So, I was going to respond to this conversation earlier, but something didn't really feel right about the way it was all going so I left it.
Aaron, Harvey and others present a compelling argument. We know that every project is constrained by a limited set of resources, and it is the ability to juggle those resources to match business needs that is part of what makes a good professional programmer. In this way, arguments about limited ROI and simply having more important things to do dominate the discussion and lead to people thinking HTML validation is a nice-to-have that doesn't really matter. What finally popped up in my mind - inconveniently moments before I was going to bed, as usual - was that we're simply aiming too low. "We are what we repeatedly do. Excellence, then, is not an act, but a habit." The risk we run, when we regularly aim for merely Good, is that we become someone who cannot ever deliver Great. This applies to everything, not just HTML validation or even programming. A fundamental aspect of greatness is attention to detail. The difference between the house designed by a good architect, and the one done by a great architect, is not in the size or the shape or how pretty it looks. It's in the myriad of different tiny ways it helps the people who live in it achieve their goals. For a web application, HTML validation is part of that. Precise, well formed HTML is easier to maintain, easier to debug and easier to use for numerous situations from screen readers to screen scrapers. Every app should have perfect HTML validation as a goal. It should almost never be the highest priority (maybe if you're doing w3c.org or something), but it should be there in the task list because without it, and without comprehensive unit tests and lint checking and code documentation and every other little element, it can never be great. The real world imposes limits on us, a number of these tasks, and others, may never be realised in a given project, but if you start day one of the project saying "Reality sucks, lets aim low" then you've already sold yourself - and your client - short. You'll never deliver a great project that way. The next time you start a project, don't compromise on the vision and don't forget the little details. Make sure you put the priorities where they should be - this is not an excuse to mess up delivering the value the business needs - but add the tasks in anyway, even if they're at the bottom of the list. Every project you do should be your best project yet. Every project you do should exceed your own expectations. When you put these tasks in your task list you are saying "I'll get there, I'm so good at this that I will get to the point where nothing would make this project better than having 100% valid HTML everywhere". The world will never have enough great programmers. We need you. Don't sell yourself short. Regards, Richard. -- NZ PHP Users Group: http://groups.google.com/group/nzphpug To post, send email to [email protected] To unsubscribe, send email to [email protected]
