Wednesday, August 13, 2008

Hello Team Fusebox

I've been very busy lately with many projects so my blog has been quiet, though I hope Bill Shelton, myself and Marc Esher (among others) can pimp some of this stuff we are working on soon. I did want to take a minute to make mention of the announcement that was made in the cfConversation's round table. It's now official, I am the new developer in charge of the Fusebox core files. Check out the announcement for my brief statement, also check Sean's blog (like you haven't already). Expect me to think out loud a lot on my blog about new features and changes to Fusebox in the future. I do want to thank Sean for entrusting me with the core as well as his kind words, I appreciate the opportunity.

In all honesty Fusebox is in a happy spot right now and aside from bug fixes I am not planning on touching the core much right away. I've heard some rumblings about getting full support for Fusebox 3 and I'd relly like to hear why, and see how many want this. Over time as I get more comfortable with the core files I might try to achieve full Fusebox 3 compatibility but that will take time. Intially I think I am more interested in experimenting with speeding up the core, small enhancements to the no XML features, and providing more definition to layouts (more on this soon this is very much an incomplete thought). So what do you hate about Fusebox?

13 comments:

Anonymous said...

Congratulations Adam! It's nice to see that Fusebox is being left in such capable hands. As long as you're asking about possible enhancements, I just wanted to say that I'd love to see controllers that are singletons so dependencies can be injected into them.

Marc Esher said...

Does this mean we'll be seeing Eclipse plugins donated from adam/kroger to fusebox core?

Anonymous said...

Congratulations!

Anonymous said...

I'd like to see an update to the API for custom lexicons which would allow those lexicons to be created as CFCs intead of include files... Not as a requirement at least at first, but as an option - so the namespace could point to a cfc and then the CFC would have an init method that sets its access to the framework and individual methods would represent the various tags available in that lexicon. So <ns:tag yadda="x" /> would execute something like ns.cfc.tag(attributes,children,parent) ... just thinking out loud really :)

Anonymous said...

Updating the core is secondary to current documentation. I know this argument has been stated many times over the last 8yrs. But for example, the fusebox wiki does not mention that XFAs accept parameters, or that the RELOCATE accepts an XFA property. That functionality has been around for a year now?

Also, the example apps are good, but they do not get updated when new versions of Fusebox are released. There is a "getting started" video by Hal Helms which is nicely done, but it's for version 4.1.

If you've ever read through the forums, people care more about getting stared and best practices, more than anything else. IMO, people want to learn in this order:
1. "How to" videos
2. Example apps
3. Tutorials (Web/PDF)
4. Advanced features (plugin/CFC)

Alpha releases of core files are fine, but maybe what is needed, is a "core example" too? Any tutorials made could build upon the basic application.

For example, you could easily demonstrate how plugins are used with a user tracker. It simply records what was clicked into a database. So, you could download the Core example, and drop the plugin into this example and it would work without modification.

I understand not everything could work that easily, but the idea is that a Core Example shows a progression of complexity.

Anonymous said...

This is exactly the direction Hal and I were trying to push things when we created the first "Wegot Widgets" app. The idea was to have a common reference app so everyone knew what they were talking about and/or building upon. FWIW...

Congrats, Adam--good to see you in The Chair.

Anonymous said...

Congrats, Adam.

As you know, Fusebox is the lifeblood of Razuna, our open source digital asset management.

I send the opinon of others, that documentation and tutorials should be given a boost.

Kind Regards,
Nitai

Anonymous said...

Hi Adam, good to have you in the big chair!

I'm a big fan of the no XML flavour of Fusebox. The things that I would like to see added are:

* the circuit permissions you have in FB XML
* routing (like rails)

If you're running a poll, then I'm personally not interested in backwards support for FB3.

Unknown said...

Count me in as a yes vote for FB3 support. Our shop has a load of FB3 stuff, and we haven't bothered with FB 4+.

Anonymous said...

+1 for no backwards support (forward ho!)

+1 for more docs, current sample apps showing 5+ features

We just 'standardized' on Fusebox so we have 5-6 people hear ramping up on Fusebox!

Anonymous said...

I agree with more docs, quickstart guides, and tutorials for FB 5.5.

Anonymous said...

I want to see FuseBox5.5.1 plugin for Eclipse

Anonymous said...

Okay, so I realize that there are a lot of voices here saying "we don't need backward compatibility for FB3"... however... just a year or so ago I was working for a shop in Portland OR that was still on FB3 because that's when they converted from not having a framework to begin with... and honestly, I think all those folks who spent their hours working on those apps deserve an easier upgrade path.

So to take the burden off of Adam and the rest of the Fusebox team, I've done it for them. It turned out to be easier than I think most folks were expecting.

You can find more info here: http://ontap.riaforge.org/blog/index.cfm/2008/10/31/Release-of-Upgrade-Path-for-Legacy-Fusebox