Thu, 19 Apr 2001 22:53:18 -0500 in <mailman.987738512.19013.python-list at python.org>, Chris Gonnerman <chris.gonnerman at usa.net> spake: >From: "Mark 'Kamikaze' Hughes" <kamikaze at kuoi.asui.uidaho.edu> >> There are two reasons not to: the practical, and the philosophical. >> The practical side is that people would have the urge to start >> fiddling with the source, and I *really* can't spare the time to even >I won't be. I don't even run the same version of Python on all my computers >(upgrading them all is a lot of work). If I had source, yeah, I'd play with >it... probably find a few bugs. If you don't want to deal with the diffs, >ignore them, but you may be ignoring real bug reports. Refactoring doesn't >excise all the old code, after all, not to mention the old bugs. Or you could just offer equally real bug reports in this strange language called "English", and let me deal with my own code. This is the most efficient process for everyone, since then you don't have to understand my code, and I don't have to understand yours. Thus do happiness and tranquility prevail. >(BTW why do people talk about refactoring like it's a new thing? I've been >doing it for years, but didn't have a cool name for it...) Likewise, but A) Fowler's book gave a common language and some more-or-less precise steps for each kind of refactoring, just as the Gang of Four gave a name and some more-or-less precise steps for the design patterns that most good OOP programmers had already applied, and B) as OOP became more widespread and IDEs became more powerful, refactoring got easier so more people could do it. And the reason it matters is that the code changes so much that a diff against an old version is much harder to integrate than in up-front design methodologies. >Compiled Python files are >inefficient, though, since the bytecode changes with each version. This is true, and an irritating defect. I'm very seriously considering doing the releases as binaries for Linux and Windoze and maybe HP-UX if that's possible (are there any MacOS users out there? Other platforms I should care about? *BSD should be able to use Linux binaries...) This would solve the entire problem. A Python obfuscator would work, as well, but I'm not aware of any. Is there a good reason the PVM bytecode isn't upward-compatible? >On the other hand, available source means possibly useful bug reports, more >meaningful feature suggestions, I definitely disagree on those points. I find English bug reports and stories more readable than even Python, and Python's the most readable language I've seen to date. > and less hassle from people with varying >Python versions. That's the only thing that would even make me consider it, but the other consequences are just unacceptable, and it's little enough trouble to add another few lines to my build script. -- <a href="http://kuoi.asui.uidaho.edu/~kamikaze/"> Mark Hughes </a> "I will tell you things that will make you laugh and uncomfortable and really fucking angry and that no one else is telling you. What I won't do is bullshit you. I'm here for the same thing you are. The Truth." -Transmetropolitan #39
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4