Re: [boost] [serialization] performance comparison
Martin Slater writes:
Do you think online code generation is useless in general?
Pretty much so yes. And even if it was non-commercial so that the very real confidentiality concerns that Rene raises that would prevent it being used anywhere I have worked were not an issue the fact that it cannot be integrated into the build process easily and lack of source code so the progress of the project is at the whim of any bugs would deter many from using it or even considering it.
Its possible to encrypt input before sending it and then decrypt the output. Class names and member names can be hidden that way. We haven't yet been able to provide support for the build integration that you mention, but we've thought about it and hope to deliver something there. In my opinion it is a question of which approach will be able to overcome its weaknesses. Besides the performance question mark over B.Ser, there are some disadvantages associated with downloading. Boost has gotten bigger and bigger and downloading/ building/installing it is not as fast as it used to be. Our model is more the once-and-for-all model. We build and install the software and that's that. We also don't take a big chunk of space on the hard drive. Brian Wood www.webEbenezer.net "For Christ also hath once suffered for sins, the just for the unjust, that he might bring us to God..." First Peter 3:18 _______________________________________
code so the progress of the project is at the whim of any bugs would deter many from using it or even considering it.
Its possible to encrypt input before sending it and then decrypt the output. Class names and member names can be hidden that way.
So by the time I do that I've had to write a parser to read and write the middle files at which point i'd assume it would take very little time to write the actual code generation, and frankly it still would not circumvent the need to NDA's and other agreements to be in place.
We haven't yet been able to provide support for the build integration that you mention, but we've thought about it and hope to deliver something there.
Anything that involves the build process contacting a webserver is unacceptable as there is just far to much that can go wrong inbetween that would mean a build cannot be done.
In my opinion it is a question of which approach will be able to overcome its weaknesses. Besides the performance question mark over B.Ser, there are some disadvantages associated with downloading. Boost has gotten bigger and bigger and downloading/ building/installing it is not as fast as it used to be. Our model is more the once-and-for-all model. We build and install the software and that's that. We also don't take a big chunk of space on the hard drive.
And I've yet to see any indication you support even a fraction of what boost serialisation does, versioning and a fully general system for user defined types and archives the first things that come to mind, speed is only a part of the equation and not something I have run up against with boost serialisation that a small amount of work on a more custom archive has solved (which was similar to the work done that is now in 1.34). That's not to say I'm not interested in helping to find places where boost.serialisation could be faster, hence why I'm in this discussion.
Brian Wood www.webEbenezer.net "For Christ also hath once suffered for sins, the just for the unjust, that he might bring us to God..." First Peter 3:18
Please, enough quotes. Lets keep the discussion in the real world. Martin
participants (2)
-
brass goowy -
Martin Slater