Many of you are aware that we have been down for over two days while attempting to cut-over to a new membership (subscriber) system.
A Bad Start
The whole process started badly when the developers forgot a critical issue (that I had documented in four different places) and wasted four hours.
However, things seemed to be getting better and, as at quitting time last night (Wednesday), I was still optimistic we would be good to go this afternoon (Thursday), as the developers assured me.
Still, I also knew that developers are notorious—kind of like boatyards—for overestimating progress. So this morning I got up early and took a personal look at what they had achieved and quickly realized that they were nowhere near as far along as they claimed; in fact, they were less than half done. And, worse still, what they had done was full of data inconsistencies and bugs.
Not a situation that we were willing to subject you, our valued members, to.
So, long story short, we pulled the plug and have reactivated the old system.
I then had a conference with the CEO and lead developer to lay out what we expected on the next attempt:
- Testing code before deploying it.
- Pre-testing data transfers.
- Reading and working off the specification.
This should have been an easy set of goals to meet since I had documented everything they needed to know—back in another life I managed big computer accounting system cut-overs—but they felt that Phyllis and my expectations were too high, so they quit—they did promise to return all our money, but we have not seen it yet.
The result is that we are back to square one and out a huge amount of time—the biggest loss here, since we could have used that time to create new content.
Just so you know, we did a huge amount of due diligence on these people. But then, I have done the same on boatyards and still got terrible results. It’s no guarantee. In fact, this whole experience feels like the worst possible boat project you can imagine…and we all know how bad that is!
As to the future of this project, right now Phyllis and I just need a rest—I have at least 200 hours in this. And there are articles to write and it’s nearly time to go sailing, so nothing much will change before the fall.
Our apologies to all of you who were inconvenienced by this. This was the longest time that AAC has been unavailable, ever…by a factor of 10.
This Will Not Hurt You
Let’s close on some good news. The old system, despite being a bit antiquated, is, and always has been, amazingly robust. And that, like with boats, is the most important thing. The even better news is that we had not in any way changed the old system, so it has not been compromised by this.
That said, it will take some 24 hours for all the account changes that should have processed on our end while we were down to catch up. (Normally it would not take that long, but the part of PayPal we need to access to make this happen is down—it never rains but it…).