Where i mostly disagree is when you talk about a complex system someone mad in war3, a great deal of skill or a complex project inst required to have a successful project. I think a project well done, that can catch enough interest simple or not, can be successful. But doing the project is hard enough, and even harder to have an project that will have a high success rate. But I cant talk to much on the topic as I had a successful project my self.
No I was referring to someone proposing a very complex system. By this it could be a "realistic AI that learns your behaviour and changes its play style to best counter you, including chatting with you and adapting to random events automatically" kind of thing. I have seen people post that as a goal for WC3 projects yet the idea itself has never been successfully implemented in massive budget real games. The idea sounds great but a "reality check" would bounce it back.
There are certain signs one can use to get a good idea on project success.
How well written is a project proposal? If it consists of only 2 sentences that give virtually no detail next to a rough idea then the idea is still too early to even think of getting a project going. Although it could be developed into a viable project most cases will be failures as they were just day dreams or spontaneous ideas with no one to follow them through.
How well thought out is the proposal? If the proposal has nonsense logic like a pyramid scheme or lacks any form of reality check then it will obviously fail in its current state. Such projects might be able to be made realistic with some revision but most will give such outlandish ideas that it is doomed to fail from the start. An example would be a MMORPG where everything is made up of atoms with ultra realistic physics, sounds cool but such simulations are beyond the capabilities of computers for many decades to come. It need not even be so alien, it could be someone describing a absolutely massive MMORPG with movie like graphics covering an entire earth sized planet which is possible with modern technology but clearly not with the resources at the guy's disposal.
How technical is the proposed project manager? If the manager lacks any understanding on the proposed subject, refuses to learn and says "oh I am the idea guy, others do the work" then there is really no hope for the project to get anywhere. An example would be someone proposing a building design that floats free-standing in mid-air and can change shape and self-repair if damaged, sounds awesome but anyone who has any understanding of physics knows that it is impossible and so should he.
How reliable is the proposed project manager? The biggest single failure of most projects is not viability, knowledge or even resources but the determination of the guy (or guys) in charge. If they go then chances are the project goes with them. This is the fate that almost every WC3 map has suffered as the map makers have long gone and map will see no maintenance ever. Kids or young adults unfortunately are often not that reliable. I myself am an example of this where I was maintaining the SC2 map Undead Assault 3 but due to university had to stop. Even though the map is heavily played on BattleNet (several dozen games a day) and I made all resources available to the public for further development, no real work has been done since I had to stop since people are conflicted what they want and no one has expertise to continue. If a project manager is not in a position to dedicate the time required for most of the duration of the project then chances are it will fail.
Anyway back on topic...
You do not need actual office space to develop software. You only need the tools and the labour. With the internet this is especially the case since you could employ people from home or even on a contact basis to fulfil labour requirements. Most tools do not need office space either since software and workstations can be setup at home. Communication can also be done with VOIP and other free real time services.
Why all the effort when an office would still work? Well offices are very expensive, and the savings can be passed on in the form of extra development resource. Also indi developers will likely not want to move to a physical location just to get some help with their project. They will however seek some help which is where decentralization helps.
So what could you do? Well next to consultant work managing projects you could also offer them labour resources for development. Additionally you could inherit any maintenance work from the projects if you have some form of share in the developed project and so keep them future proof (fix bugs, release on new platforms etc). You could also manage labour resources by a job based system where you advertise for specific jobs (such as part of a game engine, an artwork etc) and when it is fulfilled the worker receives a pre-determined payment. As an organization you could also have an IOU system where you promise to pay for work done if the project ever becomes profitable (the earnings from the game first go to cover the IOUs before any profit is considered).
