XML.com Redux

January 1, 2017

Tim Bray

Tim talks about why we decided to relaunch XML.com, and what the focus of the new-old site will be.

Wow, it’s been a while. XML isn’t the new hotness, but it’s useful and being used – a decent way to make a living and get stuff done. That’s what XML.com wants to be about: tools and techniques and stories around getting things done.

Today’s acronyms are DITA and HL7 and FHIR, but some of the oldies are still goodies; our esteemed editor Lauren Wood is doing nicely these days by, among other things, improving U.S. Federal legislation with XSLT.

We’ll cover the new stuff and the classics. But first, a question:

What’s XML For?

The official terminology is “an XML Document”, and that ought to be a clue: XML is about documents.

Documents are built around organized sequences of words, and have a beginning, a middle, and an end. XML was invented by a bunch of publishing-technology geeks who understood these things.

Documents are where memory and knowledge lives when it’s not in wetware between people’s ears. Without over-reaching, documents are near the core of what makes civilization possible. The ones that matter deserve to be treated with respect: structured and re-used and revised and linked and stored with care for a long, long time. That’s what XML is for.

Now, there was a time when XML was the only portable data format, independent of operating systems and databases and authoring systems programming languages. (Twenty years later, it’s still hard to believe that it took till 1996 build such a thing.)

Today, we’ve got lots: JSON and YAML and CBOR and Protobufs, to name a few. Those are the good ones, many others have vanished unmourned. And those are what people should use for moving data records across the Net. But we’ve learned that documents aren’t just a basket of data records; and also that data records mostly shouldn’t care about things like ordering that are fundamental to documents.

Anyhow, XML is about serious documents: Legislation, health records, academic publications, policies and procedures, bibliographies and theorems.

Which is to say, stuff you expect to get paid for working on.

We’d sort of like to get paid for working on XML.com too; for details on that, see the Contribute page.

XML.com’s History

I grabbed the domain name within a couple of weeks of the Working Group settling on it, which was within a couple of weeks after James Clark proposed it. Jon Bosak grimaced and said “I guess that’s how it works”, then snapped up the .net and .org.

What happened was, I got together with O’Reilly and the now-departed Seybold publishing-tech empire and we ran XML.com at a tidy little profit for a while. We had loads of fun, and even a gossip columnist, the fairly-skanky Xavier McLipps. After a while it was just O’Reilly and me, then we all got busy and XML.com lay fallow and mostly unloved.

I might have forgotten all about it, only we kept getting emails from one Chinese operation and another offering to buy the domain name, for a price that kept going up and up and up; the highest quoted was US$50K, but some less-euphonious names have gone for much more. So someone was placing value on it.

Anyhow XML is still a thing and the site’s barrenness made me sad, and Lauren agreed to pull the pieces together, and O’Reilly agreed to let us stage the legacy stuff (thanks, Mac!) and only about six months after we’d planned to launch, here we are.

Thanks for dropping by! Of course we’ve got a feed, so you don’t have to come to us, we’ll come to you. Well, unless you have something to tell the still-numerous and quite-well-funded XML community; in which case you have to come to us and offer to write something. We’re even planning to pay.