Gleanings by Bradypus

It's a funny business, business. How many times have you come across a 'best of breed' program but been unable to persuade business enterprises to use it because it's in the public domain, and not from a commercial source? OK the Linux bores can stop shouting at the back and make room for those of us who have been promoting all those wonderful PD SGML tools (and the people who wrote them).

Same thing with genuinely open standards. It has taken years for company bosses to accept them and even now we get resistance whenever a new one comes along.

Take XML as an example. Every time Bradypus talks about XML to senior executives in business the first question they ask is 'who supports it'. Good question, but what they really mean is 'does it come from Microsoft or IBM because I will need to hide behind an industry giant if things go wrong when I adopt it'. For these people, it is not good enough that James Clark and Peter Murray-Rust and Henry Thompson are producing great tools - no, it has got to be Microsoft and Sun. Oh, and before I bet the company, IBM and Netscape also have to say 'we love XML'.

So it's good news for the SGML people who will have to roll out the technology in commercial companies that commercial products are appearing even before the standard is complete. All sorts of players are building it into their products. Microsoft is already active and Netscape is getting serious (what took them so long?). DataChannel has launched what it claims is the first XML Viewer. A first? Well, Bradypus says, only up to a point. Anyway you can play with it at http://www.datachannel.com/.

If you do visit the DataChannel site the first thing you will be able to do is read all about the bruhaha at W3C over Tim Bray's involvement with Netscape while he is still an editor of the XML specification. This issue brings into the open the whole question of commercial versus public interest in the world of open standards and the W3C's new policy regarding editors of specifications is welcomed. It is important that big commercial players participate in creating our standards but it is equally important that no company, or group of companies, steals the process. Fairness is the key.

Tim Bray, along with Jon Bosak, has made a massive contribution to the process of bringing XML to birth. It would have been sad to lose him over a conflict of interest. This is why Bradypus welcomes the W3C's brilliant solution: Jean Paoli and Tim Bray working along with Michael Sperberg-McQueen as co-editors! How about that for fairness!

There is also an upside to all of this. This may be the straw in the wind that signals Netscape's intention to behave. Just as we knew Microsoft was serious about SGML standards when Jean Paoli joined them, is not Tim's connection with Netscape a good signal? They certainly need all the help they can get catching up.

Despite the growing vendor support for XML, we still need to break down the prejudice against technological ideas and products which come from the non-commercial sector. There are many reasons why the prejudice exists: business is very suspicious of anything it does not understand, it's especially suspicious of the non-profit motive; business is wary of academics and sometimes with good reason; most of all, business doesn't like dealing with people and organisations it will never control. Yet it's business that is the loser.

What we need is an aggressive programme to sell the benefits of technologies like HyTime, DSSSL and XML to the top business decision makers. Sponsored by the likes of the W3C or the SGML Open consortium or the GCA, it should mount seminars, press campaigns and media coverage to spread the message. It should also provide house room at technology shows and trade fairs for the public domain and shareware software angels of the SGML world. We will know it's happening when some October we see JADE and JUMBO and their like on a big stand in the technology halls at Frankfurt.

Here's a newsletter Bradypus can warmly recommend: WOT is published free, every two weeks, with news and opinions about developments in web open technologies (WOT, geddit?). WOT is completely vendor-independent. Like Bradypus, its editors, Peter Judge and Christopher Ogg, believe the future of the web technologies lies in open systems, inter-connectivity and convergence. To subscribe, send an email to subs@wot.co.uk, with the message 'subscribe WOT' as the subject.

Have you read the HTML 4.0 draft yet? Have you commented? If not, why not? Yes, Bradypus knows it's long and, dare he say it, tedious. And sure, it's not as much fun as XML but it matters. Point your browser to http://www.w3.org/Press/HTML4, follow the links and share your thoughts.

Talking about freely available software this announcement appeared in Usenet News comp.text.sgml:

A new version of the SGMLDEV product suite from Copernican Solutions has been released. This build includes the following products: DSSSL Developer's Toolkit (DSSSLTK) A common API for Java/DSSSL environments; Grove Toolkit (GROVETK) A reference implementation of the dsssl.grove package; Java SGML Parsing Interface (JSPI) An SP-based Java SGML parser for processing SGML documents into groves. For more information and product download access, see: http://www.copsol.com/products/.

Well that's great news, but remind Bradypus, who was probably asleep at the time of the first release: is this stuff freeware, shareware or commercial software? It helps if posters keep us sloths fully informed, we miss so much in our somnolent world.

A final thought from Bradypus, whose tribe evolved to fill a slow moving niche: why have we been joined in our niche by the makers of SGML editors? Have you noticed that Adept and GriF and AuthorEditor et al have changed very little since they first appeared? Why is this? In the time Adept has been around WYSISYG word processing has been through uncountable generations, quadrupled its functionality and changed our working life. Yet Adept still offers little more than it had when it first appeared. Same story with GriF and the others. How come, guys? Bradypus has a long wish list of missing functions all based on requests from real paying customers. Drop a note to have them shared with you. Better still, read the current 'Design Question for an SGML Editor' thread in comp.text.sgml.