Hey Jack, thanks for the feedback.

So, I feel like there are 2 ways to do a launch:

1) Waterfall – Do a lot of unproven work upfront and HOPE that you connect with the right kinds of users and get a ton of signups on day #1. But, uh oh, look at all that traffic. And, damn, now people are seeing errors because the backend hasn't been scaled to handle that much traffic. So now we've got 5,000 signups on day #1 but they're getting a shitty experience and we've might've blown our chance to get them to try it again later when we can handle that much traffic. "I know it shat the bed last time, but it'll be better this time I promise…"

2) Agile – Do enough work to get those people in the door who are excited enough about the product’s potential to take a chance. We won't get everyone that COULD'VE signed up, but we'll get the passionate users and we'll be able to handle the traffic, give everyone a good experience, and build off that for next time. Folks who didn't sign up the first time keep hearing about Inline from users who had a good experience, and when they visit next time we've opened things up even more and put more time into the marketing pages because now we're no longer guessing – we know what resonates with users and what to focus on.

So, we went with #2! We got 500 adventurous folks to signup yesterday, were able to handle that amount of traffic, and are excited about their feedback and excited about what's next!

And yep, great call, I'll get on those hover states ;) Thanks again Jack!

Hey Nathan, I’m not able to reproduce that bug right now, but when we spun up a couple of new Heroku Dynos earlier I did see what you're talking about. I'll keep an eye on it and hopefully it's just a Heroku cache thing.

Hey Michael, I’d be lying if I told you we did extensive testing in Firefox. :/

BUT, now that we’ve opened things up, I’ll add it to the list and start looking at it right away. Thanks for the feedback!