Falling Icicle
Adventurer
Wrong on how DC were set. As stated in one of the first issue of Dragon covering 3e, the 3e skill DCs were based upon what WOTC called the "commoner standard" which is how difficult it would be for someone with no bonus from race, ability scores, class, skill, etc, to perform the action in question.
That may have been their intent, and with some skills, it is true that you don't need a lot of ranks, or even any, to have a decent chance of success. But there are many skills where that is not the case. Opening an average lock, according to their list of examples, is DC 25, making it flat out impossible for someone with a 10 ability score and no skill ranks. Even with 5 skill ranks, they'd fail on anything but a roll of 20. Then there are skills which rely on opposed checks, like listen vs. move silently, and many opponents have such high bonuses that if you don't have max ranks in the skill, you might as well not bother trying.
This is off-topic, but has anyone compared the DC table in 3.x to Next? http://www.d20srd.org/srd/skills/usingSkills.htm#difficultyClass. The DCs in Next are typically 5 points higher than their 3rd edition equivalents (with the exception of nearly impossible which is 5 points lower). This is quite bizarre considering that characters in 3.x could achieve much, much higher bonuses than characters in Next can. This makes accomplishing tasks much more difficult in Next, especially when you consider that skills aren't even going to be in play for some groups. I've said it before and I'll say it again, they need to reduce the suggested DCs by 5 across the board, and this isn't just based on comparisons to 3rd edition, but also the math as it exists now. I don't think it's right for an average person to fail an "easy" action nearly half of the time.