I chose Egypt as an example because it was just brought up in the thread, I could have probably chosen any other civ as proposed in the opening post. 2 UU's don't achieve anything spectactular for Egypt, so I'd phrase it the other way round. Why force two UU's for Egypt in particular? It's about the design of the civ regarding gameplay, not a simulation. (That I btw. would find highly debateable even in the case of Egypt where we put 4000+ years of history and countless different states into one civ).
That proposed Swordsman for example. How do you make it distinct? Is it a defensive focused Warrior (boring as Archers/War Chariot do that much better), an better Swordman (bad, because now all three of Egypt's Uniques are spread out over the whole tech tree), an earlier/moved Swordman (now you are competing with Rome and the Iroquois for a military civ). You see we already have problems of being distinct, now think of all the other uniques and you see that the task is quite impossible. Sometimes, less is more.
I'm getting the impression that you're arguing against the entire idea of unique units, given that we're talking about a hypothetical unit yet to be fully designed and you've already said that it won't "do anything spectacular". Am I way off-target?
There simply aren't enough types of standard units in the game that one can declare a civ "in competition" with another civ because they have the same UU replacement. At this point, there might even be more replacements than there are units to replace(?). Swordsman in just one of those units that is likely to see numerous replacements.
And I don't understand what your points about Wonders is. I do agree with you, but fail to see how that adresses my point?
PS: I'd argue that Egypt doesn't have a strong UA, but a 'beginners' one. It's fairly powerful on the lower levels but becomes weaker and weaker the higher you go, especially due to the fact that good players manage to play without wonders. And I'm not sure that you need to strive for perfect balance in civ between the civilizations, the game is too set up randomly for that in my mind. But of course I'd avoid creating overpowered combinations as well. That however doesn't mean we need to assign a useless or bad unique unit for balance. That's simply not fun at all after all...
I'm not sure what you mean by "good players manage to play without wonders". Good players can do without a great deal. I suppose you're framing that within the context of playing on higher difficulties. Most of the poeple playing on high difficulties are simply using a scripted path to victory. It's an exercise in efficiency, as well as the ability to shamelessly exploit AI weaknesses. The chief reason to not build wonders on Diety is that it's got a low chance of paying off, not because a player is so good that he doesn't need to build them. Wonder have been written out of The Script.
I can appreciate playing in a hardcore siege mode where you're struggling to survive a non-stop onslaught. Every bullet has to count and one slip-up can cost you everything. That's kind of what Diety difficulty is for. What I don't agree with is the notion that efficiently executing a predeterimined path to victory on Deity makes someone a better or more experienced player than someone who plays Civ on King, but plays it as a 4X game, not a turtling siege game.
The reason I went into all of that is because it is actually related to our discussion. Namely, I propose that "optimal" design is not necessarily the same thing as "good" design.
Optimal civ design means that you only give people uniques that compliment established core strengths. It means that the only good unique unit is one that replaces whatever is already the ranged-unit-de-jour of that era. Even then, you should go with a unique building instead, since its benefits are passive and consistent throughout the game.
Good civ design means that you try to bestow the sense that playing a particular civ feels unique and distinctive from playing other civ's. It steers players into experimenting with styles of play they wouldn't normally consider viable. Of course, the Diety player will simply judge any civ that isn't compatible with the Diety Script to be "weak" or "UP" and avoid it. But that doesn't mean the civ is badly-designed or is only for beginners.
That's one man's opinion, anyway.
) you think of civ design when making these like I did above.
) but it seems to me that the Bruk might be better served as a Lumbermill UI...perhaps +1
when next to another Bruk, +2 
per tile and somehow work uniquely with an archaeologist.
