We have already had a similar discussion years ago, when the official SRD was Paizo's own PRD.
A question asking about preferences between AON and the PFSRD already exists, but it doesn't bring to the table an important factor that has become evident to me in the past week: the PFSRD is full of errors.
So many errors, indeed, that people from the Pathfinder RPG Discord server (which has a channel where you can bring AON errors to the AON crew and they fix them) came to consider RPG.SE an unreliable source, partly because some of our answers include developer opinions from the Paizo forums without clearly labeling them as developer opinions (I don't think we can solve this part of the problem) and partly because many answers here rely on PFSRD errors.
While I realize that those people are not the judges of how this site works, it still rings an alarm bell inside my head. I've been speaking to those people, they know what they're talking about. I believe them when they say that RPG.SE is a poor place for Pathfinder questions and I'd like this to change.
What can/should we do?
Here is a list of the main pros and cons of the PFSRD compared to AON. Some are minor, but some worry me.
Pros of d20pfsrd
- easier navigation
- pages with lot of content work well with the browser search
- has archetypes comparison tables
- hosts 3rd party content
(I think that 3rd party content question and answers should still link to the PFSRD, but double checking is a both a must and a chore.)
Cons of d20pfsrd
- part of it is taken traight from the D&D 3.5e SRD, even when PF has made changes (copy-paste from previous SRDs that never got touched?).
- sometimes, the manual reference is wrong.
- due to OGL issues, some feats, traits, archetypes and so on have been renamed, for example Wayang Spellhunter becoming Metamagic Master, leading to players taking the same trait or feat twice.
- due to OGL issues, options limited to a certain region or deity don't list the limitation.
- I have personally found a trait that does not exist in the referenced manual (after some research, it appears to be unlabeled 3rd party).
- same problem with the Still Spell Metamagic Rod.
- The combat page states that a tied up creature is bound, and therefore helpless, despite no such rule existing.
- recently, the archetype pages have been made anew, but there are still links to the old pages around and sometimes the content is different (a friend making a PG stumbled upon this last week).
- the grappled rules in the Combat section said that grappled creatures could not cast spells without somatic components, then an errata come out and this is no longer the case. Before the errata, PFSRD decided to copy that rule to this page, most probably for ease of convenience. Now they fixed the Combat page according to the errata but they explicitly refuse to change the other page (I have been told that they argue that there never was an errata to the grappled condition, and of course there never was since the part about casting spells was not there to begin with), and several players got the rules wrong from the PFSRD.
- takes advice given in forums by developers and treats it as if it was official errata (for example Patrick Renie saying "we have come up with an errata for white-haired witch" turning Constrict, Trip and Pull to swift actions instead of free, but no such errata actually existing to our knowledge).
- mixes 3rd party content and 1st party content (for example, the alternate rules system is full of 3pp rulesets but also has the occult classes - this is correct, but it makes a lot of people think that occult classes are 3pp).
- automatically generated links bringing to unrelated pages (any mention of darkness linking to the same-name spell, the "regeneration" word in the ioun stone description letting player thinks that it works like the same-name spell and so on).
- the curators of the site don't listen to feedback (not necessarily in a malicious way).
I would like to thank Grand Druid Zollqir on the Pathfinder RPG Discord server for their list of PFSRD problems, most of which made it to this question.