I agree that the mapping and movement part of the game seem like an excessive extrapolation of dungeon-crawling rules, and that things work relatively fine if just replace the scale with something else and keeping the caves 11 hours direct march (2-3 days with exploration) from the keep. What bugs me about the existing scale (much more than travel speed, which honestly often felt off in TSR-era D&D) is that 6000 yards, much of it with a road, is too close to have a major hub of brigand activity, plus several man-eater monsters, for the threat to be only starting as 'minor issue the keep is willing to pawn off to adventurers.'
Regarding the sparsity of town detail, I think there's room to disagree. Potentially based on what you expect from the towns. Justin Alexander
posits that Gary "
primarily saw cities as the hub around which adventures were based: You came back to the city to get supplies and hirelings. You left the city in order to have adventures." I can't find it, but in another post I think he highlighted how the description of the keep rapidly provides for the players all the information they need to use the town for that purpose -- the main drag is set up with all the different locations an adventurer would need to know about roughly in the order they would be useful -- Gatehouse where the guards can provide the most basic information on what's happened since the party last visited, then stables, then shops and chapel (for healing) before the inn, with a separate inner baily full of the rich and powerful people which can literally be cordoned off until the PCs finish the quest and earn enough renown to deal with such people (and need future adventure hooks). I think that's a not-unreasonable conjecture about the purpose/logic. It's also worth noting that this is intended as an introductory adventure, where maybe it is okay if the guards are 'the guards' and don't have a backstory or wants and dreams. More importantly, that is the time in your (players and DMS) gaming career when too much detail and diversion from the intended adventure (the Caves) might lead to people getting stuck on side plots and not get to the exciting adventure or something.
I don't know. Whenever I've run it, I've given townsfolk names and personalities and plots going on ahead of time, but only revealed them when the players press. It didn't seem hard, but a pre-made write-up (perhaps in an optional extra pamphlet labelled 'adventures in town') would have been helpful for a new DM.
What I rather more dislike about the module is the incentivization structure it teaches with meeting strangers. There are <spoilers>
three specific instances when they run into people wanting the them to help out, offering to help out, etc.:
the hermit, the town cleric, and the lady (medusa) in the jail cell). In each case, the party has to decide whether to trust the person/help them out/accept their help or not. In each case, it is a bad idea. Not in isolation, but I think this did contribute to the murderhobo reputation of that era of gaming, as it at least kinda telegraphed that doing otherwise was a bad idea. In my own games, I tend to give them a random chance of being hostile, neutral, or helpful.