From a literary angle it may seem shoddy for JKR to have given a minor character the same last name as an important existing character, because that generally Isn't Done. On the other hand, avoiding the re-use of names is simply a concession to the reader in the interest of minimizing confusion, not an inviolable rule of writing. In reality the world is full of people who share a last name without being related, some of whom live in the same town and neighbourhood and go to the same schools as each other. As such there's nothing implausible about Mark Evans's last name, no reason for it to wreak havoc on anyone's suspension of disbelief.
Nevertheless, there are a lot of people out there who are disillusioned with JKR right now -- something Jo herself obviously anticipated, given the wry, apologetic tone of her answer. The question I want to ask is, are our expectations of JKR reasonable?
It's easy for us as fans to pick up on minute details of the Potter universe (though we argue and disagree even about those, so the answers aren't always as obvious as they may seem), because all we have to do is read what JKR's written and think about it, and many of us have ample spare time in which to do both. We aren't producing the books, feeling the pressure of millions clamouring for the next one, trying to work as quickly as possible yet maintain some level of quality, and at the same time trying to juggle writing with family and social life. The wonder to me isn't that JKR makes mistakes from time to time; it's that she hasn't made even bigger ones.
Of course the easy answer -- or so it might seem -- would be for JKR to hire some obsessed fan to be her continuity checker. But she can't afford to do that, for fear of spoilers leaking out. Even the best-intentioned, most highly paid assistant in the world can't be trusted to be completely discreet. (Heck, if you knew what was going to happen in Book Six right now, could you keep it all in without exploding? I know I couldn't!) No, the safest thing for Jo to do is exactly what she's doing: keep the whole thing in her own brain and locked up with her private notebooks, and tell nobody, not even her own husband or daughter, what she's written until it's published.
Oh, you might say, but once the manuscript goes to the publisher, her editors should catch those kinds of things. Well, they probably do catch some mistakes (although it appears they're also capable of finding mistakes that don't exist and actually leading Jo astray -- i.e. the Wand Order problem in the first edition of GoF). But professional editors sift through hundreds of manuscripts a year, not just JKR's latest opus -- they don't have the time to devote to obsessive fact-checking and speculation about the HP books specifically, the way that the fans do. So it's inevitable that they, too, are going to miss things.
From my own experience in writing, and talking to other writers, I know that you can have a calendar, and diagrams, and lists of names, and all kinds of little memory-joggers scattered around your desk while you're working, and worry the whole time about getting every detail right, yet still end up making a mistake which the reader thinks is ridiculously obvious. I've even made mistakes that none of my readers picked up, but which appalled me when I read the story over again a year or so later. From comments Jo has made, it seems that she's already noticed a number of Flints in her own work, even apart from the ones the fans have pointed out, and genuinely wants to go back and fix them eventually. But can you imagine the uproar if she were to take time away from writing new material to make corrections to the earlier books now? Obviously, that project will have to wait until Book Seven is done.
Even if Jo could somehow miraculously live up to her readers' expectations and get all the dates, moon phases, character names, mathematical calculations, and other details right in an epic seven-book series that covers thousands of pages -- if every loose end were tied up and every question answered -- what would we fans have to talk about? All right, that's a bit facetious, but -- there is a unique satisfaction in coming up with a neat and well-reasoned explanation for something that would otherwise appear to be an obvious inconsistency or mistake.
The best example of this is in Sherlockian fandom, where (as in HP) you're dealing with a single author, a large and complex canon, and millions of obsessed fans. In fact, there are few characters in literature whose popularity and worldwide recognition compares with that of Sherlock Holmes -- only time will tell if Harry Potter can hold onto the same kind of status.
What do Sherlockians do when their favourite author has obviously screwed up? They compose essays. They hold debates. They write books -- fiction, non-fiction, even encyclopaedias -- to supplement and explain the various inconsistencies in canon. And they play the whole thing as a game, in which the rule is that you can never, ever explain anything away by saying, "Oh, that was just Conan Doyle's mistake." The universal understanding is that Sherlock Holmes was a real person, Dr. Watson was his biographer, and Sir Arthur Conan Doyle was just the literary agent. So if Watson tells us something that makes no sense or clashes with some other stated fact in canon, the onus is on the fans to figure out how to get all the pieces to fit.
And you know what? It's fun. The more obvious the error, the more ingenuity it brings out in the fans as they try to explain it away. And Conan Doyle (here I go stepping out of the Game, and I beg pardon of my Sherlockian friends) made some real whoppers, let me tell you. You think it's a problem that JKR gave an incredibly minor, only-once-mentioned character the same last name as Lily had before she was married? That's nothing. Conan Doyle forgot the first name of his own narrator. Yes, you heard right -- he had Dr. Watson's own wife call him James, when it had been clearly stated in the very first Sherlock Holmes story that Watson's name was John.
If Harry's scar were known to migrate from his forehead to his leg and back, if Hedwig mysteriously vanished between Book One and Book Two and was never seen again, if Hermione inexplicably started calling Harry's best friend "Rob", then HP fans would have good reason to think that JKR wasn't paying attention and didn't really care. After all, that was what happened to Conan Doyle -- he resented Holmes's popularity, and only kept writing the stories because the fans and his editors wouldn't let him stop. Fortunately, Holmes and Watson had already developed a life of their own, and the fans were prepared not just to resign themselves to these errors, but indeed to embrace them with enthusiasm and pretend they weren't really errors at all -- an activity which is still merrily going on, more than a hundred years after the Sherlock Holmes stories were written.
In JKR's case, all the evidence seems to indicate that she loves her characters, she's wholly committed to the story she wants to tell, and she's doing the best she can under very high-pressure and distracting (believe me, if you have a preschooler, you know about distracting) circumstances. Yes, she has made mistakes, and will doubtless make a few more along the way. But considering the amount of pleasure she's brought us, and the sheer scope of the story she's telling, I think that we fans can afford to cut her some slack.
That doesn't mean ignoring the errors. It doesn't even mean refusing to admit that they are errors (although, as I said above, denial can be fun). But I don't think anything good is likely to be accomplished by grumbling among ourselves about how JKR's a careless writer and if we were writing the series we'd do a better job yadda yadda. Especially since the former claim really isn't true and, in 99.9% of cases, the latter isn't either.*
*I will really look forward to seeing the fabulous original epic fantasy novels produced by the remaining 0.1%, though...