MozillaZine

A Reader Calls On MS To Adopt Mozilla's Rendering Engine

Tuesday September 14th, 1999

Dave Usher of Tiny Communications, a new free-access Internet service in the UK, writes, "Inspired by the recent discussions on MozillaZine I've written an open letter advocating that Microsoft adopt the NGlayout (Gecko) code for their IE rendering engine."

An interesting idea, IMO, which would preserve standards compliance on all platforms while allowing browser makers to compete in the user-interface arena. Instead of putting out competing (and incompatible) rendering engines, why not rally around a single rendering engine which could help guarantee standards compliance by virtue of its Open Source nature and cross-platform architecture?


#26 Reference implementation? Impossible

by Anon

Wednesday September 15th, 1999 11:06 PM

You are replying to this message

Dave Usher wrote:

`Gecko can then be the refer[e]nce implementation of the standards and both Microsoft and any other Mozilla developers would have to ensure that it kept up and fixed all the bugs to make sure it complied 100% with all the standards.'

<HAL>I'm sorry, Dave, you can't do that.</HAL>

*giggle* ... <cough/>

As I was saying. You can't do a reference implementation of HTML. Reference implementations of a specification only make sense if the specification was intended to be interpreted in only one way -- HTTP, for example, or TCP/IP.

However, single-minded interpretation is the very *antithesis* of HTML. HTML was deliberately designed so that it could be rendered in a huge variety of ways.

Some of these variations in interpretation are the result of the device: a terminal-based browser is going to render the META description of a page differently from a search engine, for example. And some of these variations are the result of individual choice (of browser, or browser preference).

Take a simple example. Let's say that Gecko indents bulleted lists (ULs) and IE's engine doesn't. (Without either engine handy, I don't know if this is the case, but it's just an example.) Now, if IE adopts Mozilla's rendering engine, some people will start assuming that UL == indentation, because that's how it works on `both' browsers. So they'll start using UL to indent text. And when you complain to them that it's not following the standard, they'll say to you: `what do you mean? Gecko *is* the standard'.

As other posters have noted, having both Mozilla and IE using Gecko will only encourage Web site developers to think that `the way Gecko does it' == `the way it is'. And those people who use browsers other than Mozilla and IE (and who are *entitled* to use those other browsers) will be worse off.

-- mpt