operator, the browser automatically > puts a space between the paragraph's ... how can I stop > this behavior? anyone know? Thanks to all responses ... I now know that this is not the right newsgroup (comp.infosystems.www.authoring.html is) ... and to use \
instead of \

... -- ---------------------------------------------- Fred Phelan Polymer Composites National Institute of Standards and Technology Gaithersburg, MD 20899 ---------------------------------------------- e-mail .... fred@poly2.nist.gov Frederick.Phelan@nist.gov ---------------------------------------------- Newsgroups: comp.text.sgml Date: Thu, 15 Feb 1996 09:48:16 -0600 Message-ID: <9602151548.AA21411@fly.HiWAAY.net> From: Len Bullard \ Subject: Notations Does anyone know of a site where I can find formal public identifiers for AVI and WAV notations. Docbook does not include these. Yes, we know to use SYSTEM identifiers in lieu of public identifiers. Len Bullard From: davep@ACM.ORG Newsgroups: comp.text.sgml Subject: Re: Correct SGML comments and other correctnesses Date: 15 Feb 1996 15:40:44 GMT Organization: ACM Network Services Message-ID: <4fvk5t$9jp@hopper.acm.org> References: <4ftr74$99b@solaris.cc.vt.edu> Reply-To: davep@ACM.ORG In article <4ftr74$99b@solaris.cc.vt.edu>, jbazuzi@vt.edu (Jay Bazuzi) writes: >However, Arena sometimes thinks comments like this are unclosed: >\ >And I've seen people say this is correct. >To what internet resource can I refer to get the final word on the >matter? Where is TFM I can R for such answers to SGML questions? That comment declaration is "closed" (I assume that means "is a correctly formed comment declaration") and anyone or any program saying otherwise is wrong, unless you've really been meessing with your SGML declaration. TFM is ISO 8879; it's copyrighted so is not freely available on the net. The relevant material is 10.3 Comment Declaration t91] comment declaration = mdo , (comment , (s | comment)*)? , mdc [92] comment = com , SGML character* , com No markup is recognized in a comment, other than the com delimiter that terminates it. I trust ISO won't complain about my quoting that little piece. In the reference concrete syntax, mdo is "", and com is "--". That's what you can muck with in the SGML declaration if you really want to. Hope this helps Dave Peterson SGMLWorks! davep@acm.org Newsgroups: comp.text.sgml Date: Thu, 15 Feb 1996 11:12:40 -0600 Message-ID: <9602151712.AA31848@fly.HiWAAY.net> From: Len Bullard \ Subject: RE: Hypertext Theories Well, it seems I've underestimated Mr. Barger. I have read his posted description of his work at Northwestern with Roger Schank, et. al. He has indeed participated in some interesting work in hypermedia. Those who are interested in his views on that work and those with whom he worked should take five minutes and peruse his posting at http://www.mcs.net/~jorn/html/net/askjorn.html They will find there a wealth of information that will give them a better understanding of the directions and topics of this thread as influenced by Mr. Barger. Len Bullard Newsgroups: alt.etext,alt.hypertext,comp.publish.prepress,comp.multimedia,comp.publish.electronic.misc,comp.text.sgml,comp.text From: noring@netcom.com (Jon Noring) Subject: --> Press Release from OmniMedia on CDA of Telecom Reform Act Message-ID: \ Reply-To: omnimedia@netcom.com Organization: Netcom Online Communications Services (408-241-9760 login: guest) Date: Thu, 15 Feb 1996 16:47:26 GMT Sender: noring@netcom3.netcom.com FOR IMMEDIATE RELEASE Date: February 9, 1996 Contact: Jon E. Noring Phone: (801) 253-4037 E-Mail: omnimedia@netcom.com South Jordan, Utah "Statement by OmniMedia Regarding the Telecommunications Reform Bill" "Its Effect on Electronic Book Publishing" Yesterday, President Clinton signed into law the Telecommunications Reform Act. Included in this legislation is a very controversial and dangerous provision, the so-called "Communications Decency Act or Amendment" (CDA). The ACLU, along with several other civil rights and net advocacy groups, such as EFF and EPIC, and joined by several businesses, are all plaintiffs in a lawsuit just filed in Federal court to rule the CDA unconstitutional. The CDA is clearly an egregious assault on our Constitutional rights and deserves to be overturned. The ramifications of the CDA to civil liberties and rights are complex and wide-ranging. It is not the intent of this statement to outline all of these complexities and ramifications as they are many, and is being done by others. Rather, we will primarily focus on the effect of the CDA to electronic book publishing here in the U.S., particularly with regard to direct marketing of titles over the Internet. However, before doing so, it is necessary to give a short summary background about the CDA itself so its probable impact on electronic book publishing will be easier to see. In essence, the CDA makes it a felony, punishable by up to two years in prison and a hefty fine, to either access or to disseminate, on any public electronic network, material which is determined to be "indecent", and that "indecent" is defined by "contemporary community standards." Not only is "indecent" still legally vague, the few court rulings define it as being much broader than "obscene" and includes speech which the courts have determined to be wholly protected in many other venues. In addition, recent court rulings strongly imply that "contemporary community standards" are defined at the local level. For example, a literary work which is acceptable in California may be considered indecent by some local community elsewhere in the U.S. This means, theoretically at least, a person in California, or even outside the U.S., who violates the "decency" of some small community in the U.S., by simply distributing some material on the global Internet, which in his or her hometown would be considered acceptable, could be extradited by that small community to face felony charges, and to be judged by a jury from that community and not their own! This small community could hold the rest of the United States and even the world hostage to their over-broad view of what is "indecent." Not only is it grossly unfair to any rational thinker, it is dangerous to all of our civil liberties as it gives too much power to any small organized fringe group, including religious groups, to trample on the rights, practices, and desires of the silent majority in mainstream America. And, to make matters worse, the obsolete Comstock Act of 1873 was extended by the CDA to include all electronic networks, making it a criminal act, punishable by a 5 year prison term, to disseminate *any* information about abortion on the Internet, even in private e-mail. This, combined with the vagueness of "indecency", would stifle any discussion on legitimate contemporary topics such as abortion, birth control, AIDS prevention, GLB issues, scholarly and health-related aspects of human sexuality, etc. Under CDA, even some court rulings cannot be distributed over the Internet since they contain one or more of the "seven words" which the "Pacifica" court ruling determined were "indecent". Thus, the "Pacifica" ruling itself is banned from distribution on the Internet since it lists these "seven words"! And let's not forget the "Roe vs. Wade" decision on abortion -- it, too, cannot be legally distributed on the Internet, at least by the strictest interpretation of the CDA. Clearly the CDA is a gross violation to our First Amendment right of Free Speech as well as our socially-accepted right for all citizens to be able to access and obtain information, including court rulings, in the most convenient way existing at the time. Free exchange of information is one of the cornerstones of democracy. The effect of the CDA, should it prevail in the courts (which is fortunately doubtful), would also be a major setback to the new and rapidly growing electronic book industry here in the U.S. It might also lead to the eventual domination of publishing by Europe and Japan, though this is admittedly more debatable. Right now the U.S. enjoys a commanding dominance in publishing. Clearly, in the absence of CDA or similar restrictions, the future of electronic book publishing is very promising. Already we are seeing significant growth of the electronic book publishing industry; OmniMedia is one of several pioneering electronic book publishers, and new ones are coming "online" every month. Some of the major print publishers are beginning to venture into this area as well. As technology advances, particularly in high resolution and inexpensive flat screen displays, it is only a matter of time, some say in as little as 15 to 20 years, when more electronic books will be sold worldwide than their bound paper counterparts. In addition, the continual upgrading of our "Information Superhighway" to higher and higher speed and capacity makes it not only possible, but also very attractive (to both publisher and consumer alike) for electronic books to be directly marketed to the consumer via the Internet. This is already being considered by the multi-billion dollar entertainment industry for the sale and distribution of audio recordings and movies (note that the CDA puts a damper on this.) Any rational thinking person would see the immense value in this means of distributing information and entertainment of all types. OmniMedia and a few other publishers now distribute and sell e-books directly over the Internet via the World Wide Web (WWW) and are finding it profitable because it is very convenient for the consumer and significantly reduces the overhead costs of production, marketing and distribution, which, with free- market competition, the savings will pass on to the consumer. It does not take a rocket scientist to realize that the CDA would ban the electronic distribution of a large fraction of the books that are published today (it would not be surprising if over half of all books would be deemed "indecent" by the "seven words" criteria alone, not to mention the more intangible and uncertain aspects of what else constitutes "indecency", and by whom.) Therefore, the CDA has a profound chilling effect on the use of the Internet to distribute such material and, in practice, will not allow this method of distribution to mature in the U.S., to the loss of both the consumer and the publishing industry. Instead, publishers will still have to rely on the traditional way to market both paper and electronic books which is much more inefficient and expensive. Ultimately, it is a major blow to consumers who, in a CDA-free world, would be able to purchase e-books at lower cost than at present and who would have the convenience of rapid delivery of any titles available on the market right to their home. It would also greatly benefit publishers in ways that would take too long to discuss here. The CDA is a damaging blow to electronic book publishing here in the U.S. (As an aside, in a similar manner, many mainstream magazines would also be banned from Internet distribution. The rapidly escalating cost of high- quality "glossy" paper is making Internet distribution extremely attractive. But of course the passage of the CDA puts a stop to this as it does electronic book distribution over the Internet.) One could argue that the U.S. publishers could suitably encrypt or password protect their e-books allowing them to be distributed on the Internet and bypass CDA restrictions. But the current CDA is not so forgiving -- the CDA makes no distinction in what form the information is distributed. If the content of any e-book is deemed to be "indecent" by some small community as mentioned above, it makes no difference if it is encrypted or written in Swahili, or even if it is distributed on an electronic network where only adults are allowed access since the CDA covers *all* public electronic networks, no matter how subscribership to the networks is configured (the CDA does this, of course, to "protect our children", who must be so cunning as to be able to break into any closed network and to decrypt messages even the NSA could not decrypt.) And all the while the U.S. publishing industry is restricted to marketing books the "old-fashioned way", our European and Japanese friends will not be so restricted. They will enjoy the ability and advantages of marketing directly over the Internet in their own countries and so will clearly dominate there as well as developing the infrastructure to do so, getting a large technological jump over U.S. publishers. How this will effect the current supremacy the U.S. has in publishing is debatable, since it can be argued that it is still a level playing field -- the U.S. publishers can market their e-books over the Internet outside the U.S., and the rest of the world has to market their e-books the "old-fashioned way" within U.S. borders. But it is hard to imagine that the greater freedom and flexibility enjoyed by the Europeans and Japanese, as well as getting a technological jump, will not eventually translate into gaining market dominance and hurting the U.S. publishing industry. And they are formidable competitors: the Japanese domination in the U.S. market for consumer electronics and automobiles should serve as a warning of what could happen in the publishing field if the CDA is not overturned or significantly amended as soon as possible. Our competitors are looking for any crack in U.S. resolve to fully compete against us in the international marketplace, and the CDA is an obvious major crack. Billions of dollars and thousands of jobs are at stake here, and this was not even considered when the ill-written CDA was silently slipped into the Telecommunications Reform Act without any public input before Congress. And let's not forget the U.S. consumer, who will also be hurt by the CDA restrictions as pointed out above. Thus, OmniMedia urges Congress to trash the CDA and pass sane and publicly debated legislation, and not only because it would be good for U.S. business and consumers, but because it is morally the right thing to do. We support the right of parents to protect their children by providing parents with the necessary software to block any Internet information that they don't want their children to view. This is easily done. We do NOT support taking away the parental right to determine what their children should and should not view, nor do we support taking away the parent's own rights to Free Speech on the Internet. The current CDA does what the silent majority in America does not want: allowing some faceless person in some far removed community to determine what parents in mainstream America should and should not allow their children to view -- it is government intrusion into the family at its worst. It also takes away the parents' own rights to Free Speech, holding them and the entire U.S. public hostage to an insane piece of legislation, and to a small minority, whose stated purposes for pushing the CDA was to "protect our children" and to "stop child pornography" (what American doesn't support these laudable goals?) But the profound and chilling ramifications of the CDA go light-years beyond these stated purposes and suggest that the *real* motives of this small but dedicated minority were something that mainstream Americans do not share, nor want. One day "our children" will grow up and find themselves without Free Speech protection as well as fewer jobs, all because of the "wisdom" of the CDA. Without guaranteed Free Speech in all mediums, it will only be a matter of time before we won't be able to protect our children, either. Free Speech is a cornerstone of democracy, and without it, our other rights, including the Freedom of Religion, are in serious peril. And when these rights are gone, and the U.S. is a dictatorship, it will be *our children* who will be sent to the Gulags. I don't know about you, but I'd rather have my children see the "seven words" over the Internet (which they hear all the time at school anyway) than to see them live in a totalitarian or theocratical state. The CDA is a bad piece of legislation that does not serve the public interest. It does not add anything useful in the legitimate fight against child pornography and online harassment using the Internet (another claimed reason for its existence) -- the proven and Constitutionally tested laws we now have on the books have been shown to be adequate to deal with the problem when law enforcement is motivated to enforce them. And as it should have been made clear by now, the CDA is bad for business and very bad for U.S. economic competitiveness in the world. Jon E. Noring President and Owner OmniMedia, publisher of electronic books -- OmniMedia Electronic Books | URL: http://www.awa.com/library/omnimedia 9671 S. 1600 West St. | Anonymous FTP: South Jordan, UT 84095 | ftp.awa.com /pub/softlock/pc/products/OmniMedia 801-253-4037 | E-mail: omnimedia@netcom.com ------------------------------------------------------------------------------- Join the Electronic Books Mailing List (EBOOK-List) Today! Just send e-mail to majordomo@aros.net, and put the following line in the body of the message: subscribe ebook-list From: jenglish@crl.com (Joe English) Newsgroups: comp.text.sgml Subject: Re: SGML Parser, YAO, SP, help please.... Date: 15 Feb 1996 10:30:38 -0800 Organization: Tagheads Message-ID: <4fvu4e$8kb@crl2.crl.com> References: \ <4fu0kn$oh9@news.cerf.net> In article <4fu0kn$oh9@news.cerf.net>, Whatis \ wrote: > >You want to be able to query the parser for that data? I take it you >want to be able to, programmatically, navigate through a parsed >document, moving along and down tags. > >I've seen commercial libraries that do that, but no free ones. Cost [1] can do that. It builds a complete parse tree from the ESIS event stream output by sgmls, and can navigate to any node. There are query primitives to select a node's parent, all ancestors (bottom-up or top-down), younger and elder siblings, direct children, and all descendants. You can also navigate directly to any node by taking its "address", and I plan to add support for the HyTime 'pathloc' [2] and 'groveloc' [3] location models too. STIL [4] is event-driven, but inside a callback you can access any node's parent and its list of children, by which you can get to any node in the tree. SGMLSpm [5] only provides access to the ancestors of the current node, though it should be possible to reconstruct a complete tree in application code. --Joe English jenglish@crl.com [1] Cost: \ [2] If I can find out what to do with negative axis markers in the second dimspec... [3] If I can find out what this *is*... [4] STIL: \ [5] SGMLSpm: \ From: sgml@parka.winternet.com (Copernican Solutions Incorporated) Newsgroups: comp.text.sgml Subject: ANNOUNCE: Join Midwest SGML Forum Online. Date: 15 Feb 1996 19:28:38 GMT Organization: Winternet Corporation, Mpls, MN Message-ID: <4g01h6$n4@blackice.winternet.com> Yes, that's right--join online. On the Midwest SGML Forum home page there is a line to an HTML Forum that allows you to join the user group online. Your information will be e-mailed to the MWSF contact for the on-line service and the information will be forwarded to the Treasurer. You will be invoiced for the membership fee. Upon receipt of payment, you membership will start. You can access this service at: http://www.winternet.com/~sgml/mwsf -- ============================================================================== R. Alexander Milowski http://www.winternet.com/~sgml sgml@winternet.com Copernican Solutions Incorporated (612) 825 - 4132 From: Michael Hilton <100016.3262@CompuServe.COM> Newsgroups: comp.text.sgml Subject: Print Formatters Date: 15 Feb 1996 19:15:11 GMT Organization: - Message-ID: <4g00nv$pdv$1@mhafc.production.compuserve.com> Does anyone have any recommendations for a print formatting engine? Initially for HTML documents but ideally also for any SGML application's documents. i.e. something that you can fire a document at - it then generates a PostScript file with contents list, index, title pages, headers & footers, etc. Thanks, -Michael -- Michael Hilton From: sgml@parka.winternet.com (Copernican Solutions Incorporated) Newsgroups: comp.text.sgml Subject: Re: SGML Parser, YAO, SP, help please.... Date: 15 Feb 1996 19:38:39 GMT Organization: Winternet Corporation, Mpls, MN Message-ID: <4g023v$139@blackice.winternet.com> References: \ <4fu0kn$oh9@news.cerf.net> In article <4fu0kn$oh9@news.cerf.net>, Whatis \ wrote: >In article \, >Jaime A. Jofre \ wrote: >>I need info about SGML parsers. I got SP and YAO so far. Sp I got to >>compile, but not YAO. Basically what I want to do is give a parser a >>data file in SGML format along with it DTD and then instead of just >>parsing the data file, I also want to be able to query the parser for >>that data. I thought I could put hooks in the SP API but I haven't >>got any good docs. The same feeling was for YAO (which seemed >>simpler...), but I haven't been able to compile the system under UNIX >>or NT. Any clues would be greately appreciated. > >You want to be able to query the parser for that data? I take it you >want to be able to, programmatically, navigate through a parsed >document, moving along and down tags. > >I've seen commercial libraries that do that, but no free ones. SP provides full access to every part of an SGML document--including the prolog declarations (DTD). This information can be accessed through the EventHandler class and the subsequent Event subclasses (StartElementEvent, etc.). Take a look at the CopyEventHandler in spam. It should show you an example of a handler that accesses and can reproduce the exact SGML that was input to the parser through the event handler API. BTW, SP provides the most complete API I have seen. The only other one that I can think of that may come close is the MarkIt parser--but SP is free and MarkIt is not. -- ============================================================================== R. Alexander Milowski http://www.winternet.com/~sgml sgml@winternet.com Copernican Solutions Incorporated (612) 825 - 4132 From: jorn@MCS.COM (Jorn Barger) Newsgroups: comp.text.sgml Subject: Re: Strongly-Typed Hyperlinks Date: 15 Feb 1996 12:33:32 -0600 Organization: The Responsible Party Message-ID: <4fvu9s$sv@Mars.mcs.com> References: <9602132124.AA02210@fly.hiwaay.net> In article <9602132124.AA02210@fly.hiwaay.net>, Len Bullard \ wrote: >>For an example of a WWW system that uses link types in a similar way (to >>clarify where the user can look for different types of information), see: >> http://www.ils.nwu.edu/~e_for_e This is Roger Schank's hyperbook, Engines for Education. I was a senior research programmer at ILS from 1989 thru 1992, and was literally present at the creation of this link system. I'm coauthor with Schank of the 'Ask Tom' tech report. The linktypes used by 'Ask systems' were originally: background, results, context, examples, warnings, opportunities, alternatives, and indicators. The way these were chosen was by starting with a much longer list, including every sort of link known to AI, and then merging and generalizing categories until they'd fit neatly on a computer screen, without being too 'busy' for the corporation executives it was supposed to impress. Unless Chip has fixed EforE, you wont see these linknames unless you have a graphical browser-- they're written on gif-buttons without ALT text. EforE was apparently a non-hyper book that was run thru ILS's 'indexing mill'-- a dozen or so employees trained to index video clips (mostly) according to certain schemes. The original plan for the eight linktypes was that they'd be connecting separate anecdotes presented on video-- interviews with an accounting consultant, in fact. The last I heard, they'd never actually tested whether sorting the links gave any advantage over presenting them in a flat list. (I'm sure it doesn't.) When you use this system to connect arbitrary subsections of a work of non-fiction, what you get is *no added value*. No one wants to stop at the end of each subsection (page... paragraph) and read a dozen possible directions they might go next. (You can easily confirm this yourself by trying EforE. It's readable if you go straight thru, but not if you use the typed links.) >>This system uses link types to collect specific follow-up questions about >>a page under general question categories. This organization is intended >>to: >> - allow users who want to ask specific questions to quickly locate them, and Heh. Use the table of contents... it's your only hope! >> - to provide a set of general types of questions that less directed >> users can scan to see which, if any, they want to ask (and then to >> show under these general questions the specific instantiations of them >> that the system can actually answer). Yeah, that's *just* they way they conceptualize it!!! In English: - If a reader gets to the end of a passage, and feels no inclination to continue, she may browse a list of questions sorted into these eight categories. - It's imagined that she'll actually have a preference *among those eight categories* of what 'sort of direction' she wants to go. - Given this inclination, she can read just the questions in that category ("I'm feeling in a sort of Opportunities frame-of-mind...") - If she has a particular question, she can 'quickly' search for it by classifying it according to this scheme, and looking at just those questions. It may or may not be there. [...] >The system I encountered at this site organizes links into categories but >this appears to be the common static organization of local TOCs. Not at all. It's likely the most chaotic linking on the Internet. (Did you have graphics on?) > It also >states that dynamic linking is not implemented on the Web version but is on >the CD. And it takes something like 8 Mb to run, too! ;^/ > Can the links be organized for the Web dynamically using HTTP or >other stateless protocol? Please correct me if I have misunderstood, but >what I saw at the site seems to confirm the assertions that strongly typed >links require a stateful protocol. Could you elaborate more on how the ASK >system can be implemented fully on the WWW using HTTP? What else is >required? Well, as the sole programmer of the first Ask system (and the sole indexer, as well), I assure you that they *can* be implemented without state memory. \ is a brief look at EforE (called "Engines of Stupefaction"). See also \ for a much longer view of the origins of Ask Tom. j -==--- . hypertext theory : artificial intelligence : finnegans wake . _+m"m+_"+_ lynx http://www.mcs.net/~jorn/ ! Jp Jp qh qh best-of news:alt.music.category-freak ! O O O O ftp://ftp.mcs.com/mcsnet.users/jorn/ Yb Yb dY dY ...do you ever feel your mind has started to erode? "Y_ "Y5m2Y" " no. From: jenglish@crl.com (Joe English) Newsgroups: comp.text.sgml Subject: Re: --> Press Release from OmniMedia on CDA of Telecom Reform Act Date: 15 Feb 1996 10:41:27 -0800 Organization: Tagheads Message-ID: <4fvuon$8su@crl2.crl.com> References: \ Summary: The LOCUSTs are coming again! In article \, Jon Noring \ wrote: Newsgroups: alt.etext,alt.hypertext,comp.publish.prepress, comp.multimedia,comp.publish.electronic.misc,comp.text.sgml,comp.text >Yesterday, President Clinton signed into law the Telecommunications Reform Act. >Included in this legislation is a very controversial and dangerous provision, >the so-called "Communications Decency Act or Amendment" (CDA). > ... with no Followup-To:. Guess it's going to be another one of *those* months for comp.text.sgml :-( --jenglish@crl.com at least he left out comp.infosystems.www.* ... From: sgml@parka.winternet.com (Copernican Solutions Incorporated) Newsgroups: comp.text.sgml Subject: ANNOUNCE: Presentation submission for Midwest SGML Online. Date: 15 Feb 1996 19:32:01 GMT Organization: Winternet Corporation, Mpls, MN Message-ID: <4g01nh$og@blackice.winternet.com> If you have an interesting paper or topic that you wish to talk about, you may now submit a presentation request on the Midwest SGML Forum home page. This information will be routed to the requested chapter(s) of Midwest SGML Forum and you will be contacted for further information. Vendors: You may submit requests here as well. Vendor demonstrations are allowed as long as they do not dominate the user forum. -- ============================================================================== R. Alexander Milowski http://www.winternet.com/~sgml sgml@winternet.com Copernican Solutions Incorporated (612) 825 - 4132 From: sgml@parka.winternet.com (Copernican Solutions Incorporated) Newsgroups: comp.text.sgml Subject: Re: ANNOUNCE: Presentation submission for Midwest SGML Online. Date: 15 Feb 1996 19:33:23 GMT Organization: Winternet Corporation, Mpls, MN Message-ID: <4g01q3$oj@blackice.winternet.com> References: <4g01nh$og@blackice.winternet.com> In article <4g01nh$og@blackice.winternet.com>, Copernican Solutions Incorporated \ wrote: >If you have an interesting paper or topic that you wish to talk about, >you may now submit a presentation request on the Midwest SGML Forum home >page. This information will be routed to the requested chapter(s) of >Midwest SGML Forum and you will be contacted for further information. > >Vendors: You may submit requests here as well. Vendor demonstrations > are allowed as long as they do not dominate the user forum. > > It always helps if I put the URL on the announcement... http://www.winternet.com/~sgml/mwsf -- ============================================================================== R. Alexander Milowski http://www.winternet.com/~sgml sgml@winternet.com Copernican Solutions Incorporated (612) 825 - 4132 From: Al Hickey \ Newsgroups: comp.text.sgml Subject: CALS Australia Conference '96 Date: Thu, 15 Feb 1996 12:57:46 -0800 Organization: Directorate of Naval Logistics Policy Message-ID: <31239E4A.7EBD@navmat.navy.gov.au> The CALS Australia '96 conference "Beyond CALS" will be held at the Adelaide Convention Centre, North Terrace, ADELAIDE, South Australia on 1-3 April 96. For further Information please see: http://www.navy.gov.au/cals/cals96.html From: "W. Eliot Kimber" \ Newsgroups: comp.text.sgml Subject: Re: DTD for *incorrect* HTML Date: Thu, 15 Feb 1996 13:23:06 -0900 Organization: Passage Systems Inc. Message-ID: <3123B24A.2BA0@passage.com> References: \ <311E2EAB.14CC@passage.com> \ Bernd Kreimeier wrote: > > In article <311E2EAB.14CC@passage.com> "W. Eliot Kimber" \ writes: > > Joe Armstrong wrote: > > > > > > Is there a DTD for flawed HTML. > > > > \ > \ > > ]> > > The w3c SGML declaration for HTML (3.0) says SUBDOC NO. Don't you > have to change this for your flawed_html.dtd as well as the w3c > HTML.DTD variations (strict, recommended)? To my understanding, > SUBDOC NO reflects the limitations of current HTML web browsers. > Perfectly complying, both Mosaic and current Netscape display minor > garbage at the document's start. I don't know what SUBDOC would have to do with this issue, but in fact, I want to appologize for my post, which was a bit snotty and made in haste under the influence of insufficient sleep and excess stress (no excuse). My proposal was not intended to be that serious, although given that most elements in HTML are not containers, it almost works as is. You'd also have to make the content models of all the containers ANY as well for it to actually allow you to parse anything. One unsolvable problem is that browsers will process pages that don't start with \ or something that could imply it by the normal rules of SGML markup minimization, so it's actually not possible to handle all *browsable* HTML files using this sort of highly-permissive DTD. Separated from any critique of HTML as a DTD design (which we need not rehash), this issue touches on the SGML without DTDs thread, as in many ways, that's the way HTML is treated. As both DynaText and Panorama demonstrate by example, reasonable (but not complete) parsing of SGML documents is straight forward *if* the containment hierarchy is obvious. More about which in the that other thread. -- \