I guess that's abstracted into fort level - some holdings in the game do have very low defenses.
Well, you have a low fort level when you have none of the following upgraded: wall, keep, gate (if castle), arsenal (if republic capital) and perhaps some more. This happens in holdings that focus on generating taxes or troop productions or are simply lleft unupgraded.
When something doesn't have a proper keep and wall etc. — you can call it unfortified, or not properly fortified, to be precise. The most basic barony starts with like a wooden Saxon hall / manor house with some fence around it. I'm not sure if cities start with at least a wooden palisade or if you have to build even that. However, how much lower can it get than that? A field full of nomadic tents would probably have sentries posted around it at least and be more defensible than a low-level civilized village populated by settled agriculturists.
Also the idea behind pagan defensive attrition is that the land is unwelcoming to foreigners and you can die at every step. Kinda runs counter of the idea of a poorly fortified place.
Besides, tribals lived in wooden forts anyway,
motte and bailey style. Which is not much different from low-level holdings managed by rural lords and castellans under European kings.
As much as I'd love a more complex vassal system - especially equal-tier vassals, multiple vassalage etc. - I don't think the new features are going to be much more complex than regular vassals, with the noted exceptions.
Yes, of course. I don't think equal-tier or multiple vassalage will happen any time soon, if at all. If anything, I'd expect a courageous and inquisitive modder to find a way to game the system and make it happen, then perhaps it could be included by the devs into an official patch, if at all. But I'd especially like to see England throwing its weight around the isles without rising to the empire tier (notwithstanding the Wessexes' bloated titulature and Henry VIII's rhetoric about, 'the imperial crown of this kingdom' — let's see Spain or France concede that to them, forget the HRE), Slavic dukes having other dukes as vassals, German families of hedge herzogs loosely headed by one family head etc., and Pomerania as a vassal of Brandenburg rather than directly the HRE.
Besides, I think even already the way things are now, in those periods in which the heir of England bent knee to France as Duke of Normandy or Aquitaine or something else, he should be put in the game as a vassal of k_France and not k_England. However, in such a case care would need to be taken to avoid HCA messing up English succession when the heir happens to be a French vassal. (Of course, double vassalage would be better here.)
That is to say that they will be in your colour blob, with outsiders unable to declare on them.
Tough luck.

I think tributaries should be attackable, although with a high chance of their suzerain coming to their aid if not prevented.
My personal opinion is that vassalage should always be an alliance mechanic of some sort, because that would be an easy way to represent equal-tier vassalage, and to represent raiders being able to attack border-counts while kings were occupied elsewhere. I doubt the system will be changed for CK2 though - perhaps in an eventual sequel.
Actually, if you raise vassal levies through the Military tab rather than the map, it does look more like an alliance. Especially if you scroll down and pick individual vassals with their portraits and troop stats to mobilize. Kinda like with a merc band. Not
that much different from a vassalized merc band or holy order.
I don't think I was playing while distance penalty was a thing, but it does seem like a shame they got rid of it. It seems like there should be an incentive to land your dynasty as independent rulers for your distant realms (particularly when inheriting another kingdom), to keep the power in the family, without having the silly system as now exists of Denmark owning e.g. Hungary quite happily.
The problem with the distance penalty was the even people who were your biggest fans according to the opinion variable still had a high chance of revolting against you. And they could do it while bordering a completely overwhelming neighbour with a holy war CB. The latter is why there are restrictions in the game right now concerning white peaces or any peaces in general when some holdings are occupied by either the liege they're rebelling against or the foreign attacker. This is because it was possible for them to lose a holy war pretty much before you could arrive with your ships.
Around the same time, the faction system was introduced to put an end to individual declarations of war by vassals. The problem with that was that those guys could draw all your other vassals in, successively, and also their own allies. That made quite a lot of people, and with all your vassals in revolt you only had your demesne troops and mercs to rely on (that was before retinues, which came around the same time as the faction system, i.e. Legacy of Rome). The artificial independence of the rebelling vassals made them ripe targets for your opponents. You were unable to be in so many places at once, and realms disintegrated easily.
The dominating sentiment back then was that realms shouldn't be so unstable. It was justified to a large extent because normal European monarchs seem to be experience Golden-Horde-style successions. Surviving successions was a science until recently — see the succession guide in my sig? It comes from before the faction system was put in place and is a good picture of how tough things were back then. Not that managing factions is a piece of cake or a boring thing to do.
This also concerns me because I like to play to convert games to EU4, and that gets messy when realms have random lands all over the place.
Shouldn't be happening in CK2, either, from the POV of realism, unless we're talking about maritime countries like Aragon, Naples or Italian republics. Those guys would typically spread out quite a lot, but normally you'd need to appoint a vassal or some sort of administrator for somewhat remote holdings. Even Croatia, which bordered on Hungary, had a constitutional viceroy (the
Ban of Croatia, all the way from 1102 to 1919). Managing the Netherlands as a possession of Spain or Austria was a huge deal. Keeping together two or more Scandinavian crowns wasn't easy, either, or even trying to rule Ireland as England.
Therefore, I believe being King of Sweden, Castille and Greece or something equally patchy should be way more difficult than it is now, with particular incentive to appoint a brother or son to rule there and be a permanent ally. (Equal-tier vassal mechanic could come in handy in such situations.)
However, having baronies all over the place inside a noble's home kingdom and even neighbouring countries was pretty normal at some point. Except we'd need some sort of multiple vassalage to make it worthwhile, otherwise we'd be stuck with HCA (non-inheritance from abroad) and county wars. In any case, if you see a patchwork in England around 1300 AD, the map is probably still not as funky as it would have been in real life. Same for France or the HRE.
That all sounds way too advanced for Crusader Kings, as amazing as it would be to have. I'm imagining there will be some increased attrition across the board in winter, and some modifiers for different units in different seasons.
Well, I'd definitely love to see the effect of desert sun or tropical heat (jungles) on European-style heavy cav. Even cataphracts who were born to the Med climate were soaking in their own sweat. Now put them on the Sahara or in a rain forest...
My main problem in many strategy games. The timescale is abstracted so that it is unfeasible to say your armies should be at home during winter - if you decide to attack in Spring, it's unlikely you'll even have your armies all set up in enemy territory by the time winter comes, let alone have sieged a holding to hole up in. My main hope is that you can use sieged holdings to avoid excess attrition in winter (I'm not sure if this currently exists?)
Yeah, even in turn-based games like TW this is a problem. But when a battle or fort assault takes 2-3 months, which is not realistic, then sticking to winter penalties for the sake of realism doesn't make sense. In can rightly feel unfair because in real life it doesn't take 2 months to resolve a battle. Moving through a province also takes longer than you could do it in real life. Real-life armies don't move at 200 km per month. This is not a problem when attrition is also heavily abstracted. But when you get winter penalties that last precisely from December to April, or something, then the excessive precision of one part of the mechanic compared to the whole system is not going to help realism.
Except sieges, as in when you finally arrive and aren't currently assaulting but just sitting on the siege. Then yeah, normal December to March/April winter would make sense.
But not where movement or battle comes into play.