Editorial: NGLayout in Communicator 5.0 - Truth and Consequences
If you have no context for the last paragraph (have you been visiting our site regularly?), NGLayout is the code-name for the new layout engine, actively being developed by Netscape and the Mozilla Project, that will (when completed) attain full compliance with HTML 4.0, XML, DOM and CSS1. The Web Standards Project is an organization of designers and concerned developers and citizens rallying around web standards and their application in web browsers and other browsing systems. The WSP recently placed a petition on their website calling for Netscape to delay the release of Communicator 5.0 until NGLayout can be completed and integrated into the system.
With that introduction, on to my feelings about this state of affairs...
Demanding that Netscape hold Communicator 5.0 until NGLayout is completed is a misguided effort that does not take into account software development cycles and the financial consequences of altering them. The goal of the Web Standards Project should be the promotion of the use of web standards across browser implementations.
Instead, I found two disparate statements of mission on the site. From the front page,
" Our mission is to stop the fragmentation of the web, by persuading the browser makers that common standards are in everyone's best interest."
And from the FAQ,
"Our major goal is to get full support in the 5.0 browsers for Cascading Style Sheets 1 (CSS-1), the Document Object Model (DOM) and XML."
The "I Want My NGLayout" petition does meet this second goal. The problem is that this second goal is not logically derived from the WSP mission statement. A correct progression from the mission statement, one that does not introduce arbitrary variables or cause potential conflicts, would read, "Our major goal is to help browser makers obtain full standards compliance for Cascading Style Sheets 1 (CSS-1), the Document Object Model (DOM) and XML in clients as soon as possible."
However, with the addition of the "5.0" clause, the interim goal has a ruinous effect on our favorite browser company: it insinuates itself into Netscape's marketing strategy and actively attempts to manhandle Netscape's decision-making power over their development process and resources. Consider this statement from a member of the WSP: "They'll be a lot better off than stepping into the ring [without better standards support] and getting clobbered -- which they will." And from the WSP web site, front page: "Whip Netscape into shape."
The WSP's petition and the comments by some of its members in the newsgroups amount to nothing more than the bullying of a company that has shown us its intentions very clearly in its recent development efforts. The Mozilla project has clearly stated that NGLayout would fully support CSS1, XML, DOM, and almost all of CSS2 (if not the complete spec) by the time of its first release. This is a statement that has more weight behind it than statements from Netscape in the past - the source is now open, and the resulting scrutiny by designers and other coders is much greater. NGLayout will have these standards, and it's just a matter of time before it makes it into Communicator.
If the petition was more than a publicity stunt, they would be haranguing Netscape over Communicator release 4.5, the next major release of the browser. But it isn't, and the net result makes Netscape look reluctant, closed, and contrary -- when in fact the opposite is true.
If they state that out of respect for the browser companies' development cycles they settled on a 5.0 compliance timeframe, then as an extension, out of respect for Netscape, they should accept a timeframe and release schedule that allows Netscape to complete its stated goal of full compliance without causing serious financial disruption. They should also keep in mind that because of anticompetitive monopoly tactics by Microsoft, Netscape has had to pursue a marketing course for Communicator in which it receives no revenue from its programmers' hard labor. The WSP should consider tempering their stance, and focus on working with Netscape to create a browser "by the people, for the people".
What's Really Involved
The act of "integrating" NGLayout is far from the simple task that people may believe. As some developers will tell you, it essentially comes down to a rewriting of the front-end of the browser (the front end meaning the menus and chrome and windows that house the layout engine). NGLayout is not just a snap-in replacement for the old layout engine; it handles many processes in a fundamentally different manner, and this new paradigm has to be accounted for. Considering all the work that was involved in getting the FE's to their current state, that is quite a daunting task (even if a good portion of the code can be reused).
Add to this the fact that the current development cycle started well before any outsiders heard of NGLayout. It seems wrongheaded to have Netscape throw away hundreds if not thousands of man-hours on viable work so the engineers can start the task of integrating a layout engine that is not yet complete. See my statement in the responses in this talkback to read a clearer explanation of this.
Yesterday, however, I read the news articles about Netscape's marketshare eroding, and more became clear.
The Kangaroo Court of Public Opinion
Netscape can be seriously hurt by this movement, even by its misrepresentations. For example, Microsoft touts the technical superiority of its IE4 client (unfounded), and designers, rallying around the desire for NGLayout in 5.0, constantly point to IE's gains of marketshare, claiming that these gains are a result of technical superiority and closer adherence to standards. And they follow up by saying that they will ditch Netscape like a broken two-dollar wristwatch if it doesn't meet their demands in - and this is the kicker - not a timeframe, but by an arbitrary release #. These same designers, calling for full compatibility of Communicator (as is their right), are neglecting to focus the same attention on IE, which may be further along in CSS compliance but far from the finish line. They step around the fact that IE's gains are, if anything, a result of MS's anticompetitive practices in the marketplace. They ignore the fact that Communicator is winning customer satisfaction polls against IE. They neglect the fact that IE is a technical morass (betas of which can eat hard drives for lunch), that it is bloated to the point of obesity and nearly impossible to uninstall completely. That it creates directories you can't delete. That it insinuates itself into your system like a virus (in the case of Windows95, like a virus feeding on a parasite). That it updates Operating System files without warning, and makes your OS run like molasses. That it sits in your active memory like a sleeping bear, to make its startup faster than the competition. These same designers, when called upon for statements in articles about Netscape's status in the community, take the opportunity to badmouth the company, with the idea that this browbeating will whip Netscape into submission.
The whole point of this petition comes down to nothing more than an attempt to intimidate Netscape into doing something that may be a mistake for them financially. From the newsgroups and web forums we hear intimidations and dire warnings and threats that seem to utterly disregard the gains that Netscape is making in becoming fully standards compliant (despite the call-to-arms on the WSP website). Outsiders have decided on Netscape's marketing strategy for them, and if Netscape doesn't comply, it'll be whipped like a dog in the press and in product reviews. And Microsoft will not only use these statements in their press releases, they'll use it in court, as well.
These are realities that Netscape cannot ignore.
A Horrible Truth
And this is why, when I survey this sorry scene, I have to say that Netscape's not acceding to these demands would be a mistake. When you have a gun pointed at your head by a design community with an itchy trigger finger, and you have a press that has no self-restraint (that quality applies to more than just the political press, it seems) your best bet is to do as they demand.
I can only suggest that Netscape bow to their demands. Put the work currently going on in Mozilla into version 4.7 (as someone else in our forums mentioned). Let the beta cycle of 4.7 hang on until NGLayout is ready for proper beta testing. Then release 5.0 betas on the public. If the WSP is seriously concerned about 5.0 being compliant, it won't mind waiting until the code is ready. And in the meantime Netscape can get some user testing and feedback for the many great new ideas appearing in the current Mozilla builds.
Netscape has clearly shown their willingness to work with the design community in reaching the goal of full standards compliance in their browsers. This fact is indisputable. The WSP has shown their ability to work with the Mozilla team to reach this goal. This is a noble stance, and the WSP should regroup, and focus on their mission: encouraging and supporting browser makers in their efforts to become standards compliant.
Got a response? TalkBack!