Why 100% good isn't neccesary.

Scoble says here that he is getting yelled at (and if you look for it, you find it pretty easily) for supporting a "crappy" format. This is asinine. as he points out, as a user you don't care what the Spec says. People have been bitching about Perl for longer than most people remember, and still it is very widely used. Why? becuase it works. People have been complaining about Windows forever as being technologically a copy cat, buggy, and crashing all the time. Last I checked a vast majority of the world who uses computers uses Windows (personally, my house has a Mac, 2 Windows machines , and a Linux box....). Many people declare MySql to be inferior to PostgreSQL. I think MySql is more popular. C++ might be a "superior" language, but C seems to work fine so lots of people keep using it.

At the end of the day people will use what works, not what is "the best". Always keep this in mind when developing a product. In software this has been called the 80% rule. In Science, it is making sure that you cover the most common use cases and knowing the limitations where the enzyme/kit just won't get it done.

Overall, getting to the 100% best solution probably isn't worth the effort. You will make so few people happy (vs what 80% got you) and you were probably late to market. So, your late to market, you blew more money on developing it (and thus may need to charge more if you have made a habit of working this way on all projects) and only a small percentage of the people on the planet actually care about the little bit that you added to the end.

The people yelling at Scoble don't get it. Their opionion, while loud, is uninteresting. Ship something, then you are relevant. Carping is useless. I mean, you do have a right to complain, but if you are just going to gripe you shouldn't expect to be taken seriously.

I tried to use this to explain my C average all the way through college, but my wife hasn't bought that excuse.