[Project_owners] A MozDev Mission Statement?

Pete Collins pete at mozdev.org
Wed Sep 17 10:59:04 EDT 2003


> Could you reiterate those things 'we all know' just for the sake of 
> those like me that don't?

Sure. 'we all know' == "all the stuff we've been talking about in this 
thread".

> You want to make a GRE/MRE using a particular build of Mozilla? Is this 
> an attempt to make a GRE/MRE but without the mozilla module 
> ownership/review process? If so, I don't think there are that many 
> backend developers for that to be realistically possible.

Actually, no. GRE/MRE is a much more ambitious endeavour.

- A generic apprunner.
- Platform agnostic app icons
- Renaming .jar file to .moz files
- XUL apps running on their own thread.
- Ripping RDF out of the chrome registry
- etc

Lot's of "Constitutional Convention" type of stuff.  Huge time investment.

What I envisage from this discussion is something Mike Potter mentioned 
about two years ago at the last Mozilla developer meeting at Carnegie 
Mellon which is a *lot* easier to implement.

A stripped down build/distribution of Mozilla or Phoenix that has all 
the essentials to run a XUL app.  Everything else has been stripped out.

So after installing the MADBird core, you could run Mozilla and get a 
simple window that loads mozdev.org or some page directing you to a list 
of xul apps to install. This (installer) window needs nothing more than 
a url bar to navigate to your project so you can install your xpi.

Your xpi's can simply be run from the command line.

./mozilla -chrome chrome:/mypackage/chrome/

./mozilla.exe -chrome chrome:/mypackage/chrome/

You can provide (via xpi) a simple shell script or .ico file or both so 
users can quickly and easily launch your app.

This is a very doable unambitious first step. Mozilla is already a 
working system that most of us are happy using as long as it stays 
reasonably stable.  So it makes the most sense to build upon that.

Take what is *doable* as the first step. Then things can evolve in a 
direction that improves the Mozilla as Platform environment/paradigm.

I don't think anyone on this list has the time to work on the more 
ambitious MRE/GRE.  We just need to use the good stuff in Mozilla that's 
already there like many here suggested.

Hosting on Mozdev is a good idea. If there is no *real* interest in 
getting MADbird to fly, then it becomes yet another abandoned project. 
If there is sufficient interest, then it can perhaps go the route that 
spellchecker did. After it became mature enough and a proff of concept. 
it got checked into the mozilla tree.

The Mozilla tree has a much higher bar for any commit made (and 
rightfully so). This has the tendency to slow development down to a 
crawl. That's actually why Phoenix was an actual fork. So some work can 
actually get done.

Since MADBird is really a build or distro, then it makes sense to do it 
on Mozdev. Heck, i'm rolling out different custom Mozilla distros for 
clients almost every other week. It's not a big deal.

The idea also needs time to incubate and grow. So that's what Mozdev was 
designed for. Plus this idea is for the Mozdev communities immediate 
interest. So having the instant access provided by Mozdev will help 
facilitate the initial development process.

If MADbird turns out to be a DEADbird, then there it is. Remember it's 
real easy to have an idea or start a project. It is a *lot* harder to 
follow through and finish it.

Am I on track w/ the ideas expressed in this thread?

--pete


-- 
Pete Collins
www.mozdev.org
www.mozdevgroup.com




More information about the Project_owners mailing list