TechEd, Open XML, and HDR Photography

While being at TechEd in Orlando, FL, last week, I had lunch with Doug Mahugh and we talked about the upcoming ODF support in Office 2007 SP2, the new features in the Open XML SDK, Altova's new support for Open XML diff/merge in DiffDog, creation of Open XML from StyleVision, and data integration and mapping for Open XML in MapForce, as well as various other XML-related topics.

We also talked about some other industry topics and finally came to chat about HDR (high dynamic range) photography. Doug sent me a few links to some of his recent photos, and this one impressed me the most.

I couldn't help it and had to get the software the same day. However, as I had left my Canon SLR camera at home for this trip, I wasn't able to test-drive HDR imaging until I got back home today:

Obviously, this isn't a particularly exciting scene - I just shot from our balcony towards the end of the cul-de-sac. I used an automatic exposure bracketing of ±2 and loaded all three images into Photomatix and then played with some settings in the tone-mapping to create some vibrant and surreal colors.

But I still like the result quite a bit - it makes me want to go out and take some HDR photos of Marblehead harbor and experiment with other local scenes where the high dynamic range can come into play nicely.

Creating Open XML (OOXML) Spreadsheet Documents

As Office Open XML (OOXML) gains more wide-spread adoption and popularity - and since it is now an ISO standard - developers will be interested in how easy it is to create Open XML documents directly in their applications, e.g. spreadsheet documents that are compatible with Excel 2007. Most approaches require quite a bit of hand-coding and worrying about the actual OpenXML specifications, but what I want to show you today on the XML Aficionado blog is a way to use MapForce to auto-generate all the source-code (for example in C#) that will produce the desired .xlsx document so that you can integrate it into your applications (and use it royalty-free within your organization).

I will use a very simple example to demonstrate how you can turn some raw sales data in an arbitrary XML format:

into a pretty business graph in Excel 2007:

For such a simple use-case you could, of course, simply open the XML file in Excel 2007 directly, but I am only using a simple example to illustrate the process. The true power of this approach is that you can easily work with very complex data in a visual and intuitive manner - and that you can auto-generate the source-code to implement this as part of your application to automate such processes.

So let's open MapForce and insert the XML data file into our working surface where we are going to define the mapping:

Next we are going to insert an OpenXML spreadsheet document into the work surface of our mapping project - we can either insert an empty spreadsheet, or we can use an example document that we have previously created in Excel to indicate what sheets and what data ranges or labels should be receiving our data:

Now it is time to define how the source XML data should be mapped to the target OpenXML document. This particular mapping is just one example - MapForce lets you map between any combination of XML, relational database, EDI, flat-file (e.g. legacy text files), and OpenXML spreadsheet documents. In our case we are going to convert from start-date/end-date ranges in the XML source to months in the OpenXML document and from states to regions:

Once you've defined the whole mapping, this is how your project will look in MapForce - note that underneath the blue-gradient working surface the "Mapping" tab is the one that is presently selected, because I've just defined my mapping between the input and output files:

To test my mapping - before I auto-generate my program code, I can click on the "Output" tab underneath the working surface, and MapForce opens up Excel 2007 embedded within the same application frame to show me the result that is produced by my mapping:

This Excel table is then used to produce the graph that I showed earlier.

Now I want to auto-generate code in C# for my data integration project that will automate this generation of Excel 2007 OpenXML documents, so the next step is to check the code-gen settings to ensure that I generate it for the correct development environment - in my case Visual Studio .NET 2008 - but MapForce supports many other environments and can also generate code in C++ or Java in addition to C#.

OK, now we are ready to generate code. All that is required is using the corresponding command on the File menu, and all the source-code files are placed in a designated output directory, and the corresponding solution file for Visual Studio is generated as well:

The auto-generated source-code can now be integrated into any application and can be used royalty-free within your organization to automate the creation of Open XML (OOXML) spreadsheet documents.

If you would like to experiment a bit more with this example yourself, you can find all the files used here in the MapForceExamples directory when you download the free 30-day evaluation version of MapForce.

Also, keep in mind that you can use Excel 2007 files (or any other OpenXML spreadsheet documents) in MapForce both as input and output files, so you can create data integration applications and mapping or conversion code for any possible scenario that involves OOXML spreadsheet data, XML, EDI, or relational databases.

Microsoft/Yahoo/Facebook rumors

The blogosphere is buzzing today with rumors on Microsoft intending to buy just the Search portion of Yahoo and then buying Facebook. It's a strange world we are living in, but some argue that the move appears to make sense in a twisted sort-of way.

This should be an interesting week...

WorldWide Telescope - pretty, but not revolutionary

Microsoft Research has launched a public beta of the WorldWide Telescope (WWT) this week, which has generated considerable buzz in the blogosphere - mainly because über-geek blogger Robert Scoble stated that it made him cry when he saw a preview earlier this year.

I just downloaded the beta version myself and it is indeed pretty. Think of Google Earth, but looking outward at the universe rather than at our planet here. You can scroll and zoom and explore and see the night sky in much more detail than most people have ever experienced in a planetarium. It has a detailed database of astronomical objects, including stars, planets, constellations, and galaxies. You can use a search function to find any celestial object, or you can use the locator pane that points out noteworthy objects in your current field of view.

And it comes with great guided tours - slides, pictures from different wavelength images, and narration - that give you expert insights into little corners of the universe you didn't know about. And it allows you to control your actual telescope to zoom in on the same object you are viewing on your computer. And it's free. So it's a nice educational tool, no doubt.

But does it make me cry?

Hardly, if you consider that it has all been done before. For the avid hobby astronomer such features have already been available for quite a while. Starry Night Software does exactly what the WWT does, i.e. it lets you explore the night sky and provides guided tours to various astronomical events, controls your telescope, and it has far more features than the WWT. The only difference is that you have to buy Starry Night on a CD/DVD and install it on your computer, whereas you use WWT like you are using Google Earth: with a thin client viewer and all the data resides on the web.

The one thing I do like about the WWT is that any tours created in the software are, of course, stored in XML format. Microsoft hasn't yet published the specifications or schema for them, but I was able to create a short tour myself and then edited it further with our XMLSpy XML Editor.

So what is so innovative about the WWT that it warrants such a buzz? All it does is apply the thin client viewer plus cloud database approach to an astronomy application. That is certainly not revolutionary - I'd say it's not even original. Why you need a research lab to do it, is beyond me. It sure is a pretty application, but it simply doesn't deserve the hype and attention.

Open XML is now an ISO Standard

The official press-release came out of the ISO offices on April 2nd and Open XML (OOXML) is now an ISO Standard with the official designation IS 29500.

Microsoft issued a press release today, and states that 86% of all voting bodies and 75% of P-members approved the standard - both measures being above the needed thresholds of 75% and 66.7% respectively.

See also the following blog and media reactions today:

• Brian Jones: "The end of the file formats war"
• Jason Matusow: "Open XML is and international standard"
• Peter O'Kelly: "Ecma Office Open XML Document Format appears to win Approval as an ISO Standard"
• Computerworld: "Unofficial vote tallies show Open XML being ratified"
• CNet: "Unofficial document: Open XML passes ISO vote"
• ZDNet: "Microsoft wasting no time rejoicing in its OOXML win"

There is also an interesting story floating around that Norway allegedly seeks to reverse its Open XML vote to No - but that seems to be irrelevant given the high margins that the tallied outcome has over the minimum requirements for approval as a standard.

To get an early start working with Open XML (IS 29500), check out Altova's support for Open XML in our XMLSpy XML Editor.

You are also invited to read all previous articles on Open XML on this XML Aficionado blog - especially my January 30 tutorial post on Content reuse with Open XML and XSLT.

Download and install the driver for ... Windows Vista?

I received this curiously strange error message today, after Windows Vista on my desktop computer came back from a blue screen:

It seems that Windows Vista is asking me to download and install a new driver to run … Windows Vista???

That's the same error message format that you sometimes see when a graphic card or other hardware device malfunctions and Windows recommends that you download and install a new driver for that hardware. But it appears that in this case Windows has a problem running itself...

Microsoft's $45 billion bid for Yahoo

Microsoft has announced just now that they are making a $45 billion bid to acquire Yahoo.

This is directly from the Microsoft press release:

Microsoft Proposes Acquisition of Yahoo! for $31 per Share
Transaction

valued at approximately $44.6 billion in cash and stock; provides 62 percent
premium to current trading price for Yahoo! shareholders; combined entity to
create a more competitive company, providing superior value to shareholders,
better choice and innovation for customers and partners

“We have great respect for Yahoo!, and together we can offer an increasingly exciting set of solutions for consumers, publishers and advertisers while becoming better positioned to compete in the online services market,” said Steve Ballmer, chief executive officer of Microsoft. “We believe our combination will deliver superior value to our respective shareholders and better choice and innovation to our customers and industry partners.”

“Our lives, our businesses, and even our society have been progressively transformed by the Web, and Yahoo! has played a pioneering role by building compelling, high-scale services and infrastructure,” said Ray Ozzie, chief software architect at Microsoft. “The combination of these two great teams would enable us to jointly deliver a broad range of new experiences to our customers that neither of us would have achieved on our own.”

The Microsoft board of directors sent this offer to the Yahoo board of directors and the letter accompanying the offer is part of the Microsoft press release, too.

If this goes through, it would indeed be a curious transformation of the web. There are two "major" web sites that I rarely ever visit: www.yahoo.com and www.msn.com - imagine how appealing a combination of them would be...

For more blog reactions go to TechMeme.

Binary and OOXML office formats - interesting news

Brian Jones has a nice post on (a) making the documentation for binary Office file formats available more easily and (b) mapping from binary formats to OOXML:

The binary formats documentation will be available publicly by February 15, 2008 and the file formats will also be subject to the Microsoft Open Specification Promise.

Regarding mapping from binary to OOXML formats (and back), Microsoft promised to start an open source effort on SourceForge for that purpose.

This also made it onto TechMeme today...

XML 2007 (or not), IE8, and Google on the iPhone

XML 2007 ended in Boston today and despite planning to go to the conference today, I unfortunately missed the event due to some time constraints at work. Elliotte Rusty Harold blogged from the conference (thank you!), and he was asking if nobody else was reporting live from the event. Well, I was planning to do so, but couldn't. Sorry.

The one talk that I had really wanted to go to was by Irina Kogan (IBM) and Dr. Nick Nagel (Altova) who spoke on "XML-Driven Database Design and Information Retrieval" this afternoon - fortunately the presentation slides can be found here so I can read up on what I missed.

In other interesting news today:

• Bill Gates semi-announced Internet Explorer 8 to bloggers
• There is now a "wicked cool" iPhone version of the Google Mobile website

I've already played with the Google interface on the iPhone and it is really nicely done. I get Google search, GMail, and Reader all nicely integrated and with a slick iPhone like UI.

Google, Microsoft, and Yahoo - interesting synchronicity

I could not help, but notice an interesting synchronicity between the various announcements and news clips about these three firms in the last 2-3 weeks or so:

Sep 9, 2007	Microsoft fails to win ISO approval for OOXML. A review of detailed country comments does, however, show that they are likely going to succeed in the next round in March 2008.
Sep 16, 2007	Yahoo launches Mash - a new social networking site designed to compete with Facebook.
Sep 17, 2007	Google adds slide-show/presentation application to Google Documents in an effort to increase competition with Microsoft Office.
Sep 24, 2007	The The Wall Street Journal reported that Microsoft is in talks with Facebook to acquire a 5% stake in the social networking site.
Sep 27, 2007	Microsoft announced an updated Search capability in the Live Search engine. Incidentally it is also Google's 9th birthday.
Sep 30, 2007	Microsoft unveils its answer to Google Docs called Office Live Workspaces.
Oct 1, 2007	Yahoo announced a new Search Assist function to improve Yahoo Search.
Oct 2, 2007	Steve Ballmer speaks in Europe and says that the craze for individual social networks such as Facebook risks being exposed as a "fad". UPDATE: Robert Scoble responds that Steve Ballmer doesn't "get" social networking.

Office, Social Networking, Search, Office, Social Networking, Search, ... — is it just me, or is there some kind of pattern here?

And it all seems to revolve around online advertising platforms. Hmmmm.

Microsoft releases updated Live Search

Microsoft has announced today that the updated Live Search engine includes a 4 x increase in index size, significant enhancements in the core algorithms, increased focus on query refinement, and a new web data extraction model.

More details about these technical improvements are promised to appear in the LiveSearch Blog over the next couple of days.

I've tested the new Live Search today, and it indeed produces more accurate results than Google: e.g. searching for the term "XML Editor" on Google produces a whole bunch of irrelevant web pages and entries on the first page of the search results - including websites like Peter's XML Editor that aren't even supported/maintained anymore. Ironically, the #1 best-selling XML Editor and de-facto industry standard tool - Altova XMLSpy - is not even listed on that first page of results.

In contrast, the same search for the term "XML Editor" on Live Search will produce only highly relevant entries, including Altova XMLSpy as well as some competitors that are indeed maintaining their software and keeping it current.

It may have taken Microsoft a while to get Live Search right, but that's simply how they work:

• The first version of Internet Explorer wasn't the killer, but subsequent versions kicked Netscape's butt.


• The first version of Visual Studio wasn't damaging the Borland developer tools, but future releases forced them out of the market.


• The first XBox wasn't hurting the PS2, but the XBox 360 is showing Sony who's boss, or rather who's Master Chief.

Extrapolate from those examples to the field of search, and I'd say now is a good time for Google to start to fear Microsoft...

Microsoft in talks to buy a stake in Facebook

The Wall Street Journal reported just now that Microsoft is in talks with Facebook to acquire a 5% stake in the company. The deal is said to result in a valuation for Facebook of $10 billion.

Both Google and Microsoft are said to have been in negotiations with Facebook at one time or the other, but Microsoft has an advantage at the moment, due to an exclusive agreement with Facebook to deliver ads to Facebook users that was signed last year and runs until 2011.

Question: why wouldn't either Google or Microsoft want to buy all of Facebook instead of just a 5% stake?

Answer (according to the Wall Street Journal):

"Microsoft has considered trying to buy the company outright, but people familiar with the matter said it's unlikely at this time. Facebook founder Mark Zuckerberg has steadfastly kept his company independent with the goal of eventually taking the company public. In a round of negotiations last year, Mr. Zuckerberg rebuffed acquisition approaches from Microsoft, Yahoo and others."

See also my previous article "Facebook: it's not just for kids anymore..." on my XML Aficionado blog.

Also seen in Valleywag and TechCrunch today.

UPDATE: New posting in the WSJ Deal Blog has an interview with Bo Peabody, the 36-year-old founder of Tripod Inc. - one of the first social networks, all the way back in 1992.