MyISPHatesENWorld, your solution is not without its merits, but I ultimately still favour any solution which reduces the incidence of DM fiat to a minimum, or if possible, can do away with it altogether.
The issue here is not so much that the DM cannot apply the houserule you have proposed (impose a cap on how high a skill check may go at any 1 level), but that he should not have to.
The issue here is not so much that the DM cannot apply the houserule you have proposed (impose a cap on how high a skill check may go at any 1 level), but that he should not have to.
I'm not suggesting DM fiat or a house rule, any more than the people suggesting that this article be retconned out of existence or nerfbatted to irrelevance before the final copy is published in the complete magazine are suggesting a house rule. I'm suggesting the article and future articles like it be used to provide more options for creating a character, and if something breaks on the back end, they fix what breaks on the back end, not stop providing options.
Do bear in mind that there may be people who subscribe to dragon but may not frequent these forums (or the one at Gleemax), and as such remain unaware of any potential problems these background traits may pose.
"Think about the children!"
It does not help that the article is prefaced by the designer proclaiming how insignificant these bonuses are, and his comments can be construed as an outright challenge to try and break them. And indeed, some players just might start stacking these bonuses, thinking them harmless as purported.
These are harmless, I'm going to stack them all? So dump an entire article becuase someone "just might" start stacking these bonuses.
For every astute DM who realizes the implications of allowing these background traits, there are likely many others less experienced who do not. DnD should be designing an internally consistent ruleset which players and DMs alike can readily just pluck out of the book and use as is without having to playtest it first to find out if they are okay or problematic, rather than flooding us with a ton of options containing both useful and crappy material, and putting the onus of sieving out the good from the bad wholly on the DM's shoulders.
Despite allegations to the contrary, I consider myself, and the DMs I play with to be astute and experienced DMs. One of them is allowing the players to retcon our choice of these onto our existing characters, adjusting the fluff if we want, and I'll be allowing the same (it hasn't come up yet with the other). And the inexperienced DMs I've played with in the past have picked up on real problems quickly enough that I know that the, "Think about the children!" reasoning doesn't hold up.
Posts claiming that this entire article is broken and unbalanced seem to indicate that individual DMs and WotC do a far finer job of sieving out the good from the bad.
As such, I advocate nipping the problem in the bud. Likewise, I see no way of justifying to my players why the character who probably doesn't need an intimidate bonus gets one (the one with 10 cha in your example), while the character who would likely appreciate such a boon (ie: the 18cha dragonborn) can't get one.
I also advocate nipping the problem itself in the bud. But these backgrounds or untyped bonuses attached to them (or possibly soon to be typed "background bonuses" or whatever) themselves aren't a problem. There need to be more ways to achieve viable, good and optimal skill proficiency and backgrounds and untyped or newly typed bonuses provide that. The "potential problems" to use your term, are with how some high things might stack, not what things stack.
The player with a 10 in CHA and trained in Intimidate (such as a fighter or a 12 CHA Dragonborn fighter) does need the bonus, specifically, he needs it more than the 18 CHA Dragonborn (say a Warlord) whose bonus, as he is "optimized" is already, um, optimal. But, since I'm not advocating a house rule, you wouldn't need to justify it.
As an aside, I think sooner or later, people are going to have to face up to the fact that shifting to 4e because they didn't like all the splatbooks in 3.5 was a pretty crappy plan.
You can only write once, "Pick two skills and add them to your class list or pick one skill and add it to your class list and pick a language." And that doesn't fill up much of your writing quota, or make the background in any way useful for anyone that has a class related to their background.