Re-read your email. Only recently have I been on bigger projects. Even so, I'm usually the main developer for a given code-base, and others are on other projects, so no ones ever looking at the code next to me. Secondly, the requirements + code-base change so much, that documenting is a waste of time, when at any moment you could wax a class, and thus destroy 20 minutes worth of documentation time.
On the 2 projects I was one where I worked with a team of developers, we were in constant contact, either in person or over AIM, and neither had problems reading eachothers code that was comment-less. We were using ARP, so it helped communication; when I updated a command, he knew where to look, and vice-versa. Admittingly, the little commenting we did have, which we couldn't live without was at the top of the file for Views; it merely reported where the app the View was and what it represented since we had a lot of forms. That was it, though, and I felt things went just fine. Hanging on the Flex list, apparently my projects, even the big ones, are to the scope level that others are, although, the more I use Flex, the more it feels like. The arguments we have over ViewHelpers in Cairngorm, for example, clearly illustrate the differences. If you have an insane amount of Views with a lot of developers working on the project with no guarentee they'll ever be in communication, I can clearly see the need for commenting, at least in the ViewHelper (class that implements functions of the View without you the developer having to know anything about the View). Same for Delegates that handle a lot of business logic and data-mangling coming from the back-end. So, I guess I've just never been on those types of projects. I swear it's felt like it in some of them, but it always felt manageable to me, no matter how many Views/Commands we had. ARP or Cairngorm always helped me and my team, and scope creep was more easily managed. As to your last question, your are apparently in one of the few companies that do those pratices, and apparently believes they work. ----- Original Message ----- From: "ryanm" <[EMAIL PROTECTED]> To: "Flashcoders mailing list" <[email protected]> Sent: Wednesday, December 21, 2005 6:09 PM Subject: Re: [Flashcoders] Faster code? > What I want to know is who is paying for #2? Very rarely do I ever get > time > to comment anything, and even if I did, the code I commented changes the > next day, rendering the whole endeavor fruitless. > > ...not to mention most service work is always a custom job/rewrite anyway. > Wow... that sucks. I hope they're small projects. Without good a reusable (and well-commented) code base, most of the larger projects I work on wouldn't even be possible. I have, at minimum, at least something like the following before every method (or sometimes groups of methods) in my classes: (spacing is jacked, but forms a box in the Flash IDE) // ######################################################################## // // ## Function: ShowMessage ## // // ## Parameters: ## // // ## Message msg ## // // ## ## // // ## This function determines how to display a message object. It ## // // ## accepts a single message object as a parameter and returns nothing.## // // ## ## // // ## Not all messages are displayed, certain types of system messages ## // // ## are used to close connections, accept or decline private chats, ## // // ## etc, and may or may not display any text to the user on receipt. ## // // ######################################################################## // I guess it really depends on the types of projects you are doing. But if there is even a chance that I'll have to come back and edit a class later, I leave myself comments next to *any* optimization so that I know when and why I put it there. At my current job, there are actually new entries in my comments, like this: // ## Modified:Ryan Mattes 12/20/05 Reviewed:Abraham Saldana 12/21/05 ## // Any code that goes into production has to follow a code standard, be reviewed by my peers, and have a record of when and why I went in and changed it. The upside being that if something breaks, everyone knows who broke it and when, making it very easy to find and fix, and if someone new comes in, they can easily figure out what the code does and how to use it (they're actually crosstraining some C# devs by having them read my source). What I want to know is who pays people to write code by the seat of their pants anymore, and why would you want to? ryanm _______________________________________________ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders _______________________________________________ Flashcoders mailing list [email protected] http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
