Evolution of technology - a perspective

Those of us who have tried world-building, be it for a story or for some other purpose, have run into this problem at some point or the other. How would the environment affect the evolution of technology?

For this discussion, I will focus on aircraft, mostly because it is one of my favourite areas. Also, what makes it ideal for discussing this topic is that aviation as we know it developed a little over a century ago, and most of its progress is recorded.
What is discussed here is the hypothetical scenario where someone or a group is approaching the development of aircraft. How will they approach it, realistically?

Where do you begin?

Imagine a civilization is developing aircraft for the first time. Where would they start? One of the first things most people would look to is birds - provided birds exist in this hypothetical scenario, of course. They're everywhere, and they're obviously flying, and they make it look easy. This is why we have stories of this sort - Deadalus and Icarus, as well as various fairy tales.
Assuming this doesn't work, what would you look to? The most obvious answer would be gliding - like eagles, for example. Remember, it is entirely possible to glide for hours if you can launch yourself off from a height.
Lighter than air craft is another possible option. These are hot air balloons, blimps, and airships. This requires some understanding that air is a fluid, and that things can float in it. This is probably not exactly obvious to someone who hasn't encountered this before, but at least you can expect scientists to be aware of this.
Lastly, there are fixed wing aircraft. This is the type we use most commonly - with fixed wings providing lift, with thrust provided by engines. The speed of the air moving over the wings creates a pressure differential, which creates the lift necessary, in the simplest terms. The concept is easy enough to discover - try moving a plate of some sort through the air at a slight upwards angle through the air. However, identifying the accurate relationship between speed and lift, and the drag generated in the process, may take time and a lot of experimentation.

What limits it?

Material

A major problem that most early attempts at flight encountered was the lack of suitable material. In general, you want the material in question to be light and strong. Modern aircraft are mostly made of alloys (mostly aluminium based, or titanium based) and composites which meet this requirement. Most models are made of Styrofoam, or balsa wood and spruce wood, and plywood and carbon fibre may be used for specific components. Birds have 'hollow bones' - actually, it doesn't have any marrow inside, but is supported by bone struts to help it take the stresses of flight. Unless you can find material that will do the job, flight would be impossible.
In general, you will need high strength for the wings and wing attachments, engine attachments (if there are any), control surfaces, and joints in ornithopters (bird-like aircraft). You'll also need high strength for internal components of engines (depends on the type of engine).

Power

Another main limiter is power. Early attempts at powered flight were made with steam engines. These tended to be on the heavy side, which was a problem, since the engine had to produce enough power to lift its own weight and the weight of the air-frame off the ground. Lifting a higher weight needs more power, which needs a heavier engine - you get the idea. So, you're going to need a power source with a reasonable power to weight ratio.
There are more ways to produce the power necessary than just engines. You can use some form of jets (again, depends on material and someone actual getting the idea), rockets (Me 163 Komet, for example), or even an electromagnetic catapult to launch it into the air. Of course there is nothing stopping you doing the same with a catapult or a crossbow, but I am not sure the pilot will appreciate it.
In a fantasy setting, you can always manipulate the airflow around the aircraft in question directly, which will likely reduce the power requirement drastically (airbending, in Avatar the last airbender, and Legend of Korra, for example).

Controllability

Figuring out how to control a plane with no prior knowledge about it can be tough. The Wright flyer used wing warping - that is, using cables to warp the wings and control surfaces to make it turn or otherwise respond to commands. Most modern aircraft use a variation of the same - instead of warping the entire wing, they move dedicated control surfaces, like the rudder, elevator, or ailerons. An ornithopter can differential flapping and moving the tail to control itself.

Other Challenges

Size scaling

A bee does not fly the same way an eagle does. Directly scaling up things does not always work when trying to develop an aircraft. For one, if you scale length up by two in all three dimensions, the area quadruples while the volume is multiplied by eight. This is a problem because lift depends on area, while weight depends on volume. You can be in major trouble if you don't realise this (running off the end of the runway without lifting off, for example). This can happen if you're developing the entire field as you go along - people have made incredible obvious-in-hindsight mistakes in similar situations. Another is that the characteristics of airflow change completely based on the size of the object flying - but the exact mechanics are enough material for an entire article on its own.

The risk

If an automobile loses control, especially at low test speeds, you can expect the driver to walk away with relatively minor injuries. Not so with aviation. It's one thing if it doesn't take off, but if the beginning of your flight involves a jump off a tower or a cliff, things can get nasty if the aircraft doesn't actually fly. The same situation can happen if the fragile test aircraft breaks up in midair for some reason. In this case, unless the person inside can fly independently of the aircraft, or is damage resistant, he's likely to suffer some form of grievous injury if not death on contact with the ground. How the characters handle this risk - whether they test it themselves, investigate the possibility of unmanned tests, or delegate the task to others can give some insight into their character.

In conclusion, the development of a new technology is a long and involved process. The presence or absence of an element can drive things in a completely different path. If they had strong, light materials and powerful engines in the renaissance, who knows, we may have started aviation with ornithopters. If it was just the material, gliding may have become quite popular a long time ago. This availability, on the other hand, could have caused other problems that we could never have anticipated. Who knows?

I haven't even begun to cover this immense topic here. If you would like to read it, I can do a much more in-depth, technical article on the basics of flight, with a better explanations of the many things I have mentioned in this article.

I hope that helps!
Not actually an avian,
Falcon-15-X-C