This week NVIDIA, the maker of Video cards, announced their new FX 5800 Graphics Card. It's specifications are startling and will continue to bring things that were impossible to the realm of possible and perhaps even common place in the world of visualization. Technologies like Windows Presentation Foundation (WPF), which is the basis for the InfoMesa project, will exploit this technology on delivery.

"The Quadro FX 5800 graphics card offers unprecedented performance and scalability to rapidly visualize and interpret massive data sets that until now were unattainable on a workstation graphics board. Offering up to 240 CUDA(TM) programmable parallel cores and the industries first 4GB of graphics memory, the Quadro FX 5800 graphics card is ideally suited for oil and gas exploration, medical imaging, styling and design, and scientific visualization. Other advanced features of the Quadro FX 5800 graphics card include:"

• Interactive 4D modeling with time lapse capabilities
• Massive memory bandwidth of up to 102 GB per second
• Fill rates that exceed 52 billion texels per second and geometry performance of 300 million triangles per second
• Support for next-generation OpenGL and Microsoft DirectX 10 applications
• Advanced multi-system and multi-device visualization environments with Quadro G-Sync II

The location for the December Release (called Phase 2) is currently at:

http://cid-794708049c7ae9c2.skydrive.live.com/browse.aspx/TechnologyDemonstrators/InfoMesa/InfoMesa%20-%20Phase%202%20|5Due%20in%20December%202008|6

You will see the work on Assets and Associations is progressing. The work on Annotations will arrive next week as well as a number of regressions that need to be fixed.

In general, the goal of December was Persistence. Storing what we show visually in a database, in a simple and predictable fashion. We are on target to deliver this in early December.

There will be a new build of InfoMesa available this Friday. It is still a little rough, but since it's due in December, I wanted to get it out in the wild. Annotations are under construction. From the very beginning we knew that InfoMesa could not afford to allow annotations to belong "inside each element" - forcing all elements to do their own annotation work. We knew that annotation and explanation needed to be part of the overall experience and "fabric" of the application.

What are Annotations?

Annotations are simply "call outs" or explanations that are important for an element's function, appearance or mechanism. They allow quick learning without reading whitepapers and boring journals (although you can link those to the annotations as well). InfoMesa is not the only one trying to deal with Annotations, but I do think we are bringing some ideas to the table that are worth considering...

Here are some examples of annotations in the current computing space...

This one is a Flickr annotation - labels and arrows are positioned inside each photo or picture that identify people or relationship. Notice the Hoops (around the Dove) and different colors for arrows and how text is handled to allow contrast.

Here's another...

These are notes in the Adobe Photoshop application (which also supports Audio Notes). By using layers inside of Photoshop's "fabric" the annotations do not corrupt the photo or image --- this is critical.

Here's my last non-InfoMesa example,

...the classic 3D annotation problem, which will complicate things - as rotations and viewpoints will change where the annotations need to be displayed.

This is not an easy problem to solve.

InfoMesa Annotations

This image shows the basic issues to overcome by having annotations "connect" to the elements, but not be part of them (like Photoshop layers do, but less manual). The three annotation lines in this image show the challenges we face: Image Resize, Scrolling Regions and Clipping, and 2D (the Canvas itself) to 3D surfaces that can be rotated and zoomed. This means unlike PowerPoint, the end points are not static and it's not as simple as VISIO, or even InfoMesa Associations which simply follow an element around as it is moved and resized. If we are pointing to an image (in this case, an image of a tumor) then if that image is resized, the *point* that was be called out must change and the line must adjust accurately, so that the science is not wrong or misleading or confusing.

The second problem, that of clipping, is simple to explain, but hard to solve. If we point to a sub-canvas that is being scrolled and clipped by a scrollviewer (like the Gene Expression Data) we need to handle "hitting" the extents of the clipping region appropriately. The line is not part of the clipped region, so we need to handle this ourselves.

Finally, the last image on the right is a Molecular Viewer showing a protein in 3D space. An annotation point is connected to the golden Beta Sheet item. As the protein is rotated and zoomed that 3D point will change (we must store the 3D point of interest, but then recalculate that point in 2D and reconnect the line to it during modification of viewing angle and distance.

Needless to say, it will take a while to solve this. If you have thoughts or ideas, please feel free to drop me a line...

Look for a post on Friday about the new code...

http://instanceofidea.blogspot.com/2008/10/adapting-infomesa-to-social-computing.html

Back in the summertime, I met with Mike Conway and a number of the people at RENCI (Renaissance Computing Institute) regarding our work on InfoMesa in Raleigh, NC.

Mike has a Blog here which has some great visualization material.

Who is RENCI?
The Renaissance Computing Institute (RENCI) was launched in 2004 as a major collaborative venture of the University of North Carolina-Chapel Hill, Duke University and North Carolina State University. Supported by the state of North Carolina, RENCI focuses on finding solutions to complex, multidisciplinary problems. The institute is a virtual organization that includes facilities at the three Triangle universities, East Carolina University, UNC Asheville and UNC Charlotte. RENCI brings together academia, government, industry and world-class computing and technology resources to find innovative solutions to these problems. Our programs span a breadth of fields, but much of our focus is on two critical multidisciplinary issues: how to better predict, plan for, and mitigate disasters and how to improve healthcare and the overall health of North Carolinians. In all our efforts, we strive to create an atmosphere that embraces innovation and bold approaches to problem solving so that North Carolina will thrive in the 21st century knowledge economy.

The RENCI Mission
The Renaissance Computing Institute, a multi-institutional organization, brings together multidisciplinary experts and advanced technological capabilities to address pressing research issues and to find solutions to complex problems that affect the quality of life in North Carolina, our nation and the world.

When I visited I was shown a number of different environments for working socially and dealing with large displays of visualized information.

Here is a Video of their work: ftp://ftp.renci.org/outgoing/CollageWPF/infomesa-videos/infomesa-scr-video-001.avi

Other Videos are stored here: ftp://ftp.renci.org/outgoing/CollageWPF/infomesa-videos/

Mike and I are continuing to share ideas about InfoMesa - stay tuned - these guys are going places.