The dream part is not the hard thing, it's the grind that people are stuck at almost all day every day. The reality is that if you have a dream, that's fine but if you are not willing to work for that dream then it is not a dream but just an idea. What you should do instead is to make sure that you end up with working hard towards finding the path and then taking those steps in order to achieve it.

I personally never really do that, did it once in my entire life and created a project with friends last year, wasn't a big success, but at least we recouped our money back (well I did and a friend did but third person didn't). This means dreams can come and go, it's the grind that matters.

firstly you need a 'thinktank' of people in business, multiple industries and economics that can come together and think about what niche service/economy/system is missing from the world.. before even thinking about coders

it requires people in different business disciplines to know about different industries enough to know whats missing from each industry to then have idea's of what missing niche needs filling the most.(what would actually be useful)

you need economics people to factor in the different ways to make your niche valued for utility(get people to want to use it and treat it as worthy of using) and also to make it valued to then make money out of to pay for those involved and also incentivise those using it.

you also need business people involved that actually use the idea. so that they can alpha test it in the real world and point out what features/benefits/utility are needed for it to work.

EG
many idiots planned to make medical record blockchains. but had no one from the medical industry involved who could point out what data/format/structure or message packets between peers should look like/contain.
they also had no medical experts involved to then actually alphatest it on in the real world to see if it was actually useful/needed in the medical industry. .. hense why those projects failed before they even begun...

.. but once you have an idea in mind and have an idea of a niche that needs to be filled
THEN and only then you can get the coders in to then turn the paper idea into code. and then trial it out.

as for how to combine the team. well thats a discussion to have with the team. as each person has their favoured communication portal. so its a conversation you have later about how you want everyone to talk together collaboratively.

as for blockchains. well thats one option. there are others.
there is also 'ball&chain' whereby instead of a ledger which always keeps every record for eternity. what can also be done is after X months/block. the algo 'milestones' historic data.
(imagine it like a snapshot of a UTXOset from the point of view of a specific date that is fixed)
parse out the 'spent'/useless data from this snapshot and 'balls' up the useful stuff. and at a later second mile marker. snips/prunes the chain at the first milestone, to only keep the ball of that first milestone.
EG at block 210,000 ball up all unspents of 0-210,000. then at 420,000 spin the chain of 0-210,000 and just have the ball of unspents of 0-210,000 attached to the chain of 210,001-420,000. whereby the network has had blocks 210k to 420k to audit, check and verify that the ball of 0-210k is indeed good/acceptable.
then repeat at every milestone gap of 210k blocks.
thus future users have less to 'initial download' to get upto date. and data is not eternally locked/present even if spent/used

there are other options aswell as this. that can be used.