Today I delivered to Simon and team the first “beta” version of the new tool. Here’s what it looks like, actively monitoring various RSS and Facebook feeds:
Content curation view:
Content curation, in dark mode:
Content curation on mobile, where article content and images are hidden:
Feed management:
Editing a Source:
I’ve also published the first public version of the Laravel package I created to make all of this work, called laravel-news-harvester. It’s on GitHub at https://github.com/CommunityInfoCoop/laravel-news-harvester along with instructions that other organizations (with a bit of Laravel knowledge) could use to get this tool up and running:
Here’s the full text of my delivery email to Simon:
Simon,
I’m pleased to deliver for your review and testing the initial working version of the new “local” dashboard tool that we’ve been planning.
You can access it at a temporary URL here:
https://bloomfield-local-412kmtwp.chrishardie.net/
username: **********************
password: **********************On your self-guided tour, make sure you don’t miss:
Dashboard
https://bloomfield-local-412kmtwp.chrishardie.net/adminContent Curation (including messing with the filter menu)
https://bloomfield-local-412kmtwp.chrishardie.net/admin/news-itemsSource management
https://bloomfield-local-412kmtwp.chrishardie.net/admin/sourcesFeed management
https://bloomfield-local-412kmtwp.chrishardie.net/admin/feedsFeed creation
https://bloomfield-local-412kmtwp.chrishardie.net/admin/feeds/createNote that all feeds are currently refreshed hourly. You can initiate a manual refresh of RSS feeds using the refresh icon on the right side of the Feed listing.
Some things that are not yet in place:
* Fine-grained permissions and roles (the foundation is there, just need to create the interface to manage)
Some quirks and disclaimers to note:
* “Dark mode” will probably be on by default after you log in; switch it using the top right menu under your initials
* To get things going quickly I auto-created Sources for the imported Feedly feeds, named the same, and also put all of the CrowdTangle groups under a single Source. See below for questions on how better to handle.
* If things get slow at any point as you move around the interface: I used the $5/month cheapest server type to set this up on. It should remain fast but if not we can upgrade to a server type with more processing power.You may want to reference:
* Wireframes: https://bsucapstone.chrishardie.com/2022/03/17/wireframes/
* Data model: https://docs.google.com/spreadsheets/d/1_AICOehFzsj21_AnqZ4VEBZnU9YoL9436A-G7Ol3m-Y/edit?usp=sharingI’ve also already packaged up this functionality for others to potentially use, as a public software repo on GitHub:
https://github.com/CommunityInfoCoop/laravel-news-harvester
(In theory someone with a bit of Laravel familiarity could use the instructions provided to launch this tool for themselves, although in practice I suspect we’d want to work closely with an organization setting it up for the first time to get feedback on the process.)
Collecting feedback: my hope is that you can start gathering any and all questions, bug reports, requests, ideas, comments, etc. in one place (Google Doc or Sheet? Email message?) and then I can process those in bulk. That said, if you encounter any blockers or big issues/questions/etc. feel free to let me know right away.
Some specific questions/discussion points for you:
* What URL do you want to use for Bloomfield’s instance of this tool – maybe local.bloomfieldinfo.org?
* How should we handle Source creation for imported Feedly feeds, and for CrowdTangle groups/pages?
* How should we license this Laravel package? I’d suggest MIT (https://mit-license.org) but want to revisit that with you.
* I should have the tool auto-delete old news items after they reach a certain age. It can be really old if you want – years? – just need to pick something.I’m sure there are plenty of nuances I’m leaving out and I’d be glad to walk through this with you, but wanted to get you access so you can start playing around with it yourself. Let me know if I should create some additional user accounts for others at this point.
Thank you,
Chris
I also followed up to provide access to an example Google Sheet where we can track feedback and bugs as users test:
It’s exciting to be at this milestone and I’ll look forward to working with Simon and team to gather and address their feedback!
