Ideas for a computer science Master's Thesis project?

Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
Number of replies: 38
I'm shopping for (trying to make up my mind, that is) for a Master's Thesis project, and would like base it on Moodle. My work already involves a lot of Moodle specific customisations, so I don't need financing, just a direction smile

One thing that I already started was a metadata based material repository. There exists a few of these, but my angle was to provide a dead simple upload (no required fields, just the file), extract some of the metadata fields semi-automatically from the material (google way, but obviously on a miniscule scale only) and then emphasize the re-structuring and labeling/tagging of the material. A bit like flickr meet's wikipedia... just a tiny bit smile But it got buried under all that metadata.

Another, a bit smaller, thing is an export to a visual format. There was a small scale implementation of this announced here in April, but I'd like to take it much further, exporting a Moodle course to a full standalone web site, with the more interactive parts as separate backup's/downloads and so forth. Maybe some of that metadata stuffed in here and there too for academic sugar.

The "some small projects" page mentioned Moodle wide search. There might be something there, just need an angle other than the obvious practicality, that's the pain of these academic things.

Also the My Moodle is being developed, maybe refactoring the forum posts and other Moodle resources for a personal take on the personalisation/blogging/My Moodle...

As it's my Thesis I can't co-operate in the actual code. I'm not pushing it to anywhere, but would it not to be a complete waste of time (other than graduating of course, which is a definitive plus smile ).

Edit: I posted on this forum as I'm looking for a coding task, not social innovation evil
Average of ratings: -
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Gavin McCullagh -
This is probably way beyond crazy.  It's also possibly not the sort of thing you're looking for in MSc thesis.

I've found that a lot of our lecturers are just taking their course notes in PPT format and banging them up online for download.  It's better than nothing but it's hardly making the most of Moodle.

I was kind of thinking about whether there might be any nice way to draw something like S3 into Moodle such that one could produce presentation notes from within Moodle.

Can anyone think of a way to do this?

Gavin
In reply to Gavin McCullagh

Re: Ideas for a computer science Master's Thesis project?--off topic reply

by Chardelle Busch -
Picture of Core developers
Gavin,

Please have a look at eXe.  It's an OS program that creates SCORM packages.  The developers (great guys) have been working closely with Moodle to ensure compatibility, etc.  It simple, free, and creates great interactive presentations.
In reply to Chardelle Busch

Re: Ideas for a computer science Master's Thesis project?--off topic reply

by Gavin McCullagh -
Looks extremely nice.

However, what I'd like, if at all possible, is something which could be used without local workstation install. 

So, for the sake of argument you could go into Moodle, click "add a resource" and you would pick something like "slide presentation".  This would then walk you through creating an S5, Scorm, etc. presentation.

I know that's a lot to ask but this was a pipe-dream after all smile

Gavin
In reply to Gavin McCullagh

Re: Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
Integrating Eric Meyer's S5 would not be too hard, as it's not an editor but a collection of HTML/CSS/Javascript tricks to mimic a slide show with the Web browser. The actual editing of the "slides" would have to be done with an HTML editor and a bit of manual coding.

It can be demoed (is that a word?) in Moodle by uploading the .html + .css + .js files that would implement an S5 presentation to a course file space and then linking to the HTML page.
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Michael Penney -

Hmm, the Moodle lesson module already does slide shows, with optional menus, branching or linear, streaming media narration file (in 1.6), and dependencys. Examples: http://wcln.org/cms, http://hop.humboldt.edu.

Why another tool to duplicate the lesson module is already capable of? Esp. when static slides shows are made much more effective as learning tools by embedding questions in them (see Dr. Richard Mayer's Practice Principle).

On the other hand, better integration of the quiz question pool with Lesson (a process under way) would be very nice.

Another nice way to make lessons easier to author we've been kicking about is lesson authoring from withing Word/Open Office via tools already available for blogging from Word.

In reply to Michael Penney

Re: Ideas for a computer science Master's Thesis project?

by Gavin McCullagh -
Hi,

sorry.  In my ignorance, I had not seen the full potential of the lesson module.  Would it stretch as far as to allow a full screen (powerpoint-esque) show?

Gavin
Still a little wet behind the ears when it comes to using Moodle from a teacher's perspective.
In reply to Gavin McCullagh

Re: Ideas for a computer science Master's Thesis project?

by Michael Penney -

Hi Gavin, full screen is not built in yet, it shows in the browser interface. So that would have to be something you build insmile.

The code the quiz uses for secure quiz mode could probably be used for this, put lesson into a full screen, toolbarless browser and you have full screen.

Lesson contains a link back to the course in it's optional left menu, and at end of lesson, so students can get back out of the presentation when they are done or during it (or you could put a button in full screen mode to return to the course).

Lesson in 1.6 will also have stylable buttons, an optional progress bar, hiding the left menu until the first run through, and we're also working on a print function like book's and export stats to excel.

In reply to Michael Penney

Re: Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
The Book module works quite nicely for static slide shows also. I took Gavin's request as to build a visual slide builder of some sort, with drag'n'drop boxes, dynamic box connectors and so on. The transition effects themselves people could do without...
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Gavin McCullagh -
What I'm looking for is totally unreasonable but I thought I'd mention it. 

I would like to be able to tell lecturers that they can bin Powerpoint and use XXXX within Moodle which will produce a slide presentation they can use in their lectures and give nice notes visible within Moodle at the same time.

Gavin
In reply to Gavin McCullagh

Re: Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
Do you mean with the visual effects and all? Or with the dualism of the "screen stuff" and the "projector stuff"? I mean what characteristics from Powerpoint are you thinking of, instead of just "plain" HTML slides?
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Gavin McCullagh -
smile  I'm really not sure now.  Sorry to be so vague.  I'm sure there are some lecturers who will demand everything powerpoint has and it simply wouldn't be possible.  Then again if a reasonable presentation could be created (as say S5 can), it might tempt a lot of them to be able to use something simple that would work for both.

Perhaps the other guys are correct and the lesson is enough, I just need to spend some more time with it.

Gavin

PS I've been trying out the OpenOffice SWF export and it's pretty nice actually, except I can't yet see any way to go backward in the presentation smile
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Michael Penney -

I've always hoped for a 'drag'n'drop' perhaps flow chart like authoring system for lesson.

It can do linear or branching or linera/branching presentations, but currently the lesson authoring system is all linear. Something like Audun Hague's javascript course layout tool for authoring and arranging branching for lesson would be awesome and we would help someone working on this all we couldsmile.

In reply to Gavin McCullagh

Re: Ideas for a computer science Master's Thesis project?

by Michael Penney -
Lesson in 1.6 also imports Powerpoint HMTL files, though this could use further work (various versions of office generate different css/js/html, which causes problems). It would be great to have a way to convert PPT directly to lessons, or to convert PPT to Flash automatically.
In reply to Michael Penney

Re: Ideas for a computer science Master's Thesis project?

by Gavin McCullagh -
It certainly would be great.

OpenOffice Impress can open PPTs with reasonable reliability.  I believe it can then export to flash but I can't say I've used this feature extensively.  Certainly this would be a very nice way to do it.  I wonder how printable such a flash applet would be.

If it could be done without need for OOo that would be quite something.  There is an OSS tool called ppthtml but I don't think it's really up to the task.

I like the sound of a Powerpoint HTML import.  One of the problems is that Powerpoint is only in Office Professional (AFAIR) and many students don't have it.

Gavin

In reply to Gavin McCullagh

Re: Ideas for a computer science Master's Thesis project?

by João Almeida -

Hi,

You can use OpenOffice Impress which is the equivalent of Microsoft Powerpoint - for free! It can output to Flash, but better still as an idea for a Master Thesis would be to write a filter or converter from OpenOffice Impress directly to Moodle's lesson or book format. The icing on the cake being that OpenOffice's file formats are ... well, open smile.

Stretching the idea farther why not write a filter or converter for all OpenOffice apps that make sense? I believe that opensource LMS Ilias already uses OpenOffice to create course content.

João

In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Bernard Boucher -
Hi Samuli,
I take my chance to the Samuli power programming lotterywink

I think we need a versatile block or module that will ( in IMS terminology )
2.7 Reduce Content in Learning Path Based upon Learner Profile and do 2.10 Adaptive Learning Delivery .

We need also to don't overcomplexify the authoring process.

To address these two issues at same time, I have imagined and started to create and test a knowledge based material repository that will permit from simple authoring to generate a lesson with many branches that should deliver adaptive learning.

The knowledge is based on a modified ( 11 levels ) Bloom's Taxonomy.

The challenges are:
  • testing the correctness and completeness of the 11 knowledge levels
  • programming interlevel correspondance to support students when they will have problems with higher levels concepts the system will route them to a lower level auto-evaluation exercices until they mastered that level.
  • choosing the most appropriate form(s) of the generated material:
    • lesson module
    • and/or the SCORM
    • and/or the Reload editor
    • ...
Here are some posts illustrating my words: 1, 2, 3.

That kind of tool should differentiate Moodle from others LMS.

All others Moodle functionnalities must be available and untouched by that add-on.

I hope I got the winning ticketwink

Thanks for the offer to the Moodle community and good luck with your thesis,

Bernard



In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Jussi Hannunen -
The two most interesting (to me smile) developments in Moodle future plans are Web Services and "some repositories integrations". I have no idea how far along the official developments are, so it would be pointless for me to list what I have been thinking about Moodle & Web Services. I would hazard a guess that Martin would welcome all serious help.

Your material repository idea could be a type of an integration between Moodle and a digital repository like Fedora or DSpace.

AJAX methods have generated some buzz; a master's thesis might be suitable vehicle to test them out. Edit-mode of the main course page re-done with AJAX might be nice. An AJAX chat-client comes to mind too.

That's also pretty much the list of my current Moodle interests, so be sure to be in contact, if any of those look like winning ideas. I'm trying get going some undergraduate student projects on them at TAMK.

In reply to Jussi Hannunen

Re: Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
Currently I'm kind of thinking of doing a very limited repository, then some expoting/integrating to Moodle modules to achieve different goals. I'm also thinking of wrapping the thing under the theoretic framework that Bernard was talking about.

One of my goals here is to get to know the theoretical and technical mechanisms of repository integration better. Then maybe trying to get some money to do a project for a larger scale repository integration etc.

About the AJAX: it would be a full circle for me, as when I first toyed around with the idea of actually graduating (several years ago blush ), I talked with my professor about doing the MSc Thesis about "improving web application usability with asynchronous data transfer". One thing was to reconsider the traditional MVC design pattern to comply with the fact that this thing (later also named as AJAX) would require for some of the logic to be replicated on the client. Best use for this would be deletion: no need to reload anything, just hide/remove the line with Javascript and then tell the server to delete. Error handling is quite tricky though.
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Bernard Boucher -
Hi Samuli,
even if I have not get the winning ticketwink, only writing about it is a good therapy for my old brain! ( I hope the post will not be too long )

I just found an english reference about the taxonomy that I want to use: Copilot.
( Christian Barrette ( an old moodler ) and J.P Regnault , 2 Québec teachers started that research 15 years ago )

Last year Timothy took time to read the french version now that will be easier for those interested.

Here are some screenshots:

The 13 ( not 11 ) relationships grouped in 4 levels.

.

Primitives forms and examples:



.

The others possible forms of presenting knowledge from the primitive form:

.


The idea here is not to reproduce Copilot in php or to replace any part of Moodle. The idea is to use Copilot relations to express and store some form of knowledge in Moodle.
From these relations, it will be possible to generate quiz, lessons, glossaries,...

The work of the teacher will be simplified: not creating lessons or quiz, just defining relations with simple text autoring or with a nice interface as Copilot provide.

For the student side: lessons or quiz with different types of feedback in case of misunderstanding:

1 - In a particular domain, the system will be able to go one level down in relations to help the student understanding higher concepts. ( adaptive )

2 - The system will be able to find examples ( if they exist ) of succesfull attempts of the student on an other concept using the same relation in an other domain or in another course. Then the similarities of that successfull atempt with that current one not succesfull to help understanding will be presented to the student to help understanding. ( adaptive and personnalized, using only recognition memory )

3 - The old attempts of the preceding paragraph may be use to simulate a pseudo Socratic dialogue with the student, encouraging him to recall the usage of that relation in an other domain.

If he recall them the socratic approach will ask to identify the elements of the correct attempt and to find correspondances with the currect missed attempt.

If the student don't recall then the Socratic will present some old student succesfull attempts with one with the same relation use and the others as distractors showing reals attempts but with others relations.
( adaptive and personnalized, using more evocation memory )

Have a good cogitation,

Bernard

In reply to Bernard Boucher

Re: Ideas for a computer science Master's Thesis project?

by Timothy Takemoto -
Hi Bernard,

I had a read.

It is still a little difficult to visualise what is going on. I guess that one would need to try out the software.

A couple of thoughts:

1) The example being used is an explanation of the software itself. I guess that Copilot is particularly suited to this task. The software itself is a tool for the logical analysis of a body of learning or knowledge, and so has a logical heirarchical structure, well suited to explanation in a logical heirarchical way.  I wonder what would happen if it were applied to the explanation of something woolier, like "The Devolution Movement in Canada" or "Developmental Psychology," or another field where there are often many conflicting theories, and not too much in the way of an overall structure or much structural depth. (This may be a problem of the social scicences. At psychology conferences there has been general lament that psychology often fails to build upon itself but rather often presents a proliferation of theories, that end up of being taught as facts.)

2) I teach English mainly. The list of verbs understood by copilot at then end of the above document, forms an interesting logical categorisation of verbs in English. It would be interesting to use the appendix (pp 52-56) as a vocabulary list, or a structure for teaching vocabulary of intermediate level verbs. If I were to do that, then I would also be teaching the structure of Copilot. And as mentioned above, Copilot may be good at teaching the structure of itself. Hence, perhaps, Copilot is well suited to teaching intermediate level verbs? 

What are you using it to teach?

Please would you send me the MS Word (OOo / rtf) form of the appendix? I had just created a list of verbs for explaining origin, for my "Tourism English" class. Looking at the copilot list of verbs, it is an excellent, categorised list of verbs for explaining. And as such, very useful to those who must explain (guides, your operators, intercultural communication specialists).

This is a rather strange way to wish to use Copilot, but...I may be able to make a Japanese translation if it is okay by you to use the list.

Tim
In reply to Timothy Takemoto

Re: Ideas for a computer science Master's Thesis project?

by Bernard Boucher -
Hi Timothy,

                  I am happy that you take time to read the user manual.

Some comments:

  1. The idea here is not to reproduce Copilot in php or to replace any part of Moodle. The idea is to use Copilot relations ( or verbs ) to express and store some form of knowledge in Moodle.
  2. From these relations or verbs, it will be possible to generate activities for the students ( quiz, lessons, glossaries,... )
  3. Anything that you are able to express with copilot verbs will be usable by the generator.
  4. The work of the teacher will be simplified: not creating lessons or quiz, just defining relations with simple text autoring done with a simple wordprocessor. 
  5. For the student side: lessons or quiz with different types of feedback in case of misunderstanding.
  6. Here is a french link showing the usage of copilot in philosophy.
  7. I teach process control and automatisation wich require electronic and some programming notions.
  8. Sorry I don't have the source of the pdf document. With copy and paste in text mode you will loose the format of the document but you will get the text.
To make it more clear let me suggest an example in a domain that you master:

if you post or email some ( 10-20) of your questions about Tourism english class I will show you the text required to represent the knowledge of your questions and Visual Basic evil will generate some questions with feedback in a GIFT format  with that text.

.

.

Have a good cogitation,

Bernard

In reply to Bernard Boucher

Re: Ideas for a computer science Master's Thesis project?

by Timothy Takemoto -
Thanks very much Benard!

I will get back to you with some questions. I don't think that they are going to be much use. But I will get back to you. I will make the table myself. It is an excellent list of descriptive verbs!

Tim
In reply to Timothy Takemoto

Copilot and Codrive

by Timothy Takemoto -

From the above and private correspondence, I have reached a certain (not neccerality correct) understanding of part of the functionlity of Copilot. Here it is.

I think I see how analysing a body of learning using Copilot allows a teacher to define relationships between concepts, so that Copilot can then create questions, answers and feedback about the body knowlege. Copilot can probably do other things too.

Copilot achieves this by defining a matrix of terms with synonyms, and active and passive forms. Hence if a teacher of moodle were to set up the relationship (this is only an example)

Who created moodle? Answer = Martin Dougiamas

Then Copilot would be able to create lots and lots of questions (about 20 or more?)
1) using synonyms
Who made moodle?
2) passives and passive synonyms
Who created moodle?
Who was moodle made by?
3) (perhaps) using antonyms
Who definately did not destroy moodle?
Who definately did not annihilate moodle?
3) (perhaps) using passive antonyms
Who was moodle definately not destroyed by?

It can also create feedback and perhaps even wrong answers automaticallly. Or something like that. I hope I am not misrepresenting copilot too much. I think that for people teaching content (non-language) courses this functionality of copilot is definately useful.

If I find that if I askk the same question then students are inclined to cheat so I am always creating at least one variation, often more. Copilot creates the variations for you. For content courses, it is definately an interesting, time saving idea. Copilot can probably do other things too.

In my language teaching situation however, the list of verbs is in itself a body of study for students.

This is also probably, or definately, part of "the dark side of the moodle," in that it is a way of automating testing rather than extending socially constructive activity.

In my situation I can not rely on my students understanding the matrix of verbs that Copilot uses. I would not therefore be able to use software that takes knowledge of the copilot lexical matrix for granted. But along the same lines, it would be nice if there were functionality at a lower level of abstraction to do a similar task. Lets call this (non existant) software "Codrive". All it takes for granted is that answers have a subject and a predicate.

For example (a real example) I have a bit of text that I want my students to read. (This happens to be a very non-socially constructive bit of text, which is perhaps appropriate)


Text I want students to read:
I am a student and my job is to study. Company employees are paid a salary. Students are paid credits. So, to me, credits are like money. Credits are the reward I get for studying.

Then I have a question:
What do students and workers have in common?

I must make some answers. I can do this in the normal way. But it would be great if I could define:
1) A list of subjects
People who study and people who work
Students and workers
People who work for a company and people who study something
Employees and students


2) Perhaps -- if I am happy with 'tricky questons' -- a list of inappropriate subjects. I don't usually do this but, for the sake of this example.
Students and people that study
People that work or have a job


3) A list of appropriate predicates
pay something to their employer or school.
think that credits resemble money.
like to get money for doing what they do.
get a reward for the money they earn.


4) A list of inappropriate predicates
receive something in return for their effort.
are given something in exchange for what they do.


Then, using "Codrive" I could press a button and it would automatically create a myriad of random permutations of the above. E.g. using GIFT notation one random question would be

What do students and workers have in common? {
~Students and people that study are given something in exchange for what they do.
=People who study and people who work like to get money for doing what they do.
~People who work for a company and people who study something think that credits resemble money.
~Employees and students get a reward for the money they earn.
}

I would like codrive software. Ideally it would be part of moodle, or an extension to the GIFT format. If an extension to the GIFT format then perhaps it might look like this in codrive-GIFT form

What do students and workers have in common? {
=S=People who study and people who work
=S=Students and workers
=S=People who work for a company and people who study something
=S=Employees and students
~S~Students and people that study
~S~People that work or have a job
=P=pay something to their employer or school.
=P=think that credits resemble money.
=P=like to get money for doing what they do.
=P=get a reward for the money they earn.
~P~receive something in return for their effort.
~P~are given something in exchange for what they do. }

Adding the feedback also requires a bit of thought. But it could be done. And it would save time.

Socially constructively speaking, is Codrive hell warmed up?

It is where I live. I must make about 80 questions. Back to the grindstone.

Tim

In reply to Timothy Takemoto

Re: Copilot and Codrive

by Bernard Boucher -
Hi TimTakwink,
I reconized your truncated photograph.

You gain a great understanding of Copilot principles and usages.

Before going deeper in the dark side of Moodleevil, a small recall : all these suggested features ( generating quiz, lessons, glossary, wiki, pseudo Socratic dialogue, ... from relations ) are only additions to Moodle.

These replace or cancel no others activities or functionnalities of Moodle and nobody will be forced to use them.

These should permit to the student to make more "formative auto-evaluation" exercices at his own pace.

For the teacher, the authoring time saved could be accorded extending socially constructive activity to direct student support ( online or face to face ) or to others activities juged usefull by the teacher.

The matrix of terms should be considered as an optionnnal feature selectable by the author when it juged it appropriate. In your particular example of language teaching, it may be avoided if the synonym replace the real verb ( more difficult for the student ) , but it can be usefull if the synonym appear in a list of choices along the real verb to give some chance to the student that don't remember the real verb but only one of the synonyms. All that depending of the intentions and the choice of the author.

You suggest a 2 lines text as input. Any text, images, schema ( electrical, organigram, grafcet, ... ) can be used as a student's stimuli to start his reflexions.
That is why the stimuli is not forced to be limited to Copilot verbs. In case of schemas no verbs are present.

After the stimuli phase the author, if he want to gain benefits from the generation, must express the relations represented in the stimuli ( text, image, schema ) with Copilot verbs. The generator itself is not intelligent, it is not able to extract knowledge from the stimuli.

In your example:
I am a student and my job is to study. Company employees are paid a salary. Students are paid credits. So, to me, credits are like money. Credits are the reward I get for studying.

Using some Copilot verbs we can express the relations in your example.

But I will continue only tomorrow!

Thanks for your interest and have a good cogitation,

Bernard









In reply to Bernard Boucher

Re: Copilot and Codrive

by Timothy Takemoto -

Dear Bernard,

I am not sure what copilot schema are but the ability to define not only singular right and wrong answers, but also matrices of right and wrong answers would be very nice.

As you say, the ability to offer synonyms (perhaps has hints/clues) would only help the learner.

A tripartite split of the problem into subjects verbs and objects would also work in my case.

E.g. the following would produce a whole variety of correct answers.
Workers and
students/people who study/university students/college students/
      are rewarded with/ get paid/ earn/ recieve
          the academic equivalent of money/ credits/ points that count towards their graduation.

This week I was too un-genki to make my weekly quota of questions. Fortunately someone else (Nathan) will be making some for me this weekend.

I think that there is a lot of positive things to be said about the dark side of the moodle, and as you say, they taken nothing away from the functionality of Moodle as it is.

Yesterday, for instance, one of my students wrote in saying that they did not want to post to the forums under their real name, and they are within their legal right to refuse. But alas there is still no way of making forum posts anonymous, even as an additional option. I will have to try and hack that again.

Would you like to write a book called "The Dark Side of the Moodle"?

Do you have a flickr account?

Tim

In reply to Timothy Takemoto

Re: Copilot and Codrive

by Bernard Boucher -
Hi Timothy,
                  sorry for the delay to complete my previous post, a small work overload here.

1 - I don't have an flickr account but yours look very nice.

2 - No time for the book, like you sometime I lagged from my students in preparing learning  activities. I really need the text authoring  stuff with the generator.

3 -  For the legal  right to refuse to post to forums  using the real name,  I don't know if it is better to smile or to cry in such situation!

4 - Now for the core of the thread:

I tried to express your 2 lines text with some copilot relations.

First one sentence by line:

I am a student and my job is to study.
Company employees are paid a salary.
Students are paid credits.
So, to me, credits are like money.
Credits are the reward I get for studying.

Second the relations:

study        is an example of    job
work        is an example of    job
money      is an example of    paid
money      is an example of    reward
credit       is an example of    paid
credit       is an example of    reward

student      perform                study
employee   perform                work

paid           depend on            job
reward       depend on            job

credit        depend on            study
money        depend on            work

Third the definitions that should be added ( like on your flickr site in english and japanese to give usefull feedback )

Job      is defined by   ...
Study    is defined by   ...
...

Fourth I renew with an "old friend"  Prolog to try to express and navigate in relations:

job(student,study).
job(employee,work).
paid(employee,money).
paid(student,credit).
reward(X, Y) :- paid(X, Y).

I hope Mark and dfwiki group will choose to integrate copilot relations in the authoring part of the dfwiki. If not I will have 2-3 weeks of spare time next april, students will be in stages, to work seriously on it at full time.

Bye,

Bernard


In reply to Timothy Takemoto

Language learning and Socratic dialogue

by Bernard Boucher -
Hi Timothy,
                 I just found a link about language learning and socratic dialogue.

In section 4 ( Dibex ) they explain much more clearly than me how knowledge can be represented and how to generate something ( socratic dialogue in that particuler case ) from it.

Similarities with my suggestions:
knowledge representation
feedback based on the type of error
interlevel ( from abstract to correction statement )
pseudo socratic dialogue
...

Differences:
grammatical knowledge ( vs bloom taxonomy )
generation of tree of questions ( mcq )  with appropriate distractors in place of socratic dialogue ( choosen by the teacher )
...

Have good cogitation,

Bernard

In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by David Cockayne -
There already exists a very basic but functional repository plugin done recently view this thread .  I think your idea of a very simple 'just upload' is along the right lines, there already exists many underutilised repositories in institutions that demand vast quantities of metadata to be added by staff, this is a big killer to new contributions.

There already exists a lot of data that could be extracted to use for searching, eg course title, course category even topic title and labels from where an object is being utilised in a real course.

A way to index the files .doc .ppt .xls .swf and questions in quizes would also I think be a useful contribution. If you take a look at moodle's stats page you can see there is already a goldmine of content over a million resources and even more quiz questions, that can not currently be searched through.

Good luck and let us know what you finally decide on smile
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
Fast forward from October 2005 (!) to April 2008: still not finished with the Thesis surprise

But I do have a topic: "Finding classic web application refactorings - Case study on Moodle course management system"

My goals are:
1) Graduate
2) See goal 1.

So the end results will most likely benefit only a very small subset of the Moodle user community (read: me), but I'll post a link to the finished Thesis PDF anyway as soon as it's done.
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
Phew... So, I've finished a first draft of my Thesis with the word "Moodle" on the topic. For masochists, I've uploaded the draft here (1.26 megabytes). If it's up to me, I won't touch it again smile
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by Mauno Korpelainen -

7 Conclusions

"The architecture of Moodle is quite incoherent and it was probably beyond the
reach of just a sole developer to improve it in the scope of this Thesis."

Incoherent? Compared to what? wink

Warmest congratulations, Samuli! smilesmilesmile

Iso urakka takana joten muista nyt nauttia kesästä ja kivoista hetkistä perheesi parissa. Onnittelut upeasta tuotoksesta!!!

Mauno

In reply to Mauno Korpelainen

Re: Ideas for a computer science Master's Thesis project?

by Tim Hunt -
Picture of Core developers Picture of Documentation writers Picture of Particularly helpful Moodlers Picture of Peer reviewers Picture of Plugin developers
Incoherent, meaning that it does not look like what they teach about software architecture in undergraduate computing classes. However, it is pretty good compared to other large, real-world applications that have been developed and modified over a period of 9 years.
In reply to Tim Hunt

Re: Ideas for a computer science Master's Thesis project?

by Samuli Karevaara -
Well, I didn't think anyone would read that clown I thought that no-one can possibly crawl through my 106 pages of drivel. I forgot that it's perfectly possible to skip straight to the Conclusions smile To be honest, the Conclusions and Results are the most rushed parts, though of course they should be the most thought-out Chapters. Just ran completely out of steam and time there...

Tim is right, the incoherence refers to the "academic purity". I had to have some excuse why I didn't come up with anything spectacular wink (and to be fair, I remember that certain OU developers had some rather harsh comments about the architecture in the past, too evil ) I think that Moodle code is very nice in places, and work-in-progress in some places.

Moodle criticism aside, I think the main results, the 14 refactorings aimed at classic web applications, are quite solid, albeit somewhat obvious in places. But then again, so are the Fowler et al. ones here and there smile

Mauno, kiitoksia kehuista! Hyvää kesää itsellesikin!
In reply to Samuli Karevaara

Re: Ideas for a computer science Master's Thesis project?

by David Horat -
Take a look at the Moodle Student Projects wiki page. Actually we are focusing on Google Summer of Code proyects. Maybe you can base your Master Thesis on a project you make for the Google Summer of Code. You still have time, because the submission deadline is next monday. Take a look at it. smile
In reply to David Horat

Re: Ideas for a computer science Master's Thesis project?

by Libby Ferguson -
Hi,

I am a moodle user and also looking for a computer science master thesis topic about web application development. I find this interesting discussion and I wondering if I am able to use one of the project for new developer as my thesis project? I am interested in enhancing the Report feature in moodle.
Or maybe is there any other suggestion that I can use for my project?

Thank you so much.
In reply to Libby Ferguson

Re: Ideas for a computer science Master's Thesis project?

by Stuart Mealor -
Hi Libby

Enhancing the standard reporting ability within Moodle would be a very valuable project indeed !

I think there was some discussion about a better report system - so check the Roadmap and see if it's in there ? I seem to recall it was awaiting funding at one stage, so perhaps it has not progressed.

Business using Modole certainly find the basic reporting within Moodle to be a little limiting - so we would be happy to test and provide feedback if you do take this project further smile

Stuart
In reply to Stuart Mealor

Re: Ideas for a computer science Master's Thesis project?

by Libby Ferguson -
Thanks for the idea Stuart, I still have to talk to my Professor first to see what he thinks about it. Will gladly let you know if I'll be working on this smile