Bradley Horowitz at The Next Web 09 courtesy of DailyM

After Jeff Jarvis had his take on Google we had somebody from the inside – Bradley Horowitz.Horowitz was the man who advised Yahoo to buy Flickr and after redefining his role there he made them acquire Delicious. Previously he had formed Virage (video categorization engine) and sold it to Autonomy after dropping out of his degree at MIT Medialab where he worked on image categorisation technologies.

It was no wonder with his background that metadata, context and categorisation formed an underlying theme to this talk. It was good to see these foundational elements of online user experience given the focus they deserve.

Moore’s law and the aim of Google
He stated that Google’s main perspective on technology was thinking big as a means of innovation. With this philosophy they have changed the world as we know it and their impact is massive. As silicon valley realised Moore’s law to be true, Google imagined a world where bandwidth, size and storage was not a problem. In doing so they created  applications such as GMail.

Google’s aim is to provide excellent search to enable users to store and find everything. It is to solve one of the world’s biggest problems -that of information overload, its loss and its retrieval.

Ubicomp is here
He told of everyday devices that have achieved ubiquity – that can now record your entire life digitally. Ubiquity is here.  The mobile phone is everywhere.

The problem as he saw it was that you can record everything but you don’t get another life to review it all. The challenge is harvesting metadata and defining context to give meaning to what we do.

So how do you use the information to a useful end? Horowitz (and Google) knows that the big problem is that we are dying from the start. Moments evaporate from the start. This fact drives Google to be fast. Latency is a key focus because time is of the essence. By being fast they give back the time to the world by speed of its services.

Be aware of our human condition
He told of, meditation, sleep, walking the dog that gives him perspective. A very pertinent point was that technology needs to adapt and enhance the human life. He asked how do we solve attention management? The moments of life that need revisiting amongst the morass of spam and junk we all wade through.

A key observation was that metadata is as important as the data itself. Even from biosignals. He stated that Wetware (using people’s minds and bodies as systems) will be as important as software or hardware.

Audience at The Next Web 09 courtesy of DailyM

Image understanding is an amazingly complex process if you expect to use machines to do this. Using humans enables it to be easy to create understanding around cognition of imagery. People can do this easily to give meaning to images and Flickr shows that by a simple means to ensure that the people can tag easily.

Though I appreciate his statement I feel we have along way to go before the wisdom of crowds can be used effectively here.

He stressed that a few dedicated curators can create value for the majority. Indeed, the collective intelligence and interestingness creates a relevance metric that goes beyond just page rank algorythmns.  We see this in blogs, twitter streams and facebook communities but the curators still control the quality. That is  an important point I feel, and one that Google obviously realises.

Then he briefly showed a mental model that reflected the Google approach to data, starting at signal and working upwards.

• enlightenment
• wisdom
• knowledge
• information
• data
• signal

From a signal you may reach the pathway to enlightenment I guess, but it certainly has a clarity to it that is admirable.

The web is broken
What we did get from this talk was a refreshing personal perspective on the world of technology and how it affects us. He iterated that the conference had talked a lot about the current web, Twitter and real time search but he urged us to think about the real future.

In closing Horowitz stated how the browser is a bad model. It is not the optimal system to deliver the next web, the model is terribly broken, latency, code storage and caching are all frailties that have consumed countless lifetimes of development time.

He asked us to think about the ‘what if’ scenarios, take the long view and where the world is trending. Solving the problems that we expect to have tomorrow.

Create space in life
In many ways when somebody talks from Google you know they will not divulge what they do exactly, or certainly not their next step. They, like other successful companies, keep their products shrouded in mystery (ironic in this culture of open source and collaborative working don’t you think?).

With a slight nod to those who were wanting to know the next move he stated  it will be Google Voice. A way of leveraging voicemail into the digital life and Gmail.

Closing the circle completely, he asked ‘how much do you want to invite tech into your life?’. He reminded us to create space in our lives.

That coming from Google was a surprise – or was it really? Their aim after all, is to make our lives easier to manage in the information age. Surely that can only be a good thing.

Tweet This Post

A recent article by Alex Iskold brilliantly captures the separations of where we imagine semantic search should be and the reality. Even if it were trying to knock Google off a top spot, what he highlights is that it would be an unnecessary exercise. Google does its thing very well. Few would argue with that. Alex suggests that semantic search should do something completely different…

To really showcase semantic search, these companies need to come up with innovative UIs that will help users to understand the power that is being put at their fingers.

I also think it should display results differently, the interface should encourage an exploratory experience and allow lateral thought to occur during research. Iskold states we should move away from  the search box, as this is the wrong type of input for a user.

Search Patterns

Peter Morville is currently producing a book on search patterns and this will also highlight how facetted navigation enables the user to experience a different search journey. He has made the slides available here on Flickr.

If you look at the core of what should be shown is the relationships between items, that give relevance to the user’s query. Somehow a system needs to be designed that will reveal elements that have relationships and connections. Ontologies could be made, linking different data sets as if they were relational databases.

If a user found truly relevant and accurate information around an entity then a business objective could be fulfilled. Really targeted advertising would occur. Users are often more forgiving of adverts if what they see is relevant.

Images as a search device

But the challenge is in the interface, how do you convey a fluid, ‘noosphere‘, visually? It has to be visual because the variety of content types are so different, and to scan and associate quickly – images could allow instant recognition.

Blaise Aguera y Arcas is an architect at Microsoft Live Labs, architect of Seadragon, and the co-creator of Photosynth

Recently Mike Laurie wrote a post about the use of video, virals and such like and he ended the post highlighting the Photosynth software developed by Microsoft. The video shows the potential of this software and it’s impressive. One quality is the ability of a system to collate images from Flickr that have been tagged with recognised terms and build those images around a 3D model.

Another element is the resolution of some of these images is 300 megapixels allowing a user to zoom right into them and read or see the contents.

Photosynth showing Flickr images mapped to 3D models of the subject

If you think about this collation of files that have been tagged by a massive variety of users (from Flickr in this instance), it seems to be a big step to Tim Berners-Lee’s vision of the semantic web.

Sure these are only pictures, they are not documents, however when you think about the metadata in the file and how this can be organised to conform to a widely accepted mental model then this is really exciting.

An interactive mental model

Imagine an interactive concept model around a physical object, you could extract the related items around this and draw relationships between inter-linked entities. I recently designed a taxonomy for a science magazine, and it had to encompass every type of science from physics to psychology to civil engineering. How would you draw relationships between these fields?

Well, what I like about this model is that this would be a visual representation of a knowledge landscape (in this case using images) that could easily be with video, audio and standard web pages. It would also be a 3d representation that would encourage digital discovery.

At the moment it is perhaps too flexible for your average user, but give it time. As more digital natives reach maturity and form the majority of the browsing public, this interface will not phase them at all.

Google’s repository of human entered queries

We have a vast amount of data against document types about relationships between one and another, search engines have log files where keywords and most relevant results are displayed.

Around any search, even if it is a mathematical equation, there is a physical object that can be related to it. Be it the theory’s creator, university, or even the theory itself the physical entity (or even known concept) could form the basis of the visual model.

If we truly want to move towards a semantic web than this type of interface would offer a rich, interactive and flexible approach to showing layers of detail that would encourage digital discovery and serendipitous finding.

Producing a list of most relevant links is still a compromise to what we could display to our users. It could be far better to show a knowledge landscape for each query where there are paths to other areas of knowledge and layers of related data that can be sorted by a series of user interactions.

In a way Microsoft has produced a microfiche for the 21st century, the difference though is the librarians that have tagged it are now the users of the system and they create the content. The machine has enabled the creation of something that is entirely user  generated but it will also help in organising this huge potential of harnessing the world’s knowledge.

Consider how Google are also integrating wikipedia entries to their maps and geo-locating photographs. By using the map, as mental model in this case, they are merely super imposing extra data types directly onto the two dimensional base.

Perhaps this is the greatest challenge to interaction designers and visual thinkers. Visually represent the knowledge available from an interlinked network of sources that are authorities around a subject area. As Alex Isold points out we are far from the solution yet.

Tweet This Post

A year ago I was involved in a major restructure of 7 major websites. Each had a new taxonomy and controlled vocabulary created. A clear vision of the direction of each site was drawn up and site maps and wireframes produced. The one problem we had, was there wasn’t a generic metadata schema that was adhered to. It was a combination of ad hoc, legacy tags. Some originated from the SEO team and some from the developers and database administrators on each site.

The problem was, to enable users (internal staff and external consumers) to be able to find what they needed and to discover content serendipitously, the metadata needed to be rigorously structured and adhere to recognised standards.

Various kinds of metadata were important to the sharing and re-use of articles, and this post will highlight these areas and also illustrate how we incorporated these into the metadata schema.

The importance of metadata

The lack of metadata in a file, adversely affects search engine retrieval but also the working efficiency in collaborative teams. Data is lost, assets are not published and effort in creation is wasted. Ensuring effective metadata implementation in the documents we create and publish results in gaining competitive advantage in the search domain but also an increased efficiency in our working practice.

Without metadata management intellectual property rights become eroded, and liability increases. Files such as an image, PDF, video or audio all need to be tagged to provide the user or employee a method in finding valuable content.

A set of Metadata Standards should govern the implementation of consistent and uniform metadata architecture. Consistency in metadata is important to enable information sharing across an organisation and to make optimal use of document management tools which rely upon this.

For metadata to be effective, it must be incorporated into the workflow from creation to publication. This emphasises the importance of any content producer making a concerted effort to synchronise their information management.

There are significant contributors in metadata standards and after analysis the following organisation’s standards formed a basis for the metadata schema.

These are:

• Dublin Core Metadata Initiative (DCMI)
• International Press Telecommunications Council (IPTC)
• Publishing Requirements for Industry Standards Metadata (PRISM)

These standards are evident in the following formats that create types of metadata:

• Adobe Software’s Extensible Metadata Platform (XMP)
• Exchangeable Image File Format (EXIF)
• Resource Description Framework (RDF)

To aid in classification of file format types we used the Internet Assigned Numbers Authority (IANA) to give us a definitive list of file types. Its list, the Multipurpose Internet Mail Extensions (MIME) covers various formats that were used and may be used in the future. All other taxonomical types of labels (countries, regions etc) we looked towards the ISO for their classifications.

There are four main areas of metadata in terms of semantic grouping.

These are:

• Descriptive metadata – Describes the contents of a file
• Administrative metadata – Data that can not be retrieved or inferred from the content and pertains to management of the content within a system
• Rights metadata – Asserts the ownership of the content, who owns it and who may distribute it and usually pertains to the usage of the document.
• Technical metadata – data about the physical properties of the content

All four areas have specific use for different reasons. The descriptive quality helps the item properties to be found either by search or by a user interface element on the page. This can be online and also offline in an application such as Adobe Bridge.

Drafting a schema

The schema was designed to incorporate the four main uses of metadata. These different facets to metadata result in different uses throughout the workflow of a publishing process.

These are revealed through tools the Adobe suite of applications and in the content of the websites where the ability to find our information is paramount to the quality of the user experience. The importance of classifying different aspects of our content is becoming increasingly important as new technologies based on the XML platform come into fruition.

The properties are unique and their use is dependent upon the content of the resource that is being tagged.

The list below outlines what we thought was of most use to be tagged from the moment content was created to the moment it became published.

The image below shows exactly how these fields are applied in practice.

Risks of not managing your metadata

Unless a unified metadata strategy is initiated there are risks that;

1. The data communication between enterprise applications will be flawed. Thereby negating any efficiency that may be gained through hardware upgrades, workflow will be inefficient, content will reside in silos .

1. The investment of designing taxonomies needs to be implemented at a technical level and this requires taxonomy management and data architecture. Not following up the investment of a metadata implementation by employing a data architect or someone responsible for metadata management is a critical weakness in the enterprise’s information management.

1. Little or no use can be made of existing content and it often presents a valuable commercial resource. As the file has substandard metadata we immediately lose our content but we need to profit from its creation and value. Without a way of implementing a schema, content is effectively lost as soon as it is created.

With so many content creators working daily, the management of this information is our most important challenge. The content needs to be found easily, both internally amongst colleagues and externally amongst users.

The future

Certain properties in the schema pave the way for future search technologies. For instance entity type (such as brand, product, location, event) are complex in their variety. However this label allows that complexity of context to be stated and clarified.

Perhaps its for this reason that I see metadata being the foundation of semantic search. Only with a rich metadata schema that incorporates several different facets will we start to enjoy highly advanced searches over content that has inherent relationships. The challenge for interaction  and interface designers is to design the interface to be intuitive and allow searching in a unique, non-text field way, that is more exploratory than is possible at present.

Tweet This Post

Tagging images at source enables a standard to be reached when all images are used during a production process, both for print and online. If they are tagged properly, valuable metadata can be captured inside the JPEG file that can be read by applications later, either on or off line.

Optimising images enables;

• · speedy retrieval of images from within an image archive
• · resource sharing across different departments or companies, allowing a greater library of images available to all
• · Metadata inclusion within the files themselves. Enabling a higher level of findability when the files are used online within websites

To ensure that this occurs, guidelines should be followed by the image creators and those who manage the administration of these images. As tagging images can be highly subjective the guidelines drawn up here are designed to be quick and simple to create within the Photoshop File info panel.There are four main fields that need descriptive information to ensure images hold metadata that will provide SEO benefit and enable the files to be found easily.The fields are;

• · OBJECT – What the image is of. Is it a person, animal, logo, building or machinery?
• · LOCATION – Where the photograph was taken, this is an entity type that helps searching for specific images around a subject area by specifying geographic place. For images such as graphics, this descriptor will not be necessary.
• · DATE – The date on which the image was created.
• · REASON – The purpose for the image being created and what the image illustrates (seen in the description field below). This is another entity type.
• · KEYWORDS – This enables searching for the image within an archive and externally

The image above has been taken for use in a magazine and online. As it is a generic image of food there is also a good chance that it could be searched for. Once the image has been downloaded from the camera it is here where the initial information needs to be added via an application, preferably Photoshop.Using the File Info command a box will appear like the example shown below. Within the first option, Description, all the metadata can be added that will aid SEO and will be maintained when taken into other applications such as Fotoware. The first option in the menu (description) uses the metadata to populate other areas, such as IPTC info areas further down the option list.

1. Document title field

This is effectively the OBJECT attribute mentioned earlier. It must be a concise description of the image contents in as few words as is possible. This will also form the JPG file name and for this reason has important SEO implications. In this example the image is called rhubarb_diced.jpg. The author’s information can also be added under this field.

2. Description field

This is the area that holds information about the LOCATION and REASON for the image’s existence. It also holds information regarding the DATE. The description should be targeted and concise. The description field should not be used as a further way to inform the art editors on the picture’s usage. This should be done in a separate document as all date within file info refers to metadata and will remain with the image after it has left the production desk.

3. Keywords field

For search engine optimisation and search through the Fotoware’s online interface – FotoWeb. These keywords are really important as they define the ease in which a user can find the appropriate image.

The image above shows a user searching for images that have been tagged ‘Langley Vale Farm’ and Fotoweb has returned 640 results. To narrow the search the categories list on the left hand side enables a user to drill down into results that have been put into relevant sections.

Clicking on ‘Land’ enables the user to see the 18 images that have the keywords associated to them. Clicking on a thumbnail allows the user to preview the image and also to examine the metadata. Here we can see that the term ‘Langley Vale’ has revealed the image of the farm on the Langley Vale estate.

4. Copyright field

It is important that the correct copyright information is displayed and Photoshop can help ratify if a copyright license agreement is in place. The © logo will appear next to the file name of the Photoshop file. The file info panel will allow the creator or the picture editor to establish what copyright information is in place.

Where possible, images should be tagged following the what (OBJECT), when (DATE), where (LOCATION) and why (REASON) of the image. Where images have rights management issues keywords and copyright information will help give a minimum standard to the tagging of images.If this data is within the JPEG of an image then there is a much higher chance of the images appearing in a search engine results page (SERP). Tagging images this way now, also helps in maintaining a standardised approach to tagging of images that will ensure placement in SERPs in the future as crawlers become more adept at reading metadata from different file types.

Tweet This Post