Closed tamarandom closed 2 years ago
There's a small issue there. CommonsStack and the TEC use the same discord bot for Praise.
To change the profile picture, we would need to deploy our own instance
Good to know. It sounds like the kind of thing we will need to do eventually.
Open question: what are the consequences to deploying an independent TEC praise bot?
There would be a hosting cost for the bot, and I think there might be some differences in how the praise gets handled for both CommonsStack and the TEC. (I think the praise goes to a spreadsheet, where griff separates them into separate Praise Quant sheets)
Tagging with Next Sprint to that this comes up for discussion in our Sprint planning now...
@Vyvy-vi @kristoferlund can we look at what it would take to deploy an independent Praise bot for the TEC and where in the Rewards System plan that would make the most sense to implement.
We would need an active worker, which can be available on any VPS like DigitalOcean for around 5$/month or something. The hosting process should be straightforward, however we might need our own spreadsheet credentials(*I'm not sure).
With regard to the new Rewards plan, and the backend approach to it, I think the bot would remain like it is however it would store the data in the backend via an API request(I guess)
@Vyvy-vi Do you want to have a micro-service architecture? Since my main stack is java + spring framework, I don't see a problem in running discord and telegram bot as components of the same application.
That can be done π€
we will want to consider doing this in telegram as well
Do you want to have a micro-service architecture? Since my main stack is java + spring framework, I don't see a problem in running discord and telegram bot as components of the same application.
Considering how the new interactions API works, we could probably cross-link these neatly. (Interactions API allows you to run the bot entirely on an HTTP server) π€
Do you want to have a micro-service architecture? Since my main stack is java + spring framework, I don't see a problem in running discord and telegram bot as components of the same application.
Considering how the new interactions API works, we could probably cross-link these neatly. (Interactions API allows you to run the bot entirely on an HTTP server) π€
Thanks @hurek @Vyvy-vi ! Looping in @kristoferlund so it's on his radar too.
I'm sensing this is feasible. I'm out of my depth with discussion on the technical architecture and advantages or disadvantages so look forward to the end solution following those discussions!
Do I understand this correctly? We consider migrating the current praise bot over to a java/spring architecture and separating the TEC and CS bots? The upgraded bot would run on an HTTP server accepting calls from the outgoing webhooks of the Discord Interaction API?
That to me seems like a good idea no matter how we move on with the rest of the praise project.
Yes!
Chatting with @Vyvy-vi and here are the nest steps as we see them:
If we decide to move forward with separating the TEC praise bot from the CS Praise bot, the next steps would be:
We can do as decribed. Or, we can wait until MVP of new praise is online by end of nov. Then, the bots need to be separated as part of the upgraded Rewards System anyway. Or rather, it wont be the same bot running on TEC as CS until CS also (hopefully) upgrade to the new praise system.
@kristoferlund That is the question! My thinking is that as long as it is relatively easy we may as well just do it now rather than wait another 6-8 weeks (at least). From what I understand so far, it is relatively simple and we should probably just make a plan and execute. Do you think there is compelling reason to hold off? There very well might be something I am not thinking of!
If it's easy, let's do it! Or rather, @Vyvy-vi, if you say let's do it, let's do it.
Hey @Vyvy-vi @kristoferlund I think we should close this. With the other priorities, this one is pretty low on the list. It will happen along with the Praise redesign and development. Any reason not to close it now?
Close it!
I think we can maybe close this for now :+1:
ππ½ What is this issue about? Give as much detail as you can.
Is is currently the Commons Stack one. See here:
Even though the Praise was received for TEC and not CS.
β¨ What does success look like? (i.e., the criteria to meet to call this issue done)
π§π»ββοΈπ§π»ββοΈ Who should be part of the Advice Process? (People who should be consulted about this issue. Those who will have valuable input.)
@iviangita @Vyvy-vi @krisjpi @kristoferlund @hurek (<-- finally got rdfbbx's github handle!)
π Are there links or docs related to this issue?
π Who should be in the loop about this issue?
@GriffGreen