Re:Clarity over brevity in variable and method names
by MichaelGG on Monday September 10, 2012 @06:37PM
(#4502711)
Generally, virtual machine or interpreted languages have such symbol information available, native compilers tend to not include such things in the binary.
It is not that big of a deal though. On platforms like Java or .NET, you can run an obfuscator to strip out names and leave confusing looking strings.
But it only increases reverse engineering complication very slightly. If you're trying to protect a small secret (like a key validation or special algorithm), you're out of luck, even if you strip symbol names. If you're trying to prevent wholesale ripoff of your app, legal action is more effective.
Making your code "worse" is a terrible tradeoff for approximately zero gain.
Re:Clarity over brevity in variable and method names
by Evbn on Monday September 10, 2012 @11:10PM
(#4503563)
In fact, shipping symbol names makes it easier to prove copyright violation.
Re:Clarity over brevity in variable and method names
by tptacek on Wednesday September 12, 2012 @09:10AM
(#4510439)
Not generally, but sometimes, depending on the language. That said: it is very easy to reverse engineer most compiled code regardless of the language; it is only a constant factor harder than understanding the source code in its original language.
Re:Clarity over brevity in variable and method names
by espinchi on Monday September 10, 2012 @06:31PM
(#4502680)
In some programming languages/platforms (e.g. Java), the product that you distribute can indeed be reverse engineered in a way that most original names are kept. But there are tools to obfuscate the code before turning it into something executable.
In some others (e.g. C), the executable that's distributed pretty much talks the language of the underlying machine, so the original code structure and naming are not preserved.
(Of course, this assumes that you distribute your executable instead of keeping it inaccessible to others, for example by running it on a Web server.)