2008/8/7 nlif <[EMAIL PROTECTED]>: > While it is very good to know that it's relatively easy to develop Wicket > components, bear in mind that management (at least mine) is more easily > convinced when presented with a wide selection of 3rd party component > libraries, since that provides an alternative to allocating time and > resources of our own developers. Thus, for them, the issue is decided more > an economical merits, then on its design/architectural ones.
Your company should concentrate on what it does as its core competency as that will bring you the most value for time invested. Based on past experience with many companies, I can most glibly and universally sum this up as: Don't write a ticketing system unless you sell ticketing systems. You are presumably building web apps because you think you're quite good at it (or perhaps will be), and you're worried about working at the right level of abstraction to achieve good productivity. You're concerned that Wicket might be at too low a level of abstraction compared to JSF, because JSF has a plentiful array of off-the-shelf components that you think will let you work at a higher level of abstraction, and therefore you'll be more productive with it. It's a nice idea. It certainly looks tempting. Unfortunately, it just isn't the case. Why is that? Go and read Joel Spolsky's article on leaky abstractions[1]. Right now: http://www.joelonsoftware.com/articles/LeakyAbstractions.html Good to have you back. Here are just a few reasons why pre-built components in JSF are not the answer. At some point, normally just after you've completely wedded yourself to a component, someone important will want you to change something that on the surface should be trivial. At this point, you will need to unpick the entire component and figure out how it works, and change it. This will be hard. You will probably introduce bugs. Unless it's a component with a lot of distinct regions of complexity, it will probably be so hard that you may as well have developed the code yourself from scratch (in either JSF or Wicket). Reading code is harder than writing it. Anything remotely complex will need to you restyle it all to make it fit in with the rest of your web pages. This will likely be painful unless the component developer has a clue. Nine times out of ten, it will take you so long to find the component you need, test it works in your environment, make sure it does what you need, make sure it probably does what you might need, discover it doesn't, find another component that does, sort out the licensing, file a purchase order for it, etc. etc. that you could have developed something in Wicket that did exactly what you wanted in half the time. It seems to be the case that if the component is sufficiently complex that you think you will save time/money by buying it in rather than building it, it doesn't do what you want. There are only about five or six truly universal components that are applicable to almost everyone. These are: tree, tree but in a table, sortable & paginated data-driven list, date picker, modal pop-up window, AJAX auto-complete drop-down. I can't think of any others, but there might be a couple. Wicket has all of these. Which you'd know if you'd bothered to look at the examples, which are live and prominently linked from the site. Sorry to sound harsh, but how much web development are you going to do? Hmmm? Eight hours' worth? Go use PHP or JSP or DHTML or whatever. Your use case isn't complex enough to be having this discussion. Eight months' worth? What? You're going to make a decision without investing a day in each option at least? Are you crazy? > How many Wicket components are there, and how mature are there? Are there > tables with sorting, filtering, scrolling, paging etc.? Are there > tree-controls with all the typical tree-functions? Is there Ajax support, as > well automatic fallback for non-javascript browsers (and what about comet)? Come back here when you have real questions that you can't answer for yourself in ten minutes. http://www.justfuckinggoogleit.com/search.pl?query=wicket+tree http://www.justfuckinggoogleit.com/search.pl?query=wicket+ajax http://londonwicket.org/content/LondonWicket-ListEditor.pdf etc. etc. etc. Alastair --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
