When posting example code or filing bug reports based on a real production app, it would be helpful to have some way to change the table and column names to not potentially give away information about the internals of the app. Doing it by hand without breaking things is time consuming. Does anything automatic exist? Ideally it would use real English words so they are more easily referred to than random text strings.
As long as you don't use real data, I don't see what the issue is. Most apps are fairly obvious based on the requirements. ie CRM system = (customer name, address, etc...) or (customer name, addressid, etc.. with some address table with parts of the address, etc...). By knowing your schema I have no idea how you implement your app. Generally without the stored procedures/program code it would be hard to steal any intellectual property. Even if you were the NSA or something (InternetIP, PacketHeadingID, PacketDetailID, TimeStampID). Even with the structure of the tables I still would have no information on how your system to log all the internet traffic actually works. I also wouldn't know anything that is logged.
I don't know of anything off hand to do what you are requesting, but I would think it is fairly easy to write a script to do it on your own. Look at the table columns and datatypes and call text columns "TextColumn1", int columns "IntColumn2", etc. and build a table of substitutions, then perform the substitutions globally in the script file. I would think this is a fairly easy Python/Perl/PowerShell/Ruby/VbScript program.
I agree that there's no real need to do so, but if you feel that way, take a look at anonymizers, usually used to protect the data and not the schemas, but you could easily apply those approaches to schemas as well.
See this paper (which is the description of this framework) especially page 8 an onwards for different anonymization methods, although replacing column names for static strings might probably be good enough anyway.