Talk:Summer of Code/2007

Indication of a font not having a glyph for a requested character

> Indication of a font not having a glyph for a requested character

The following proposal is, for the reasons in the comments appended below, not useful

The "comments appended below" do not appear to be present.

For example, the suggested project could be regarded as for a software module named text_and_font_compatibility_tester which would read a Unicode plain text file and would read a font file and would report back as to whether the font could supply all of the glyphs needed by the text.

No comments have been appended giving reasons as to why such a module would not be useful.

I wasn't the original poster, but ... that effort is a matter of less than a day's worth of work. While it's not a useless function, it doesn't seem like it's worth worrying about very much or funding an SoC project to get it.

--

Educational games

Our federal aids preventation institute has created some games for windows which were quite sucessful. see [www.stopaids.ch]

About the educational games about HIV and AIDS - I would like to participate in it, but do you have some more specific ideas, what it should look like?

twext

WiXi

twexter on a wiki could be called WiXi.. Alexander Gelbukh has agreed to mentor, in case no one else wishes to mentor a student.. is there any way the WiXi project could be included in ideas for 2007 OLPC SOC? Duke 16:45, 21 March 2007 (EDT)

FAQ document

There is an interesting FAQ document.

http://twext.cc/faq.html

The twext system looks very interesting.

glad you like it.. updated http://twext.cc 
now aims to work with wikis.. details:
http://wixi.cc

Question about word order

If the English sentence I have bought a book. is translated into German, the word order becomes something like I have a book bought.. Thus the word order in different. How does the twext system deal with that please?

Having looked further at twext I am thinking that the twext system would simply put the English version as "I have a book bought." with the English words arranged in the same word order as the German words. This is because such a text would have the German words larger and the English words smaller and maybe in a different, lighter colour: the intention being to help a speaker of English to learn German. This would be good as it could help the learner get a feel for how the German language expresses things.

"help the learner get a feel for (construction in new language)"
exactly.. "I have a book bought" may sound either incorrect or poetic,
but "I have a book bought" fairly well communicates idea intended by the 
"correct" construction of "I have bought a book".

a flaw in languange methods may be too much stress on "correctness"
when "communication" may be a much higher priority.. (with repeated
meaningful "communication" in various contexts, "correct" grammar gets acquired)

further, there are two elements to the twext system that can help 
manage variable work order in various languages:

1. chunks can be variably identified, (meaning the twext method can 
parse word-for-word, phrase-by-phrase and/or full sentence).. thus, no single
"chunk size" is correct; variable chunk sizes adapt to various comparisons 

2. mut or "marked-up twext" can evolve to map characteristics,
including grammar characteristics, and communicate such with variable
formating (ie color, face etc)

Questions about tagging the text with grammatical information

Verbs

Suppose that someone whose own language is English is trying to learn Spanish using the twext system. As the text is in plain text strings, it seems as if it would be possible, if someone so desired when preparing some bilingual text using the twext system, to augment the English words with grammatical information. For example, if the English word have appears then that someone could express it, if he or she so chose, as have(1,1) so as to indicate have as in first person singular or have(1,2) to indicate have as in first person plural or have(3,2) to indicate have as in third person plural or have(3,2,m) to indicate have as in third person plural masculine. The third person plural is the same for masculine, feminine and neuter in English, though English uses gender only for people, animals and a few other items like ships anyway, yet the tags could be used with English words if the word translates into the language being learned differently according to gender. Would this be possible using the twext system?

absolutely.. note that the twext method separates each chunk on
a new line.. thus, a simple mut or "marked-up twext" language can 
be developed to add meaning to any chunk of TEXT or twext.. added
meaning can clearly include grammatical mark-up

Nouns and adjectives

I remember once reading some way into a book for learning German and was somewhat concerned that at one stage it was suddenly proclaimed that the way to say some particular thing was more complicated than explained up to then and in fact all of the nouns thus far used had been of feminine gender. I felt that I would rather have had the fact that the nouns being used were all of feminine gender at that stage had been mentioned before I started that stage, not in the stage after that stage, as I had then to unlearn some of what I thought I had learned rather than have reached that stage knowing that that only applied to feminine nouns. I am wondering quite how that would intercat with learning a language using the twext system. If a speaker of English is learning Spanish, would it be better for the English words in a twext page (smaller and in a lighter colour) for nouns and adjectives to have gender information of the Spanish words in parentheses after them, so that the speaker of English has that information available when learning?

the "twext" translation (smaller, lighter, betwixt the TEXT to learn) 
has limited space (often the twext translation wants more characters)
so adding characters (ie gender info between parenthesis) to communicate 
grammar may be less effective than formatting the twext (and/or text) 
with mut variable font, color etc.

also relevant is the output.. twext is designed to work on paper as well 
as on computer monitors.. output via computers enables all kindsa rich 
linkages upon demand.. for example, if a user wants MuT grammar analysis,
they could filter other possible details to focus on a grammar presentation

Distributed e-Voting Software for the OLPC

My name is Ignaico Vergara and I'm searching for a mentor for a mini-democracy enforcement software over OLPC plataform. It is hoped that this instant micro-democracy can be useful as a coordination means in network contexts. Any intrested in mentoring me, please send an e-mail to ivergar1@uc.cl

SoC account content

[1]

OLPC description

One Laptop per Child

One Laptop per Child (OLPC) is a new, non-profit association dedicated to research to develop a $100 laptop—a technology that could revolutionize how we educate the world's children. This initiative was first announced by Nicholas Negroponte at the World Economic Forum at Davos, Switzerland in January 2005.

Our goal: to provide children around the world with new opportunities to explore, experiment, and express themselves.

Please note that the $100 laptops—not yet in production—will not be available for sale. The laptops will only be distributed to schools directly through large government initiatives.

Note that most of the work OLPC needs done needs to be done as part of other existing projects, though there will be some OLPC specific opportunities as well. We expect work done in various projects should conform to the licensing policies of that project; work on OLPC specific projects will likely be GPL/LGPL/MIT, depending on precise circumstances, to be determined at the time.

Application template

Name:
Email:
Phone:
Address:

Organization:

Previous open source projects:

Summer of Code project interest:

Link to proposed development idea and timeline:

Here may be some of the ideas for Google Soc

http://wiki.olpcnepal.org/index.php?title=Extra-Curricular_Activities http://wiki.olpcnepal.org/index.php?title=JotPad http://wiki.olpcnepal.org/index.php?title=Basictivity

Using the camera as control device

A nice SoC project would be to build an API to use the XO camera as a game input device. A few ideas come to mind:

• Detect one or more regions of distinctive color (a ball, a sheet of paper) and track its position as an X-Y input device (possibly even Z if you can take size into account). User could then move the object in space to provide input to a game.
• Detect motion of the image background so as to infer the motion of the XO itself. It would be a substitute for a tilt-swivel sensor and could be used to play games like Marble Madness and Super Monkey Ball.
• Detect motion of the image background so as to provide simple dead reckoning position tracking. This would be unbelievably crude but it might be possible to use a down-facing camera to make some sort of hide and seek game enabled by the XO.
• Detect faces and/or hands like the Sony EyeToy. Face tracking can provide a nice input model.
• Capture a background with no user involved, then notice the changes as the user moves in the frame to return an outline of the area that's "different" (this is the "Shadow Garden" model).

It's unlikely that all of these ideas would be implemented, but a high quality subset of them would be nice to have.

The API must include Python support, and be capable of operating at "reasonable" frame rates (preferably at least 10 fps).