On Dec 4, 11:45 am, Thadeus Burgess <thade...@thadeusb.com> wrote: > There are many parts of gluon that I look at and go... wtf. > > And read it again, and go... wtf. > > Only to find out the wtf comes from another module that has another > wtf in it that comes from another module.. > > If the chain of cohesion was commented, gluon would be much much > easier to learn.
...coupling, while sometimes necessary, should _always_ be intentional and transparent; Implicit chaining has bit many smart PhD developers I've know thru the years - they would _insist_ the system they built didn't do xyz, until I reverse engineered, and showed them the model - "Ooops! How did that happen?" This has happened to me enough times over 35+ years, that I am CONVINCED that explicit is better all around... "wtf" is manual reverse engineering (hard, hard work!); the only things that there are reasonable reverse eng. tools for are / seem to be Java (good, but....) --- all other things should have explicit, transparent, easily accesible structure. "wtf" is not a method; it shows _something_ is missing (some amount of this is inevitable for initial orientation, but then....) - Yarko > > -Thadeus > > On Fri, Dec 4, 2009 at 11:24 AM, Joe Barnhart <joe.barnh...@gmail.com> wrote: > > > > > Always the contrarian, I have a completely different view on > > comments... > > > Comments are the things programmers write intending to make the code > > clearer, but they wind up being WRONG most of the time. Why? Because > > the code changes and the comments don't. > > > Many studies support this. Comments that are misleading and wrong are > > the norm, not the exception, in a typical software project. The more > > effort put into comments, the more likely they won't be changed as the > > code underneath evolves. "After all," the programmer thinks to > > himself, "it's just a small change and the comments are so pretty and > > well structured -- nobody will be confused by such a small change in > > the code." > > > It's far better to write the code in as clear a way as possible and > > let the code itself, along with carefully chosen variable names, BE > > the documentation. In my day job we have 500,000 lines of Smalltalk > > and not one single comment. Our "project" is small -- only about five > > people have ever written code on this system, but we still manage to > > handle our own code as well as suggest changes to other folk's by just > > reading the code itself. > > > On Dec 3, 5:28 pm, waTR <r...@devshell.org> wrote: > > >> Comments in the code I see as the Achilles' heel of this project at > >> the moment. > > > -- > > > You received this message because you are subscribed to the Google Groups > > "web2py-users" group. > > To post to this group, send email to web...@googlegroups.com. > > To unsubscribe from this group, send email to > > web2py+unsubscr...@googlegroups.com. > > For more options, visit this group > > athttp://groups.google.com/group/web2py?hl=en. -- You received this message because you are subscribed to the Google Groups "web2py-users" group. To post to this group, send email to web...@googlegroups.com. To unsubscribe from this group, send email to web2py+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/web2py?hl=en.
