Why? It may well be the case, but certainly not necessarily so. Heck, none of us really know how many testers there are. We know there are 126 whose names were announced by themselves on the forums. Could be others from the forums who chose not to say so, but I figure that wouldn't be many. We also know we started with about 20 in the first test group AND at that time WT said there were a total of "hundreds" and that those from the forum were a small percentage. So, even taking "hundreds" as not meaning more than 200, we have added over 100 to that from the forums and we don't know if they also added more who are not.
This also doesn't count any internal testers (not counting any machine testing).
I'm all for WT giving more information and doing so more often, but I simply don't see that the negative views of what they are doing with the product itself means all the things some seem to believe. Simply because I have no problem seeing other reasons for it. I'll remind you that there have been a number of people chosen for TREG who made similar complaints to yours. Then they got one, were able to see how it worked, how complicated it was, etc, and they have taken a quite different view. Oh, they may still say they think it is fine to ship now - but they also can see why WT may not think that is a good idea.
While they may exist, I don't know of any tester who has commented here, on Slack, or to me privately that WT should definitely be shipping. They complain about a number of things (generally about updates). Or frustration that it gets delayed again. But not that WT is wrong not to ship.
That was to another treg tester. I'll point out a variation of the point I made above - you aren't the only person with such experience developing or shipping products. We have them in Treg. Some of them said the same things as you - until they learned more about the TB during TREG.
Sure, you "know what you are talking about", but only in the general sense. Maybe in the majority of such cases. But no one without a TB knows a lot about this particular product and even the treg members know only so much. I can't tell you how many problems exist, how common they are (as in how many have a given problem and how often it shows up). Oh, I know how often something is mentioned in Slack (not all testers post there and some rarely do), but that is only a small subset of all testing and even if we only count those, the info we see isn't usually going to go into the detail that the same people send in to WT.
Consider, someone says, "My TB wouldn't jump to my other device". That doesn't tell you anything about how often it happens. May have been just that one time. Or may have happened more than once that day. Even if they say, "I keep having trouble with jumping to different devices", that doesn't really tell you how bad it is, does it? Because that statement could mean it happens every week, every day, or multiple times a day. All we would know is that it isn't a one-time or very rare thing.
I try to keep a list of those with problems. But I only put people on that list if the problem they reported keeps reoccurring - but that may be as little as weekly. If someone says something only happened once, I don't list them. Could just be an electronic fluke. But often I can't tell from the reports. But WT would have gotten more details from them.
There is no need to say the firmware is "horribly broken" because it isn't. But it does NOT follow that it is time to ship. There is a vast number of in between possibilities.
None of us knows if WT is drawing the line in the right place or not. I think I could say that WT doesn't either - and that isn't a criticism of WT. It is simply a recognition that there is going to be a lot of conjecture about the negatives of releasing too early vs too late. There are very popular products we have now which can be traced back to similar concepts years ago - but those earlier ones were a flop. Maybe the technology wasn't good enough. Maybe they didn't make it good enough before shipping.
It all goes back to where the risk is. We have the risk of $100 invested in a TB. And we can respond to that risk by getting our money back now or sticking with it. I decided from the start that I was going to let it ride no matter what. WT has the risk of millions of dollars in investment. If they screw up, either too soon or too late, they suffer. So I'll leave it to them to decide when the time is right - and hope they made the right, or at least a good enough, choice.
And a computer is "just" a machine that gives ones and zeros.
I don't recall the exact names, but I know some fairly recent treg additions were highly critical. As for the comments you read, yes, they love their TB. But what matters is you don't see them saying that automatically means they think it should ship.
True. So what? Because there will always be issues, it does not follow that you ship now. After all, your logic would have applied to the very first firmware version too!
A company doesn't ship based on individual needs. They ship based on a much bigger picture. If the "[" key never works, but for some reason you can say, "But I never type a "[" so ship me mine, that would be silly. No company is going to leave out a standard capability just because some individual doesn't use it.
As for the various cases, a number of them have been posted here, by testers and by WT.
I've reported one that may be unique to me - a tendency to get a C instead of T (on Dvorak, the C in right above the T). There is no boundary for it to adjust. If there were, it may solve that problem. But maybe that particular character pair is not one that others have reported and thus, even though it would help me, WT has not seen fit to add it (I do know changing the boundary is not as simple as it sounds - there are some complicated interactions). Now, if these changes in the firmware involve an easier way to handle boundaries internally (once they make it available, it is easy for the user), maybe that and other boundaries will become available. But they clearly are not changing it just for me or it would have been done years ago.
I've had times jumps didn't work right - though they usually did. There was a time there were a lot of switched characters on certain keys. Didn't happen all the time, but it happened way too much to send to the general public. I think that one may be gone. Or at least it is so rare that it is more likely to me user error.
But none of us can tell you how common these or other issues are in general.
No, it is not. It is their product, their investment, their decision. Not to mention no one outside a company is going to know enough about a product to make an informed decision.