ive been working on a world now for small bit and was wondering how do any of you guys start world building a setting   im sure others might like to know to 

i tend to start with an event (a war or a treaty) or a forced constant(something that people take for granted but know is unstable such as peace or magic access)

The first question is how do I limit travel.
Let's say I make it easy and have a baron's Highway, quick and reasonably cheap if you choose a taxi carriage. 
go from one side of the small barony to the other.. how can that be limiting.. it only stops on those limited numbers of settlements on the highway.. 4? 3? 5?.
maybe wagon roads, to rough for delicate carriage wheels.. But who owns the bridges.. and do you need a permit to travel.

But this is only one barony surely others.. but if this is a winding river valley.. and the river is too great for a bridge.. it may mean you have to travel by keelboat or commercial raft. 

Quote from Toddy_Shelfungus >>

The first question is how do I limit travel.
Let's say I make it easy and have a baron's Highway, quick and reasonably cheap if you choose a taxi carriage. 
go from one side of the small barony to the other.. how can that be limiting.. it only stops on those limited numbers of settlements on the highway.. 4? 3? 5?.
maybe wagon roads, to rough for delicate carriage wheels.. But who owns the bridges.. and do you need a permit to travel.

But this is only one barony surely others.. but if this is a winding river valley.. and the river is too great for a bridge.. it may mean you have to travel by keelboat or commercial raft. 

 thank you i never even thought of this 

I start with a theme, like “Slavic folklore” or “Italian Renaissance,” then build a starting town and a few starting adventures. I focus on local locations, quests, and NPCs. Then as time goes on, I branch out. Anything bigger is too big to care about, if I try to bring in a massive history or super powerful heroes it just takes the focus off the players.

Mine started with a single idea for a location which I bulked out. As I expanded on the location I had picked (for arguments sake, a town) I started to think of the area around it - so the farms, and the roads, trade routes, and subsequently towns to trade with, and then added a coastline to give ports for the trade goods to come and go from. Then I added small settlements along the trade routes (a days travel apart, started from inns to sell rooms to traders, expanded to permanent settlements with minor trade with the caravans). Then I added some history to the world, and that further developed the map (for example, the gods once attempted to wipe a certain advanced civilization off the map with a tsunami, which caused an incredibly deep bank of sand dunes, which itself turned into an ever changing desert), then added more features to the map on top of the things which were added for the history, and then started assigning names based off the historical information I'd created. I took any sections of the map which were mostly empty and added mountains, forest, moors and swamps for flavour and am currently building from there.

I have been steadily writing up ideas for adventures, most of which are one-shots in my setting, and then adding the places which would fit the adventure to the map - if an adventure is set in a dwarven mine, I add a dwarven mining settlement to the map. If it's an abandoned mine, I add an abandoned mine, an ancient road leading to it, and a reason for its abandonment.

My focus is on making the world feel real enough that:

A: The players feel like it would exist without their being there
B: Events have been happening before the adventurers arrived, and have had consequences.

I am also (to make things more difficult for myself) building the world at least twice, to allow for the prospect of time travel within the world. The history I've added is ancient and unalterable, and I intend for the adventures to start at time A, and then the party travels (inadvertently, possibly without knowing) forward in time by about 50 years. The land is much the same, but there are some important new settlements.

If you're not sure where to start, then start small. Make a town that's fairly isolated in the wilderness, come up with immediate surroundings; forests, rivers, caves, dungeons, etc, then populate the town with npc's and a few plot hooks, decide in general what these npcs care about or believe in, hammer out a few details there, then let your players explore the area for a few sessions. You worldbuild more the more your players push out into the world, eventually shifting from them playing on the micro scale of your world to the macro scale. 

Best way to worldbuild is a little at a time, not all at once. 

I find that using the website world anvil is extremely helpful in fleshing out worlds. It has a ton of options for the tiny specific details that you normally wouldn't think of.

I personally built my word off of what kind of world I would want to live in taken to an extreme. I love deserts and heat so I started with making the planet one big desert, but I also added variety by looking up different desert animals that i could combine with creatures not native to that region. I then built ecosystems around those animals, creating a food chain (I am a big biology nerd :,)) and adding plants that could grow in that kind of environment. Then I focused on the races, building characteristics that would benefit the people living there. I found some cool landmarks that I thought would be interesting to add and created separate regions with different climates, still fitting the hot desert vibe. Mostly, I build from creature designs first but you can honestly start with anything. Food, transportation, weapons, ore/metals, climate etc are some good ones.

Quote from cynophobia_ >>

I find that using the website world anvil is extremely helpful in fleshing out worlds. It has a ton of options for the tiny specific details that you normally wouldn't think of.

I personally built my word off of what kind of world I would want to live in taken to an extreme. I love deserts and heat so I started with making the planet one big desert, but I also added variety by looking up different desert animals that i could combine with creatures not native to that region. I then built ecosystems around those animals, creating a food chain (I am a big biology nerd :,)) and adding plants that could grow in that kind of environment. Then I focused on the races, building characteristics that would benefit the people living there. I found some cool landmarks that I thought would be interesting to add and created separate regions with different climates, still fitting the hot desert vibe. Mostly, I build from creature designs first but you can honestly start with anything. Food, transportation, weapons, ore/metals, climate etc are some good ones.

That sounds interesting. I recommend Frank Herbert's Dune, especially the appendix on ecology, for inspiration if you haven't already read it.

Quote from CharlesThePlant >>

If you're not sure where to start, then start small. Make a town that's fairly isolated in the wilderness, come up with immediate surroundings; forests, rivers, caves, dungeons, etc, then populate the town with npc's and a few plot hooks, decide in general what these npcs care about or believe in, hammer out a few details there, then let your players explore the area for a few sessions. You worldbuild more the more your players push out into the world, eventually shifting from them playing on the micro scale of your world to the macro scale. 

Best way to worldbuild is a little at a time, not all at once. 

Can't +1 this enough.  It's also echoes the advice given in the DMG.  Here's my current mode: What I do is usually set levels 1-3 in a very limited geography.  Either entirely within a single medium to large city, or small town or fortification from which they make a very limited excursion "into the wild."  I drop hints at "larger things" at play via an NPC encounters/gossip, things flying overhead or overhead in passing, scraps of different stories and odd pieces of evidence.  When we're done, I literally poll the group.  I provide them a list of elements they've been exposed to and themes they could play with (war, heists, ladder of power climbing, exploration, mysteries, crime fighting, peacemaking, etc).  As well as environs (ice, jungles, deserts, etc.) and monsters/adversaries (pirates, dragons, devils, oozes, ghosts ... keeping to broad categories) that would entertain me to put the party up against).  Etc.  I don't have my survey locked in it yet it's something I develop.  

Anyway from this list of key terms, they get to pull ten from the lists total (some may want to go to specific types of places and don't care what they encounter, some want to come against particular types of adversaries and don't care where, others just want to steal and con their fences for maximum profit).  Then I have them use mentimeter which will generate a word cloud, it's a fun tool and free at least at the level you're average DM would employ it at.  I use this word cloud to determine the plots/themes (more important) and figure out from there how to construct the setting (not as important as a lot are led to believe), making sure to more heavily weigh the "bigger words" and use the "smaller words" for side quest opportunities.  

When I first started DMing, yeah, I covered my kitchen table with I dunno 16 or 25 pieces of quad ruled graph paper and drew whole continents and kept three ring binders of varied geopolitical entities ... and never really used it.  I've learned a lot about writing and storytelling since then, admittedly skewed toward more experimental forms that call into question classical styles of presentation in fiction, as well as opened up my mind to this important concept called "feedback" when engaged in collaborative work.  I believe what I do now is a superior mode to the "DM auteur theory of world building" because it's a heckalot less labor intensive, grants player agency while still granting the DM the role of arbiter of that agency, and literally gives me stuff I hadn't thought of which, for me, is probably the greatest benefit I get from TTRPGs in contrast to the sort of writing I do professionally and recreationally.   

I might start with the question.. why?

That is... Why build a whole world from scratch when there are plenty of options already in existence to work within.

I'm not saying don't build a world... But start with why. Will it be significantly different in some way? Or do you feel restrained or overwhelmed by existing lore for FR, or Greyhawk or Tamriel?

When I use forgotten realms... As a dm... It is my own personal copy of FR. Maybe I want neverwinter to look like it did a hundred years ago... In my world, mount hotenow flowed more northward.

Also, consider the group. If my players all spent months of their lives living in skyrim virtually, I would rather have them adventure thru Riften and Windhelm, places they know already. It takes considerable effort to get the lay of the land in a imaginary world... Perhaps familiarity would be a boon for yours players

I usually start by what genre I'm making a campaign for. If it's meant to be a fun adventure then bright colors, probably sunny days, and Spring/Summer time for setting. If it's horror then dark misty days, cemeteries, and horror themes like the undead/gore/creepy things. I also like digging through fantasy concept art to get ideas for cities and stuff.

“I wisely started with a map.” ― J.R.R. Tolkien