As the non-tech person on the team, the greatest value I see is how fast it got us up and running. it was like a dream come true when jaigouk showed off how we can see live transactions appear in real time on our dashboard, making tracking activities a breeze, above all, it make proactively predicting when we should reach out to our users possible, which helped a lot in helping some of our users along in their transaction.
Maybe I missed it, I'm not the sharpest, but I don't quite understand what about meteor in particular made it easier or better than Sinatra + some frontend JS framework.
I'm in no way saying it's not, I just, from the article, didn't gather why the interviewee really thought it was better? Like what in particular is so much more awesome?
Primarily the ability to load data straight from Mongo DB without writing any APIs. Don't get me wrong, you can certainly implement the same things with other tools, but Meteor already bundles a lot of building. We pretty much "watch" live data on the server and display it as it comes. Front end JS helps us do calculations and apply labels as we see fit. The best part, we just change the front end and Meteor takes care of watching and updating the data.
But I'm not sure what you mean by a Meteor app with a native client. Do you mean a native client that uses the DDP protocol (http://www.meteor.com/blog/2012/03/21/introducing-ddp) to communicate with a Meteor back-end? I haven't heard of that yet, but PiJS (http://pijs.io) uses DDP to do interesting things with the Raspberry Pi.
Hi Kirill, in the article you state "would not build production pieces on it yet"
Can you point out what are the downsides and pitfalls?
We are actually trying to build our MVP completely on Meteor
I think we will post it up in a few weeks. The whole thing was built so fast, the dash looks incredibly ugly - think bootstrap without any styling. It's border line command line colors :) But, it's working out so well that we haven't spent resources to pretty it up yet. I think in 1-2 months we will need to redesign it, based on all the pain-points we have learned, and that's when we will implement a stylish and efficient dash, ready to show.