I thank you all for the feedback.
Addressing the main topic, we generally work by the philosophy that the realism should be implemented to the degree that it further the immersion instead of hurting the player. Of course, since the player varies greatly from case to case, this is arbitrary, but we try to judge what will feel limiting to those with mainly roleplay in mind.
Immersion, to me atleast, is also largely dependent on a logical coherence between cause and effect. It encourages you to think in-character, rather than game mechanics. Additionally, if there's no consequence and no priorities to be made, it can quickly turn into vanity. You could use the recent implementation of the encumbrance system as an example. If carrying heavy encumbrance made no differentiation between carrying two times your normal limit and ten times that, there's little reason not to - you could say the game engine somewhat encourages you to - but it foils the experience of an in-character logic.
I think Icono has hit the nail on the head quite well. I realise that the direction we are moving in might not appease all - but that would be utopian. What benefits one style of play is most likely going to hurt another. Our aim has always been for the immersive, atmospheric, deep roleplay. Saying that we aren't true to DnD makes little sense. The premise of DnD has always been to largely let groups define themselves what they want the game to be. On the other hand, you could say we are true to the Ravenloft principles of roleplay, but first and foremost, we are true to our own.
Regarding the loot, as a note, the loot rates haven't changed within the last month at least. The experienced change could either be more players adventuring or the factor of randomness. The items pulled from the stores haven't been pulled from the loot tables either. They remain the same.
The loot rate hasn't even really changed for the last year - but a bug that was causing it to generate too much loot was corrected. When the bug was corrected, it quite possibly left the settings on specific treasure spawns too low. We are intending to increase the loot rates, but it all takes time.
We could likely also achieve more if we allowed more freely for persistent changes and PCs taking up roles, but it introduce a plethora of other issues in it's wake. First and foremost, we can't all have characters in top posts or it would turn awkward. Second, it'll take a lot of time implementing the changes it would cause - likely much more time than we have at hand. Third, it introduces an increased problem of timezones. If the Burgomaster of Vallaki was in the European timezone, all US roleplay revolving him would be impractical if not halt completely. Fourth, it can, and often has, left the world and people in a void when the player behind a central PC (or even NPC) feels to take a break or have real life issues taking him/her away from the game. Fifth (and I promise, final for now) it has to fit into the greater balance of the game. It might be that a powerful mage could turn the Vistani camp where newcomers arrive into a burning hotspot of devilish beings, but it would break the experience for most everyone else. Someone could burn down the Lady's Rest, but it would cause us to loose the great roleplay generating aspect it has, being a central meeting point.
All that said, we are not against persistent changes, but it has to be done as a cooperation between what's in character, the current world and what plots that currently are running, but also with the general experience and balance of play in mind.
I'd suggest, rather than remove all items for a new player to save for and aspire towards purchasing, consider instead allowing some items sold that allow something to be worked and saved towards before that level of total immersion where you know IC crafters or become one. It gives the new player something to look forward to and doesn't detract from crafting aspects since the /best/ will always be crafted.
I think this is an excellent idea
It all takes time implementing though, but I'll have it in mind. The issue was, sort of, that we couldn't just make craftable items more powerful than those available in stores, lest it would become all too powerful.