Peter J Milanese said: > > The lack of error checking on the server side means better > performance in my opinion.
Does it? Or does client side error checking use server resources as well? Lets suppose that we want to validate a new message that is entered into a thread on a messageboard. Then we need to check that the thread we want to insert the message into actually exists. If we want to do that client side we need to open a transaction, query the threads table, insert the new message and commit. If we check it server side we just create a foreign key to the threads table, insert the message and if it gives a foreign key error we handle the exception. Which one would put a bigger load on the server? Naturally running an expensive regex on the input to make sure the message does not contain any cross site scripting code is something you would do on the client, but validating *relations* is something you should do in your *relational* database. > I would > have to agree that error checking does belong > on the client side (at least from my experience), and it can be a > lot more productive as a database, and not a > workflow tool. As an application developer, I don't trust the employees to be able to enter data correctly. That is why I validate the input they send me. As a DBA, I don't trust application developers to be able to validate relations correctly. That is why I validate the input they send me. Jochem -- MySQL General Mailing List For list archives: http://lists.mysql.com/mysql To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
