[1] here's a hint: don't say "i can do XYZ, trust me" -- signal to them you can do XYZ by blogging, speaking at meetups, portfolio of projects, etc.
You then get to choose to counter or walk away. If you're a good negotiator (which you're not, since you've said the first number), we'll probably end up at 0.9Y.
Meanwhile, Joe Bloggs comes in, and doesn't say a number. The recruiter offers 0.8Y, and Joe finally says "I was thinking 1.2Y". Joe, being a savvy negotiator, ends up with at least Y, and maybe 1.1Y.
It doesn't matter what Y is, you always have a chance of doing better than your goal if you speak last. You lose that advantage if you speak first.
You can also always try for the DITF[1].
So you start by lying to the interviewee? That's the beginning of a great relationship.
Which reminds me of a question I have been asking myself lately. What do you do with your portfolio of "hacky" code?
For instance, recently I wanted to experiment porting an open source application to iOS, to give the old iPhone 3G I had just lying around a new life. There were certain parts of the existing codebase that weren't well structured to meet the requirements of the iOS APIs I had to fit in, but I managed to squeeze in the changes without any major restructuring.
It is code I am not proud of, code I would never write in a professional setting, but it was the perfect direction for my requirements and constraints. For that, I am really happy with the result.
Ultimately, I determined that the 3G just didn't have enough processing power to meet the expectations I had, so I don't have any interest in taking the project further or cleaning up the mess I made. However, even in the state it is in, it is code that is most likely useful to others. But at the same time I don't feel it reflects well on my abilities.