Ofness

Public Art Works for 160 € (Euros) a year

Proprietary digital art collections make me mad.  As do all the art resources online that cost money.  It’s my open source, information should be free, mentality, I guess.  I understand that everyone needs to make a living, and that it takes time and resources to put together and maintain a digital collection.  I don’t have an answer to this paradox.  But I didn’t say I did.  Nevertheless it makes me mad.

Here is an example: The art-public.com online library. As described on its website it “focuses on the role of art in urban projects.  Recent creations and works in progress are organized by category and illustrated with examples from around the world, with an emphasis on the principal cities of Europe.”   The website also claims that “all subjects feature examples of artwork with commentary and color photo illustrations.”   A list of the types of works included is at the end of this article.

With respect to the main subject of Ofness, the searchability of the art in the collection, the following information is provided.  “Information is stocked in a database structured around three index fields – artist, location and commissioning sponsors – which provide access to representative articles including practical description of artwork (artist, title, year, location…). This information is complemented by a collection of images. Each work of art is illustrated with up to six color photos; commentary is provided in several languages.   Through the SIGAP [System of Information and Administration of Public Art].system, the database can be accessed via keywords.”

The enhanced content, the articles, biographies, links, and data organization sound good,  Valuable material indeed.  But it is only as good as the metadata.  And the site doesn’t tell us much about its SIGAP system.  A search online revealed no such system.  I did turn up a SIGAP site, but it is “a non-profitimaking organization for serious ufologists,”  http://www.sigap2.talktalk.net/   So how good is this sites system?  How many keywords?  Who created them?  Are the keywords drawn from any standard controlled vocabulary?  Can they be mapped to other systems?

Moreover, not one single example is provided to illustrate the usefulness of the collection, its searchability, or its contents.   This may just be a product of poor marketing, and a lack of understanding about the best standards for digital collections.  But I maintain that this should not be occurring with substantive collections in this day and age.  Especially when they are not free.  Finally I can’t help noting the contradiction between the claim that “Each work of art is illustrated with up to six color photos” and the assertion that it contains more than 8000 works of art and 5500 photos.” (I note that a far lower number appear in a pop up box on the site, but assume the larger number is correct.) They can’t both be right.

The kicker for me, however, is the scope of the collection versus its price.  As of  2004, the site claims it has more than 8000 works of art and 5500 photos.  It sounds like a worthwhile and comprehensive project.  But the price?  “Annual membership is 180 € (Euros) V.A.T. incl.”   Wholly apart from the exchange rate here in the US, that seems pretty steep to me.  Accordingly I can’t review the collection itself.   It sounds interesting though, and regret that I can’t even take a peek.

This is the list of the types of public are that are included:

• Public Areas (civic squares, urban paving, ornamental pools, fountains, waterfalls)
• Street Furniture (benches, fences, gates, entryways, streetlights)
• Murals
• Light Works
• Roadway Infrastructures (highways, roundabouts)
• Engineering Structures (footbridges, tunnels)
• Utility Structures (water towers, ventilation shafts, power stations)
• Underground Areas (underground car parks, pedestrian walkways)
• Memorial Sites (commemorative monuments)
• Historical Heritage Sites (stained-glass windows)
• Industrial Buildings
• Shopping Centres and Business Centres
• Administrative and Institutional Buildings
• Educational Buildings
• Buildings for Social and Cultural Use
• Hospitals
• Sports Centres (stadiums, swimming pools)
• Parks and Public Gardens
• Playgrounds
• Rural Territories and Nature Sites
• Open-air Museums
• Major Urban Projects and Land Use Planning (new city and new district development, regeneration of old areas, restructuring of inner-city districts, regeneration of low-income neighbourhoods and housing projects, redevelopment of industrial sites and buildings)
• Public Transportation (tramway, subway)
• Municipal and Regional Public Art Policies

Pig in Bath England Photograph by Marcio Cabral de Moura

As an aside I have to note that there are probably vast numbers of pictures of these works.  With geotagging they will probably all be available to the interested searcher without too much difficulty.  If the mash-up doesn’t exist yet, I suspect it is on its way.  So what this site offers it the compilation of the images and related material.  Only as good as the metadata that organizes them.  I can’t render a verdict on that.

In lieu of images form this locked up tight site, above is an image of public art from Flickr.  Thank god for Flickr.

A Mini History of Pre Digital Image Collections

Ancient History

Prior to the development of digitization techniques and the internet, image collections, typically slides, were organized by individuals who used the slides for teaching, or by individual holding institutions, often in unique or idiosyncratic ways.  The systems for organizing these slides were relatively simple and had limited access points.  While there were common elements, there were many differences, both in the character and depth of the organization.

Photograph by Night Owl City under Creative Commons License 2

Recent History

Since digitization of images emerged radical changes have occurred.  Many associations involved in this areas, of professionals who handle these collections, of institutions who house them (including colleges, universities, museums, and archives), and of researchers who use them, turned their attention to the ways that digitization can be harnessed.  Individually and collectively they have developed planning procedures for large-scale conversion of analog images to digital format, systems for organizing and managing these images, websites for sharing them, and protocols for exchanging metadata.  There are a host of open source and proprietary tools supporting these many efforts.

Why Metadata Matters

The ability to actually retrieve and use these heritage images ultimately depends on metadata.  Andrew Wray put it well when he said that metadata is the “invisible glue that hold the system together and makes the massive amounts of digital information usable.”  In the case of images, access is even more dependent on metadata.  Text files, even if not indexed, tagged in XML, or otherwise cataloged can always be subjected to a full text search.  Images, on the other hand, are basically useless without added metadata.  While direct (or content-based) image search technologies, which use complex algorithms to determine the subject matter of digital images, exist this technology is still in its infancy, and there is general agreement that it could never replace the detail and scope of individual cataloging for heritage images.  The creation of meaningful metadata is thus one of the chief challenges facing heritage image collections.  This is a challenge for both technology and for the development and management of human resources.

Utopian Image Access - WorldPic, Perhaps?

The Dream

The potential usefulness of well cataloged images seems enormous. Heritage Images, particularly, are of great, if not essential, value to scholars and researchers. Images provide information that cannot be communicated by text and they transcend many of the barriers of language. Images can open up avenues of thought and research previously unavailable, not just within the narrow category that the image may be typically associated with (whether that be art history, architecture, geology, or some other area) but across disciplines, and into wholly new areas.  While cross disciplinary studies are not new, ready access to the whole store of images would greatly facilitate and stimulate further work by artists, economists, historians, and doctors, among others, independently and collaboratively.

Women at work on bomber, Douglas Aircraft Company, Long Beach, Calif. (LOC

Leaving aside the myriad barriers posed by expense, expertise, technology, and legal considerations (including copyright), and the differing views about priorities and cost allocation, most interested parties could probably agree on the following as an ideal:

The Ideal for Image Access

Universal, scalable, trustworthy, sustainable, and permanent access to all heritage digital images from the publicly accessible to the privately held, including those from all museums and cultural institutions, archives and libraries, private collections, and publishers, using a system of organization that maximizes entry points and interconnections between objects and relevant texts, to facilitate retrieval for all imaginable purposes, through interfaces which have the capability of searching across all platforms, providing high resolution images, thumbnail browsing, copying and reuse for non-commercial purposes, and links to contextual materials, with continuously updated information of the applicability of worldwide copyright status and contact information for obtaining permissions.

WorldCat 1.2 Billion Item Available here

In the Meantime…

This ideal is more utopian than anything, but it provides a context for looking at what exists, and what is evolving.  Having an ideal toward which we strive is critical, even if entirely impractical at present. It seems to me that the steps that are being taken now should at least consider the ways in which they will impede or facilitate movement in this direction.  We must keep the big picture in view.  The broad access to books and periodicals we currently enjoy grew gradually out the hard work of collaboration across borders and in the face of differing practices.  We take for granted our ability to search books and periodicals by subject in a unified catalog, WORLDCAT.

I’d like to imagine that we took for granted a similar level of access to images.   Any proposals for a good name?  Is there anything you would add that I have overlooked?

Playing Art History on Artigo

How it works

Artigo is another site that, like Labelme, Peekaboo, and steve.museum, is geared toward using the collective intelligence of internet users to gather data that will improve image search functionality.

Artigo is most like steve.museum in being semantically oriented. As at steve.museum, you provide your own tags for the image that is presented. Artigo’s image database contains 15,000 images and is set up at the University of Munich. But there are several refinements that make Artigo game-like, and arguably more effective in obtaining valid results.

Each game session is timed; the players have 5 minutes. The number of pictures reviewed is a function of how quickly the players agree to move on to the next one. Generally you see 5-7 images per game. There are TABOO words, ones that have already been associated with the image. These appear below the image in RED. Players must dig deeper to find appropriate tags.

The more taboo words there are for an image the more points you earn for the tags you provide. Most importantly your tag is accepted only if you and your partner have both provided it, thus weeding out the chaff.

As you play a countdown of the seconds remaining appears at the top, and your points up to that point on the bottom right. You can see how many words your partner has selected, they appear as BLUE dots on the left, but not what they are. But when the game is over, each image is shown again with artist, title and date identified.

You also see the tags given by your partner and yourself, and any that agree are shown in YELLOW. They are the ones that you get credit for. Of course you get your final score as well.  Artigo also offers a monetary inducement to play. “Each month, the player with the highest score of the month will receive 50 euros!” says the site. My level of play is way too low to ever win. The highest scorer this month to date has 10,510 points. I can only imagine a 24/7 game going on with numerous players. The interface is simple, the explanation straightforward, the game interesting.

Background

The creators credit “Luis von Ahn whose ESP Game inspired us to create Artigo” and refer users to his talk on Human Computation, which is really interesting. A paper relating to the work of Luis von Ahn’s and his collaborators further elucidates the principles. I should also note that Peekaboom is now called Sqigl, has a slightly different interface and is bundled with 4 other Human computation games, three of which are image related, at I’ll have to get back there before reporting further.

Thoughts

One of the key benefits of these types of games is that they are so much fun that no financial inducement is necessary. I wonder whether the Artigo monthly payout actually promotes greater use, and whether determining this is a part of the project. Unfortunately Artigo doesn’t provide any information about how the project is going, about resulting papers, etc. Nor do they even provide an email address to ask further questions. This, as I’ve noted elsewhere, is frustrating.

Thesaurus of Graphic Materials 1, 2, & 3

The image to the left is a screenshot of a partially expanded, but truncated (”A” to “R”), hierarchy of the first (of 26) main subdivision of the TGM1: Subject Terms “Activities”.  The list spans A to R.  This screenshot is not from the, most user unfriendly, albeit official, interface of the The Thesaurus for Graphic Materials (TGM), which is maintained under the aegis of the Library of Congress.  Rather it is from a free vocabulary site, TemaTres Vocabulary Server which is much easier to navigate.  That’s why it’s here, to give a meaningful sense of its scope and contents.

As described on the official site:

“The Thesaurus for Graphic Materials is a tool for indexing visual materials by subject and by genre/format. The thesaurus includes more than 7,000 subject terms and 650 genre/format terms to index types of photographs, prints, design drawings, ephemera, and other pictures. In 2007, the subject and genre/format vocabularies, previously maintained separately, were merged into a single list and migrated to new software, MultiTes.  Other minor changes are clarified in the links below. For questions about the thesaurus, contact TGM editors at: tgmed@loc.gov.”

The merger of the subject and genre/format terms in a single thesaurus simplifies things in some respects but it leads to some confusion about these areas, which are really distinct.  It is worth noting that the genre/format term list can still be downloaded separately, but the subject terms cannot.  The logic of this choice evades me.  While a single list may be beneficial for site users in searching the LC assets, there appears no reason (except possibly for some additional maintenance work)not to also keep each list available separately.

The official TGM  site, however, contains only a blind search function for the new combined TGM, on the one hand, or the ability to download either a text of an XML file of original TGM1: Subject Terms (which it what I’m really interested in here) or TGM2: Genre/Format Terms. It isn’t clear at the present whether the downloadable files will eventually reflect the merger of the two lists or not.

While blind search boxes have their place I think they should always be accompanied by a complete list of terms which allow the hierarchy to be navigated.  The lack of such a tool on the official site it frustrating.  But there is an alternative.

At TemaTres Vocabulary Server one can find both TGM1 and TGM2 in both a purely alphabetic listing or in an expandable hierarchical list, each of which in their own way allows for a much more satisfying and thorough search of terms in its own way.  I discovered this source only after reaching my frustration level while trying to strip everything but the alphabetic terms from the XML, in preparation for a comparison of the TGM1 with other subject vocabularies.  Thank goodness!

Here is a A list of the highest level categories of the TGM1, as set forth on the TemaTres site.

Activities

Audiovisual materials

Behavior

Body parts

Children playing adults

Concepts

Economic & social conditions

Insignia

Materials

Mental states

Natural phenomena

Objects

Organisms

Patterns (Design elements)

People

Periodicals

Physical characteristics

Physical geographic features

Signs (Notices)

Smoking paraphernalia

Sounds

Sports

Staffs (Sticks)

State medicine

Supernatural

Surplus commodities

Surveillance

Swimming

Symbols

Tapping

Teeth

Tenure of office

Testing

Textiles

Therapy

Tiles

Time

Tires

Tobacco products

Toys

Transportation

Travel

Twins

Undertaking

Vacations

Vehicles

Vessel components

Veterans

Violence

Wages

Water use

Weather control

Wigs

Woodwork

Wounds & injuries

Writing materials

Other Observations

My brief exploration of this hierarchy revealed four levels to this hierarchy, but I can’t rule out the possibility that it goes deeper in some cases.  One limitation of these version of the TGM is the discrepancy that may exist between it and the LC version.  On my vistit on 07/23/06, the “About” note on TemaTres listed a date of creation of 04/07/07, with no indication as to how frequently it will be updated.  So you may miss a few terms here, but you’ll have a much better experience.  Actually, when you consider the poor interface on the LC site, you would likely miss even more terms on the official site.

My thanks to TemaTres and its MUCH better presentation of the TGM. This tool makes it easier to contemplate my plan to compare it with other visual material subject vocabularies.

26 More Vocabularies

Finally I can’t help noting that TemaTres offer another 26 vocabularies in the same format as well as other resources, how to, and FAQ’s.  Included are a mammalian taxonlomy and the Unesco Thesaurus. More on the TemaTres Vocabulary Server and other vocabularies down the road.

TASI - A Great Portal to Image Collections

One of the best.  The TASI list of digital image collections is big (800 collections) well organized, and has a short abstract of each collection including  ……  It is also browsable and searchable. It is heavily weighted with British and Canadian collections, which makes sense since TASI is a British institution, but it has a broad range. Over 800 digital collections are listed organized into 12 classifications (lArchitecture (62), Architecture (11), Art (1),Art and Design (182), Entertainment (52), Geography (134), History (261), Literature (62), Religion (41), and Science (112))  I have no idea why there are 2 architecture classifications.  You will have to figure that out yourself.

Here is a view of the info page for an individual collection which gives a good idea of its scope.  Note that this is only one of the many valuable resources on this site relating to image searching.

TASI also offers great guides for the following topics: Finding images online.

* Finding current events images
* Finding art, architecture and design images
* Finding historical images
* Finding scientific images
* Finding maps
* Finding images on Flickr
* Images in blogs and wikis
* Photo sharing sites

TASI has much more than this about digital images and collections but you’ll have to visit to scope the rest out yourself.  I’m really trying very heard to stick to SUBJECT MATTER IMAGE SEARCH.

Nice feedback,BMA!

A current exhibit at the Brooklyn Museum of Art (BMA) called CLICK! is a great example of the creative role that museums can play in utilizing the web to involve users.  Click! is a multi stage project.  First, there was an open call for photographs for the exhibit on the theme of the “Changing Faces of Brooklyn.”  Then an open call for evaluation of the submitted photographs online was issued.  Finally an exhibit at the museum of the top 20% of the 389 photos submitted was assembled.  But that isn’t all. There is also an online exhibit of all the photographs, and multiple statistical views of the voting.  These results include very interesting data about the relative ratings of the raters (self-identified in 5 categories from naive to expert), along with a host of other data. I found this information as interesting as the exhibit itself.

Begging won’t help.  You’ll just have to visit the site yourself to find out how the ratings of experts compared to the ratings of the non-experts.  

The site has lots of other great features, including an intelligent commentary, a look into the process of mounting the physical exhibit, the comments of raters about the photos, a blog, a podcast, and, novel to me, a print-on-demand catalog! Of course, there are also the photos themselves.  There are some terrific ones here - well worth the visit.

CLICK!’s Image Retreival

What makes this exhibit interesting from the standpoint of image retrieval is that the results of the rating process are made available to everyone.  Probably because I had recently been visiting some of the research focused CBIR sites I couldn’t help contrasting this level of information disclosure to the minimal effort given to providing similar data to the user/contributors to the CBIR data collection sites i am familiar with.  Granted the purposes are different, and the CBIR sites may want to keep the raw data of their research private.  But some level of sharing would be interesting.  Perhaps they would do well to take a lesson from BMA.  On the other hand, even the Click! exhibit could have done more. As far as I could see, there was no opportunity on the CLick! site for the raters (or for that matter the photographers) to tag the images.  That would have generated some interesting results as well, and perhaps would lead to more visibility online.

Historical Note on these Images

As I’m a Brooklynite, and as this site is about the subject matter of images, I will add a historical note about the SUBJECT of three of the submitted images.  This is the new Richard Meier building at the intersection of Eastern Parkway and Plaza Street (now denominated “1 Prospect Park”in a new application of tagging) which is just a short walk from the BMA.  The first photo, from the CLICK! exhibit, appears at the top of this post, courtesy of the photographer, Elyse Taylor.

Whether you love the Meier building or hate it, and I know folks of both persuasions, it is worth noting that the site of this building, on an especially beautiful and prominent corner right on Grand Army Plaza, which centers on the Soldier’s and Sailor’s Memorial Arch (Brooklyn’s answer to the Arc de Triomphe) was a blacktopped parking lot until sold by Union Temple (which has occupied the plot next door and owned the Meier building site since circa 1925).

The original plan for the site where the Meier building is situated called for the construction of a sanctuary for Union Temple.  Unfortunately, shortly after the erection of the Temple House came the stock market crash of 1929.   The site remained undeveloped for 90 years a a result of the depression of the 30’s, WWII, and then the massive migration of people out of New York City to the suburbs during the fabulous 50’s and 60’s.  Here is a rendering by the architect of the original design for the original vision for this site.  Below is a view of Grand Army Plaza from the roof of Union Temple on a glorious fall day.

I’m a big fan of different points of view.  Yours are welcome too.

Why Ofness?

My interest in the issues around retrieving digital images evolved from my interest in art, text, the intersection of the two, a project, a penchant for organization, and a library science degree.

Several years ago I learned Photoshop with a particular project in mind. After working on that project for a year and a half, I had a new idea. This idea was to create a large rose, using one of my watercolors, but to make an image of the rose exclusively out of words. The original inspiration for this was my discovery of micrography. Not the micrography that is about photos of microscopic things; rather the micrography that has been practiced for hundreds of years by Jewish artists. Jewish micrography creates images related to the texts from which they are made.

For example, as shown here, the text of the book of Jonah is used to create the boat on the water,the whale, etc.  The text is written in calligraphy but in a virtually microscopic form. From what I understand this art form was an outgrowth of the prohibition against making images. An online exhibit of micrography including the above image can be found at the site of the Micrography Exhibit Jewish Theological Society" href="http://www.jtsa.edu/prebuilt/exhib/microg/index.shtml" target="_blank">Jewish Theological Seminary

My initial idea was to create the rose using the genome of the rose expressed in the letters A, T, C, G. But as i researched online I found so many texts that refer to rose. I gradually came to appreciate, in a way I never had before, the rich symbolism of the rose, which dates back as far as recorded history. My online research continued; I purchased books on the subject; and I found references to roses everything: newspapers, films, television. I even quite coincidentally stumbled over references to roses in books where one would not expect them. I found some wonderful references. Then my concept shifted. I decided to use the texts I found most interesting to create the rose. This project too took a long time, but ultimately I completed it and was very satisfied.

Fast forward. A few years later I decided to got back to school in Library and Information Sciences. I had thought i was a pretty good online researcher.  I discovered how little I knew and became even more involved in general exploration online. Whenever I explored a new database, website, or library, I would search for “rose.” It had become rather a habit. This was actually quite useful. It was a controlled experiment for comparing (in a rudimentary way) the usefulness of these resources.

My interest in art and art history led me to research the intersection of copyright law and online access to images. This led to deeper exploration of museum and other digital image collections, where i found such a wide variety of practices. On and on, from one area to another. I looked into DAM’s and had the opportunity to use TMS while volunteering at an art museum. A paper on access to images in scholarly journals and books took me deeper into the problems involved in providing meaningful access. I ended up immersed in the metadata resources used to classify art images, While I learned alot about the many kinds of information that can be preserved in metadata, my interest, as a layman, was primarily with subject access. More on why in another entry. Ultimately I learned alot about the English based terminologies and schemas, and ultimately introduced to the concepts of Ofness and Aboutness.

Digital Asset Managment Systems

Digital Asset Management (DAM) systems have become a necessity for large image collections, because large digital collections simply cannot be managed and the digital resources used efficiently when the digital files are stored in a simple hierarchical file system. DAM’s are systems that make the digital assets of a business or organization more accessible to all those in the organization, and sometimes to those outside of the organization. There are both commercial and open source DAM’s, off the shelf systems and customized ones, small desktop DAM’s and large institutional ones. In some cases the server is located and maintained on site, but there are now also available vendors who provide scalable off site storage.

DAM technology grew out of the growing need to find, access, share, repurpose, and cross reference, extensive digital assets. While the major DAM’s share many features and characteristics, specialized features to satisfy the needs of different clients types are available. Some DAM”s have a special focus on images, others in rich media like video. Large businesses requires more workflow and end-to end solutions. Organization with extensive image collections such as libraries, archives, and museums require DAM’s that can both handle the projects for digitizing existing items but may have less need for features important to fast paced businesses.

The websites for DAMs do not reveal many specifics about the cataloging of records, and what features might make this process quicker. To get that sort of information one probably needs to schedule an appointment with a sales representative, and then be ready to cross examine him (or her). But even without specific knowledge regarding this, it is probably safe to say that competition in the field will lead providers to stay on top of technical developments and to be ready to incorporate improvements to their systems.

Internet Users Contribute to CBIR Research

CBIR (Content Based Image Retrieval) is an extremely busy area.  A wide variety of approaches are being explored.  All of these approaches, of course, require a great deal of data obtained from individuals.

Peekaboom screenshot

One of the tools being used for this research is, not surprisingly, the internet.  Researchers have developed websites encouraging individuals to contribute to the data they is needed for their research. Peekaboom is one example of the approaches being taken to image identification. Here users log on to be matched in competitive play with others in a game that involves revealing and identifying parts of images, and giving clues. With a nice and easy interface and a good challenge, the site has obtained a lot of data, and has created some Peekaboo addicts.

Labelme is another example of this use of the web. As in Peekboom, researchers are trying to build a large collection of annotated images.  Labelme. however, is not a game like Peekaboo, and it is thus not as much fun.  Visitors to the site trace the outline of as many different objects from different images as possible and label each one with appropriate terms.  This includes labeling parts of objects.   There is a guide explaining how to handle overlapping objects and simple explanation for how it works. With some additional software one can also view the images in 3-D and even print out a Pop-Up version of an image to create a 3-d model.  You can even upload your own images for similar treatment.  These perks add to the attraction.

As you work with Label Me it is  interesting to reflect on how you identify “objects.”  Some things are straightforward: a car, a house, a road.  But others are not.  where do you assign the boundaries of a leafy tree?  Would you identify a reflection in a window as an object? If is clearly identifiable? If it is an amorphous shape? The shadow of an object?   But the wisdom of the crowd as to what is label-able is a part of the data, i suppose.

I do wish that there were more information for the user/contributors available on these sites about the data that has been collected so far.  Some preliminary analysis of the data would be nice to have as well.  From my perspective such feedback is part of the incentive to participate and might provoke comments for the users that would actually be helpful to the researchers.

Definitions of “Ofness” - Not

The word “Ofness” (hereafter simply Ofness), previously unknown to me, became a familiar word as I delved into the subject of image subject retrieval. The clearest discussion of its meaning that I have found online can be found in a wonderfully illustrated book produced by the Getty: Introduction to Art Image Access by Sara Shatford Layne, Patricia Harpring, Colum Hourihane, Christine L. Sundt. The term is used pervasively in art and art history journals.

It was thus rather startling to me when I discovered that the word “Ofness” cannot be found in standard reference works.

Meanwhile the word appears in lots of other places. A search at Amazon for Ofness turned up 144 books, including the examples below.

Philosophical Foundations for a Christian Worldview by J. P. Moreland and William Lane Craig (Hardcover - April 2003) Excerpt - page 136: “… obtains. The important thing to note is that propositions have intentionality-ofness, aboutness, directedness towards an object. In fact, it is because …”

The Cambridge Companion to Wittgenstein (Cambridge Companions to Philosophy) by Hans D. Sluga and David G. Stern (Paperback - Oct 28, 1996) Excerpt - page 100: “… extremely general ones about intentionality, the property of “aboutness” or “ofness” whereby one part of the world

Truth and the New Kind of Christian: The Emerging Effects of Postmodernism in the Church by R. Scott Smith and J. P. Moreland (Paperback - Nov 8, 2005) Excerpt - page 181: “… beliefs and emotions) have intentionality, which sim- ply is their ofness and aboutness. For instance, a feeling is about some- thing; …”\

Making it Explicit: Reasoning, Representing, and Discursive Commitment by Robert B. Brandom (Paperback - Nov 1, 1998) Excerpt - page 336: “… of which expressions are properly interpreted as expressing attributions of ofness or aboutness in the intentional or semantic sense. …”

Reclaiming the Center: Confronting Evangelical Accommodation in Postmodern Times by D. A. Carson, Douglas Groothuis, J. P. Moreland, and Garrett DeWeese (Paperback - Nov 9, 2004) Excerpt - page 96: “… is just a construction. On this view, intentionality-a mental state’s “ofness,” “aboutness,” or “directedness” toward an object (even if the object …”

The following online references have NO entry for Ofness: Wictionary.org, Dictionary.com, Thesaurus.com, Visual Thesaurus, Random House Dictionary, Webster’s, Roget’s Thesaurus, OneLook Dictionary Search, Bartelby.com.

A general Google search for Ofness reveals NOT a single listing.

Ofness is invalid in the the WORD spell check function, though it has now been added to mine.

I did find Ofness in Wordie, a site where members can share favorite, interesting, and unusual words. But it was on only one list, that of Made-up Library/Information Science Words.  No further comment.  “Aboutness” was also on this list. a disgruntled library science student perhaps? It was this entry that actually sent me on my search for this word.

A closer look at these 144 books revealed that many were listed only as the result of some truly appalling and uncorrected OCR scanning. But many, especially in the area of philosophy, were entirely valid. The plot thickens. Is this the context in which the word was first used?

A search of Google Scholar retrieved 188 works, which like the Amazon results fell mostly in the categories of philosophy, art, information science and lousy OCR scanning.

Unfortunately I’ve been unable to search the OED. I’m still looking Forward to the day when one can purchase access to a single entry for a modest fee. Surely there is a cost point where this would be a win-win. Perhaps I’ll check this out, if I remember the next time I’m at the library. Meanwhile input from those who do have access is welcome.

Clearly Ofness is a term that has a long history and is important in certain realms of thought, if not in common parlance. Granted that general dictionaries are not going to contain every highly specialized word, but Ofness is such a useful word. Is there any other word in English that captures the meaning of Ofness? I cannot think of any. Are there any in other languages? It seems to me that it could be much more widely used. It’s the kind of word that makes you think.

Its absence from dictionaries and other reference works may account for the fact that the domain names “Ofness.com” and “Ofness.org” were still available in early 2008. For this i am grateful, but still puzzled.

I shall continue to follow the trail of Ofness and relate any new discoveries as they occur.

Selecting a DAM

Selecting a DAM requires careful evaluation of the institution’s needs. Among the selection factors are:

• Who are the users? How do they work? What is their expertise?
• What purposes is the DAM expected to fulfill?
• What types of file formats are used?
• Who will support the DAM?
• What will be done with the files?
• How granular will the metadata need to be?
• Is there a need for retrospective digitization?
• What is the size of the files to be saved?
• Does the database(s) need to communicate, or be integrated, with those in other institutions or organizations?
• Is collaboration with other institutions necessary?
• Will any of the database be made available online?

Tailoring DAMs for different situations

Museums, archives, libraries and similar institutions, are relatively slow growing, need extensive and flexible metadata fields for the large controlled vocabularies, do extensive digitization of existing collections, and may have some needs like commercial enterprises.

Large dynamic corporate settings, such as those involved with television, media, publishing, advertising, architecture, and film also have to deal with retrospective digitization projects, but they also have to deal with the continuing onslaught of new images which digital technology and computer networking have made so easy to create and share.

Other digital image collections, those of individuals, including professional and amateur photographers, entirely new collections associated with smaller institutions, need far fewer features than the others.

The complexity of image searching by subject matter or content.

In contrast to searching for textual material, searching for images requires bridging the gap between the visual world and language, a very complex process. Human beings are able to do this with great accuracy, but our understanding of how this happens and our ability to program computers to do the same is still rudimentary.

The process of retrieving image meaning based on syntactical image features, without the use of words, is a far more technically challenging task than creating algorithms for searching in textual material. Efforts have been and continue to be made in this area, which has come to be known as Content Based Image Retreival (CBIR). “The techniques, tools and algorithms that are used originate from fields such as statistics, pattern recognition, signal processing, and computer vision.” CBIR systems use a variety of approaches: color, color proportions, layout, shape. In some cases users select images and the software finds similar material. CBIR is used in some practical ways these are limited. It is a technology that is still very much in the developmental stage

The alternative to CBIR, language based image retrieval, which can take many forms, is the only really useful game in town. This means that somehow subject classification has to be done by human beings.

The Impediments to Language Based Image Searching

There is no dispute that the best way to search images is by the creation of relevant and thorough metadata. But there are a number of factors unique to images and existing image collections which makes the creation of such metadata difficult to accomplish.

First there is no single controlled vocabulary applicable to images, and no single organization principle for the classification system. The Getty Art and Architecture Thesaurus is the primary resource for cataloging this type of material. The Cataloging Cultural Objects (CCO) manual prepared by the Visual Resources Association, which is still in draft form, is intended to supplement Getty by providing “guidelines for selecting, ordering, and formatting data used to populate catalog records.” Both are works in progress, and explicitly recognize that catalogers need to draw on other sources for additional subject matter terminology, in particular the Library of Congress. Nevertheless, they are not in widespread use in the way that the Library of Congress catalog is with respect to published material. The incorporation by reference of other more specialized vocabularies, which may themselves not be authoritative leaves the whole subject to inconsistencies in the future. In any event, they have little or no bearing on other sorts of collections, such as those online, or in a commercial environment.

Equally important is the fact that even when a controlled vocabulary exists, subject matter classification is costly and time consuming because it must be done by an individual knowledgeable about both the image as well as the classification system used. Art librarians, art book catalogers and curators come closest to meeting this requirement in the areas of art. But thorough general image cataloging, and even art image cataloging, can be more difficult than book cataloging. As one professional in this area observed:

“the image world is just too complex. There are too many ways of approaching an image (from the standpoint of an art historian, a historian, an architect, a studio artist, a botanist, a medical professional, an urban planner, a political scientist, on and on). And really the image is often a surrogate for a thing. A surrogate for an emotion, even, let alone an “object” or an art work. And there can be so many things going on in an image that mean different things to different people or different disciplines. It’s just endless. Trying to harpoon that Moby Dick of a shared catalog is becoming an impediment to vibrant and accessible vehicle(s) for academic images.” email exchange on the Art Libraries Society Discussion List [ARLIS-L@LSV.UKY.EDU] November 7 -9, 2006.

The Simplicity of Searching For Printed Matter by Subject Area

Subject matter access to digital text documents may not be perfect, but it has long had the natural benefit that comes from having its content expressed in the same form as the standard search method: words. The use of keywords to search within digitized texts is simple and straightforward and, though limited, relatively successful. It is also fairly easy since librarians have been cataloging monographs, periodicals, and other published material in sophisticated subject matter classification systems for over 100 years and for over 30 years in a digital format. Subject matter searching within the MARC records can thus be done for virtually all published material. In addition, those records are accessible in single location, Worldcat. It is not necessary to visit numerous different sites, whether virtual or real, to find books on a select subject matter.

Since the vast majority of published materials are available in multiple locations around the world, a searcher can probably find what they are looking for.  With Google, and now Yahoo, having undertaken the massive digitization of the all the world’s published texts, the full texts of copyright cleared texts will become a reality, at which point keyword searches of all of this material will be able to be accomplished from one’s desktop. Here is one selection from a Google book search for “rose symbol.” Interestingly, this book, Advertising as Communication by Gillian Dyer, appears to address Ofness from the advertising perspective. I’ll have to get back to it.

Tags: Advertising, Google book search, MARC, Search, Text, Words, Worldcat

Challenges of Text Based Image Retrieval

The challenge for language based image retrieval is thus to find a way to do more quality classification, more quickly, and at no more expense. A tall order: At present a number of course are being followed to one degree or another include:

• Standardizing the controlled vocabularies in use,
• Simplifying the process of classification,
• Enlarging the number of human beings doing the classifying,
• Facilitating interoperability between systems, and
• Coordinating collaborative collection searching.

Many different groups and organizations have an intense interest in Text Based Image Retrieval. Among these can be counted developers of Digital Asset Management systems, Visual Resource managers in the many businesses already mentioned, Librarians, Archivists, Curators, and Academics involved with digital collections. Last, but not least, the general public, which has only the algorithms of Google, and Yahoo, Flicker and the rest to depend on.

Members of the Art Libraries Society (ARLIS) and the Visual Resources Association (VRA) are actively involved in collaborating on this issue which is critical to their fields. The Getty Thesaurus and the VRA Core 4.0 guidelines, both recognize the need for the controlled Getty vocabulary to be used consistently, to be supplemented by input from specialists in other areas, and suggest the use of the Library of Congress subject headings for subject classification. But these guidelines, like Dewey and the Library of Congress Classifications before them, can not be forced on any institution. There will have to be a period of education and acceptance, which the internet will presumably help to spread. In one lively discussion on the ARLIS listserv it was asserted that there is “some agreement that there is a need for both folksonomy and a structured approach.” But the mechanism for accomplishing this and other goals is still very much a work in process. Among the ideas being bounced around were “academic Wikis,” “OAI with some minimal interface/download tools,” “a scholarly version of Flickr,” “the CCO website,” or “STEVE.MUSEUM.”

How other interest groups are handling subject matter classification within their own organizations is not clear. Presumably in business the costs involved must be justified by the bottom line. It also seems likely that these organizations have developed a classification system tailored to their collection and needs. Consistency with other systems is probably not a factor. It would be interesting to get more information about this, and to see if the other concerns are even on the radar screen.

One result of the digital revolution that merged as an essential tool in the effort to improve access to digital images (among other things) is Digital Asset Management Systems. Many large companies are competing for business in this area, and more and more businesses, and non-profits with digital collections are finding that such systems are a necessity.

What is the subject of “Ofness”

The subject of “Ofness” is the subject of images, and more specifically the challenge of retrieving images by subject matter.   Ofness refers to one of the two elements or ways of describing the subject matter of an image as first articulated by Irwin Panofsky. The other is Aboutness. More on this distinction later, but for now suffice it to say that the simplest aspect of Ofness should be easy to identify and less subject to judgment.  It refers to the things, the objects depicted. (e.g. horse, man jar, table, sky, stream, television, etc,)  I have to note that I can’t help being tickled by Panofsky’s name.  It seems to refer itself to this subject which he helped to explore. All About (Pan) Ofness (of). I will have to seek out the meaning of the “-sky.”   Meanwhile I’ve been unable to locate a single picture of Panofsky on the web, quite startling in this context.  So I fell back on a book cover for an illustration.  It’s the cover of a paperback dating from 1955 of Panofsky’s Meaning in the Visual Arts

The widespread digitization of visual material is a relatively new phenomenon.  It’s use has rapidly accelerated as the technology improved, as the cost of memory plummeted, and as the networking of computers (including large local networks as well as the internet) have generated a widespread demand for it.  The technology to find images in this haystack of haystacks, meanwhile, remains rudimentary at best.

This site concerns the rapidly growing need for effective and useful tools for retrieving digital images, particularly by the subject matter or content of the image, the challenges involved in this area, the current state of tools and technology, ongoing research, and personal experiences and musings on the subject.

Searching for image files by other characteristics, such as creator, accession number, date of creation, copyright status, file size, exposure data, or any of the other of the many categories by which an image can be classified, while interesting, is not the subject of this site.

Having found many interesting sources on this topic, both online and in actual books and journals not available online, during the course [I originally wrote "curse"] of my explorations, and having found no website that pulled all these resources together, it seemed like creating my own would be a good way to organize what I have learned, and might do someone else some good. Hopefully as time goes by the disparate threads of interest I follow will coalesce into a natural organization and others, more knowledgeable than I on this subject, will contribute their comments, thoughts, links, articles, etc.

A partial and evolving list of topics that will be covered here include:

• Categories for the Description of Works of Art  (CDWA)
• TGM (Thesaurus of Graphic Materials)
• CCO (Cataloging Cultural Objects)
• LC Categories (Library of Congress)
• Iconclass
• AAT (Art and Architecture Thesaurus)
• CBIR (Content Based Image Retreival)
• Steve.museum
• Folksonomy
• Flickr
• Irwin Panofsky
• SKOS (Simple Knowledge Organization system)
• Illuminata ( A Proquest Product)
• VRA (Visual Resources Association)
• ARLIS (Art
• Dublin Core
• Photoshop
• Taxonomies
• ALIPR (Automatic Liguistic Indexing of Pictures)
• CLUE (Cluster Based Image Retrieval)
• SIMPLitcity (Semantics-Sensitive Integrated Matching for Picture Libraries)
• MINDat.org
• Memorynet.org
• Terrrgalleria.com
• DAM (Digital Asset Management)
• CONTENT dm
• MDID (
• Oxford University Press
• LUNA Insight
• Dspace
• VRA Core 4.0
• IRIS
• Extensis Portfolio
• Archivsion
• ARTStor
• IOS (internatiuonal Organization for Standardization)
• Chenault
• Garnier Thesaurus Iconographique
• AACR (Anglo American Cataloging Rules)
• V&A Subject Index for the Visual Arts
• Controlled Vocabularies
• Visual Image User Studies
• EAD
• VISION project (HW Wilson)
• Revised Nomenclature for Museum Cataloging
• Google Image Search
• CLiMB
• Union Catalog of Art Images
• WebSEEK
• CIMI (Computer Interchange of Museum Information
• Artcyclopedia
• DIAP (Digital Image Access Project)
• SASKIA Ltd
• Flamingo