Is this functionality a bug?
Having skimmed through this thread once again, I'm leaning towards changing my earlier opinion (which was that it's ok to use a branch name in a descriptive manner, even for the "default" fastest completion.) (Now) I think it would a good idea to name publications in a consistent manner. In other words, either: game in time by author or: game "alternative branch" in time by author The first form would indicate the "official world record" version, ie. the fastest completion of the game by any means, within the rules of the site. The second form would be any other TAS of the same game. The branch name in quotation marks always uniquely and unambiguously defines that particular branch (such as "100%", etc.) and its presence indicates that it's not the "default" WR. (It wouldn't be impossible for an alternate branch to be actually faster than the "official WR". It just means that it uses something that disqualifies it from being the official WR, eg. starting from a savestate.) Descriptive tags ought to be limited to the movie classes.
In BizHawk 1.5.3, when I make and save a RAM Watch .wch file with ColecoHawk or SMSHawk and then load it up again later, all of the data types reset to Signed (though their sizes are kept). GBCHawk's RAM Watch gives this error when attempting to load a wch (but loads it anyway) and this error when you try to save said wch (but saves it anyway). Also, in Atari2600Hawk, something I did caused me to get this error while trying to save a watch file and the wch did not save. I haven't been able to reproduce it though.
On that note, how do we use the appropriate categories: Foregoes memory corruption & Corrupts memory?
This means that this hypothetical being ought be able to somehow affect the initial RAM state of the console as the game starts. Is this even physically possible?
Yep, exactly. Just like when you press the stop button with your mouse.
waitToNext() to get the same thing as pcsx.suspend.
I'm more concerned about the fact that this run actually does use glitches: pause tricks to prevent damage knockback and to reset boss invulnerability periods.
I'll suggest a function like emu.quitscript(), to directly stop the script. Currently, if you use multiple script(by using something like require'myscript'), you can't easily stop the script, despite using coroutine.yield() but this isn't a good pratice if the script already use coroutine for other means.
Also, sometime you need somekind of realtime timer to perform specific task, without doing a "busy-wait" function inside lua. At some point, I did a pcsx.suspend(int cycleToWait) function to suspend a lua script(without suspending the whole emulator) for a x number of cycle, but there's probably a better way to do this.