Empire. The game is also known as Nine Man Morris, Mill, Mills, Merels, Merelles, and Merrills in English.
This game is still very popular in some place of the world. Having this type of game in GCompris brings the opportunity to expose children to games they won't find otherwise. It would complete our board activity besides chess, oware, connect4, ... This idea has been proposed by Phanindra Ramesh Challa.
Checkers is a board game in the spirit of the chess game. It would be nice to have one beside our chess game.
With the new Qt Quick port a learning activity for an Android's (or iOS based) virtual keyboard could be added. This would have to be aware of the special constraints of the virtual keyboard logic on mobile devices, such as necessary confirmation by "Return" special handling of non-ASCII characters (german umlauts e.g.) etc.
Some of the classic keyboard or writing activities of GCompris are difficult or not at all to port to the virtual keyboard logic of an Android device. New ideas will be collected here: ...
Its a classical in terms of learning letter, an image is presented to the children and he must click on the keyboard the starting letter of the image (like Apple = A). At this stage we don't check the case.
At the first stage of this activity, all our images with the voices are introduced to the children. He can see the associated image, word and letter and click on them to hear the letter and word spoken.
Add a new type of memory in our memory type activity in which the children is propose card with upper and lower case and he must associate them.
A set of letter is spoken, the children must write them in the correct order. As the level increase, there are more letters and the spoken delay between them decreases.
A word is displayed and/or spoken the children must re-type it. Then the word is spelled letter by letter and the children must type it.
A letter is written and/or spoken. Some words are displayed, the children must find the word or the words in which this letter appears.
Several words are displayed, the children must find the letter which is common to all of them.
The children is proposed a set of letter and an entry area. He must enter all the letters in the alphabetical order in the text area.
A text story is displayed, the children must read it. Once done, a number of questions about it are are asked. The texts may be simple and written by ourself or taken from public domain famous texts.
There are several levels depending on the text difficulty. Also at first the text and the questions are visible, then when the children has completed the reading the text disappears and the questions are asked.
A text story is displayed, the children must read it. Once done, another text, related to the first contains holes and must be filled by the children.
A sentence is proposed with the words shuffled. The children must put back the words in the correct order.
A sentence is proposed in which spaces have been removed. The children must add then.
Like we have an activity to learn letters, we must have one to learn phonemes in each language.
We could make an activity in which a phoneme is spoken and the children must find the word which includes it from a list of words.
An activity focused on grammar questions and reading. The idea is to provide a simple wiki like syntax to let a teacher propose this kind of exercise. For example one exercise could be: Where is the [little|big,red] cat (here the question is find the opposite. In  are the answers to propose, the first one is the good one).
A set of activities in the area of language learning. These activities should help a children learn or discover a foreign language. To achieve this, we can base this work on the shtooka project which provides a large set of recorded words.
We already have an in progress version of an activity dedicated to language learning. It is named _lang-activity_ and is organized in lessons having each a bunch of related sentences with their voices counterpart. When they are all presented to the children some exercises are proposed to help the children remember the sentences.
Extend GCompris core to include the concept of a target language for that this can be done from a single place. We currently have the ability to set a target language in some relevant activities, this must be changed to use the new target language facility.
In order to create reading, writing or language learning activities we need to have a well formatted list of words for each language. Some work has been done as seen in this page: Wiktionary_to_XML.
We already have an activity to learn how to read on an analog clock. This could be enhanced to learn how to tell the time in a given language (like 10 past 4).
At the moment, the system pulls the list of letters for letter games from a single line in the po where all letters are added in a row (abcdefghi...).
This works ok for languages which only use monographs (like a b c d e) or precomposed complex letters (like ď ť in Slovak) but it's problematic for languages which use di or trigraphs which are not precomposed (a digraph being a combination of two letters which is seen/treated as a single letter). This affects a lot of languages - even though not all potentially regard them as individual members of the alphabet (like German sch, English ch).
Nonetheless there are languages where these are handled rigidly as single units, such as Welsh ll, dd, ff or in our case (Scottish Gaelic) bh, ch, dh... and where the absence of a distinction between (for example) b (phonetically close to /b/ and bh (phonetically /v/) leaves a huge gap in the system.
We must extend GCompris to support this by adding delimiters between the letters in the po file, e.g. a|à|b|bh|c|ch.
We can make a new section aimed to introduce children to environment care. Some ideas of activities are:
One simple activity to show the major color components of the white light. One idea is to make a cloud that rains by clicking on it and a raising sun. It will create a rainbow. Maybe a zoom to each individual drop can be done in order to show the incoming white light and the major outgoing colored beams.
The idea here if to let children take care of a virtual garden. The children have a garden and some plants in it. They must take care of it with actions like :
Help the children understand the different concept used in photography, focal, focus, iso, ... We could let the children take a picture and simulate the rendering given the children's virtual camera settings.
The morse code maybe a funny way to discover a lot of concepts. And it would be really fun to be able to discuss with Tux this way. It can be done in 2 side, Tux talks in morse and the children have the alphabet and must report the text. On the other side, the children write a message to Tux (it is a little bit more complex to implement).
We already have an electricity activity, it works fine but it could be improved in a number of ways:
Having a goal would be interesting and not necessarily too complex. For example we could ask the children to:
In the Gtk+ version we had a nice animation activity. Children or teacher may be interested in publishing the creation of the children in the animation activity. A Web export is a nice feature to have.
Let the children record her voice and to associate it to a given animation frame. i.e. a car moves along a picture of a road and create a voice over of "the car drove down the road".
We have a nice owale activity. But it is hard to understand because we don't see the computer moves. An animation would help so that children see and understand what's going on.
Our photohunter activity does not have enough levels.
We already have a lot of activities to discover how to use the computer. We also need to expose the computer internals to the children. Perhaps an activity to virtually cable all the main component of the computer would do it. Starting with mouse / keyboard / PC / screen at first level to the processor and memory at the higher level. For each component, a description must be given to explain what it is.
A network is an interesting topic to expose to children.
For example we coud create an activity in which we see some computers (school.net, gcompris.net, mycomputer.local) and a switch on the screen with their services (ping and http). Then from a simulated console the children can enter commands like 'ping school.net' and the network packet moves over the wires. For a web page, we can see it on the disk of the computer school.net (the children can edit it), then the children can start a simulated browser and load http://school.net/mypage. In http mode, we see the GET request going out of the browser, to the web server, to the disk and back to the browser with the file attached. The services can be started/stopped by the children but to make the game fun they maybe stopped rendomly to let the children fix the network.
Some ideas on how to improve the heart of GCompris:
It would be nice to have some sort of educational network game that would involve a team competition, collaboration of sorts.
There brings different types of activities:
Activities to make children understand the workflow by some small decision making constructs. such as .