Richard Davies wrote: The UK has a good crop of technology pioneers in cloud computing - for example ElasticHosts, FlexiScale, Flexiant, OnApp - and also some strong government initiatives such as G-Cloud.
We will have to see whether this kind of technical leadership converts into swift mass-market adoption or not.
When we at Antarctica start talking to a potential customer or
partner, they say, "That's Tim Bray's company? So this is XML-based
data visualization?" And we have to say, "No, it's ordinary database
visualization. But because it's modern software, there's a lot of XML
in the plumbing." And in this there's a lesson.
A lot of people who care about XML spend a lot of time
thinking and worrying about where in the real world of applied
technology XML is getting traction.
If you read the trade press and listen to the
prognosticators, XML is a landscape of ground-breaking initiatives,
industry buzzwords, and architectural upheaval. Here are a few
examples:
The many layers of Web services technology are going to lead
to the rebirth of EAI.
The Semantic Web is going to teach my refrigerator to talk to
my local beer vendor when my supply runs low.
Microsoft's .NET platform is leading to a new dawn of
productivity and interoperability for developers.
Executives and prognosticators alike are telling us that XML
will achieve "zero latency" or "real-time business" or "customer
focus" or whatever the buzzword of the day is.
But something's wrong with this picture. None of the above
has much to do with the way we're using XML, and none seems to have
much to do with the way people I talk to in the real worlds of
manufacturing or financial services or publishing are using it either.
When I talk to people about how they're really using XML, I
detect a few patterns, but they're not the ones I read about in
business publications. For example:
At Antarctica, we wanted to experiment with a Flash client.
We had our server, which normally sends out an HTML picture of an
information space, send out an XML description, which was pretty
easy. Then we programmed the new Flash 6 software to parse the XML
and draw the map right on the desktop. That was harder, but it all
worked fine; and the Flash maps looked great.
At a major Web portal, they're assembling their search result
and news screens by shooting XML messages off to a half dozen
business partners, getting more XML messages back, extracting
information from them, and gluing it all together for the portal
screen. They didn't write any schemas or organize any task forces,
they just e-mailed examples around until everyone was happy with
them, and went to work.
At Antarctica, we're always receiving new information that a
customer (or prospect) wants mapped. We used to get tab-delimited
files and Oracle dumps and all sorts of other old-fashioned, awkward,
fragile interchange formats. With every month that goes by, more and
more people say "Sor I could just dump that for you in XML, if you'd
like." Of course we'd like. It simplifies the internationalization
issues, and when something goes wrong (and something always goes
wrong) the XML makes it obvious where the breakage is, so you can fix
it.
The Government of Ireland is working really hard to build a
coherent, efficient Web presence for all its citizens. After looking
at a dozen different portal architectures, they've decided the way to
go is to build some service-oriented infrastructure, define some XML message interchange
vocabularies, and get out of the way of the individual departments.
It's a bold stroke, and it just might work.
What common threads do you see? I see a tendency to
improvise, the application of a lot of ingenuity, and an assumption
that anything in the world of computing can be made to talk to
anything else in the world of computing.
Here's another pattern that I see, kind of a subversive one.
There seem to be two kinds of XML-based initiatives out there. In the
first, they don't bother too much with the niceties, they just focus
on getting some things put together and happening, they make up and
refine the messages as they go along, and they've been in production
now for six months. The second kind takes a more carefully structured
approach, builds things from the schemas out, worries a lot about
choreography and data modeling and semantics, and is still in the
planning stage, with the revised schedule calling for deployment two
quarters from now if things go well.
Admittedly, I'm exaggerating a bit; but not all that much.
Meanwhile, the biggest story in the XML world is happening
just off the radar of the prognosticators and executives. It's called
RSS, and it's a simple format for pumping the content of dynamic
information sources around. It was invented for use by the legions of
webloggers, but it's mainstream now; I no longer surf to the New York
Times or the BBC or MSDN, I subscribe to them, and when something
changes, I get a nice little summary and decide whether I want to
check it out.
RSS has never actually been blessed as a standard, and its
development has been fraught with nasty personalities and politics.
There are competing versions, and the next-generation version
probably won't be called RSS. But it's changing the world, and it's
based on XML, and it's coming from a direction that nobody's looking
in. Stand by.
About Tim Bray Tim Bray founded Antarctica Systems, a pioneer developer of data visualization technology. In 1989, Tim cofounded Open Text Corporation, where he developed high-performance text retrieval software and in 1994 introduced what would become one of the first commercial Web search engines. In 1998, he co-invented Extensible Markup Language (XML). More recently, Tim was named to the InfoWorld Hall of Fame and Upside's Elite 100.
Reader Feedback: Page 1 of 1
#2
Abbas Shojaee commented on 18 Dec 2003
Dear Tim
I agree with you that there is limited focuses (possibly 2) for using XML in current IT opinion.
Using XML for defining complex data structures is great but not enough.(Using XML for data transformation and making integrated systems and .... also are not so important or deep.
I guess new perspectives and worlds are coming through using XML in OO data explanation. Have you ever heard about "Object Transparency"?
Regards
#1
jay_sdk commented on 7 Aug 2003
This very much reflects what I see and experience in the field ( work for the services dept. of a enterprise software company ).
* No webservices, pretty much
* XML/XSLT used for display primarily, or passing XML docs around for different apps talking to eachother (with the XML "invented" as one goes along, and either with or without schemas - in the latter case, the schema usually comes after the fact, generated off an actual XML doc.)
Abbas Shojaee wrote: Dear Tim
I agree with you that there is limited focuses (possibly 2) for using XML in current IT opinion.
Using XML for defining complex data structures is great but not enough.(Using XML for data transformation and making integrated systems and .... also are not so important or deep.
I guess new perspectives and worlds are coming through using XML in OO data explanation. Have you ever heard about "Object Transparency"?
Regards
jay_sdk wrote: This very much reflects what I see and experience in the field ( work for the services dept. of a enterprise software company ).
* No webservices, pretty much
* XML/XSLT used for display primarily, or passing XML docs around for different apps talking to eachother (with the XML "invented" as one goes along, and either with or without schemas - in the latter case, the schema usually comes after the fact, generated off an actual XML doc.)
Oh, and virtually no .Net at all... (all Java)
Can you bring services from the cloud to your customers faster and have them adopt it with ease of use or bring the power of bundled services to the fingertips of your clients without creating new rigid ‘apps stove pipes'? Do you want to prevent your business running away to publ...
OCZ Technology Group, a provider of high-performance solid-state drives (SSDs) for computing devices and systems, on Tuesday announced the Z-Drive R4 CloudServ PCI Express (PCIe) flash storage solution, designed to accelerate cloud computing applications and reduce operating expe...
Many organizations have embraced, or are considering, the benefits of cloud computing – speed, flexibility, increased expertise, shared workload, reduced costs, etc. The benefits are many – but so are the risks. What are the threats to cloud security? Which parties assume respons...
In August 2011, SHI Enterprise Solutions (ESS) division launched the SHI Cloud, offering reliable and cost-effective industrial-grade cloud computing platforms. That same division achieved an 82 percent increase in revenue over 2010.
SoftLayer Technologies on Tuesday announced the immediate worldwide availability of SoftLayer Object Storage, a redundant and highly scalable cloud storage service that allows users to easily store, search and retrieve data across the Internet, with optional CDN connectivity, or ...