Design goals for a new edition are essentially a list of things that are wrong with the previous edition, plus a list of things that must not be changed while trying to fix that.
1. Create a new edition of Pathfinder that's much simpler to learn and play—a core system that's easy to grasp but expandable—while remaining true to the spirit of what makes Pathfinder great: customization, flexibility of story, and rules that reward those who take the time to master them.
2. Ensure that the new version of the game allows us to tell the same stories and share in the same worlds as the previous edition, but also makes room for new stories and new worlds wherever possible.
3. Work to incorporate the innovations of the past decade into the core engine of the game, allowing the best rules elements and discoveries we've made to have an integrated home in the new system (even if they aren't present in the initial book).
4. Forge a more balanced play environment where every character has a chance to contribute to the adventure in a meaningful way by allowing characters to thrive in their defined role. Encourage characters to play to their strengths, while working with others to bolster their place in the group.
5. Make Pathfinder a game that's open and welcoming to all, no matter their background or experience.
#1 says PF1 is too hard to get into. I agree entirely. That is it's greatest flaw by far. The classes that get some fiddly little thing every level, on top of archetypes that change arbitrary elements of that progression, on top of feat chains and all the other options, it's just stupidly complicated to understand how to turn an idea into a character by reading the books.
Liking system mastery is kinda weird in 2018 though. google exists. For 3.0, in 2000, I understand system mastery, it was a big deal in the 80's and 90's, and the prosaic books of yore made searching for stuff less bad, but it's a stupid idea now. Everything needs to work because people just look it up, and the books are way too full of math that most players don't interact with, they just google the answer.
So I'm gunna say that last bit is bullshit and the more they stick to it the worse the game will be. Which, kinda sucks in a game with lots of options, if none of them are particularly good or bad, well, that makes choosing painful, rather than fun. See also CRPG changes in the last twenty years. Way less choices, way more impact behind each, that's the current zeitgeist, put it up front rather than hidden behind twenty garbage choices that people bypass with a net search.
#2 ... PF1 didn't really support the stories in the modules they made for PF1, so continuing to support the PF1 module stories and also many more stories, uh, OK. Like, it's still all XP for killing things and then, uh, we're not supposed to slaughter the entire town before setting off to the dungeon? Money is power but not greyhawk the whole planet? Kill the magic mart merchant in his sleep and win? Pfft. Bullshit, they're not even trying.
#3 Keeping all the good splatbook stuff. Which, they seem to think is "alchemists". Here, designers seem to keep falling into the trap of believing people like the concept of "a character throwing bomb potions", rather than people just like effective characters. There are people who play Monks because Ember looks cool, but mostly they end up not liking RPGs because their GM didn't throw enough unique powerup artifacts at them to compensate for being a Monk, while the "system mastery" people just don't play Monks.
Which is to say, designers often get the completely wrong idea about what is "good", when trying to keep it. This seems, uh, fully on display with PF2.
Like the thing with getting rid of wands, and then replacing the function with something else. People don't like the emergent game properties of the function! There's a huge problem with every fight being tightly balanced where if one PC doesn't show for the session you have to change all the fights so the party can win at all, it's so much mind caulk to overlook that all the time, and smaller fights just being totally pointless.
More than that though, if you went to see your favourite band and they asked you personally what your favourite song was and then just played that same song all night, for hours, eeew. It doesn't work, you need little fights for those big drag out fights to feel good, and for little fights to be a valid use of game time you can't have unlimited post-fight healing. It's just bad. People express that as "I don't like wands of CLW", but it's not the wands!!!!!!
Lots of little changes in RPG editions are like this. 4e D&D was so much this all over the place. People can't tell you they dislike healing, they like healing, they like being fully healed, just the game that results from everyone being fully healed all the time is terrible, so they tell you they dislike wands of CLW, but also like being fully healed, and the response to that in PF2 completely misses the point.
And it misses the point because PF modules have always had bullshit hard fights everywhere to compensate for the wand of CLW effect, which worked and people liked within that rules complex, but for a new edition it's just a bad idea and they do not hear that.
#4 Is apparent in the playtest. They tried to give you strong mechanical reasons to play within your character's niche, and support others within their own niche. As you can see in this thread, some don't much like that, especially when one class' niche is "healing", and other classes like Monk still don't have a niche that works at all. Plus, the numbers on what monsters put out make some of it dysfunctional.
@GâtFromKl, I think that's what they mean. So Rogues should have a niche they shine in regardless of system mastery, but they'll be better still if you get gud (or google the answer like a real person). I'm not sure they'll get far on it, but that's what they're aiming for.
#5 ... hmm. I think they mean they want to be less racist and stuff. That's a fine life goal.
PC, SJW, anti-fascist, not being a dick, or working on it, he/him.