Nothing is stopping you from playing a character you feel is flavourful, even if is not necessarily optimized in the sense of the word.
The problem with the builds in dragon (this goes all the way back to when it first appeared for 3e, such as the cringe-worthy fighter/wizard/spellsword/palemaster build before 3.5) is that they are for most part woefully underpowered, yet wotc continues to misrepresent these builds as perfectly viable.
The problem with this comes when people start copying these builds, thinking that they are playing an optimized PC, then start getting frustrated when their characters start dying like flies (because they are too weak to hold their hold, and will just end up being a liability to the party and dragging everyone down).
Having DM'ed before, I can say from personal experience the frustration of trying to dissaude my players from trying to replicate builds from said dragon articles, because I know fully well that they suck (part of the perks of being a regular on gleemax back then
), and wouldn't last 1 encounter in my games, yet my players keep insisting on playing them because he (mistakenly) believes that those builds are optimized (his rationale was that the build involved 2 base classes and 2 prcs - it had to be viable to some extent).
Wotc should learn to start advocating simple yet good and effective builds which actually work, rather than traps (ie: stuff which sound good on paper but stink in actual gameplay) like great-weapon fighter or paragon multiclassing. It is like you are already lame in one foot, and are intent on shooting yourself in the other!
In the very least, I don't see why it would be impossible for wotc to offer articles detailing flavourful, yet optimized builds which are fun to play and can hold their own. The wizards dnd forum is chock-full of them, and the agreement we signed when we registered means that wotc reserves the right to essentially plagiarize the information wholesale. I myself have played optimized builds which possess deep and immersive personalities, and are a blast to roleplay. So I can say that they are far from being mutually exclusive.
Same goes for this article. I can virtually guarantee that anyone trying to follow the gish build (after making allowances for the myriad of glaring rules inconsistencies) will be gnashing his teeth in frustration, because of one inescaple reason.
You can't roleplay if you are dead. And it is not the most flavourful character which lets you stand toe to toe with a dragon but cold hard stats. The most beautifully roleplayed PC is just another corpse littering its lair if it lacks the capability to take hits, deal damage and do all the stunts required/expected of him/her.
This is a serious bug, not a feature. All the multiclassing it does serves nothing but to dilute its effectiveness. If anything, this article is simply testament that multiclassing just sucks (save for the initiate feats). There are certainly many other ways of designing a gish build which does not suck.
If you want an idea of the sort of caliber I should be expecting from wotc, here is a sample 3.5 character build posted by another member. I know this is 4e, but I feel the article displays the sort of standards wotc should be working towards.
http://forums.gleemax.com/showthread.php?t=974454&highlight=warblade+swordsage+monk+master