I've attached a comparison of various platforms that might be used as a starting point for the Adapt authoring tool (Adapt Builder?). I've added the initial functional requirements for the MVP as suggested during the workshop in Derry last week and also a handful of non-functional requirements.
This is not meant as a comprehensive review, but an opportunity for the community to discuss some of the issues around the choice of platform. Hope you find it useful.
A few extra things to consider:
- Effort to complete / maintain, time it takes to build.
- Effort to modify / need to hack at prewritten code
- Desire to work in this technology
- Stability of core updates - Let's take Drupal as an example, if Drupal does a massive core update we would then need to retro fit our system around this, whereas if we create it from the ground up we control the updates.
- One language? How easy can a developer swap between working on the Editor and Framework?
- Build tools like grunt (node), minify (php)
I have realised that most of my points add positives to a node based platform. So it would be good to get some views from a PHP or Python developer? Also Chris and Alan from Sponge have built an LMS in node. Would be great to hear your positives and negatives of using node.
Thanks,
Daryl
Thanks for this, guys. It's good to look at things from a data as well as a personal preference / gut feeling perspective.
My initial sense was very much based around reuse and achievability and I admit that I haven't been as open to node.js as I probably should have been. Overall, I feel this is a decision for the deeply technical of us and these categories help us to consider it well before embarking on any particular path.
The workshop in Derry has been very helpful and I think that the multi tenancy item is one of the key deciding factors and probably forces us to consider a 'new build' as opposed to using something that's already out there.
Node certainly has quite a bit of traction and it also fits well with the skills used on the output framework, which is a plus point. Like Ryan, I have looked at the vision and outlined approach and as long as we are comfortable we can tick the following boxes, we are free to choose whatever fits.
Key 'vision' items:
Some other practical items:
Sven
Just had another thought that we should consider - what server do we need, how easy to setup and cost of this type of server. Scalability and server costs are quite a big concern with a web based tool.
How easy will it be for a non-technical user to set up a python or node based server? PHP have the advantage of there being so many hosting services that allow easy installs and setups.
Thanks,
Daryl
I agree. We need to identify who we expect to install this, and what their existing technology stack looks like. Given that Moodle is probably the largest target application we should seriously consider using PHP as the language (but Node is great and we shouldn't necessarily constrain ourselves by third-party integration targets)
The decision then needs to be made about using an off-the-shelf product or build our own.
I think this spreadsheet certainly goes some way towards making that decision.
In line with Ryan's comment, what does it actually take to run a node.js webserver? Something I should know, I guess. Maybe I'm just testing you all :-)
Taken from nodejs.org
This simple web server written in Node responds with "Hello World" for every request.
var http = require('http');
http.createServer(function (req, res) {
res.writeHead(200, {'Content-Type': 'text/plain'});
res.end('Hello World\n');
}).listen(1337, '127.0.0.1');
console.log('Server running at http://127.0.0.1:1337/');
To run the server, put the code into a file example.js and execute it with the node program from the command line:
% node example.js
Server running at http://127.0.0.1:1337/
So in answer to your question you do not need a dedicated Webserver like IIS or Apache.
However I think the real question is "How quickly can we get new developers working on the project?"
The answer, I think is 2 part
I can only talk from first hand experience of course but getting anything white listed/installed at C&G had a lead time of about 8 months!
With banks there is a reticence to install anything due to security concerns and with some large businesses the reluctance is born from the charges incurred from working with third party IT suppliers who charge £x per kit for even the most basic update.
To be fair, this is more to do with local installs than servers though so may not be relevant. that said there are definitely barriers for even the most basic of undertakings.
I'd quite like to think of scenarios like:
"I have a hosting solution with *company name*, and I can install Drupal/Wordpress with a one click installation - it's easy. Can I install the Adapt editor on this?"
"I don't want to run this online and would prefer to have this offline. Can I do this?"
"I have a VPS server and would like to run this. Can I?"
"I'm completely new to server technology. How and where do I install this?"
"How much will this cost?"
"Maintenance and support?"
From my point of few as a developer who could use Adapt anywhere - schools, colleges, universities and businesses, I would like to worry less about setting up the server and more about creating content. I like how my VPS has easy install one-click setups - But I also know developers who aren't even using VPS servers and have shared hosting.
I wonder whether all of this could be resolved by giving documentation and a walk through on our recommended hosting companies.
Thanks,
Daryl