> It seems to me that there are two ways one can approach this subject.
There are a lot of ways you can approach this subject. I'm going to rely on an example I used before[1]. Let's assume that if you're on this site, you probably build some kind of web technology. Were you aware that a portion of your users are colorblind? Did you just do a mental double-check and go "shoot, I should check up on that?"
You didn't actively do anything against colorblind people, it simply wasn't on your radar. You can go through your entire site design, do A/B testing, deploy, get good results, and still be completely unaware of this. When someone comes up to you and says "I'm colorblind, you should add 'OK' to this green button because it's hard for me to use your site without that," what do you do?
"Sorry! I didn't intentionally mean to make it hard for you to use! I deeply regret that you can't use the site."
or how about
"Hey! Chill out! We were just trying to make a good site and the numbers say we're ok!"
Both of those answers sound pretty awful and insensitive. And yet, these are the responses to your "two ways" listed above. Probably the correct answer is "Oh wow, we weren't aware that was a problem. We're going to go fix it, can you tell us specifically what is wrong?"
Which is exactly the kind of response we need to sexism in technology. It boils down to being aware of how your actions make other people feel, and when an issue is brought up you consider it earnestly. When you figure that out, and learn to handle when you haven't, you can stop "being afraid of accidentally offending someone" (itself casual sexism) and you can actually work to make things inclusive.
One great read on this is Ellen Spertus' MIT AI Lab Technical Report[2].
[1] http://news.ycombinator.com/item?id=3772919
[2] http://people.mills.edu/spertus/Gender/pap/pap.html