you are right.some time we have to invest on such things.but we cannot spend $50 for small applications like remote controll.the cost of the system will increase.in the market there are peolpe who works with cheap modules and they are succeded in making such cheap systems.so to survive,we also have to do the r&d and findout some solutions.

Agreed, but the point is "do not go in blindly". If there is no datasheet available, think it through "is the time and cost it takes to decipher how to work this available/economically feasible". If you plan to make a million thingies the development cost is not the problem, but what about time to market.

Try this: write the company, ask for a quote for one million units and a datasheet, it might work.

I can not guarantee this, but I do believe that with some (intense?) market research you can find a documented device in your price range. The point is that when you see a cheap device that is not documented, your first thought should be "is there any other in this price range that is documented", not "this is the right price, let's try". As you see http://www.8052.com/forum/read.phtml?id=137429 I got burned too, even dealing with a top 25 semiconductor manufacturer. I keep learning the same lessons over and over again, albeit not as frequently as I once did.

Erik