[Project_owners] Getting to know you...

Pete Collins pete at mozdevgroup.com
Fri May 27 09:17:01 EDT 2005


> Ready for 3rd party developers (non-Mozilla hackers).  As far as 
> capabilities go, the Platform seems capable, albeit, with a few ugly 
> warts.  But, it gets better all the time -- that's why I still lurk 
> and that's exactly what I'm watching evolve.

I've been watching it *slowly* evolve sine 1999.

>
> I bought 'Creating Applications with Mozilla' back in 2002 and I'm 
> still waiting for Mozilla to create the Platform that the book was all 
> about.  The same thing goes for the 'Rapid Application Development 
> with Mozilla' book that I bought in 2003.  I just don't think that the 
> Platform has materialized yet, and what little there is of the 
> Platform is poorly documented and not ready for 3rd party developers.

Yup. It hasn't. It's getting there but there needs to be a huge effort 
to get it over the hump. A lot time and effort and no one to pay for it.

>
> Now, Myk mentioned http://developer-test.mozilla.org/ is being used to 
> address this problem, but from what little I've seen, developer-test 
> is a replacement for the now-dead Netscape DevEdge.  In other words, 
> it's not Platform documentation; it's more like IBM's DeveloperWorks, 
> which is useful, but not the type of documentation that I think needs 
> to be available for Mozilla to be a viable Platform.

Correct. You can't really document technology that is still changing all 
the time, with baroque API's not created for general purpose 
consumption. The platform was an afterthought and never a core part of 
architecture or focus.

>
> I will also throw out the amazing http://www.xulplanet.com/ website as 
> a good beginning source of documentation.  But, it's one person 
> maintaining the whole thing right now with very little community 
> involvement.  The forums are virtually dead, comments on the various 
> tutorial pages/reference pages seem few and far between.  I feel that 
> it's a PHP-like documentation site that just hasn't achieved critical 
> mass which doesn't make it as useful as it could be.


That could just be that the pool of MAD developers is a steadily growing 
one. That will probably change. But I can see how developers who are 
accustomed to developing on a stable platform will sometimes throw their 
hands up in frustration and say this is a royal PITA.

>
> With XULRunner on the horizon, Mozilla the Platform might be just 
> around the corner.  I don't know because I can't find a decent 
> roadmap, or even exactly what XULRunner is supposed to be providing 
> us.  I get the sense that the immediate goal is to get XULRunner ready 
> for Firefox, Thunderbird, and Sunbird, but that's not going to lead to 
> a very well thought out Platform by itself.  Again, it'll be 
> interesting to see it evolve.

I don't know what XUL runner is going to do either. From what I can tell 
it's going to be a way for FF, TB, SB, to run off the same core libs 
instead of rolling out redundant distributions for each.  The suite 
"lite" w/ different UI threads?

>
> I really don't want to whine about the state things are in because I'm 
> not doing anything to make them better.  Whining is not my intention. 
> You asked for clarification on my stance and I believe that I've 
> provided exactly that.
>

Your complaints are essentially the same complaints I've had since 
1999.  You have to realize that the platform was a serendipitous act by 
a few (mostly external) developers who were not thinking or hired to do 
browser development.

I know what the big problems are developing apps for Mozilla, as do many 
on this list.

How do you fix these problems realistically?

1. Address all the weaknesses of "The Platform"
2. Embark on an well organized path to fix them

To me that equates to a clearly stated goal, developer resources, time, 
money.

--pete

-- 
Pete Collins - Founder, Mozdev Group Inc.
www.mozdevgroup.com
Mozilla Software Development Solutions 




More information about the Project_owners mailing list