One aspect of space warfare neglected in the previous writeups is that of relativity.
On a strategic level, space warfare would take military theory back several hundred years. In the days of old, fleets of ships were dispatched with their orders in hand. It was difficult to change these orders once the fleet was launched. If space ships manage to travel light-days, light-weeks, or even light-months from their port of origin, communications between command and the fleet will become outdated and virtually useless. Like in times past, fleets will have to play hide and seek to engage each other. However, unlike the old days of naval warfare, it will be much more difficult to seek an enemy fleet in space. At sea, boats are only able to move in a plane, but in space ships can move in three dimensions. While, the area of a circle is determined by the square of its radius, the area of a sphere is determined by the cube of its radius. This means that as the distance traveled by a fleet increases, the probability space of where it can be is much larger than that of a naval fleet's. This will lead to an inability to locate enemies which will make it likely that defense and attack will be simplified. Instead of grand strategies, fleets will proceed in an unimpeded way to their targets of attack. Waiting for them, at the objective, will be a rival fleet that will fight them in orbit. The defending fleet will have the duel role of shooting down ordnance launched at the object of their defense while, at the same time, shooting down enemy ships.
On a tactical scale, relativity would also heavily influence space combat. Ships flying at or away from others at high velocities would exhibit red or blue shift. This could cause the use of radar to fail. The expected return frequency of the radar and the actual could be so different, due to red/blue shift, that the antenna would not pick up the return. Additionally, if the ships were travelling very close to the speed of light, radar would be useless, anyway. By the time the radar return from an enemy ship would be detected, it would already be on top of you. Another neglected component of combat at relativistic speeds would be that of time dilation. Time proceeds more slowly for a ship traveling near the speed of light. Humans and computers on high velocity ships would not have as much time to react the actions of slower moving ships, which would appear to move at a blinding pace.
For the above tactical considerations, space warfare at relativistic speeds would be impractical at best. It is more likely that an attacking fleet would decelerate to non-relativistic speeds before falling upon their objective (With gravity, this would be in a literal sense. Remember, "the enemy"s gate is down."). There they would meet an opposing fleet waiting for them. Also, since it would be difficult or impossible to change objectives once the ships have been launched, strategy would devolve into merely deciding the split of strength between defenses and attacks, without the ability to evolve plans with changing information.