Although it may be tempting to simply declare "No IE6 Support!" without considering the realities faced by web developers, the truth is that sometimes the decision is out of their hands entirely.
Of course, in an ideal world, we wouldn't have to support IE6 at all.
If you do find yourself needing to accommodate IE6 users, the level of support required will vary based on the specific circumstances. Are you catering to a company with a large number of entrenched IE6 installations, or is it simply a client request for broad compatibility? Perhaps it falls somewhere in between?
In my experience, I aim for a "functional, presentable-but-not-perfect" layout for IE6. Using tools like IETester helps me catch any differences across IE versions, including the quirks of IE6. My clients have generally been satisfied with this approach, as I've been able to demonstrate that the percentage of users on IE6 is dwindling and likely less bothered by minor display issues.
The key question is: how much time should one invest in optimizing for an outdated browser with a shrinking user base, versus focusing on modern browsers where most users are moving? Ultimately, the decision may hinge on the priorities set by those commissioning the work and their reasons for valuing IE6 support.
Regardless, I highly recommend using IETester to test compatibility: