ietf-822
[Top] [All Lists]

Re: Revising Richtext

1992-02-07 08:00:56
The reason I oppose separating richtext into a separate RFC is fairly simple, and rooted in a good deal of real-world experience.

When you ask people why they might like multimedia mail, they invariably talk about audio & images. However, we've actually collected statistics on how the Andrew system at CMU, one of the world's largest communities where multimedia mail has been in actual daily use for many years. In that community, the overwhelming majority of all non-plain-text mail is simply text enriched in very simple ways.

In other words, what's actually most useful, in daily life, is to be able to provide simple emphasis to your text.

The problem is, much of the world doesn't realize this. Until you've seen this in operation, it's hard to realize how much this gains you. Thus what I really want from richtext is an extremely simple mechanism from doing this, and what I want from MIME is a strong encouragement for implementors to actually implement this. Separating richtext to a separate RFC conveys precisely the wrong message, to my mind, which is that richtext is not a very important facility, and that's why I'm so opposed to moving it to a separate RFC.

As I've argued many times, I think that the syntax of a simple richtext is almost irrelevant -- in the current version, its so simple that it isn't worth arguing over. What matters is that there's one such syntax, easy to implement and widely implemented. An argument that there are better document formats in the world, or that this one isn't perfect, completely misses the point. To my knowledge, there are no SIMPLER richtext formats in the world, none that are easier to implement, and none that are more likely to generate a wide consensus. The current spec favors no vendor, etc. And I think it provides a really vital facility.

If people feel that richtext is "not ready for prime time" I would like to hear the details. I certainly haven't heard anything yet that convinces me that there are any critical flaws here. All the arguments I've heard have been philosophical with respect to the one document versus two documents issue. -- Nathaniel
<Prev in Thread] Current Thread [Next in Thread>