arcadia_update.md

From Brian:

1. What is the overall vibe between MP and Arcadia? MP Feels it is very good and we have a very nice looking app with all of the bells and whistles we could deliver in the time given so far. We need to know so we can plan for our people for next projects and work with Arcadia or other clients.
2. We have spent a week trying to massage this new data to 1) get it more current so the app could use it, 2) expand it to be more than 3 months as the app needs to be able to hit data for the past 30/90/365 days, and 3) understand the spltting up of the providerActivity data file into now two files and how to use this in the strutcure of the app. We provided the data model and the S3 file structure and it was largely not followed which has caused this reworking of the app.
3. A large amount of work performed was done so because of the app design and function changing on the fly with very little specifications provided. This is fine, but it has added to the timeframe to complete work on the project. Examples: Slide out navigation and moving controls to this new feature, DITL as a "modal" that takes up a whole page and has a large amount of the same things from the main page (should have been a SPA with tabs for all graphs), Most designs only had one view and did not address all parts of the app (all graphs, font specifics, colors, etc.) until asked for multiple times. Again, not a huge issue be has contributed to the time taken to complete said work.

From Jon:

1. Column renames forced us to find/replace a lot of different points of access throughout the app. Given that some of our initial names were more generic than those selected, this wasn’t a simple sed/awk situation. (Nor did we ask for the column names to be changed.)
2. The data files have weird content issues, all of which we’ve found are described on the Confluence page. None of them are absolutely bonkers, but we’ve had to implement (and have documented / are documenting) these workarounds.
3. The splitting of the Day in the Life file into two files. In their defense, they added this in May and asked us if it was kosher. In our defense, we didn’t see data files in that format until nearly a month later and didn’t think it wise to spend even more time creating fake data. (edited)

I’d also add that we have three days worth of questions on here, and we’re coming up on a fourth (haven’t added our ask for the possibility to re-join the DITL files together yet, want to give it my all before I do), with no response. Nick has asked that we ping him directly with “questions of any urgency” but none of these seem urgent. I mean, until they’re left until the last minute, in which case they become unnecessarily urgent.

Confluence Page with Discussion/Questions: https://confluence.arcadiasolutions.com/display/AADWO/15+June+Data+Model+Discussion