Major Roadmap Update Centers Around Phoenix, Thunderbird; 1.4 Branch to Replace 1.0; Changes Planned for Module Ownership Model
Wednesday April 2nd, 2003
In the most radical change to the Mozilla project since the late 1998 decision to rewrite much of the code, mozilla.org today announced a major new roadmap proposal that will see Phoenix and Thunderbird (also known as Minotaur) becoming the focus of future development. According to the roadmap, 1.4 is likely to be the last milestone of the traditional Mozilla suite and the 1.4 branch will replace the 1.0 branch as the stable development path. mozilla.org is also proposing changes to the module ownership model including a move towards stronger leadership and the removal of mandatory super-review in some cases. Please click the Full Article link to read the full analysis.
#185 Re: Re: Please God, NO!
Friday April 4th, 2003 7:42 PM
You are replying to this message
"One question remains: How many person-hours of work will be lost? 5%? 25%? Obviously, only a rough estimate is possible."
The source is there. You can find an answer for yourself. Count the lines of code that create front-end elements are unique to Mozilla's front-end (that don't also exist in Phoenix). Now count the number of lines of code for the rest of the Mozilla application (all of gecko, necko, the toolkit, nspr, nss, editor, jsengine, oji, printing, xpinstall, xpcom, etc.) Make some estimate about how much time it takes to write and debug front-end code compared to back-end code and then do the math. My guess is that the unique front-end code for Mozilla is less than 5% of the total codebase and I'd also guess (I could be wildly off here because I do neither) that writing and debugging XUL and JS is faster and easier than C or C++.
You also have to look at how much of that old Mozilla code provided the basis for the new Phoenix stuff. The Phoenix options window, for example, is different than Mozilla's preferences window, but I'll bet if you look at the code you'll see that some of the new stuff is based on the Mozilla preferences XUL even if the window does look radically different. So was that Mozilla work wasted work? I say no. Not only did it provide useful access to the feature for many years, but it also served as the base for the advancements that Phoenix made. So maybe a better way to look at this is to look specifically at the time that was put in to just the particulars that aren't and won't be in Phoenix at all, and are not likely to be converted into standalone apps or extensions excluding any of the underlying toolkit or even some of the implementations that have been modified or otherwise reused, repurposed or extended to create the new Phoenix stuff. You'd probably also want to exclude some of the time put into even those particulars because they provided a valuable testing implementation that helped us to stabilize and improve the underlying infrastructure.
It's a tricky question. How many hours did it take to name the menuitem "Preferences" and place it on the Edit menu. Is that work lost now that the menuitem is called "Options" and was moved to the Tools menu? And how different is that from the hours that were put into making a table lay out one way, realizing that it wasn't ideal for the standard or the real-world web and then making it lay out some other way?
You should also probably factor in the time saved for Mozilla by the handful of very useful features and improved functionality that were developed for Phoenix and later incorporated in Mozilla at much less cost (everything from the re-write of bookmarks, to quicksearch, toolbar overflow, etc.)
Is it "lost" if it gets turned into an extension and moved to mozdev.org?
Is it "lost" if it the original author thinks it was buggy and created a less buggy replacement?
If some of your goals are higher quality/stability, increased performance, and reduced size then is the net effect of making massive improvements toward those goals by whacking a feature still considered a loss?
I'm not sure what the final answer is. I personally believe that we're gaining more than we're losing in this transition. That's why I support it. There's no doubt that some very talented individuals (several of them, blake, ben, hyatt, hewitt, are the folks that created Phoenix) put a lot of time and effort into the old Mozilla front-end. Whether or not it's fair to call that lost work and if it is, then how much lost, is not as easy as your question suggested. Notice that I said above that you could find _an_ answer for yourself. I suspect there are many different answers to this question depending on how you look at it.