Open source development is driven by user requirements, available resources (mainly time, but also components), and competing projects. If you were considering the purchase of a new car, you might consider
- problems / costs with your current transportation
- benefits / improvements in the new car in comparison with your current transportation
- your goals / plans for the future
In the Nuke world, many were seduced by the improved functionality of Nuke starting around version 6.5, circa 2002-2003. Modules, themes, and other addons proliferated - but something big was lurking behind the cheerful adoption: security (or, rather, the lack thereof). A plethora of tools and solutions sprung up to address this critical need, and NukeSentinel became the standard. It has been enhanced and now provides an incredible amount of functionality to help webmasters secure their sites (not to mention, the ability to get more sleep - thanks, Bob!).
We began to look at two other areas that needed attention: SEO and WYSIWYG editing. SEO was almost an afterthought with standard Nuke (and still is, for that matter). Maybe Nuke's "customers" didn't demand it. Our customers did, so we filled a niche. But many were demanding WYSIWYG editing - unfortunately the initial "official" solution for that was a good editor that was poorly implemented, taking security back several steps in the process. A quick evaluation of possible solutions (prior to 7.7, when TinyMCE was first implemented with PHP-Nuke), considering the design and experiences of others among other factors, led to the selection of kses for HTML filtering and FCKeditor for HTML editing. HTML editing required the re-evaulation of (and changes to) the check_html function used by many standard (and optional) functions in Nuke. One of the goals was to minimize the impact that these changes had on the hundreds (if not thousands) of available addons for Nuke. nukeSEO and nukeWYSIWYG were the result.
Around the same time, Raven formalized the development of RavenNuke and recruited a dedicated team of developers, testers, reviewers, etc. to collaborate on a significantly enhanced version of Nuke that became RavenNuke. The team addresses security, performance, installation, documentation, integration and additional functionality. Most of the developers have contributed enhancements that are also packaged separately (e.g. Gremmie's gCalendar; montego's HTML Newsletter, TegoNuke Short Links, and more; Guardian's Spam Stopper, Comments administration - you get the idea). All of these improvements address specific issues or opportunities for improvement. We are now focused on improving the syndication features of phpNuke, starting with nukeFEEDS (syndicating Nuke content for the rest of the world) and soon, nukePIE (consuming syndicated content from other sites). We'll then have a major upgrade for nukeSEO (which hasn't been included in RavenNuke for several important reasons - mainly compliance with design standards). There are other teams working on different Nuke distributions (e.g. BonusNuke, DadaNuke, NukeEVO, NukePlatinum and more) - each with its own goals and niche, and each with its own take on the cost / benefit equation.
Along the way, we (nukeSEO) upgraded nukeWYSIWYG from FCKeditor 2.2 to 2.3.2. The reasons behind this upgrade are important, especially when considering the possible upgrades for FCKeditor, kses, and other important components. FCKeditor 2.3.2 offered significant improvements and enhancements that addressed the major problems and enhancement with 2.2 that impacted nukeWYSIWYG users: performance, HTML compliance, and full screen editing. We continue to monitor both requests / issues and upgrades to FCKeditor (currently at version 2.5) as it relates to the points above (problems / costs, benefits / improvements, and goals / plans) to determine if conditions change sufficiently to require a change in direction.
Right now, there isn't a compelling reason to upgrade / replace either kses or FCKeditor. Some of the FCKeditor enhancements might be significant for custom solutions, but other than the bug fixes, not much seems incredibly desirable for nukeWYSIWYG. The other enhancements (nukePIE, nukeSEO upgrade) are more important and offer better returns for investing our resources. One of the issues delaying nukePIE is that a key component (the SimplePIE script) has been upgraded significantly since the original nukePIE was developed and before we could deliver it. Since we'd hate to release nukePIE with a key component that's already seriously out of date, we're testing the new release of SimplePIE so it can be included in nukePIE.
Hopefully, this provides some insight into where we're going, why and how we'll get there. It might be time to implement a more formal way to defining our road map and projects. Feel free to comment on any of these points here or (preferably) in the forums.