Most people I've talked to dislike the incredibly unimaginative Trial of the Grand Crusader gear, but the issue I'm talking about is as old as WoW itself. For one, gear isn't very varied (and will be worse with Cataclysm.) But even if it was, the stats would just be very boring and unimaginative.
Especially for a DPS player, the various stats don't change anything about how the game is played. The decision between haste and crit is a purely mathematical one - whichever provides the most DPS is the better stat, no questions asked. The math may be difficult as in the case of armor penetration, but essentially all DPS stats do exactly the same thing.
For healers and tanks it's slightly different as there is more than one dimension they are measured on. A tank has to decide between threat, mitigation, avoidance, and health. One may choose to forgo one stat in favour of another but these decisions are rarely clear-cut. The debate of effective health vs. avoidance has been raging for years now and even if an agreement could be reached, it would very likely be that for some fights EH is better, for others avoidance. Threat has become more and more important with people gearing up, as tank threat doesn't scale as well with gear as DPS does.
Healers also have choices, although less than tanks. Throughput vs. mana regeneration is the big one, but things like crit vs. haste can make a difference in healing style as well. That said, the rule here is mostly to take the mathematically best throughput stat unless you actually run into mana issues in which case you stack some of your mathematically best regeneration stat. A similar idea works for tanks - they will have an EH set and an avoidance set and gear some threat if theirs is too low.
There are clear examples that items can be made more interesting, even directly in WoW. Set bonuses sometimes alter the way you play a character, and the decision of breaking a set bonus is often a major one. Another Blizzard game shows this even better, however. Diablo 2. Below are two high-end bows from that game. Have a look at those stats.
EaglehornTwo-Hand Damage: 45 To 313
Required Strength: 97
Required Dexterity: 121
Base Weapon Speed: [10]
+200% Enhanced Damage
+2-198% Enhanced Maximum Damage (Based On Character Level)
+1 To Amazon Skill Levels
+6-594 To Attack Rating (Based On Character Level)
+25 To Dexterity
Ignores Target's Defense
WindforceTwo-Hand Damage: 35 To 547
Required Strength: 134
Required Dexterity: 167
Base Weapon Speed: [10]
+250% Enhanced Damage
+3-309 To Maximum Damage (Based On Character Level)
20% Increased Attack Speed
8% Mana Stolen Per Hit
Heal Stamina Plus 30%
+10 To Strength
+5 To Dexterity
Knockback
There are some familiar stats on these bows, like damage, speed, strength, and dexterity. But there is also so much more that makes these 2 bows differ from each other. The mana leech on Windforce enables a replenishment of the character's mana while fighting and removes the need to regenerate mana in other ways, while Eaglehorn ignores the target's defense and is therefore very strong against heavily armored bosses. In addition to more of those unique stats (knockback, additional skill levels, stamina healing), the two items also differ in their requirements. Eaglehorn requires a lot lower investment in strength than Windforce, opening the way for a character that spens those points in more useful stats instead.
Now, Windforce was usually the bow of choice for most Amazons (if they had access to it) because of the raw damage it provides. If the raw damage had been roughly equal, however, the choice between the two bows would have been an interesting one. In WoW, one of these bows would have had armor penetration and the other haste rating. A purely mathematical choice, booooring. Armor penetration and haste are, essentially, the same stat - throughput. Ignoring the target's defense and knocking the target back on the other hand are vastly different from each other.
We know Blizzard aren't idiots (even if I sometimes call them that), making it very unlikely that they just didn't have the idea of putting Diablo-like items into WoW. The issue is more complicated than that. World of Warcraft relies far more on class and character balance than Diablo II did, and if you have a large amount of high-impact stats on items you have a huge balancing task in front of you. Imagine a hunter with a bow with knockback in PvP - that would be absolutely overpowered. That doesn't have to mean though, that it is impossible to have more interesting stats than those in WoW and still keep them balanced. Set bonuses show this - it is possible to balance unique effects, it's just not as easy to do.
The other issue - and I think this is the big one - is the sheer amount of items in WoW. All stats in WoW are numerical and scalable, making a new item is easily done by just increasing the numbers on an old one and maybe jumbling them around a bit. Unique stats make this a lot harder. Twenty unique, binary, stats at 5 stats per item allow for only 15504 items - with a good amount of nonsensical combinations in there that one wouldn't want to have. WoWHead lists 34708 items in the game at this stage, requiring a high amount of binary stats and coming up with a lot of interesting binary stats is difficult.
Once again, I can see why the Blizzard designers won't go for a full blown system of binary stats - but what's wrong with mixing some more into itemization? They seem to save them for legendaries and set bonuses mostly - and the odd flavour piece. (i.e. Thrash Blade, Ravager.) Give me a healing mace that procs Inspiration, a low damage sword with 100% armor penetration (and varying amounts of armor on bosses to make that interesting), or a bow that shoots arrows with fire damage. That's way more exciting than numerical stats.
