I'm not going to quote everything, because basically everyone has been the saying the same thing over and over. However, based on remarks specifically that ForeverSlayer and others have made, I just wanted to point out a couple things I haven't seen mentioned.
First off, what we're getting now is highly irregular in this day and age. The term "beta" has really taken on a marketing tone that has really replaced the concept of a "demo." By the time a game goes into beta today, it's become expected that it is playable with perhaps a few minor tweaks here and there. That is to say, it's become expected that the mechanics are generally locked in. This is, of course, not what we're getting here. At this point, we're clearly in what would be considered an "alpha" stage of development - core mechanics are still being worked on. Very rarely does any company release a product for open testing at this stage of the game. I think some people are overlooking that, either deliberately or not.
That brings me nicely to my second point. I feel that Wizards of the Coast has done a fairly abysmal job at communicating their motives. One big thing they could do immediately is explicitly state the game is in alpha, and what that means. Clearly, as per above, I think there's been a breakdown in understanding the fundamental purpose of these packets. However, I don't think it stops there. There really needs to be communication from WotC as to what exactly each packet is testing. I've found often I don't know (though sometimes can assume) what WotC is focusing on until someone here does the survey and reports on it. Perhaps there's some degree of single-blind testing (ie, subject is obfuscated as to the test) that is useful for WotC. I can easily see people providing input without actually testing things; after all, tinkering is a key component to the RPG hobby. Even so, I don't think it's unreasonable for them to put some of their goals for each packet directly in the "Read First" document.
There's a third point to be made - testing is boring. I don't know about anyone here, but I do not want to spend my free time doing the same thing over and over. However, especially at this early stage, that's pretty much what testing entails. In this, WotC is in between a rock and a hard place. They need to release new stuff in every packet to keep player interest up, otherwise the whole point of the public test is lost. On the other hand, what they really need (for example) is groups running the core four levels 11-15, multiple times, with a strong emphasis on (ie, only) combat. That sucked just thinking about and typing it out. However, that's what testing is. What testing is (mostly) not is playing your weekly campaign with the new toys. So unfortunately, WotC has to make a judgement call balancing player interest and "clean" testing. I would certainly agree that recent packets have swung too far towards player interest. For example, I find no reason to have high-level testing when basic class mechanics are still being altered. However, given the clamoring for new playtest documents here that crops up roughly once a month, they clearly can't release a packet that's levels 1-5 with basic mechanics changes.