As I and others have mentioned, there are issues with the TextBlade.
Some would be strictly personal. That is, no matter how good something is, it may not fulfill personal preferences. This could be something fairly universal for a given situation, such as someone who says it is tough to use when coding in particular languages. In those cases, there may or may not be work-arounds that the tester hasn't figured out or it may be that the TextBlade won't work for anyone in that situation. Or it may be just personal preferences. For example, you can find people who just personally prefer one feel vs another when typing. Problems with feel don't matter in the big picture though. You can't cater to every individual preference! Problems with feel only matter to the big picture when the feel is just not good for most people - a common problem with small mobile keyboards, not to mention some full size ones!
As testers, the problems that matter are actually bugs in the system mainly. And sometimes things that it may not offer which would be very useful. An example of this is the filter timing settings that were added after I started testing. When doing different layers, the timing is critical. Hit the space bar and are you trying to access the green layer or just type a space? Turns out the way people type varies enough that the timing can benefit from being able to be adjusted.
So, while there are things testers have reported that they would like to see, I don't think any of them are planned until after shipping begins. At least I gave a list of things and was told any that may be done would be after shipping and I've seen other testers report the same response. And we are fine with that. None of us are trying to get our personal idea added in at the expense of shipping. It is just part of our job to point out things we believe will make it better eventually.
Which leaves us with actual bugs or aggravations with the fundamental design that isn't going to change.
From my view, there isn't much to say negative about the design. I wonder about some things though. The biggest one is the choice for multiple keys on one keycap vs a separate keycap for each key, like regular keyboards have (but retain the size of the TB).
I have no idea what problems separate keys could cause. It may be the keys would be so small that they really couldn't get a good key feel. But if that could be done, I see one big advantage - it would be easier to feel errors, thus easier to avoid them. While there are edges between keys you can feel, it simply isn't the same thing as feeling a key which is totally unconnected to its neighbor! I almost never hit the wrong key by moving my finger across to a neighboring key if it is on a different keycap. But I will do so if on the same keycap. This isn't a TB error in interpretation (for me, at least, those seem far in the past). It is just easier for the finger to wander.
Of course, this part of the design isn't going to change anyway. At least not in the first version to ship.
Which brings me to just the possible
Don't want to say too much. As I've pointed out before, I think this is stuff WT should report on and in more detail (since they have more data) than I can. And because I think it is up to the tester with a specific problem to decide what to say publicly. But I can give some info, in general:
I have a list of those who have reported any problems that may not be a one-off situation. I don't count the one-off stuff because with electronics, stuff happens that you can't avoid, but the product is fine. For example, today Siri suddenly stopped working on my iPhone. I got the wavy lines but Siri didn't seem to actually hear anything. Did the usual first fix for such things by turning the phone off, waiting 30 seconds, and restarted. Everything worked. In all the time I've had the iPhone 6s, I believe this is only the 2nd time that has happened. So I wouldn't count it as something that must be "fixed" by Apple.
Some things that are reported are not obvious as to whether they are user error or TB error. So I don't count those until the tester has had a chance to make sure it isn't them.
Which leaves us with those I do put on the "problem" list. Not too many. And only a few have had recent updates. I give dates for the most recent report for each tester. At least I started including that back in June. Well, I just checked and I can say only 3 people reported still having issues in October so the odds are they still have them.
The next most recent have dates way back into August. Various updates may or may not have solved them since then. The rest go back to July or late June plus one with no date (so even earlier).
So the "not so good" stuff isn't big and they may not all still exist (it always helps if testers periodically confirm or change their status even if nothing has changed!).
The 3 recent issues cover things like lagginess, connections (these could be the same thing depending on details) and some key repeats.
The other ones mostly deal with much the same things.
My own experience isn't perfect, though I don't consider myself to be in the problem group. I've had lags occasionally. Mostly they seem to have been related to running test software because I've left it off for a long time now and gotten no lags. I also think some lagging has been because of a particular website being bad. So I'm not concerned about this.
I have had connection issues. Main example is, as part of my testing, I go to the Apple store or somewhere and pair a slot to a different device. I usually use slot 4, which normally is one of the slots that is pared with my Mac. There is a routine to follow - hold down the chord for slot 4 for 10 seconds. This is supposed to completely erase the pairing info on the TB. Then pair with the new device. Well, normally that works fine. The problem comes when I return home and try to re-pair to the iMac. It often will not cooperate. Another step to take when pairing again is to remove the old paring first. I've tried it that way as well as without removing the old pairing, but still not get it to work. I have not figured out a pattern to this. I always eventually get it to work though. And normally people aren't going around changing what they have paired anyway.
Usually, but not always, if I'm going to get a successful pairing, the iMac will show "TextBlade" on the list. You select it and the name changes to "TextBlade 4xxxx". The first number is the slot it is in. The next three are hex numbers. I don't recall if the number are added after you select it or after you enter the 6 digit pairing code.
Well, when I'm having trouble, most of the time it will start out showing the number after "TextBlade" immediately. Which means I can pretty much expect it simply isn't going to connect, at least not for ore than a few seconds. If it shows just "TextBlade" at first, it may connect fine. But not always.
So, for me, this is really the only problem that concerns me, even though it is just my testing situation that makes it show up much.
As for shipping, if they all worked like mine, I'd ship. But they don't all work like mine. And I don't have enough info from others to feel justified in saying what WT should do.