voodooking said:
So you're saying that they often cause more harm than good. Even if Microsoft can't perfect one (they have the most to gain) then it's probably irresponsible for a third party to try to make one.
I've been thinking about this as well. Don't programmers sometimes use "hacks" or poor coding technique that can register as errors in certian utility programs? That is another reason not to use cleaners.
I would say it's more that they are generally a solution in search of a problem, and along the way may create problems.
Let's just kind of put it this way. For the sake of argument we can assume there are 1 billion unique programs for Windows. There's probably more, but just for example. Now if I am writing a registry cleaning program, how do I systematically/programmatically go about evaluating the countless possible combinations of registry entries these programs might make? Do I have some massive database somewhere that has a listing of every registry entry created by every single Windows program? If so, how do I manage to cram that much info into such a small program? How often do I update this database, as new programs are created? These are just a couple of basic questions, which will branch off into several other areas if you keep following the rabbit hole down, and I've never come across any answers.
There's also the fact that you see questionable marketing tactics being used. Glary Utilities is a prime example. Maybe a year or two ago there was a pretty concerted astroturfing campaign going on. You would go to almost any web forum and every single post concerning the registry, or even some that didn't, would all have a very similar sounding message saying that the solution was to download Glary Utilities. Just my opinion, but if a company is producing a quality product, it should be able to stand on its own merits, and you don't need to resort to fake grass-roots marketing campaigns.
Finally, just FYI... It's called a kludge when a programmer makes a kind of quick and dirty solution to a problem.