December 2021 Status Update
Greetings and Happy Holidays. Here is your next installment of starbeam.one monthly updates.
I feel the need to say this each time, but yes, I am still alive. 😊 I am also a bit late this time around, having skipped the November update. Originally, I was aiming for the middle of November, but the final feature task that I was working on took way longer than I anticipated… again. 😅😭
With that, I took the few weeks to round out that final story and decided to round out to the next month, and here we are. Going forward, I will aim for the first Tuesday of every month with an update. So with that, let’s get on with the show here.
Marketplace Application Feature Complete!
I am pleased to announce that after many many months of refactoring and delays with “perfection,” the marketplace application is feature complete and ready for deployment. I know I know, I shouldn’t be spending so much time being “perfect," (and in fact refer to this rather tongue-in-cheek) but when I say “perfection” I really mean “doesn’t obviously break.”
In other words, it works on my machine and is ready for testing on another.
This last leg has been especially brutal. In addition to a few fundamental refactorings with the data access and a re-write of the identity system, the last final feature was more involved than anticipated. I had it pegged for 10 days and it ended up being 20. To refresh, my days are 10-12 hours, no weekends, so it has been an absolute slog. I finished on December 2nd and have been attending to minor tasks since then, taking a “break” of sorts.
What’s Next?
What a great question. I’m glad I had you ask me. 😄 Two days ago I had one answer for you (“Primordial Alpha”), but now that I have had some clear time to think some more, I have another one that I like better (“Playground Alpha”).
Let’s explore.
Primordial Alpha + Its Challenges
What is the Primordial Alpha? Well, I’ve gone back and forth on exactly what this is. Initially, I envisioned having a very well-secured application that is only accessible via Allowed List. However, there are some “chicken-egg” challenges with doing so.
First is how do creators onboard their potential customers? We have to Allow List them as well, and that immediately becomes a challenge. Either I have to create a mechanism that enables them to do so and/or they have to go through a process that may or may not be intuitive/rewarding/positive.
We’re wanting a good first impression here, or as close to one as possible. Considering that the primary goal is to see what breaks – which may be everything – I would rather reduce the amount of friction required to onboard users into the system for kicking its tires.
In addition to onboarding creators and their users, there are other challenges. First is legal and other parties that require access to the system for review. It’s easier to have something deployed and easily accessible when I send it to them. Sure, again, I could have sign-in/security mechanisms but historically I have never had much luck with them and end up being more trouble than its worth.
Also, those sign-in/security mechanisms are an issue of their own. Authentication into starbeam.one is handled exclusively by Oauth2 providers, courtesy of ASP.NET Core’s built-in security providers and the aspnet-contrib package (of which I am a contributor! 🎉). Each one of those goes through a review process, and the review team requires access to the application to verify it conforms to their individual established policies. Giving access to these teams has been a challenging riddle to me for some time. Again, doing a security wall is challenging and there always seems to be a problem. So what’s the best way of providing access while getting approval?
Finally, there is the issue of onramp and exit ramps with fiat currency. I am using PayPal for both, and Coinbase for onramping. There is currently no viable exit ramp for Coinbase. Well there is, but it simply does not work, and I am very irritated with that. Let’s just say that Coinbase appears to have very little to no support for their development offerings, and I struggle to see why they have such a high valuation.
So, I have been looking at BitPay instead. They offer a promising solution, but it has revenue requirements. Again, I would like to show them the working product and demonstrate what I have in mind and see what is possible.
The Solution: The Playground Alpha
So what is the solution? It wasn’t until yesterday that I thought of something: why not simply deploy the application in all its (potentially broken 😁) operational glory with the exception of “live” deposit and withdrawal functionality? Deposit and withdrawal will still be “available” but they will be “no-op,” meaning they really do not do anything consequential.
I call it… The Playground Alpha!
To clarify further, depositing will be possible via “faucet” where you can debit (fake/system-provided) money into your account, and the withdrawal will credit your account but will not actually send it anywhere.
So, everything appears to be working and is fully functional as deployed, but no actual “real” money will be utilized.
This way, the application can be viewed by all and I can start collecting all the valuable feedback immediately – including and especially what breaks. I can also start pointing dependent parties towards the location and having reviews done to further proceed with chicken-egg scenarios.
Boom. I then anticipate a 2-3 month period where I tackle obvious problems while implementing BitPay/other on/off-ramps and get this to a deployable alpha where it can be used for real in a “live money” setting, all the while building some sort of confidence in the system.
Schedule
All told, this is what I currently have in mind:
- December 2021: Playground Alpha Preparation
- January 2022: Deployment to Azure/Marketing Video Tour
- February 2022: Playground Alpha (No-op Deposit/Withdrawal)
- April 2022: Alpha (Live Deposit/Withdrawal)
Note that there is a chance that the schedule might accelerate, based on all sorts of conditions, primarily that of the quality of code and how well it reacts once under test.
And, of course, There is also a chance the schedule could slip. In addition to things Breaking Badly™, I am now officially on the clock and I do not operate well under pressure. 🙃
But, we’re getting closer here. This is feeling more and more real every day and I am having to deal with the stress of such. In a good way, I swear! 😅
That’s it for this month’s installment. Until next time, stay bright. ✌✨