I was surprised to read that Stuart McKee, Microsoft’s US National Technology Officer, declared that “ODF has clearly won” at a Red Hat summit in Boston.

Open Document Format is an XML standard for office documents. Microsoft has its own XML format, called [Office] Open XML, and fought a bitter fight to get it standardised through ISO.

What’s a National Technology Officer? The role seems to involve promoting Microsoft products to government organizations. It is in the public sector that pressure towards standards adoption has been most intense. Presumably McKee is constantly having to defend Microsoft’s position. In May Microsoft announced that it will support ODF natively in Office, and will join OASIS to work on the standard.

Can Microsoft successfully promote its own OOXML standard, while simultaneously playing nicely with OASIS and ODF? The messaging, as PR people say, is tricky. Microsoft has told us that ODF cannot capture all Office documents with true fidelity, and that OOXML is more complete and better tuned for high performance. Can it now say convincingly that Office will be a great ODF editor? And if it can, surely Microsoft is undermining its own arguments for why OOXML was necessary in the first place.

Although Microsoft is fighting to maintain market share in the public sector by introducing ODF support, it still has a problem. ODF is closely associated with Open Office, and Microsoft Office is likely to lag its rival in this respect.

The real value of XML documents comes when you start manipulating them programmatically and on the server; I would have thought it would be difficult for Microsoft to make products like future versions of SharePoint work equally well with both formats. IBM’s server products will use ODF, which I presume is why the company fought brutally to oppose the standardization of OOXML.

In other words, a lot of future business hangs on this ODF vs OOXML argument. It is remarkable that a senior Microsoft person has said publicly, albeit in a small session at an open source conference, that “ODF has clearly won.”

Businesses and developers planning their future document management strategy can reasonably ask: is Microsoft still committed to OOXML? Does the format have a future?

I asked the company for comment and clarification, but as yet none has been forthcoming.

One other thing I forgot to mention in my post on Bill Gates Tech Ed keynote: Silverlight vs WPF. Someone asked Gates whether desktop applications were merging with web applications; Gates actually misunderstood the question, which he heard as whether Silverlight and WPF were being merged together. He said:

With WPF we get to assume we have the full power of the PC; we’re not just running in a browser environment … Silverlight will probably have almost everything WPF has today, but WPF will keep getting richer and richer as we go forward.

In reality, it would not surprise me if Silverlight thoroughly outshines WFP. I realise that they share a lot of technology, in particular XAML and the .NET Framework. But in many ways Silverlight is .NET done right, from a client perspective; it delivers just what is necessary for a rich client. It also runs cross-platform, a huge asset bearing in mind the increasing market share of the Mac and signs of life in desktop Linux. I’d also suggest that Silverlight will not always be restricted to the browser – look at Times Reader for proof.

In answer to the question actually asked: undoubtedly.

My second today:

This one is a CD from HM Revenue & Customs. No reason for complacency you may say, but on balance I suspect it is PC Tools ThreatFire that is slipping up here.

For the background to this, see here and here.

A friend is in the habit of checking her online credit card statement most days, just in case.

Recently she noticed an item that made no sense to her. It was a pending payment for several hundred pounds. Pending payments do not show a supplier, just an amount that counts against the credit limit.

The card is operated by John Lewis, a large chain of department stores in the UK (though it is general credit card, not a store card). She called immediately. The customer service representative advised that it was not possible to identify the payment in detail, and to wait until it was properly itemised.

She remained anxious and called back later to request that the card be blocked.

Later that same day, Dell dispatched a computer system to my friend, but not to her address; it had been paid for fraudulently. A couple of days later she discovered that the customer service representative at John Lewis was wrong; it would have been possible to identify the payment. The delivery could have been prevented; or if the powers that be actually wanted to catch the fraudster, they could have intercepted the goods at the moment of delivery. Frustrating.

By coincidence, a similar thing happened to me two years ago. In my case it was the bank that spotted the fraud and contacted me; but I still had to intervene myself to stop the delivery of the goods. It suggests that this type of event is common.

It strikes me that it would be worth pursuing these cases more vigorously, when there is an obvious opportunity to identify the offender, if only to deter others from committing similar crimes.

Old rocker Neil Young announced at the JavaOne conference that he will issue a chronological greatest hits/archive package on a 10 disc Blu-ray set.

And here is something really new, we will be able to add content to already released Blu-ray Disc archive volumes by downloading it, whether it is music, film or vintage recording sessions, recently found photographs, or other archival materials that were located after the release of that volume. Users will be able to download any of these archival materials and they will automatically be assigned to their place in the Chronology timeline … this could potentially include content updates such as music, film, adding new photos and providing tour information. It could also provide the ability to support dynamic fan community features such as message boards, concert reviews or even enable fans to use a BD-Live donation mechanism to help support the Bridge School.

Sounds like a web site to me. There will also be a DVD set, but only Blu-ray will offer the full interactive experience, powered by Java.

Of course, Blu-ray is a handy mechanism for delivering large high-resolution audio files to the consumer; but will users take to this kind of hybrid approach? Young says we will love the sound quality; but the general public seems tolerant of almost any sound quality that is good enough; a few of us complain about iTunes music at 128kpbs, but most listeners seem happy enough.

There is a discussion on Steve Hoffman’s audiophile message board which includes many of the fans who have been waiting years for archive Neil Young material. The response is mixed:

This is cool and all, but I have to say that when it comes to music, I still want it on CD, not on any video hybrid disc. I don’t want to have to turn on my TV to listen, I want to be able to play it in my car, etc.

and this:

And now this, expecting us to buy a whole new medium just to get his archives… I don’t want to watch this stuff, I want to play it and enjoy the MUSIC, on the musical formats I already own.

It reminds me of a very cool interactive CD I have somewhere, covering Bob Dylan’s chronology; it was called Highway 61 Interactive and came out in 1995. The disk included unreleased work and a multimedia presentation covering the recording of Like a Rolling Stone. As I recall, it was not a big success, which accounts for why it was not followed by similar titles for other artists. Now it is sought-after by collectors, but not that easy to play; it uses Quicktime but thirteen years later there are compatibility issues.

I will look forward to Neil Young’s set, if I can afford it, but it will be a niche item, and most people will buy his old hits on CD or download.

I don’t envy anyone trying to sell a WordProcessor or Office Suite that is neither Microsoft Office, nor free. Corel has just released WordPerfect X4, which it is promoting as a PDF editor. Both Open Office and Microsoft Word 2007 can save in PDF format, but WordPerfect can open PDFs as well. That could be a handy feature, though PDF was conceived as an output format so arguably it is not a big deal. In any case, you may be less keen on the idea once you read online help, as opposed to the marketing blurb, which explains that WordPerfect does not preserve the formatting of most PDF documents. If it’s an honest PDF, you get something editable but possibly different:

The layout in the imported PDF may be different from the layout in the original PDF, but you can still modify text strings and create a new document without having to copy or redesign all the elements.

If the PDF contains images of text, WordPerfect uses OCR to scan the images and generate editable text. Again, that could be handy, but if you think you can use WordPerfect to open an incoming PDF, make a few changes, and send it on its way, think again.

Let me add that all my attempts to import a PDF into WordPerfect have failed. I installed the trial, and tried to open the first PDF I came across – a 20 page Forrester report. WordPerfect whirled way at 25% CPU and using over 1GB RAM (I have lots installed), eventually offering a blank document. I tried again, and it crashed. Finally, I started a new document, typed the word “Test”, and exported it to PDF. Then I tried to open it in WordPerfect – nothing. It opens fine in Acrobat. I guess something is broken in my install.

Personally I am more interested in its support for OOXML (or possibly OXML), the native format in Microsoft Office 2007 and the subject of contentious ISO standardization. WordPerfect X4 has the cheek to make itself the default editor for .docx, .xlsx, and .pptx. Again, I opened the first document I came across, which is an 8 page Q&A, very simple, no images. The good news is that it opened. The bad news: plain text became bold, paragraph spacing disappeared, and the result looked worse than the original.

Next, I tried a document with a more complex layout. This is actually a bidding card for use in Duplicate Bridge: you can find similar ones here, but in .doc format – mine was one I had amended and saved as .docx. WordPerfect opened it, but with the layout completely messed up. Graphics were lost. I tried opening the old .doc version. Better, but still not right. It spread the document over 5 pages, a shame when it is meant to be printed on two-sided A4 to make a card. Open Office on the other hand could handle the .doc version nicely; I was impressed.

That gave me an idea for a further torture test. Open the bidding card in Open Office, save in ODF format, which WordPerfect X4 is also meant to support. Now open the .odt in WordPerfect X4. It crashed.

WordPerfect X4 may have all sorts of good points as a general Office Suite, but what about this claim in the press release [PDF]:

File Format Freedom
In addition to its significant PDF enhancements, WordPerfect Office X4 now provides suitewide  compatibility with Microsoft Office 2007 files (OOXML) and Open Document Format (ODF) in WordPerfect X4. With PDF-reading software installed on more than 80% of all U.S. PCs (Source: Jupiter Research), WordPerfect Office X4 enables users to collaborate and share files more broadly and more effectively than ever before.

Hmmm.

Update

I did a bit more experimentation. It turned out that the worst case (in terms of messed up formatting) involved a document which had originally been pasted from HTML. I imagine it was a bit of a mess internally, so perhaps one should excuse WordPerfect (though users don’t understand these distinctions). I reconverted a Bidding Card document and this time WP did better. Here are some images. First, a portion of the document in Word:

Now, here’s the doc imported into WordPerfect X4. Not right, but looks fixable:

Here it is in Open Office:

Identical to the Word rendering as far as I can see. Then I saved as .odt and opened in WordPerfect X4:

Pretty vague. As you’d expect. In this excellent interview Microsoft’s developer division VP Scott Guthrie cites three revenue opportunities:

• Tools and servers
• Customer engagement leading to ad sales
• As a platform for other, presumably profitable, apps

I’m most interested in the third of these. By the way, I like Silverlight. Cross-platform .NET has been a personal interest of mine for ever. In 2002 I wrote an introductory article about .NET, and said:

It would do .Net enormous good if it became a credible cross-platform contender, say on Windows, Linux and the Mac. It would do Microsoft enormous good if it could be seen to work with the open source community in the same way as IBM does so successfully.

Six years on, the cross-platform potential in .NET is finally coming together. However, it is as a web-based runtime, rather than as a desktop runtime. That wasn’t quite what I expected back in 2002, but it is no bad thing. If Microsoft is serious about refactoring all its software for cloud services, as Ray Ozzie stated at Mix08, then Silverlight could be a key enabling technology, giving a rich desktop-like experience but in browser-hosted applications.

I was also interested in Guthrie’s comments on open source:

…people in the Linux community are much more likely to trust Novell and, specifically, Miguel [de Icaza] and the Mono Project and feel like, “Okay — if it is open source, I can get access to all the source [code]. You’re telling me that I can snap the source and build it myself if you’re not doing a good job? Okay, that’s interesting.” The higher level libraries that we are distributing — our controls and things like that — those will just work on the Linux version of Silverlight. They can take our source and use them for that.

Microsoft isn’t posting its source for the Silverlight runtime, but it is supporting an official open source implementation. That’s an intriguing distinction versus Flash, which has open source implementations none of which have taken off. Adobe has open-sourced Flex, but not the Flash runtime. However, note Guthrie’s comment:

We actually deliver the media graphics stack to Novell, so we use the same video pipeline and same media pipeline on the Linux version as on the Windows and Mac versions.

So that “media graphics stack”, is that open source? I suspect not but would be glad to be proved wrong. This point might make a difference to Linux distributions that exclude proprietary software by default.

Finally, Guthrie makes some remarks about Adobe AIR and the fact that Silverlight doesn’t have an equivalent cross-platform desktop engine. He says businesses are more interested in a “web-based model”, and observes that the full .NET and WPF stack is already a desktop runtime.

I’m not sure that this is a big deal. It wouldn’t be a huge step to host Silverlight in a cross-platform desktop application, for example by including it in a browser control. At the 2007 Mix, the New York Times folk told me they intended to do this with Times Reader. We are also going to see a number of different approaches to this problem. Mozilla is working on desktop integration for browser apps. Google shows a desktop shortcut in its introductory video for offline access. I recall Adobe’s Kevin Lynch remarking on the psychological barrier to opening a browser application when offline, as being one of the motivations for developing AIR, but there is more than one way to mitigate this.

The second day here at Qcon London was more compelling than the first, from my point of view. There were several themes (I am just talking about what I personally attended).

The first was Java’s mid-life crisis and the reform of the JCP, which I will make the subject of a separate post.

The second was how the BBC runs its web site and its coming “tech refresh”, which was covered in a late “birds of a feather” session (which turned into a drink at the bar). I’ll come back to this later, after I’ve attended the BBC’s further session today.

The third was REST. Stefan Tilkov led a track on SOA, REST and the Web. Now, the general theme of this (following on from the Fowler/Webber session the day before on the shortcomings of the Enterprise Bus) is that SOAP and WSDL and WS-* have failed to deliver and that REST is fundamentally a better approach to designing distributed inter-application systems. What’s wrong with WS-*, SOAP and WSDL? Too many standards; too complex; too brittle; too incompatible; too few free and open source implementations; leaky abstractions; hijacked by middleware vendors who have an interest in keeping technology arcane and expensive.

By contrast REST is being embraced for all sorts of reasons, ranging from purist arguments about the value of resource-based computing where everything has an URI, to pragmatic arguments along the lines of “it works, I can use it, I understand it.” However, if you poke at some of the solutions which are described as REST, it turns out that some are more RESTful than others – using HTTP as a transport for POX (plain old XML) is not necessarily REST.

I’m still learning about this stuff; but I have no doubt of the value of REST. It fits particularly well with certain types of application. It is ideal for Internet mash-ups, for example. It is a natural fit with the Web. It builds on an infrastructure which is mature, proven and scalable.

Still, I have some scepticism about the rush to REST. I recall the early days of SOAP, when it was still meant to be the “Simple Object Access Protocol”, and at the time some of the claims being made for SOAP were not dissimilar to those now being made for REST (and yes, I know you can use SOAP and REST together). Specifically, it was intended to be a simple, loosely-coupled alternative to the brittle, highly complex, tightly-coupled, arcane, vendor-hijacked technologies such as CORBA and DCOM.

Someone yesterday – I think it was Jim Webber – remarked that middleware vendors are eyeing up REST; we can expect them to introduce products that will supposedly bring resilience, security, manageability, governance, scalability, and other good things; but which will also bring baggage that may sully the REST dream.

We may be having this conversation again in ten years as we discuss what-comes-after-REST.

Even if this is the case, I expect that Agile themes like YAGNI (You aren’t going to need it); appropriate planning and appropriate flexibility; communication and team dynamics; breaking hard problems into smaller, easier parts; testability; accountability – these will endure.