Help Building a new VLE

[aTEK]

Thanks.
I understand that SIMS integration will be difficult however it appears that it will become more and more required.

Burning bridges?
In all of these instances I have alerted the IT administrator and in return he has given me access to many different areas of the school system the majority of students are unaware of such as accessing my files from home ect. I am going to create the source forge page later after me and my team have decided on a good name. Any Ideas?

[sylvio]

[..]
We are planning on producing the system as open source, written in a mixture of codes however mainly asp.net and coldfusion with html and php where necercerry. THe team also has great knowledge in Ruby and VB that can be used if required. The main plan is to use an SQL database as it is freely and easily available via many forms.
...

Your times a view of best.net ( SourceForge.net: Files ) or DotNetSCORM - Home made?

S.

[dhicks]

So the main features that are suggested are for it to be fully customisable with practically any language available.

My suggestion is that you don't want an "it" at all, but a framework of some kind that small components can be slotted into.

"students being able to access files from home"

This is easy - just use Apache HTTP server's built-in WebDAV facility.

"complete SIMS (or a similar system) integration."

Check these forums for posts on this. The main problem with "complete" integration is licensing difficulties with writing data to the SIMS database. I'd say forget it - get your system to read data from SIMS, write it back to its own database, then one day if your system becomes successful enough it'll hold all the data and SIMS will disappear.

Oh and by the way my team just picked up 2 more programmers, one with mainly knowledge in C and Delphi, and one with extensive Java knowledge.

I have a nasty feeling you equate knowledge of a particular language with software development expertise. A decent software developer should be able to pick up a new language over a long weekend - these aren't spoken languages we're talking about here, these are human-designed artefacts with a set (and generally very small) syntax and a bunch of libraries. Learning, or indeed making, a new programming language isn't difficult. The difficult part is going to be coordinating your diverse team and getting them all to write decent, coherent code, with comments, and to a given style. I would suggest that this is actually going to be impossible, hence my advice to forget about designing yet another monolithic system and aiming for a framework to support individually written components. That way, individual developers can mess up stuff, leave or join the project, get in huffs and argue with other developers, all without impacting the entire project too badly.

--
David Hicks

[bizzel]

First of all, I admire what you want to do. My first project was to create a helpdesk system for my own school, that was way back when I was in year 10. Now it never got adopted (understandably, running student code on the network is a big no-no in most places) but it did teach me a lot about programming.

You seem to have plenty of people on board at least. With that in mind, I'd suggest that your main challenge will actually be co-ordinating any progress rather than writing code. Code is easy but it's only half the story, if that. Getting a specification laid out and having people agree on the direction to take is much harder than Ruby, Python, C# or whatever language you choose. Unfortunately the specification and design mock-ups are both boring and neccessary.

That said, don't be put off totally. I've written some software that I could only have dreamed of 2 or 3 years ago. If nothing else this will be valuable teamwork experience for all of you. Good luck if you decide to go ahead.

[budgester]

First thing look sensibly at what you re trying to achieve.

Next find a problem and solve that, I recently had an issue with CMIS to EasyTrace Integration so I wrote some code and opensourced it. Got an itch scratch it.

I think a project that could do well for you would be something like broker service with plugins for the major apps.

eportfolio
^
|
Broker <-> MIS
|
v
Library Software

Plus as many other apps that you can find, e.g. AD, Cashless catering, storage, helpdesk, email.

SIF and ZIS is supposed to do this but no decent implementations so far.

Regards

Budgester

[GrumbleDook]

Wow ... where to start.

A very ambitious project here. As previous posters have mentioned the coding is only a small part of it all. Let's get down to brass tacks.

You have clearly started to sit down and work out the needs of the application(s), the method of delivery and your target audience. Now you need to seriously drill down into the specifications for functionality. Have you nominated a development manager? What model are you using to manage the devolved coding?

As previously mentioned, you are going to struggle to come up with decent methods for direct integration with the MIS. The first hurdle you will have is that you will not be allowed access to a live school MIS to try things with (and if you are then you, your IT Manager and your school are seriously opening yourselves up to major trouble). To access a demo database you would need to sign up as a developer for SIMS, CMIS and so on ... and you have to pay a development licence for this unless you can come to a very convenient arrangement. I think the MIS side of things might be worth leaving to one side for the moment and keep all your data self-contained in your VLE.

The VLE ... it is not just about moving data around ... you need to consider how staff and students will interact with the learning objects ... whether you prefer a structured model of learning (courses completed in a set order based on units of work and individual tasks) or whether you are going to work on skill building and then allow for application of those skills to be developed in a variety of ways, but at the control of the learner rather than driven by the teacher.

I would look into the history of Moodle and Bodington (the open source VLE used and developed by Leeds) and compar it to things like Blackboard / WebCT / FirstClass ... and see why each works slightly differently but with similar goals. This will help you decide which model to go for (or which parts of teh differen models to go for). It will also give you a greater understanding of the development of VLEs and how organic they are.

Have you considered following the model that a few others out there have? Take existing online tools and come up with methods of integrating them. Moodle, ELGG, Drupal, Google Apps, Twitter, Facebook, YouTube, SlideShare, etc ... this might be a better starting point.

HTH

[Face-Man]

This is certainly not a small project and I would seriously look into a small task first and then once you have a user base move up.

There are companies out there trying to put learning material into Facebook (see udutu | online learning simulations made easy )The problem with udutu is it really for college and industry learning not schools. It is their model I would follow create tools that are able to produce Scorm compliant modules. This will give teachers the tools missing from there current school chosen VLE and will be transferable to a different one. Basically don't t try and compete with what is out there, try an innovate an add on.

After all Microsoft didn't invent the PC they just saw that everyone one of them would need software.

[aTEK]

We have decided to follow the ideology of providing a completely open web based system. The main idea of this being as it will allow for any school to fully customise the system. We are going to build the framework so that there are modules such as forums ect that can be inserted into any chosen page on the site. This will allow for those with limited knowledge to also edit it and add to it. We may then eventually add a web based wysiwyg editor for changing pages.

We are going to ignore any form of MIS integration until after we have the framework complete.

[dhicks]

We have decided to follow the ideology of providing a completely open web based system.

Do you have a project home page of some sort yet, a SourceForge page or similar? I'm intrigued as to what you're going to come up with.

We are going to build the framework so that there are modules such as forums ect that can be inserted into any chosen page on the site.

How are the modules written, server side or client-side (i.e. JavaScript)? If you're thinking server-side then you're thinking like a PHP (or ASP, ColdFusion, etc) programmer - web pages, no doubt fully customisable with some kind of style sheet or template system (which will take up a decent-sized wodge of your development effort, and will still require effort and expertise on the part of the end-user to use), with insertable module sections. By the end of your development effort's you'll have created version 0.9 of yet another Moodle-like system, but with the disadvantage of a bunch of bugs and a lack of features.

If you're thinking client-side, then at least you're up to date - more "Web 2.0" JavaScript-based client-side applications seem about to take off (although I could be wrong there, of course). You might want to investigate development frameworks such as Google Gears or Adobe Air that allow you to write client-side, web-based applications that can still work when disconnected from the Internet - just the thing for children to use at home with no Internet access.

We may then eventually add a web based wysiwyg editor for changing pages.

Don't roll your own, use an existing JavaScript editor.

We are going to ignore any form of MIS integration until after we have the framework complete.

Don't roll your own authentication system, use an existing system like LDAP (use Active Directory as an LDAP server if you have to, don't go for direct AD authentication) or maybe OpenID.

--
David Hicks

[bizzel]

Don't roll your own, use an existing JavaScript editor.

TinyMCE - Home

[Face-Man]

If you want to see you competition imJack - Communicate. Collaborate. Learn. is being given free to schools

[Jack]

Right, this is what I would do.

• Straight off the bat, LDAP integration. Users can then login to your system with their domain usernames and passwords, and is one less thing to remember and go wrong.
• Sort out some different types of users and groups. Global admins, school admins (IT, music tech...etc), teachers, students...etc. See if you can automate this by getting information from OUs and things.
• Try and get your hands on an export from SIMS or EBS or another big MIS system and see if you can import some of the information into your system. Even better would be to get the data "live" from the MIS system so that you don't actually store it anywhere. This removes you from a lot of data protection rubbish and you don't have to worry about it because you're not storing any data.

Once you've got that down, you can shelve all that and leave it alone. Next comes things like making courses, uploading documents...etc, but all that will come naturally once you have the basic framework down.

You will probably find that people will be a lot more willing to talk and share ideas once you have a basic shell of a program down and they have something to work with instead of a blank canvas.

[sahmeepee]

If you want to see you competition imJack - Communicate. Collaborate. Learn. is being given free to schools

I still had someone trying to sell it to me quite recently. Seems like they've relaunched it since they were last pushing it 18-24 months ago.

God knows where the SSAT get off with all these commercial companies they keep pushing at^W^W endorsing to schools. If it's not IMJack it's some (non-free) network auditing service.

TBH IMJack looks a bit chucked together.. I'd be aiming a bit higher than that!

[Jay]

Not been able to read all of the (lenthy) posts, but If you really really are going to go for it....

MIS intergration...

1) Sims.Net (Capita) supports IMS export
2) Serco Facilty has an API to export data (IMS format too I think)
3) Pearson e1 has a Web Data Services (web services) tool to extract data in bulk. Although they write the reports upon request, they also have a "Report Writer" tool that allows data extraction at a school level.
4) Pearson Phoenix Gold has an IMS export tool (or so I'm informed)

I don't think may MIS providers permit the import of data. I've been told that Capita have a tool for it (at extra cost) but not sure on the others.

As you can see from the list, understanding and supporting the IMS formats are a must. Becta has also mandated SIF as the future mechanism for data exchange, the core data set is fairly similar to IMS core and the eduPerson LDAP schema. All the suppliers above and more are working on or already have SIF agents.

For authentication and access control, Shibboleth is the mantra in higher and further education and is gradually devolving down to school level.

You can find out more via Internet2 (Shibboleth) and Oasis (eduPerson schema, IMS plus SCORM). Edustructures are the leading provider of SIF software (Zone Intergration Server and development tools), the UK data schema is authoured by a SIFA working group.

You might want to start by building a core set of modules to facilitate interoperability and intergration and not worry about the features until later. Most technical people are happy with simple systems that "just work" and don't care about the bells and whistles so much (IMO).