db v2

[Kyle]

Its not official yet, but db v2 is in the works. I can't guarantee it can get done, but we have a good start so far with much needed updated web technologies including by not limited to another 100% ground up content mgmt system. This design using react + redux, mongo-db and running off a node express webserver with ES7 support both in the client and server. The app will be deployed in the amazon cloud so it can scale if needed on demand.

 

I expect 6-9months to finish, but I'm hoping to deploy scaled back versions sooner. The original site was written in php in just about 3 months.

 

What this means for you?

A better UI built in a single page app (no page reloads)

A better cms for our admins to create content and articles / reviews.

Quicker updates and features ongoing.

Easier to support and improve / scale over time.

 

The plan is to first simplify the site UI a little and then make what we got even better.

 

I don't really want to open the flood gates for feature requests, but I wouldn't mind hearing what you dislike and do like about the current site

 

 

Thanks!

[Rowan611]

Awesome. Thanks Kyle!

[Kyle]

Progress is going strong.

 

The work behind the work

 

Some cool things coming out of this, I created a server driven, strongly typed validation runner. Its now super easy to scale new forms and define a validation schema that checks data before going into the db. This will greatly help scale the application and keep data constant while also informing users of proper error messages. Seems easy, but honestly there are so many cases of horrible and bad form UI that its hard to find a good one.  Also, Images can be uploaded now in bulk and will eventually be stored in CDN like any true modern and highly scalable webapp. This will be able to take 10's of gigs of data.

 

The data model is switching from table based db to a document db. This makes it easier to develop on since we don't do a lot of custom queries, we don't need a tables or even to define a table scheme for any new feature or changes. We just make an adjustment to the validation schema and UI.

Graphs will no longer be bound to a measurement sample, but to a system directly. This is probably the biggest architecture change but its designed to make the systems read and feel more like reviews/articles. A system can still have multiple CEA2010 plots, graphs, images, etc but look and feel will be much easier to digest. We hope to get some more content contributors and if all goes well we can expand the purpose of the site and open it up for some user content

 

 

Most content mgmt systems use a framework starting point like wordpress. Like the original db, v2 will also be a totally ground up custom application so it will take 3-4 times longer to develop, but it will far better in the end and it will allow us to take it much farther than being locked into a technology or standard. 

 

 

Some screenies (hard to get an idea of the work, but you can at least see something...)

 

 

[Ricci]

 

We need to talk soon about some of this. I'll try to give you a call this weekend.

[Kyle]

 

We need to talk soon about some of this. I'll try to give you a call this weekend.

 

lol

 

/me runs and hides

[SME]

Kyle,

 

This looks awesome so far!  And it looks ambitious.  I'm very excited to see where this leads.

[Kyle]

Thanks, its a bit more work than I originally thought, the bulk of that is I have to be far more careful with implementation because if we're ever going to open this up to user content al-la facebook style, then everything has to be validated, double checked, and tamper proof and scalable.

[Kyle]

Right now I'm also exploring the idea of integrating Invision power board posts with systems / drivers, etc, so you can add comments directly in the website. I'm not sure if that's do-able with the IPB REST API but I would like to integrate this forum with the website a little more.