--Jeff Veen, director of interface, Wired Digital/Lycos.comThe fastest Web sites, regardless of end-user bandwidth, will be the most successful.
When I think of the things I do online -- manage my portfolio, buy plane tickets, use a search engine -- I'm looking for a page loading experience of under one second. Period. It should be as fast as the applications I use on my desktop, regardless of branding, advertising, and interface.
So there's a broadband challenge. Yes, many million more folks will have access to high bandwidth. But your Web site will need to be as fast as ever to compete.
--Tim Bray on the xml-dev mailing listI'm beginning to think that we should have written into the spec an express prohibition against land-grab attempts on the address function of the namespace name. The notion that a single URL can address the One True Schema Which Will Meet All Needs is demonstrably, empirically, absolutely wrong.
Often prefixes are chosen as a useful mnemonic, for example, using "xsl" as a common prefix for 'http://www.w3.org/1999/XSL/Transform'. In this case, it is nice to be courteous enough to maintain the prefix through transformations if possible.--Uche Ogbuji on the xml-dev mailing list
namespaces are not about semantics but about avoiding name clashes. In other words, the namespace spec is about telling 'foo:href' apart from 'html:href' and not about what 'foo:href' means within a certain element.--Don Park on the xml-dev mailing list
Today's web technology is HTML driven: content, logic, style and behavior mixed in a single file. Very nice for newbies (you learn HTML and you're done), a pain for more advanced uses.--Stefano Mazzocchi on the xml apache mailing list
XML separates data from presentation as well as logic. It also separates these from the transport. HTML files blend all of them.--Mike Dierken on the xml apache mailing list
Namespace support is an all-or-nothing thing - if you have one critical tool in the set that doesn't support namespaces, then it is likely that you won't use namespaces.--Peter Murray-Rust on the xml-dev mailing list
A few years ago, the then SGML community made a huge titanic effort and produced something called "architectural forms" which was said to do inheritance (sort of); unfortunately, only 25 people in the world ever understood it.--Tim Bray on the Apache XML mailing list
Until schemas/DTDs are capable of doing real work with namespaces, we are living in the pre-namespace era. The W3C dropped the ball on validation and namespaces, and we've been living with the consequences - life between 'eras' - ever since.--Simon St.Laurent on the xml-dev mailing list
Shortly after the release of XML, some folks, including some very important folks in W3C and its members, who had been big supporters of XML, actually got around to reading the spec, and discovered to their horror that they had, instead of DPML (Don Park Markup Language), an XML which included entities, DTDs, PIs, and assorted other baggage.--Tim Bray on the xml-dev mailing list
The real problem in wireless technology as far as content providers are concerned is figuring out the best way to display data on an incredibly small screen...--Richard Lanyon on the xml-dev mailing list
The reason the Web has taken off is that it promises data-stream ubiquity and connectivity, freeing businesses from dependence on proprietary technologies. XML is on the plate of every database company I've talked to. Because it promises a *useful* format for making their proprietary technology available to everyone in the world. Major customers are *demanding* it, even before it's fully standardized, because their *yearning* for multiple suppliers and second sources is so strong that they're calling it into existence by sheer willpower.--Lee Anne Phillips on the xml-dev mailing list
XML itself took a brilliant (though somewhat compromised) approach - defining a minimum set of tools that could be used by the widest range of users. That foundation - elements and attributes - provides much more powerful and flexible structures than was readily available before.--Simon St.Laurent on the xml-dev mailing list
If you want to simplify XML for beginners then go ahead. But why not just write the definitive introductory book that takes the core features and explains them so simply that it just seems obvious, and then has an appendix called 'other features of XML you might want to use but don't worry about for now'.--Mark Birbeck on the xml-dev mailing list
As long as human beings are the only plausible "end consumers" of these documents, their semantics will always be determined ultimately by fuzzy things like intentions and expectations. The semantic constraints I am talking about are one step away from these "ultimate" semantics; they tell you that an integer contained in a given element cannot be greater than 100, but they don't tell you why--Matthew Gertner on the xml-dev mailing list
SAX is simple but it is difficult to use in applications where access to yet-to-come elements and character data is needed.--Don Park on the xml-dev mailing list
the important thing to remember about RDF is that it is not XML. It is fundamentally an abstract model for expressing metadata. It happens to be representable using XML, but it is different from XML.--Jeffrey E. Sussna on the xml-dev mailing list
SML is not contributing anything to the greater understanding of mark-up languages. Witness the fact that the debates are re-hashes of old XML debates, not new ones.--Mark Birbeck on the xml-dev mailing list
The SGML market has flatlined, only a year and a half after XML came out -- there are virtually no new SGML products, no new SGML books, almost no SGML conferences, and everyone I know who has a big SGML system is privately talking about XML migration plans (which may take a decade in some cases). SGML development on the major document types and industry specs (NITF, DocBook, TEI, etc.) has ceased completely, and all new industry document types that I've seen recently are coming out only in XML.--David Megginson on the xml-dev mailing list
We can't afford to support or develop to secret or proprietary standards, such as currently exist in all the major browsers. Individual designers who study the tricks of the trade may profit from this lack of standards, but at the expense of their customers, who have to pay dearly for their expertise. Individual companies, especially those with monopoly dominance of the market, may profit from selling figurative razors that fit only their own razor blades, but at the expense of their customers, who may well resent it. The "spin" from marketeers is that "innovation" in standards is a *good* thing, but try selling an "innovative" bolt size to anyone in the world except, as mentioned before, a nut.--Lee Anne Phillips on the xml-dev mailing list
The problem with SGML was that almost no one except James Clark or a well-funded university research project had the time and energy to create the SGML libraries in the first place (I know, I tried to do it in Java). By comparison, I had AElfred doing useful parsing in the first couple of hours, and could handle most documents in a couple of days (adding character set support and tuning the performance took a little longer) -- it was an order of magnitude simpler to write an XML parser than it was to write a general SGML parser.--David Megginson on the xml-dev mailing list
DOM still needs some work before it can truly be an implementation-independent API. This includes having ways to hook a DOM up to an XML processor (parsing document text into a DOM tree), and setting options for validation, whitespace handling, and use of various types of nodes in resulting tree.--David Brownell
XML has become increasingly crucial throughout the software industry, as well as the Open Source community, as a nonproprietary method for storing and exchanging complex data--Brian Behlendorf, President of the Apache Software Foundation
--David Megginson on the xml-dev mailing listtraditionally, standardization has solved existing problems: everyone had trains but they could not run on the same rail gauge, or there were many different phone companies but customers from one could not call customers from another another, or every public house used a different sized glass so it wasn't possible to compare prices.
Now that we're trying to standardize in *advance* of implementation, we run an enormous risk of messing up: our industry simply lacks any real, large-scale implementation experience to guide the process, so we're just publishing our own wild speculations and stamping them as W3C Recommendations or ISO standards or what-have-you.
Far too much of current Web and data design practice is involved in making metaphorical wrenches to fit proprietary bolt sizes from every manufacturer. The business community as a whole doesn't like it. They'd far prefer a world in which they could depend on things working interchangeably. And it *will* happen. That's where the money is. Even Microsoft will be forced to go along, one way or another.--Lee Anne Phillips on the xml-dev mailing list
--David Megginson on the xml-dev mailing listMicrosoft has backed so many losing specs over the past few years (and then quietly backed away, leaving naive smaller partners dangling) that the effect of their backing alone cannot have helped XML get to where it is.
What's special about XML is that *both* Microsoft *and* the anyone-but-Microsoft camp (Sunday, IBM, Oracle, and Netscape [at the time]) backed it.
I have often wondered where this myth arose that DTDs are somewhere between evil and useless. It's baffling that this myth arises when there's nothing better than, or even as good as DTDs. Meanwhile, perhaps because of this myth, several serious W3C efforts appear to be taking giant steps backward from providing even the usefulness that DTDs do in fact provide. I have no adequate explanation for this. I'm putting it down to some new electronic form of mass engineering hysteria, in which rumors, perhaps based on casual and unguarded remarks of experts, shape public opinion in unintended ways that work against the public interest.--Steven R. Newcomb on the xml-dev mailing list
If I were paid to help people figure out XML syntax, I would be way, way, way, WAY overpaid. What companies hire consultants for is to help them understand how to exchange and process information: 90% of the complexity comes from the nature of the information they're trying to model and the business environment in which they work, 9.9% of the complexity comes from finding, learning, and integrating the software components, and perhaps the remaining 0.1% has something to do with the syntax of the markup layer (but probably not).--David Megginson on the xml-dev mailing list
I have often wondered where this myth arose that DTDs are somewhere between evil and useless. It's baffling that this myth arises when there's nothing better than, or even as good as DTDs. Meanwhile, perhaps because of this myth, several serious W3C efforts appear to be taking giant steps backward from providing even the usefulness that DTDs do in fact provide. I have no adequate explanation for this. I'm putting it down to some new electronic form of mass engineering hysteria, in which rumors, perhaps based on casual and unguarded remarks of experts, shape public opinion in unintended ways that work against the public interest.--Steven R. Newcomb on the xml-dev mailing list
I think we should shoot for a goal within the developed countries of having Internet access as complete as telephone access within a fixed number of years. It will do as much as anything else to reduce income inequality.--Bill Clinton
Jon and I and some of the others made a concentrated marketing effort starting at the end of 1996 and went shouting off in all directions about XML. It was like hurling your entire weight against a locked door that turns out not to be there. The world, more or less, said "yeah, OK".--Tim Bray on the xml-dev list
I recommend following the conventions even in those cases where a different design would be better if seen in isolation. The fact is, no website is seen in isolation: users come to your site expecting things to work the same way they are already used to.--Jakob Nielsen
DTDs and other schemata can be useful (I would have less error-checking code in FOP if I used a validating parser), but I agree that a myth has arisen that DTDs (and especially schemata in new schema languages) do a lot more than they actual do. My favourite myth is that a schema tells you what an XML document means.--James Tauber on the xml-dev mailing list
There's no good reason for XML to be hard - there's no good reason programmers should have to know the ins and outs of XML to be able to process XML documents reliably. Abstracting all this work to the 'parser' (which works by magic, right?) has made XML much more usable than its competition. Unfortunately, the contents of documents may vary depending on which parser you choose.--Simon St.Laurent on the xml-dev mailing list
The key reason for needing external parsed entities is "manageability". Large documents are not written as monoliths. They are written and reviewed in sections, with different teams often being used to create and comment on different sections. If the document needs to be translated this often needs to be done in sections as well, rather than waiting for the completion of the last page of the document before you can start translation.--Martin Bryan on the xml-dev mailing list
XML is central to the future of the Web. Apache is central to the Web today - among other things, I'm betting my new startup on XML and the Apache server - and the two need to work well together. Today's announcement makes it clear that this won't be a problem. The combination of Apache and XML is going to hasten the day when proprietary system software and proprietary data formats are both regarded as quaint antiquities--Tim Bray
XML would be altogether better if we'd just bagged external parsed entities.--Tim Bray on the xml-dev mailing list
Each year or so, the computer industry anoints a new technology as the "holy grail" of software development. The trade press happily bangs the drum, encouraging upper-management to hand down edicts outlining grand technology visions according to the pundit du jour. XML is bound to fall prey to this nonsense.--Don Box
5 years down the road when phones are using real web browsers that can hit pages written for *everybody*, where do you think WML is going to be left standing? [x]HTML obviously has a future. I'd be cautious about spending much time with WML unless you absolutely *have* to author for these current-generartion crippled 9.6Kbps (GSM) and 14.4Kbps (CDMA) devices.--Jelks Cabaniss on the xml-dev mailing list
in the end, it's simple peer pressure that forces interoperability. That's a terrifying thought for standards writers, but it's also the explanation for why scarcely any of the 18 W3C Recommendations at http://www.w3.org/TR/ have actually been widely implemented so far (CSS1, XML, and maybe DOM -- anything else?). We won't even start counting the WDs...--David Megginson on the xml-dev mailing list
Would it not save everyone a lot of anguish if XHTML 1.0 were just HTML 4.01 recast as XML 1.0, in other words without mentioning namespaces *at all*, and saving the namespace stuff for future versions (if it's proven to be viable)?--Jelks Cabaniss on the xml-dev mailing list
there is also a huge amount of communal MIS-understanding that lives in wetware. It is flushed out through the process of formalization. For instance we all talk about "links" but it is only in trying to formalize XLink that it became clear that we all used the word to mean radically different things.--Paul Prescod on the xml-dev mailing list
If your objective is to support multiple browsers, then you have to generate HTML on the server. There is no other way.--Terris on the xsl mailing list
Ignore what the spec says RDF is (particularly the stuff about metadata) -- it's really just a flexible and robust format for serializing objects in XML over the Web. In particular, you are well advised to skip any section of the RDF spec that contains the word "reification" (you won't be any poorer for having done so).--David Megginson on the XML-L mailing list
It's funny (at least to me) that HTML started out as "make this text bigger than that text" where presentation (what font, etc..) was entirely under the control of the user ... and now has devolved to yet another complex page description format. So complex that we now have a content-pure format, XML, to do the job that HTML was supposed to do in the first place.--William Volk on the WWWAC mailing list
In the 1980s, Charles Goldfarb invented SGML, an elegant metalanguage focusing on the careful composition of reusable structured data, and thousands loved it. In the 1990s, Tim Berners-Lee created HTML, a hobbled subset of SGML that completely ignored the latter's fundamental tenets... and revolutionized the world economy, creating fortunes and moving nations. As it conquered the universe, the weaknesses of HTML began to show, and it dawned on everyone that they needed something more flexible, more generic, more reusable, something more like... yes, SGML. They threw out 90% of the standard to get at that golden 10% that preserved SGML's essence, gave it a cooler name, and XML was born. Everyone was happy and had their place in the sun (except those HyTime guys).--Matthew Gertner
In certain communities, RDF is a good buzzword, apparently because metadata people can't agree on much else; RDF lets them disagree. We try our best to take advantage of the buzzwordness, but shield users from the actual RDF whenever possible.--Eric Hellman on the XML-L mailing list
The fact that variables can't vary in XSLT means that it is often easier to use recursion than for-loops, with the notable exception (I believe) of situations where you need alphabetic sorting.--Francis Norton on the xsl mailing list
The whole SGML community, after disparaging HTML for a while, finally caught on to what a tremendous business opportunity it was for them. While HTML did provide important immediate benefits, many realized that there were significant issues in how this information was to be managed in the long-term. HTML was not an ideal way to manage hundreds of pages of documents. We needed more of what SGML provided -- first of all, an extensible tagset. But some thought we didn't need all of what made SGML hard to work with. Following this thread led to XML, a mid-point between HTML and SGML led to XML.--Dale Dougherty
ID and IDREF(S) allow general directed graphs to be encoded as hierarchical documents.--John Cowan on the xml-dev mailing list
I could take (the Net) for granted, because of the clean design of what was underneath. All these people had done all that work. The important thing was that the Internet was designed so you could use it for anything. And that's also important about the Web. We should keep the design very clean so we can build anything on top of the Web.--Tim Berners-Lee
The purpose of typography is to best communicate the information contained in content. When paper was the only medium, designing the best possible representation in a fixed way was the best way to communicate the information. But now that information can be presented to myriad users in myriad ways, the best way to communicate the information is to describe a series of optimal constraints, not to focus on the best possible picture of the information.--Christopher R. Maden on the xsl mailing list
Today, nobody writes HTML that conforms to any particular standard, they write HTML that works in Internet Explorer and Netscape Navigator. If the XSL processor in some hypothetical Internet Explorer 7.0 contains a bug that requires that the fo's be ordered in a certain way, then people will write XSLT that transforms their XML documents into XSL fo's that work around that bug, no matter that the XSL spec says that the fo's don't have to be ordered in that way.--Steve Schafer on the xsl mailing list
It looks to me like the primary reasons for HyTime's lack of mass acceptance, to date, have been the lack of a general toolkit implementation, and public ignorance of the problem space in which HyTime offers solutions. Both of these problems are rapidly being resolved now, so I think the death-knell of HyTime is still being rung prematurely. People ridiculed HyTime's complexity, but then they had to come up with a way to implement extended linking. And that led to the realization that the DOM has no foundation -- it was not, in fact, an object model. And that led to the XML infoset activity. Now W3C is at the same point the creators of HyTime were, after they discovered the need for an SGML Property Set, but before they understood that the ability to express the SGML Property Set depended on yet another needed invention, an invention that turned out to be the grove paradigm. Slowly but inexorably, the XML world is re-inventing HyTime.--Steven R. Newcomb on the xml-dev mailing list
An editor which simply displays a tree diagram of XML content does not buy you much.--Tyler Baker on the xml-dev mailing list
HyTime throws out too many hard-to-understand concepts at once. The linking model is hard to understand. Groves are hard to understand. Architectural forms are hard to understand. Throwing these together in a 1000 page document is not a recipe for commercial success. It seems to be an unambiguously good thing that these concepts are now being split into bite-sized chunks in the process of transfering them into an XML context.--Matthew Gertner on the xml-dev mailing list
*My* take on the situation is that we see the difference between people `trading down' from book typesetting systems (eg Arbortext, Framemaker, 3B2, LaTeX), and people `trading up' from Netscape. I badly want a standard formatting language to typeset my XML documents, but compromising on page formatting features is simply not an option. If I was currently using HTML + Netscape, and was offered something that does better, I'd no doubt accept it gladly. But I am not in that situation; to me, in my book-typesetting persona (I have others), XSL FO as it is proposed is interesting, but not a real option.--Sebastian Rahtz on the xsl-list mailing list
--Tim O'ReillyMicrosoft actually was ahead of the curve in realizing the power of online multimedia. In 1994, when the Web started to take off, Microsoft's CD-ROM products like Encarta, their online encyclopedia, and Cinemania, their online movie reference, were ahead of the Web in providing online hyperlinked documents with rich multimedia capabilities. Microsoft even realized that it was important to provide information resources via online networks.
There was only one problem with Microsoft's vision of the Microsoft Network: barriers to entry were high. Publishers were expected to use proprietary Microsoft tools, to apply and be approved by Microsoft, and to pay to play. By contrast, anyone could start a web site. The software you needed was free. The specifications for creating documents and dynamic content were simple, open, and clearly documented.
not all great experiments succeed, and groves/DSSSL/Hytime have failed in the marketplace for good reasons, and that should guide us in building XML.--Gregg Reynolds on the xml-dev mailing list
Part of the zen of SGML (inherited by XML) is that standards exist to protect end users from programmers.--Paul Prescod on the XML-DEV mailing list
Not only is the adoption rate of new browsers not improving, it may be declining. Statistics are notoriously unreliable in this arena, but spend enough time on the Web and talking to users, and you get a sense that there isn't a lot of penetration of IE 5.0 and Netscape 4.6 out there. Too many people are still using 3.0 (and, heaven help us, even earlier) browsers. Only forcing new browsers down users' throats by bundling them with new OSs (we know who has the edge there, don't we, boys and girls?) and including them on newly shipping machines (ditto) results in any substantial movement along upgrade paths.--Dan Shafer
I have to suggest that there is a difference between HTML 4.0 (the formally specified, designed-by-experts version) and HTML (the one that's used in real life) - and do what we can to improve HTML, the real-life version. If XHTML wants to design changes, they'd do well do consider what the public will actually adopt, not try to force an odd notion of formal grammars mapping to namespaces upon that public.--Simon St.Laurent on the XML-DEV mailing list
XML serves a different purpose from a database management system. In fact, they often server complementary purposes. XML is a way to represent data that makes it easier to, in Paul Prescod's words, "move the information between processes separated by space, time or incompatibility." It's not about storing data for efficient retrieval, which is the purpose of a DBMS.--Robert DuCharme on the XML-L mailing list
Although popular software these days can't necessarily read the popular data formats from 15 years ago, the fact that many of those were binary and XML is self-describing text means that if extra steps are necessary to read in XML 1.0 data into an application developed in the year 2014, they should be trivial steps. By then, people will say "it's just a three-line Thnad (or Vroo or Zatz or Floob) script, or 5 lines of Python if you're old and gray and prefer those cranky old-fashioned scripting languages!"--Robert DuCharme XML-Lmailing list
The namespace spec had its work cut out to find some form of face-saving phraseology to rubberstamp a syntactic device that external parties had already decided that they kinda sorta liked, and therefore have it they would, and therefore have it they must.--Arjun Ray on the xml-dev mailing list
You can't swing an expired IDE drive at your local magazine rack without hitting a glossy publication extolling the virtues of XML. XML is this year's Java. Everyone is writing about it and depopulating whole rainforests with impressive coffee table sized books on the intricacies of XML, but almost no-one is actually talking about the practical aspects of using XML. Or even why you might want to use it.--Zac Belado
the namespace editors disagree on what the namespace specification "really means". It seems to me that the specification has proven too general, too flexible and has been shown to depend too much on mystical shared understanding of a "reasonable namespace" that turns out not to be shared.--Paul Prescod on the XML-DEV mailing list
The tumult and chaos of the browser wars seem to have numbed many developers into accepting the W3C's status uncritically, but I don't know how long that acceptance will last. Opening participation significantly would seem to give the W3C a lot more legitimacy heading into the future, although it would certainly increase the pressure for accountability by the W3C.--Simon St.Laurent on the XML-Dev mailing list
Today, the W3C process is not an open process, and it's not a system that ordinary people can get involved in and "change from within". The W3C process is the (rather unwieldy) tool of the dominant software vendors, balanced against the flawed personal vision and absolute veto power of its Director, a single human being named Tim Berners-Lee. The institutional structure of the W3C was not designed to serve the public interest, and even the most cursory analysis reveals that there is little reason to expect it to create standards that are optimized for serving the public interest. Everyone involved primarily serves special interests, and the public is neither invited nor involved in any meaningfully institutionalized fashion. The W3C is a software vendor consortium; people (including W3C members) who believe otherwise are deluding themselves.--Steven R. Newcomb on the xml-dev mailing list
--David Megginson on the XML-DEV mailing listI agree that there's little point in writing XML parsers any more, but the fact that they were trivially easy to write helped massively with their adoption -- the irony of the simplicity (or worse-is-better) principle is that, in the end, it conceals itself: now that there is lots of XML parsing software, it doesn't matter whether XML is simple or complex (because the complexity is hidden from the application); but that software wouldn't exist if XML hadn't been so simple in the first place.
That's why, for example, we never ended up with a single Java-based SGML parser (even though it would be just as easy to process SGML if we did have the libraries).
Open source software is about a _lot_ more than "software availability/testing" - it's providing serious competition to many of the members of the W3C with 'real' software. Unfortunately, it's a competitor that's pretty well locked out of the process, because of the cost, secrecy, and structure of the W3C.--Simon St.Laurent on the xml-dev mailing list
XML 1.0 was a guerilla project by a bunch of people who'd known each other for years and very few of whom had management that understood what it was really about.--Tim Bray on the xml-dev list
The working group never intended XHTML 1.0 to be extensible. It is a bridge. It is there to help HTML document authors make the transition to XML in a way that is backward compatible with existing browsers. Nothing more. Don't read too much into it. It will drive you crazy.--Shane P. McCarron on the xml-dev mailing list
It's difficult for a designer to anticipate all of the expected uses of a Namespace -- for example, it's quite reasonable to specify what *browsers* should do when the find an unknown element type in an HTML document, but should the same behaviour apply to (for example) to search engines and text repositories? I'd be very upset if an XML repository automatically stripped out unrecognized element boundaries when I checked in an XHTML document.--David Megginson on the XML-DEV mailing list
--Simon St.Laurent on the XML-Dev Mailing list mailing listThe W3C's closed process is unfortunate. However, that doesn't mean that outsiders should read drafts under the assumption that 'future drafts will fix everything', which appeared to be the drift of your earlier comments.
If the W3C wants meaningful public comments, it has to be prepared to deal with comments from those of use who don't have access the full set of background information locked away in members-only areas. Saying 'trust us' isn't enough. It requires explanation of underlying assumptions, at least.
In fact, if/when the true potential of XML is realized, the web will be a minor player in that. SGML is/was not the native language of the web. HTML was derived for that purpose. XML is/has the potential to be used for much more then internet publishing of information. The web has benefited from many technological contributions, many of which predate the web by a couple of decades, and most of the technology that goes into it is not at all web specific (connection oriented stream TCP communications, MIME, request/response, etc.). To think that XML is a web-only or even a web-mostly language is to miss the boat so far as to not even notice the ripples.--Blair Murri on the xml-dev mailing list
The W3C approach is probably optimal for the design and development of sample advanced technology, especially when it addresses problems two or three years ahead of current products, while the IETF approach is far better for actual standardization--John C. Klensin, distinguished engineering fellow at MCI
...there are no Open Source validating processors supporting the SAX API--David Brownell
The structure of the W3C didn't lend itself to quite the degree of freedom to contribute that the IETF does. We found it difficult to get points across and to influence what was happening.--Vint Cerf
The more I watch things at W3C, the more I feel that the Web should be driven instead by a standards organization with public accountability. Being accountable to vendors who have vested interests in bloatware (as key parts of new barriers to entry) isn't the right model.--David Brownell on the xml-dev mailing list
We were never completely comfortable with W3C acting as a standards body, with its decision model based ultimately on the personal preferences of the director. We've tended to prefer Internet standards work to be done in bodies that more clearly use an open consensus process rather than in limited-membership consortia of any sort, including W3C.--John C. Klensin, distinguished engineering fellow at MCI
For us, interoperability is crucial. Without the W3C, we would not have standardization of protocols, and individual vendors would dominate with their proprietary formats.--Ann Bassetti, Web products manager at Boeing
Because it's a PDF file you can only look at it, you can't sort through it, or search for names. You can't even add up the totals that have been given. This is a very sophisticated operation. They know very well the difference between a PDF file and a text file you can import into a database, and they've chosen to take this route.--Larry Makinson, Center for Responsive Politics
The more I watch things at W3C, the more I feel that the Web should be driven instead by a standards organization with public accountability. Being accountable to vendors who have vested interests in bloatware (as key parts of new barriers to entry) isn't the right model.--David Brownell on the xml-dev mailing list
--David Brownell on the xml-dev mailing listThe XML 1.0 specification defines what XML processors must do. It explicitly requires accepting documents with constructs that do not conform to what the XML namespaces spec wants.
That means that XML 1.0 processors must ignore namespaces. Whatever component of a system pays attention to namespaces isn't going to be called an XML 1.0 processor.
IETF defines working designs which have to be proved by implementation. Very practical. The difference is, the IETF is not a consortium. Anyone can sign up for the debate and get the details. The argument here is about W3C processes that result in decisions not justified by rationale. It creates an atmosphere of suspicion, mistrust, and fear.--Len Bullard on the xml-dev mailing list
Simple wins, complex loses. I remember avidly reading the literature about the complex experimental hypertext systems of the late '80s and early '90s, but they lost and HTML won. Now, maybe HTML was a little too far on the stupid side, but not so far that it couldn't mop the floor with the competition.--David Megginson on the xml-dev mailing list
SAX 1.0 is widely implemented because it is simple: everyone has one or two more things they'd like to see in SAX, but since everyone's one or two more things are different, SAX would have become quite complex if it had tried to accomodate all of them, and probably no one would have implemented it.--David Megginson on the xml-dev mailing list
One doesn't need to be a conspiracy theorist to identify real flaws in how the W3C does its business. As a steward of an international resource, it should be as accountable to customers as to vendors.--David Brownell on the xml-dev mailing list
I truly question the value of namespaces without association with some DTD or schema. They purport to resolve ambiguity, but only do it syntically - there is no identification of what each element is a member of. The problem is punted off to the application, which somehow must assign meaning to the URIs. In doing so, validation is thrown to the wind - we might as well redefine XML to only parse well-formed documents and toss the DTD baggage.--Marc McDonald on the xml-dev mailing list
One problem is that ISO and the W3C are both doing standardization backwards these days -- the idea of standardization is traditionally to align current practice (we all have railroads, so let's use the same rail gauge), not to invent new practice (hey, maybe someone will invent railroads in fifty years -- let's standardize the gauge for all four rails that we think they might need, and the width of the brick path for the horse in the middle while we're at it [at which point a long debate ensues about whether horses or donkeys will pull trains]).--David Megginson on the XML-Dev Mailing list mailing list
no one has really demonstrated a 'native XML database' that isn't a light XML processing layer on top of something else - whether it's a plain old filesystem (add indexing, and voila!), an RDBMS (or object-relational DBMS), or an object database.--Simon St.Laurent on the XML-L mailing list
it's a great pity that online sources aren't more elegantly citable. At the moment, there's a whole string of messes and slashes, and in book publishing, elegance and practicality need to be considered.--John Simpson, Chief Editor of the Oxford English Dictionary
Seems to me that Microsoft Windows and Microsoft Office rate at least the same amount of inspection as the plans and parts for a nuclear reactor, given the productivity impact any failure can cause on a very large number of people.--Simon St.Laurent on the "Computer Book Publishing" mailing list
Critical infrastructure software works best when it is free software. Part of the reason free software works better is that only when the source is available, hackable, and freely redistributable is there any serious chance of honest cooperation and honest competition.--Jay Sulzberger on the WWWAC mailing list
Beating on your chest and saying, "We are taking on Microsoft" is one sure way to get their attention. No one who has done that has survived.--David Smith, Gartner Group
the Internet should not be registering and blessing every corporate character set (e.g. PC code page) that comes along, and in this case (UTF-16), it should not be registering every corporate variation (e.g. Intel vs Sun byte order), because this serves no useful purpose. Specify one and only form for UTF-16 on the wire and be done with it instead of "some people do it this way but others like to do it that way.." That's not a standard.--Frank da Cruz on the Unicode mailing list
XML is your data layer. XSL is your presentation layer. The two are merged together to present your GUI on either client or server-side. As far as I'm concerned, it's an ideal application design. The designers produce the stylesheets and present the layout in whatever fashion they want. The developers code all of their heavy business logic and DB code to produce an XML structure that merges with the stylesheet. Designers don't have to code their layout between blocks of program code, and developers don't have to try and put together a layout of a GUI when that's not their thing. And just think, wouldn't it be nice if all of these content management systems like Vignette, FutureTense, and INSO all settled on XSL as their template language because its soon to be an open W3C standard instead of proprietary stuff like TCL/TK, etc that would have to be completely revamped if god forbid, you switched from that content managment system?--B.J. Weschke on the WWWAC mailing list
Not to rekindle the Word v. FrameMaker debate, but one of the things I remember liking about Maker was that it made it difficult to do stupid things. Of course, one of Word's selling points is that it makes it extremely easy to do stupid things.--Steven Champeon on the "Computer Book Publishing" mailing list
--Greg DeKoenigsberg on the WWWAC mailing listInformation architects love XML and paint bright, shiny pictures of what the world will look like in XML, because it conforms conceptually to what people think data modeling *should* be -- just like its well-beloved daddy, SGML. The other side, the actual presentation side, is *inherently* ugly, because it bears the burden of translating nice, pretty XML into usable information for *all* client implementations -- which is precisely why SGML ran aground. It was great information design, but who had time to learn to implement it in the real world?
Obviously, the stakes are high enough now, and the potential user base sufficiently large, to make XML inevitable in a way that SGML never could have been. But the fundamental problems of making well-marked data *useful* still exist.
Love them, loathe them, or anything in-between, Microsoft is a fact of life, and Microsoft's browser has at least half the market. Now that Netscape has committed to delivering full support for Cascading Style Sheets Level-1, HTML 4.0, DOM 1.0 and XML 1.0 in Navigator 5 and appears to be on the verge of doing so it's time to get Microsoft to commit to doing the same.--The Web Standards Project
SMIL Boston is an important improvement over the SMIL 1.0 specification thanks to its modular design and its integration with other XML languages. SMIL Boston provides content designers a first class language for the production of compelling multimedia presentations over the Web. The support of animations and event-based timing is a radical enhancement of the language functionality.--Jean-Pierre Verjus, Director, INRIA Rhone-Alpes
XML is not a formatting language.--Miloslav Nic on the XML-L mailing list
The IE5 XSL processor is a buggy implementation of the Dec 1998 XSL draft, but it is your only option right now for client-side processing. I would strongly suggest updating your code to work with the July 9 XSLT and XPath drafts, and using the standalone version of XT for testing, because sooner or later you must unlearn the bad habits you have acquired in trying to make your code work with IE5.--Mike Brown on the XSL mailing list
XML lets you store your documents independently of a particular surface presentation. Document presentation can be tailored to specific contexts, media and users. Documents and parts of documents can be reused and repurposed. Information can be drawn from a variety of sources without those sources needing to know final presentation. Structure can be constrained and documents validated to such constraints. Parts of documents can more easily be integrated with database systems.--James Tauber on the XML-L mailing list
There is no technical, cultural or ergonomical justification for the many keyboard layouts that we have at the moment. For instance, it is easily conceivable to come up with a single highly practical layout for the West European and Panamerican Market.--Markus Kuhn on the Unicode mailing list
compare a HTML ebook and a PDF ebook. It's like sky and earth. The HTML is in clear, good for eye viewing especially on the energy sapping computer screen. And the PDF format make you feel like vomit after reading for a while and it's realy not worth to strain your eyes; you'll prefer to print it on paper and then only do the reading.--Daniel Tay on the "Computer Book Publishing" mailing list
--Mikki Barry, president of the Domain Name Rights CoalitionThe US Postal Service does not have the expertise (either technical or policy wise) to run a top level domain of the potential popularity of the .US. Physical mail is not analogous to electronic communication.
The USPS does not have the infrastructure, or the engineering talent necessary to build and maintain a database, generate domain name service, create a directory structure, or run a root zone.
Saying that RDF is an alternative to XML is like saying that HTTP is an alternative to IP.--David Megginson on the xml-dev mailing list
Metcalfe's Law provides much of the explanation of the success of the Web relative to earlier hypertext systems like HyperCard, Intermedia, and NoteCards. They were all much better than the Web and had features ten years ago that we are still sorely missing on the Web. But the Web was universal and the other systems were proprietary. You know who won.--Jakob Nielsen
Frame is dying (I hate to admit this, but its true). It's still big in documentatation, but not anywhere else. Use Word for writing, Quark for layout. Only use Latex if you're a mathematics professor, and only use SGML if you have 150,000,000 pages of mil spec documentation to keep maintained for GM.--Laura Lemay on the "Computer Book Publishing" mailing list
Metcalfe's Law provides much of the explanation of the success of the Web relative to earlier hypertext systems like HyperCard, Intermedia, and NoteCards. They were all much better than the Web and had features ten years ago that we are still sorely missing on the Web. But the Web was universal and the other systems were proprietary. You know who won.--Jakob Nielsen
Parsed character data is text that can be analyzed by the parser for entities and markup and processed. Parsed character data should not contain any &, <, or > characters; these need to be represented by the & < and > entities, respectively. Mark-up entities will be recognized within the text.--Chris Maden on the xml-dev mailing list
DTDs are far superior to the incredibly bloated form the XML Schema seems to be turning into. Although you *do* have to go learn a little BNF to understand them, I challenge anyone to translate an XML Schema to a mental model without the aid of either paper or machine. It'd take less time to just learn BNF.--Joshua E. Smith on the xml-dev mailing list
We want to create a run time environment that supports Java, C, Visual Basic, and a new set of languages that are coming along that really deal with this XML schema issue, that really take data binding, which has been really the worst thing in development, where you have to go and find the data, find the format of the data, pull it in, transform it, and try to get it back out. Most code is written around that. And that impedance match, there are ways of taking some of the ideas that were around object database language couplings, and really getting those into the mainstream.--Bill Gates
If I have two co-operating applications of different endianness machines writing to the same log where one machine is big endian and one little endian, then the application needs to care about endianness when it's writing utf-16 but not so with utf-8.--Gianni Mariani on the Unicode mailing list
When XMI came out, I had just been studying up on UML, and I thought "Cool! I'll print out the DTD so that I can look it over on the subway ride home!" When I saw how big the XMI DTD was, I decided not to print it out--I prefer not to spend that much time in the subway.--Robert DuCharme on the xml-dev mailing list
utf-8 is beautiful. While utf-16 is an abortion and ucs4 and utf-16 are stateful, utf-8 is simple, easy to work with and is easy to upgrade to from where most systems are today.--Gianni Mariani on the Unicode mailing list
There are two main APIs for applications that process XML documents - DOM (Document Object Model) and SAX (Simple API to XML). DOM involves constructing a tree of Nodes from the XML document, and then using a simple API to walk the tree and extract information from it. SAX is an event based API designed to fire user-written code as elements etc. are encountered in the parsing of the document. SAX is very useful when the document may be too large to fit into memory.--Chris Harris on the xml-dev mailing list
A DTD is *rarely* enough to ensure the required conformance to any application of XML. That's one of the benefits of more expressive schema languages.--James Tauber on the XML-L mailing list
DTDs have at least another few years of active use in them. I suspect (though it isn't certain) that while they're less powerful than schemas, they involve less processing as well. For some situations, even 'non-legacy' ones, that may insure a very long life.--Simon St.Laurent on the xml-dev mailing list
we have not submitted XML-RPC to W3C. If they loosen up their membership rules and allow individuals to be involved without paying a fee we'll be happy to work with them. Some compromise is needed. Right now all the implementations come from either small companies or individuals, they're doing excellent work, and it would be stupid of us to hand this over to W3C without a change in their way of working. I understand that they need money to operate, but also am aware that this cuts them off from excellent work being done in the open source and scripting worlds.--Dave Winer on the xml-dev mailing list
As a representative for a company that can not afford to 'join' the w3c, I depressingly have to live behind the times at all times. I am sure they are well intentioned, but waiting for corporate representatives to lob a finished standard over the net well after they have established their corporate strategies around their advanced knowledge sure does not seem in the spirit of the whole thing does it?--Erik Freed on the xml-dev mailing list
--Frank da Cruz on the Unicode mailing listAnd how do I read your PDF if I don't have a PDF reader? (Don't say "get one" -- I'm reading your mail on a DOS PC or a PDP-11, or a Cray supercomputer.) How do I read TeX if I don't have the software? How do I read PostScript if I don't have a PostScript printer or rendering engine. But the crucial point is:
How will I read your PDF file 200 years from now, when PDF itself has been consigned to the "legacy" trashheap for the past 195 years?
Modern mathematical notation gets its variety of symbols primarily from the Latin and Greek alphabet, combined with numerous combining characters (TeX certainly had a lot of influence here). Some mathematics professors have made it their mission to impress their students with more and more exotic symbols, but this should more be seen as an eccentricity than something that helps the reader or that the publishing industry should support with a lot of energy.--Markus Kuhn on the Unicode mailing list
--Jay Sulzberger on the wwwac mailing listI find that almost every time the words "business model" appear in an article, it is an indication that the author does not grasp the subtle, and as I always add, shocking, theory of the free market. Here is the canonical use of the phrase "business model":
But I do not see how I can make money with a free software business model. Why someone else might make make a better product than mine and sell it at a lower price. Then I cannot make money with my product!
To which the answer is:
Yes. You are on the verge of getting the concept "free market".
Productivity gains from XML are in part achieved by being able to represent many forms of information in a consistent way. For example, if you develop a tool that can transform data of one type, it can also be used to transform information of thousands of other types. An editor that can edit data of one type - and enforce data types, enumerated values, and so on - can be used to edit data of another type. Tools that transfer, store, compress, parse, validate, and so on, data of one type can be used to do the same to data of other types.--Mark Birbeck on the xml-dev@ic.ac.uk mailing list
I can't count how many speeches and tutorials I've seen that started along the lines of: "XML schemas are great because they don't require you to learn a new syntax" and then continued along the lines of: "Now let's spend an hour talking about the syntax." A syntax is built on top of XML and still be a unique syntax that must be learned. RDF has a very complex syntax despite the fact that it is "just XML."--Paul Prescod on the xml-dev@ic.ac.uk mailing list
software is largely a service industry operating under the persistent but unfounded delusion that it is a manufacturing industry.--Eric Raymond
I think if you crossed greenspun and that david siegel guy and took out the splash pages and exit tunnels, you'd actually have a visually appealing, navigable website with content.--Robert Gruber on the wwwac@lists.wwwac.org mailing list
Any new architecture can only succeed if it is reachable in small increments from the current architectures. At the semantic level, we do not have current, widely deployed architectures using Dublin Core, RDF, database schemas, metadata registries, nor ontology systems. The reality is that currently deployed architectures for shared semantics are basically bibliographic models and generalizations of bibliographic models. These catalog systems are well understood by billions(?) of users. On a global basis, investment in public sector bibliographic systems and the annual revenues from commercial information services must surely exceed tens of billions of dollars.--Eliot Christian on the rdf-dev mailing list
Don't use DTDs and Namespaces within 100 metres of each other, the combination is explosive.--Michael Kay on the xml-dev mailing list
If I learned nothing from CALS, SGML, HyTime, etc., I hope I learned that our tools are just tools to be wielded by hands. Every DTD written so far is being modified by human hands, or languishes somewhere in a thousand page tome. Every mudbrick house needs patching or is buried in sand somewhere waiting for Ozymandias to return. Still, better than the caves.--Len Bullard on the xml-dev mailing list
--Paul Prescod on the xml-dev mailing listWhen you sit down to implement an XML system the last thing you think about is CSS, XSL, or the DOM. The first thing you think about is the needs of the data. If you have any volume of textual data, you will probably not get another opportunity to re-encode all of it until CSS and XSL are historical artifacts, Microsoft is a division of Red Hat, and the Web has more users than the telephone.
So you *must* concentrate on the needs of the data. You must make richly semantic markup that captures the structure of the data. And you must have human authors start to add this semantic markup to the data as soon as possible. You must minimize the cost of this markup effort.
I would genuinely like the W3C to sit down and ask if XSL is _good for the Web_. Not good for the XSL community, not good for the DSSSL community, but whether it is good for _the Web_. That is, after all, their job. If, after some serious, preferably public, pondering, they conclude that it's good for the Web, then fine. XSL can become a recommendation. If they decide that it's not good for the Web, they'd better drop it. XSL can move on to a different organization if necessary - I don't see it dying any time soon.--Simon St.Laurent on the xml-dev mailing list
In standards you have no rights, except maybe the right to complain. You still have plenty of freedom to develop systems - it just may not be the 'standard' of your choice. Like it or not, if Tim Berners-Lee decides he doesn't like XSL (seems unlikely but possible to me), it won't become a W3C recommendation. Period.--Simon St.Laurent on the "XML Developers' List" mailing list
I don't know why people upgrade or switch to IE from Navigator, but I suspect that the trade press, being well-supported in its mission by extensive Microsoft marketing and press relations, has the choice of becoming well-informed and remaining impartial, or taking MS press kits at face value, and parroting their claims in print.--Steven Champeon on the Computer Book Publishing mailing list
HTML's gross simplification of hyperlinks to unidirectional traversal was an important advance (perhaps even _the_ advance) that made the Web possible.--Simon St.Laurent on the xlxp-dev mailing list
The real question is what kind of market penetration XSL will have. I'll probably use it a lot for my personal work, because I'm used to that kind of programming and quite enjoy it; you may be right, though, that it's too far in front of the Web community -- even a brain-dead-simple thing like CSS1 is still having trouble catching on after several years, and XSL is at least two orders of magnitude more complex than CSS.--David Megginson on the xml-dev mailing list
I firmly believe you can teach almost anyone basic XSL for displaying, sorting, and filtering XML files. I do not think the same is true for VBA, VB, JScript, VBScript, Python, OmniMark, or Supralapsarianism.--Eric E. Cohen on the XSL mailing list
I wouldn't be laughing so much if they spent as much effort on the spec as they apparently did spend on all those press releases. BizTalk spec version 0.8 is so full of nothing that I wonder if the version 0.7 was on a piece of napkin.--Don Park on the xml-dev mailing list
The law of networking applies very clearly to standards, where the value of a standard (big 'S' or little 's') is the square of the number of users, so that a standard with 100 times as many implementors, for example, is 10,000 times more useful.--David Megginson on the xlxp-dev mailing list
--David Megginson on the xml-dev mailing listSAX is a very widely-implemented XML API, but it has no owner, no organization promoting it, no press releases, no copyright or terms of usage (it's truly public domain), and no formalized process for development or maintenance. (It also has no proper language-independent spec, but I consider that a bug rather than a feature, and one for which I take personal responsibility.)
Essentially, megginson.com donates a little Web space, xml-dev donates a lot of discussion space, and many companies and individuals donate the time from the people who help to discuss and develop SAX. In this regard, SAX is a lot like Linux, except that Linux is GPL'ed and Linus now holds the trademark on the Linux name after a nasty squabble with someone who tried to cash in; SAX is explicitly in the public domain and not copyrighted, copylefted, or trademarked in any way, and relies exclusively on peer-pressure and user demand to ensure conformance.
I don't see FOs as a particular challenge, but I know that some folk, particularly those with a very low-level, bit twiddling view, tend to find them scary. But then, thats probably because CS students are taught all about complier technology but aren't taught all about constraint-based geometric systems.--Chris Lilley on the xml-dev mailing list
I lived in the USDoD standards world for a long, long time, and I know that their process is completely incapable of producing anything with any value whatsoever. I have also dabbled in the IETF standards world, and have been uniformly impressed at how their process lets the cream rise to the top.--Joshua E. Smith on the xml-dev mailing list
Never underestimate the stupidity of some of the people we have to deal with.-- William A. Reinsch, U.S. Under Secretary of Commerce for the Bureau of Export Administration
XSL is DSSSL in sheep's clothing, an SGML transformation and formatting language that an ISO committee worked on for eight years, producing finally a language specification and an utterly heroic and breathtaking but incomplete implementation by James Clark that years of real world experience has shown to be incomprehensible and unusable to nearly everyone in the document transformation and formatting business. XSL is DSSSL, it is 100% DSSSL concepts and processing model, with some syntax changes and some attempt to merge with CSS.--Michael Leventhal
the most useless language features are those where the interpretation is completely up to the software. The most useful features are those that either have a well defined underlying data model *or* a well-defined processing model. XML notations are useless not because they are a bad idea but because nobody has defined the data model or processing implications of the associated system identifier.--Paul Prescod on the xml-dev mailing list
Though I believe almost nothing of the Java hype, I believe most of the XML hype.--Jay Sulzberger on the wwwac mailing list
Using another's trademark in one's meta tags is much like posting a sign with another's trademark in front of one's store--Judge Diarmuid F. O'Scannlain, U.S. Court of Appeals for the Ninth Circuit
Getting a solid XSL version 1.0 out in a reasonable time requires leaving lots of potentially useful and important things out of version 1.0.--James Clark on the xsl mailing list
-- Eric KiddNavigator 5 was a doomed product. Netscape wrote Navigator 0.9 from scratch in less than a year and never rewrote it. They moved from version to version, piling on features. Nobody ever said, "Stop! Lots of things have changed since 1.0. We need to redesign stuff!"
Things were starting to break down by Navigator 3.0. Netscape needed to step back, reassess their product, and take the time for a major overhaul. But they were under attack by Microsoft and didn't dare to skip a browser generation.
Navigator 5 was still based on the original codebase. It was full of #ifdefs, Win16 support, fake multithreading hacks and layers of useless code.
Actually, any object-oriented language should do a pretty good job of handling XML. The initial push among developers was in Java, but now there's excellent support growing in Python and Perl5 as well. Personally, I use Perl5 and Java, but I do sneak into JPython once and a while so that I have an interpreter to poke around in the Java classes. C++ is still badly under-represented in XML software, but you can always use the C Expat parser.--David Megginson on the xml-dev mailing list
My *hope* is that IE5 is MS playing with XSL as they did with IE4 and XML, and that IE6 will see a complete XSL solution..... this is assuming that XSL is finished by the time IE6 comes out. Like it or not, MS is that black body that creates a distortion in space-time.--Guy_Murphy on the xsl mailing list
The slow uptake speeds and the bugs and inconsistencies in advanced browser features constitute a cloud with a distinct silver lining: Recognizing that we are stuck with old technology for some time frees sites from being consumed by technology considerations and focuses them on content, customer service, and usability. Back to basics indeed: that's what sells since that's what users want--Jakob Nielsen
Netscape 5 will never get more users than Netscape 4-- Jakob Nielsen
--Paul Prescod on the xsl-list@mulberrytech.com mailing listThere is no boundary between formatting languages and semantic languages. As Håkon points out there is only a spectrum. More abstract -> more rendition-specific. Typically more abstraction is expensive and difficult but provides more flexibility in the long run.
HTML is at about the right point in the abstraction->rendition spectrum to allow braille, TTYs and other non-GUI interfaces to render information in a manner compatible with GUI interfaces. We cannot make a global language much more abstract HTML (though footnotes, headers, footers, etc. would be nice). We cannot make a multiple-media language much less abstract than HTML.
I can send a DocBook document and an XSL->HTML mapping across and that is more useful than sending just the HTML output. The important point is that we don't have to *agree* on the semantics in advance. We just have to agree on an extension mechanism (stylesheet, Java applet, JavaScript+DOM code, etc.)--Paul Prescod on the xsl-list@mulberrytech.com mailing list
DER and BER are examples of a philosophy of protocol development that's been largely discredited for mainstream applications: "bitstuffing". It was a design principle that bit efficiency was more important than time spent to encode or decode ... perhaps understandable for systems using X.25 networks where you more or less paid by the byte, but not on a LAN or even the Internet. Many folk think DER/BER should be the first to be put against the wall when the revolution (XML?) comes; they're that unpleasant to use.--David Brownell on the xml-dev mailing list
If you're at all serious about doing XML programming I would definitely recommend taking a look at Python. Chances are you'll find yourself using it for all sorts of other stuff as well. The only real weaknesses for XML work are a lack of true garbage collection and Unicode support. The latter is being fixed, though. Speed may also be an issue, but you can almost always fix that by redoing the critical parts in C.--Lars Marius Garshol on the xsl mailing list
Let me jump in to point out that there are many conceptual similarities between XSL patterns and XPointer. Also, the sky is blue. :-)--Micah Dubinko on the xsl mailing list
Once you've tasted XLink's Chunky Monkey, it's hard to reconcile yourself to HTML's vanilla.--John E. Simpson on the xsl-list@mulberrytech.com mailing list
One of every 500 million people, give or take a million, is likely to ever run into even ONE Chinese character that is not already among the approximately 25,000 of them encoded in Unicode today. Most learned people do not know even 6,000 characters. You could enumerate Chinese characters forever and never reach the end -- I'll grant that. However, you could also *READ* Chinese forever and never run into any of the supposed 50,000 or 80,000 or whatever thousand there are. And even if you did run into one, chance are you could look it up only to find that no dictionary exists which could tell you what it means.--Rick McGowan on the Unicode mailing list
SGML is complex and intractable. Software that works with it is scarce and often expensive, and too often doesn't work very well. Just because a giant telco firm can muster the personnel to deal with SGML doesn't make it a particularly elegant solution, except by way of comparison with approaches that use non-standard or presentation-focused languages.--Richard Goerwitz on the xml-dev mailing list
the XML standard was written with SGML prac- tice in mind. If you know what SGML parsers typically do in such situ- ations, you know immediately what the XML spec editors really meant to say. The question of whether what they actually _did_ say will work in practice is another matter.--Richard L. Goerwitz on the xml-dev mailing list
The cost of poor navigation and lack of design standards is even higher: at least ten million dollars per year in lost employee productivity for a company with 10,000 employees. World-wide the cost of bad intranet usability will grow to about $100 billion by the year 2001 unless better navigation systems are built and much stricter internal design standards enforced.--Jakob Nielsen
if I could give one piece of advice to anyone writing computer books - or instructions of any kind - it's to expunge all occurrences of the the word "simply," as in "You simply need to...," and replace each with a full paragraph of explanation.--Jan Axelson on the Computer Book Publishing mailing list
Processing instructions are useful because they are invisible to the validator. This gives us a rule of thumb of where how they should be used: when a structural instruction should be invisible to the validator. Examples include editor-specific meta-information that is invariant across document types, formatter-specific information (e.g. page breaks) that is invariant across document types and so forth.--Paul Prescod on the xsl mailing list
If the browser is so large that it requires more than, say, 30 minutes at 28.8 kbps to download, most users won't upgrade. That leaves Web builders in a place where they have to keep designing for last-generation browsers. I can't for the life of me figure out why Netscape and Microsoft keep doing this to themselves and us, but they do.--Dan Shafer
W3C specifications are not treated by vendors as standards which should be strictly adhered to: they are treated as sources of APIs which can be embraced and extended, or partially implemented. W3C does not have the authority to check, demand or expect conformance: either moral or legal. A W3C spec needs all the help it can get to ensure complete implementation: being an ISO standard helps.--Rick Jelliffe on the xml-dev mailing list
Everything that we learned in SGML is there in XML, and all the careful thought and person years of work from the Charles Goldfarb and the other members of the ISO subcommittees is the fundamental reason for XML's success. Essentially, the W3C just did what ISO was too slow at doing, and gave SGML a proper 12-year review; without the ISO baggage and the emotional attachment to the minutiae of ISO 8879:1986 esoterica, the W3C's SGML ERB cum XML WG was able to wield a sharp knife and cut away a lot of fat (though still not all of it).--David Megginson on the xml-dev mailing list
...standardization often has legal tangles. When engineers practice law, you get poor law. When lawyers engineer, airplanes fall out of the sky.--Len Bullard on the xml-dev mailing list
While DTDs are extremely useful and should be constructed as (a small) part of any serious language-design effort, they are in some cases unnecessary (for validation, full-text indexing, and lots of other things) and in other cases insufficient - DTD validation never comes close to real business-logic validation. I am near-schizophrenic these days, running around telling people that yes, they should use DTDs, and simultaneously warning them that there are situations where they fail to be either necessary or sufficient; the kind of mystico- religious attitude above does not help.--Tim Bray on the xml-dev mailing list
...the term 'meta' has been used too much recently. Some people even say there is something like meta of meta. When I began working on XML I was indeed confused with such expressions. I haven't had enough time to work on it but one way to find out what meta is would probably be to think about languages in general. What is a language? What is common between English, HTML, calculus? Also, there is the concept of notation. Basically a DTD is a certain notation of rules which you could as well be expressed in English (what would be much more difficult). Maybe linguists could help on this issue. To get back to meta meta: I think there are always multiple levels of information like object - class - class model or document instance - DTD - XML and to describe such relationships you don't even need the term meta. Inheritance, subset or whatever does it. Meta seems to be a word people use when they don't know exactly what these relationships are. But it sounds good.--Laurent Rutz on the XML-L mailing list
--Simon St.Laurent on the xml-dev mailing listOne of these days I'd really love to stop talking about what is and isn't XML, though I know it's fun, and start talking about what we can do with XML and XML-like structures, whether they are SAX event flows, DOM trees, or binary formats that build on an XML foundation.
We might even get some real work done - and it might even be fun.
It's amazing how much you can do with a search engine -- and it's also amazing how much you can't do with a search engine,--Tim Berners Lee
SGML is useful enough (over and above XML) in enough serious systems that it will not go away in the foreseeable future (which, admittedly, isn't that long in this industry).--Marcelo Cantos on the xml-dev mailing list
If we are looking to the animal kingdom for examples, the W3C is evolving like armadillos: timid animals that slowly go their own way. When faced with predators, they roll up into a ball and depend on their hard shell to protect them. It works until the predator is a speeding truck.--len bullard on the xml-dev mailing list
--David Megginson on the xml-dev mailing listThe biggest cost of Omnimark was not the purchase price (although $12K or more per seat [per annum, I think] would give most programmers reason to pause), but the cost of maintenance.
A lot of people know Perl and Java but almost no one knows Omnimark, and maintaining a system that relies heavily on scripts written in an esoteric and virtually-unknown language can be extremely difficult and painfully expensive, when it's possible at all.
Microsoft has fallen well short of its promises to support World Wide Web Consortium (W3C) standards, particularly in its selective, fragmentary, and unhelpful support for the Document Object Model (DOM). After running a copy of the final-release version of IE 5.0 for several days on dozens of sites and putting the product through literally hundreds of tests, I can say with some confidence that Web builders who have to design sites for multiple browsers are going to want to lay in a big stock of headache remedies if they add IE 5.0 to that list.--Dan Shafer
--Richard Goerwitz on the xml-dev mailing listIn reality, XML is functioning less like a "simplification," and more like a political move intended to facilitate changes that could never have been made to a mature standard like SGML.
This is actually a very old story that's been repeated many times over. (Just look at what's happened to LDAP. By the time we get all the PKI and ACL extensions in place, it's really not going to be very L.)
In the end, LDAP and XML may end up serving their constituencies better than their predecessors did. Or they may not. Frankly, with regard to XML, the jury is still out. It's not catching on nearly as fast as pre- dicted a year or two ago. And it's taking considerably more work to im- plement it than anybody ever envisioned.
Those of us who have done the work of writing XML processing software, and of making it work, have a right to say this.
The emperor may or may not have clothes.
If it weren't for the promise of backwards compatibility with SGML/HTML, XML could not have gathered the initial following that it did.--Richard L. Goerwitz on the xml-dev mailing list
A prime reason for staying with SGML is one of the usual arguments for adopting it in the first place, namely to have a platform-independent, vendor-neutral and stable medium for long term storage of mission-critical information (sorry about the corporate-speak). To me, at the moment XML is just another output format that SGML can easily filter to, but not one that I have any need to support as yet.--John Hanratty on the XML-L mailing list
the more granular/atomic the structure, the more flexibility downstream -- not just for XSL, but for querying and (yes) data interchange. For applications, it's easy to extract the structured data you want, even if apparently overly-nested, and ignore those portions of the structure that you don't... but harder (and needing more hard-coding and application-specific intelligence) to *add* structure to the source where it's no better than implicit.--John E. Simpson on the xsl-list mailing list
I have yet to hear of a single real application that needs element level prefix declarations. Not one! The PI was just fine for 99.99% of applications. The 0.01% should simply not use XML (or may need an additional layer, such as AF or a schema processor). Element declared namespaces is a solution in search of a problem. Unfortunately, namespaces have effectively killed DTD validation.--Charles Reitzel on the xml-dev mailing list
It's called XML, which stands for extensible mark-up language, and this is a progress report. When I first wrote about it some 15 months ago I was absolutely jazzed by the possibilities. I'm still optimistic, but reality is always more daunting than dreams.--Dan Gillmor
Because the DTD is not namespace aware, all it can deal with are the pre- fixes you declare (not the URLs associated with them). Since these pre- fixes are declared in the document content, you end up with a peculiar situation in which the DTD has to be written according to declarations in a given document instance, rather than the reverse. Worse yet, there is no way to be sure that the various documents being validated against a particular DTD use the prefixes correctly, with the correct URLs, un- less you make extensive use of attribute defaults - which, ironically, means we now need the DTD (probably an external one, typically with a bunch of parameter entities; so get your validating parser ready).--Richard Goerwitz on the xml-dev mailing list
*Is* there really a huge market for text management/retrieval? The history of software is littered with the corpses of companies who tried to make a go of it in that area; I know from personal experience that up to and through the year 1996, there was *not* any such huge market. Will XML change that? It would be nice to think so.--Tim Bray on the xml-dev mailing list
you want to encourage vertical portals to send you traffic, so make it easy for them to link to you: Have persistent URLs that don't change in the middle of the day or go away unexpectedly. Don't have your PR folks spam the editors of other Web sites with URLs, but do provide advance warning of upcoming good stuff that would be worth a link.--Jakob Nielsen
--The citizens of the Net and the Web live in a world without walls, denizens of a new kind of social geography. Much more than technology, this is the stunning new reality of networked computing.
We go online so often, to so many different places, and browse and download so freely, it's sometimes easy to take our own culture for granted.
But this world without walls increasingly is colliding with the off-line place, the one that's criss-crossed with walls. As these two very different countries interact, the issue of walls and boundaries becomes more intense, even bitter. And the stakes become bigger, for politics, business and for nearly every powerful institution.
The Net is entering its second generation. People who grew up on take for granted their freedom to say what they want, go where they want, download and retrieve what they want. They have little experience with walls. As the Net becomes more central, lucrative and culturally powerful, the outside world is thundering online like a great, frenzied herd. And they're throwing up walls like mad.
Just as furiously as they want to put their walls up, many of us resist. This profound difference puts the Net at odds with most of the important institutions, traditions and instincts in the world.
I'm all for privacy on cellular phones, but Congress is going about this all wrong. The way to protect privacy is not to make certain technologies illegal. Instead, the way should be to make encryption legal.--Shari Steele, director of legal services with the Electronic Frontier Foundation
VRML was born dead. It never solved any real problems. Good riddance.--Clay Shirky on the wwwac mailing list
Actually, if you look at the map, you'll find that the modern forms of Latin are the predominant language(s) in much of south-western Europe, most of the Western hemisphere (with the exception of the anglophone parts of Canada and the U.S.), and scattered parts of Eastern Europe, Africa, and Asia. Latin has changed over two millenia, but no more than proto-Germanic changed over the same period: from the simple perspective of geographical spread, they're both enormous success stories.--David Megginson on the xml-dev mailing list
XML is looking like a magic elixir in a cloudy bottle these days: cheap data exchange, document structure, and a proven cure for warts.--Uche Ogbuji
Maybe Microsoft thinks testimony and evidence are like products. Version 1.0 doesn't cut it? Announce an upgrade. Maybe Version 2.0 will be good enough to persuade someone. For Microsoft, truth, evasion and deception seem like mere gradations of the same thing: propaganda in service of the empire and its defense against interlopers.--Dan Gillmor
I find it a little scary how many people get their information on XML from Microsoft. Microsoft can't even implement XML correctly in their own products (i.e. Office 2000). They also do not differentiate between Microsoft "ideas" (like data islands and XML-Data) and W3C recommendations. I would suggest you use XML.COM or the OASIS site for a more realistic view of the state of the art in XML/XSL land.--Paul Prescod on the xsl mailing list
open DTDs are not good news for the ultra-dominant software vendors. Efforts to create industry-standard DTDs are the strategic Manhattan Projects of the ongoing struggle between information owners and software vendors for control of huge libraries of valuable commercial information. Eventually, the information owners are going to win, leaving them in a position to buy their software from the lowest bidder.--Steven R. Newcomb on the xml-dev mailing list
The problem with WYSIWYG is that it is usually equivalent to WYSIATI (What You See Is All There Is). A document has a rich semantic structure that is often poorly captured by its appearance on a screen or printed page. For example, a word may be printed in italic font for emphasis, as part of a book title, or as part of a quotation, but the specific meaning is lost if it is represented only by the fact that the characters are italicized. A WYSIWYG document shows only the final printed representation; it does not capture the user's intentions.--Jakob Nielsen and Don Gentner
W3C's standards committees should be able to make an informed decision about whether to include something in a standard that may be covered by a patent -- particularly if the patent is held by one of W3C members helping develop that standard--Web Standards Project Leader George Olsen
DTDs are essential in marketplaces in which open information interchange occurs in a multivendor environment. In this kind of situation, DTDs serve as contracts between information-creating application developers and information-consuming application developers. When information fails to be interchanged successfully (i.e., when things don't work), and if there's no DTD contract, then there's no way to tell who's responsible to make what changes in order to restore successful open information interchange. Software maintenance costs spiral upward, customers get confused and unhappy, and the atmosphere in the marketplace is poisoned. With a DTD contract in place, the reliability of open information interchange is much higher, and the entry cost to software vendors of serving any given marketplace is much more predictable.--Steven R. Newcomb on the xml-dev mailing list
There are already 101 solutions for producing server-side mark-up, with XML or with 101 other possible data formats. What we need is a solution that has a smooth flow from XML to rendered content in the browser, with transformation on the client-side allowing scalability, rather than the server. Server-side solutions will continue to move forward, it's the client-side that is in need of inovation.--Guy Murphy on the xsl mailing list
Back in 1987-88 I helped build a style sheet-driven browser (the chief author was Darrell Raymond) that became a commercial product of Open Text Corporation in 1989. It did several things that the Microsoft patent seems to cover. I'm confident that Microsoft will do the right thing and simply ignore the existence of this patent.--Tim Bray
XSL right now is too much like a woman trying to get dressed (style) at the same time she's putting on her makeup (transformation). The two tend to get smeared together and the result is none too pretty.--David LeBlanc on the xsl mailing list
It most certainly *is* the job of the standards developers to make sure we understand what they write. What is the point of a standard if nobody can understand it? Even more to the point, if what standards writers write is routinely interpreted to mean many different things by many different people, then I think the standards writers have failed their job.--Ronald Bourret on the XML Dev mailing list
XML is many things to many people. It is simple enough and generic enough that if you try hard enough, you can stretch it to fit just about any problem in computing. It may not always be the optimal approach, but it fits many problems very well.--Simon St.Laurent on the xml-dev mailing list
XML is not the answer to all the world's problems - it creates new problems, that are awfully damn interesting to solve.--Simon St.Laurent on the xml-dev mailing list
As a Certified SGML Paranoid Nutcase (CSPN) I distrust all software implicitly and therefore always prefer solutions in which the data, represented using SGML or XML, is the primary data store, with any other representations being merely transient reflections of that data for purposes of optimization and that sometimes you are forced to trust your software not to screw up your data too badly.--W. Eliot Kimber on the xml-dev@ic.ac.uk mailing list
The people making the decisions -- making the trade-off where privacy prevention and financial interest should be drawn -- are the ones who have no interest in privacy and have every financial incentive to extract the maximum amount of information about consumers. Copyright owners want registration and absolute identification so they can prosecute people who infringe. And that level of information is intrusive to everybody.--Jason Catlett, president of Junkbusters,
Anyone who says XML "belongs" anywhere in particular doesn't understand what XML is about. "XML should be everywhere! Every database should be able to import and export XML-formatted text. So should every programming and scripting language. That's the purpose of XML; to break down the barriers and to allow all kinds of data to be interchanged with all kinds of software.--Dave Winer, president of Userland Software
I predicted that the Internet would collapse every year between 1988 and 1993 inclusive. Still looks shaky to me, but you get tired of being wrong all the time.--Tim Bray on the xml-dev mailing list
"Namespaces" is technically inelegant, it also has the problem of being incompatible with a lot of other XML things. Unfortunately, though, I think it is doomed to succeed.--Michael Kay on the xml-dev
...specs are for implementors and very sophisticated users. Joe Average will never read a spec., no matter how "friendly" it is. Specifications are more like laws than they are like novels. The move to informal specs has not decreased the need for "language lawyers" -- it has just made their job harder. It has also made implementors jobs harder, which works AGAINST the popularity of the language.--Paul Prescod on the xml-dev mailing list
The Web creates a snowball effect in terms of demands for data access. One of the key technologies in this area will be XML for managing, storing and distributing information.-- Pete Fiore, executive vice president and general manager at Ardent
...it would be nice if Microsoft made a meaningful move toward open standards like XML, but history suggests that they're more likely to take the open standard du jour (ASCII, HTML, Java, whatever) and then add proprietary bells and whistles that make it nearly impossible to write stuff that can be used anywhere except on Windows.--John R Levine on the Computer Book Publishing mailing list
...a less formal spec that many people can understand easily will generally be adopted and implemented much more successfully than a more formal spec that fewer people can understand, despite the disadvantage that the less formal spec probably contains ambiguities, inconsistencies and omissions.--David Megginson xml-dev mailing list
You _can_ use XML to create proprietary formats, but unless you go to great (and foolish) lengths to encrypt your information, it's still going to be a lot easier for your competitors to figure out what's going on in your documents than it is with a messed-up binary format.--Simon St.Laurent on the Computer Book Publishing mailing list
Looking at the Office 2K implementation, though, it's basically Microsoft saving its own relevant information through XML data, rather than creating a vendor neutral solution. This has been my objection to Frontpage as well, and a common problem that I see while working as a contractor at Microsoft. Sure you can save into HTML/XML, but there's only going to be one browser on the planet that will be able to open it up and do anything useful with it. MS doesn't see XML as an interchange format for vendor neutral solutions, they see it as a database format to augment their current products.--Kurt Cagle on the Computer Book Publishing mailing list
I look forward to the day that we have an XML based full featured word processor, where the editor supplies the requisite style sheets and writers can just write to their hearts content without the need of specialized data formats. No more worrying whether or not the editor is using an older version of Word, never having to reconstruct style sheets. If and when such an application comes up, if it isn't from Microsoft then they will be running scared.--Kurt Cagle on the Computer Book Publishing mailing list
Purity of language is not a goal, it is a tool. A cluttered language is harder to learn and harder to implement; a language that is unusable (for want of implementations) or unused (for want of understanding) may as well not be designed at all.--Chris Maden on the xsl-list@mulberrytech.com mailing list
designing the namespace spec was the longest, toughest, most contentious, most complex work ever done in the sphere of the XML activity.--Tim Bray on the xml-dev mailing list
This is a very, very, very sad day for the XML Community which will be remembered for eons as the day XML complicated itself to the point where it is virtually unusable by the masses, at least as far as namespaces is concerned.--Tyler Baker on the xml-dev mailing list
Like all other non-metric units, points (all of them), picas, ciceros, cpi, dpi, lpi, etc. are a real mess and should be replaced by the millimeter as the one and only unit for every length measure in typography as soon as possible (paper sizes, column width/length/ distances, font sizes, font heights, etc.). Everything else is just ridiculous historic ballast and it boggles the mind that we still use it and waste a lot of time with converting units. Surely some old typographers will turn out to be conservative and mentally inflexible and express their irrational love for the points, but we can't continue this unit nonsense forever, even if it means breaking quite some inertia.--Markus Kuhn on the Unicode mailing list
--Gerard Berthet on the xml-dev mailing listThe biggest value XML brings is that, in the past, an app vendor was very reluctant to support other vendor-proprietary APIs to support such data exchange, not because it was not feasible, but because it was cost-prohibitive. Imagine an application having to interoperate with one or more app, each app having a couple of releases in the market, each release on several UNIX flavors & versions and MS platforms. A great deal of development resources had to be allocated just to maintain the status-quo, constantly catching up with the latest release. This model was fundamentally broken.
With XML, interoperability is simpler for both data producers and consumers, and needs only to be implemented once for all apps and platforms, clearly relieving valuable development resources to add real value to the app and eventually to the end-user, killer app or not.
There will never be a demo of XML-based technology that is substantially more interesting than that same demo based on legacy or proprietary formats. Word will not instantly become more fun and exciting to use when it exports XML.--Paul Prescod on the XML-Dev mailing list
Internal text XML entities are not logical objects in the same way that characters and elements are. They are text substitution hacks. Once the text substitution is complete, the entities go away. XSL does not provide a mechanism for generating them in the output, because they are text substitition hacks and you could just as easily generate the *value* of the substitution instead of the substitution name.--Paul Prescod on the xsl mailing list
The digital age is about empowerment and giving consumers flexibility. The music industry is stuck on an old business model that no longer works. They'll just have to adapt and come up with a new one.--Hassan Miah, Xing
Beware the temptation to think that "XML is all and all is XML". XML is a great standard interchange language. Java is a great portable applications language. Neither were intended to solve all problems. Remember that a standard mechanism for moving information between two points places NO restrictions on how the information is represented or managed within each point. There are thousands of man-years worth of effort and products in the persistence space, that should be leveraged, not replaced.--Jeffrey E. Sussna on the XML-DEV mailing list
XML is not optimized for update, retrieval, searching or anything else. It is optimized for interchange, interchange and interchange.--Paul Prescod on the xml-dev mailing list
XML is a very sophisticated exchange (import/export) format, and RDBMs are very sophisticated storage and retrieval tools. Different species, but they can be taught to play together nicely.--David Megginson on the xml-dev mailing list
No one should feel like they have to apologize for being excited about XML. However much XML may be a subset of SGML, XML has opened the door to new possibilities that SGML was just too big and unwieldy for. There's plenty to be excited about in XML, beyond the excitement that SGML had to offer. Simplification is important - strip it down to reasonably basic structures, and let people build their own solutions. That's what XML has done, finally.--Simon St.Laurent on the XML-L mailing list
One of the major reasons for the traditional cost was Uncle Sam's early adoption of SGML, and the same price-tag was applied to that as to the apochryphal $500 hammers and $100 nuts and bolts. When I first went looking for a decent graphical SGML editor in the early years of this decade, the one I liked was $6,000 per copy, no academic discount, and a salesperson so rude that it was clear my business was unwelcome unless I was ordering in bulk.--Peter Flynn on the XML-L mailing list