The following are what I hope to include for the next update. For anyone playing it already who finds the game too easy, fear not! Complexity is inbound! One of my main goals with this project is to create a very good AI opponent, who can surprise the player without cheating against them. The next update will be a big step in that direction. Without further ado, here is the plan:

AI Updates:

* Make sure the AI players aren't viewing non-spacefaring systems as threats. That's
probably holding them back.

* Maybe buff the amount of ships the AI Empires get at the start, but also give them
a cool-down which prevents them from attacking the player for the first 10-ish turns.
That would give them a buff without the player feeling cheated. They will get no more
than the player either way, who currently gets 3+ times as many to start.

* Joint Attacks: There is no reason not to have the AI consider the strength of multiple
worlds in an attack. Especially given that they can't bank reenforcements like the player,
this is the only real way to make them compete on the offensive. It will also prevent
the exploit of tying up forces that outnumber you in aggregate but are spread thin across
multiple nearby systems.

* During fleet shuffling, have the AI look for nearby worlds that are under threat from
incoming fleets, and prioritize those worlds for destinations.

^ Some of these require making the AI logic a little higher level. It is going to have to
make decisions by empire faction for most things instead of by planet, using a big picture
approach. Which will be intensive, but will allow for not only the above but also the following:

* Strategic objectives. The AI can have "objectives" it will be trying to meet from turn to turn,
allowing for deliberate conquests of parts of the map based on the situation, just like the player.

* AI personalities. Each faction, when spawned, can have a personality. Based on this, they will
prefer different objectives at different times, and approach them differently.

* Diplomacy. The AI Factions can remember who attacked them, and who attacked who attacked them,
and fuzz their logic probabilistically based on that (ideally in a way that can't be easily gamed;
for example, a flat point system would be easily gamed and exploited). Prior to the implementation
of any obvious diplomacy mechanic for the player, they can "make friends and enemies" organically,
just by behaving like an enemy or friend. I can add a more obvious layer down the road.

Non-FTL Promotions: 

* At some point in the late game (perhaps when all pirates are destroyed? perhaps randomly based on ship strength?) allow for some worlds to promote to Pirate systems. 

* Additionally, some could promote to AI Empires (certainly when there are less than 6 active, but the exact threshold I'll have to think about).

Off-Map Threats:

* Pirate raids... no reason small pirate raids can't come from off-map. Nothing huge, but
enough for the player to not feel safe leaving 1 ship on all worlds.

* An end-game boss or two. The Empire returns in force with a big fleet. Aliens invade.
All the unaligned worlds form a new empire. Something like that. Maybe all of them.
Will think about it.

Stats, baby! 

* Track a bunch of arbitrary stats and display them at the end-game, and perhaps
also when a key or button is pressed.

Progress Update: Most of the improvements to the AI are complete. It is infinitely more capable now, although it is still only making decisions on a turn-by-turn basis. Over the next few days, I'll include Strategic Objectives, Personality Types, a rudimentary diplomacy system, non-FTL promotions, off-map threats, and stats. It is tempting to release the update early with only the AI improvements, and if anyone playing the game so far wishes me to do so then I will do so, and save the other additions for 0.0.3.

Update: Pretty much complete with 0.0.2. The AI stuff is almost complete, and the off-world threats are shortly to be implemented. Taking a little time to do some optimization and also ensure accuracy in the comments (constants have changed a lot in testing over the last few weeks). 0.0.2 is so much better than 0.0.1 that I will encourage anyone who tried 0,0.1 to try 0.0.2 without too many preconceptions. There are a lot of new features and the AI is much better now.

0.0.2 is ready! I'll publish it tonight or tomorrow. Not all of my planned improvements made it in to this update (and will be saved for the next), while some I didn't plan on did make it in to this one. 

0.0.2 Change Log:

----------------

* AI is vastly improved. Will now counter attack, patrol, mass for coordinated assaults, re-enforce and defend  intelligently, and trade space for time in an unwinnable situation, among other things. It still struggles against a competent player, who should be able to win most games within 300 turns by taking advantage of the reenforcement pool mechanic. I am toying with the idea of giving that mechanic to the AI as well, which would make the game much more challenging for a competent player. Another option is to give them a higher reenforcement rate than the player, to offset the fact that they can't place reenforcements at will like the player does. I will probably include a "hard mode" toggle before I publish this update, for those who want an extra challenge, which does just that. With default settings a competent player won't notice most of what the AI is doing, unless they just watch them play themselves. The reenforcement pool mechanical advantage for the player is very powerful; so much so that even when the AI is playing for the player in auto-play/watch mode, it will win most of the time just because of that, even though it uses the pool sub-optimally relative to a competent player. I'm open to suggestions on this, as I haven't made up my mind yet. In the meantime, the starting number of pirate systems and the starting fleets for non-FTL systems have been tuned to allow AI empires to expand early, as best they can (which benefits the player as much as the AI).

* AI now has zero mechanical advantages over the player. In fact, the player has some over the AI.

* Adjusted the amount of starting ships. Player still gets more than AI. AI Empires get more than Pirates and non-FTL  systems. Non-FTL systems get an amount within a semi-random range.

* Non-FTL worlds now receive a production penalty to their shipyards after a certain point, to make them less of an  end-game slog while still being an early game terrain obstacle.

* Pirate behavior is a little more intelligent.

* Victory splash now displays some stats, although it's rudimentary and in an early stage.

* Much improved art. The ships now use sprite-like objects generated prior to a battle. Still in an early state,  and I intend to use actual sprite sheets for them at some point (or at least to generate more interesting sprite-like  surfaces for them). Systems now include planets and stars generated at the initial load of the game, which are  persistent, so that each system is recognizable throughout the run.

* Adjusted the lore a little, since I have given each system an Earth-like planet.

* Pirate raids can now come from off-map. Any system within fuel range of the map's edge is at risk. Pirate raids  from off-map are rare, but in the late game can be very dangerous because their size is dependent on the number of ships at the targeted planet.

* Auto-play system. The player can now press Ctrl+Shift+W to set the game in motion on its own. The AI won't handle  reenforcements as well as a competent player would, and it doesn't scout yet, but it still enjoys the mechanical  advantage of reenforcement banking over the AI Empires, and is capable of winning the game most of the time. I highly  recommend just sitting back and watching a playthrough at least once or twice, because it's pretty neat. This can be toggled on/off, so that if the player wants they can treat the game somewhat like a real-time-pausable strategy game instead of a turn based one.

* Ctrl+Shift+B to toggle automatic skipping of tactical battles

* Ctrl+Shift+D to toggle debug mode. This removes the fog of war, and allows the player to select planets and fleets  anywhere on the map. For players who just want to watch the AI play the game, this is a good one.

(I ran out of GUI space, which is why those are hot-keys. I'll add some menus down the road for more GUI space)

* Battles are much improved. There is now more variation, with a chance for either side to receive special  bonuses at the beginning of the battle if they are outnumbered and a small chance each round to get a bonus for  that round. This makes them less predictable, although in general ensuring you outnumber the opponent is still  the key to victory.

* Massive performance gains. CPU use in strategy mode is cut down to almost nothing, while CPU use in tactical battle  mode is much reduced.

* Hit boxes are now ideal in the strategy mode. They were a bit off before.

------------------

Some of the stuff I wanted to do for 0.0.2 is going to be delayed until 0.0.3, owing to the sheer size of 0.0.2's improvements already. I don't want to delay this update any longer, because it's so much better than 0.0.1. But, the following will make it in to the next update for sure:

* AI Personalities. Each AI will have a distinct play style.

* Promotion of non-FTL worlds. If there are less than the starting number of AI Empires left alive,  then non-FTL systems which have achieved a certain level of strength might promote to an empire.

* Diplomacy by action: AI Empires will remember who attacked them, and who attacked who attacked them, and  maybe a few other things, so that the player can influence their behavior by acting like a friend or foe.  In conjunction with AI personalities, some of them may be loyal allies while others may be back-stabbers.

* End-game boss. Some kind of end-game boss. The empire returns from off-map, the surviving AI Empires form a  coalition, a pirate empire forms, etc. Haven't decided yet.