From: Alex.Hopmann@resnova.com Message-Id: <199410310241.SAA28787@nic.cerf.net> Date: Sat, 29 Oct 1994 00:55:21 -0500 To: firstname.lastname@example.org, email@example.com Subject: RE:Re: More thoughts In article of 11:16 PM 10/27/94, firstname.lastname@example.org (Tom Magliery) writes: >I'm not disagreeing about the usefulness of a product that saves the > user >from entering tags the hard way, but at some level, the user is going to >have to be able to indicate that some certain groups of characters in > the >document are, for example, a "level 1 header", or a "paragraph", and so > on. >So much the better if they've got WYSIWYG to boot. I think Microsoft Word is a great example of such a scheme. >But at the same time, from your point of view as creator of the product, >it's important to adhere to the principle of "accept liberally, generate >conservatively". What you're doing is generating HTML. While the user > is >happily pecking away at his document, you need to make sure that the > HTML >that you're quietly representing it with in the background is always > valid. >(Or if it temporarily isn't, somehow let the user know that something's >wrong with what they're attempting to do.) Well if we define that "unrecognized tags' are ignored, I think we can generate HTML to do nifty stuff, that might get ignore, but oh well. Recently I browsed some WWW pages using NCSA Mosaic version 1.x, and guess what? No forms. But no crash, no great tragedy. Alex Hopmann ResNova Software, Inc.