You have to weigh up the cost (in terms of time and inconvenience) against the benefit.
Software tends to fall into three categories:
1) That which behaves well in this regard and therefore so long as it's even remotely useful is fine
2) That which behaves badly but is very useful (or, like Java, basically mandatory) in which case you have to go through advanced installs, deselect options and then tidy up after it. Not great and leaves a bad taste in your mouth but ultimately the effort is still worth it because what you get at the end is still worth having (or at the very least the price of doing business).
3) That which behaves badly and doesn't offer enough benefit in which case for me it doesn't even get as far as completing the install.
I can see the appeal of the model from the perspective of those being paid but for those piggybacking their products it's essentially the refuge of the desperate. Think about it - your product is free but it's so bad you're effectively having to pay to get people to install it. How about putting that money into developing something people don't object to instead? Might be a bit more sustainable.
Put that way I'd hope that this is something which will pass with time, a business model which will ultimately be seen as unworkable, that does little other than skew the installed user base numbers (after all, who cares if they've got it installed if they don't use it, hate it and hate you as a result) and damage your reputation. I just wish they'd get there sooner.
