Yes, I agree that these types of tutorials can be useful, and they should probably exist in some kind of "tutorial repository" associated with different languages. If that was the intention of the person who wrote it, then their time was time well spent in terms of how much they might help others.
Part of the feeling I get out of these submissions, though, is "Hey, our platform is just as good as RoR/Arc/x because we can do what they did this easily." I was trying to point out that such activities are not really accurate depictions of how good a given platform is, because much of what makes a language or platform good is being able to "think in the language", to develop your design and ideas as you go. Once you have a solid design in mind, you could use just about any language to implement it.
That's why I thought it would be more interesting for someone to develop something new using a given platform, and show the kind of work that gets done that way. For example, pg often writes about how good Lisp is for writing programs when you're not sure of exactly what you want to do yet. It would be intriguing to see someone use a similar approach with their platform of choice, recording the process, in order to show how flexible it is and what kinds of issues come up from the very start. For example, how much re-writing occurs when a very small change is made? Does the platform provide enough abstraction capabilities to make the underlying implementation transparent?
As I've been reading On Lisp lately, I'm finally starting to get why pg likes Lisp so much. There's just something cool about being able to write macro-defining-macro-defining macros, and it turns out that it can make the development process much more liquid. Do these platforms (like PHP) support similar levels of abstraction? Judging from "For example, to display the time elapsed since a site was added, you could use: $time->relativeTime($site['Site']['created']), now that is awesome!" I think not, because there are definitely implementation-specific details there that hamper flexibility and would be better off being abstracted away somehow.
Hey BrandonM, I am a big fan of CakePHP, the only problem with Cake is the lack of documentation, that's why i wrote this tutorial. I believe in giving back to the open source community, whether by writing code or writing tutorials. Also you make a good point about procrastination, guilty :), I will also give Lisp a try, especially that more developers are adopting it now.
Part of the feeling I get out of these submissions, though, is "Hey, our platform is just as good as RoR/Arc/x because we can do what they did this easily." I was trying to point out that such activities are not really accurate depictions of how good a given platform is, because much of what makes a language or platform good is being able to "think in the language", to develop your design and ideas as you go. Once you have a solid design in mind, you could use just about any language to implement it.
That's why I thought it would be more interesting for someone to develop something new using a given platform, and show the kind of work that gets done that way. For example, pg often writes about how good Lisp is for writing programs when you're not sure of exactly what you want to do yet. It would be intriguing to see someone use a similar approach with their platform of choice, recording the process, in order to show how flexible it is and what kinds of issues come up from the very start. For example, how much re-writing occurs when a very small change is made? Does the platform provide enough abstraction capabilities to make the underlying implementation transparent?
As I've been reading On Lisp lately, I'm finally starting to get why pg likes Lisp so much. There's just something cool about being able to write macro-defining-macro-defining macros, and it turns out that it can make the development process much more liquid. Do these platforms (like PHP) support similar levels of abstraction? Judging from "For example, to display the time elapsed since a site was added, you could use: $time->relativeTime($site['Site']['created']), now that is awesome!" I think not, because there are definitely implementation-specific details there that hamper flexibility and would be better off being abstracted away somehow.