WordPress.com



Creating the Online Collection “Snaps”Online Collection Assignment PaperLIBR 284-514 - DigitizationGroup 2Susan EdwardsStudent BStudent CStudent DIntroduction27946351930400Figure 1 - Cover of the Snaps photo album00Figure 1 - Cover of the Snaps photo album279463510160000This paper describes the process of creating an online digitized collection to represent a photo album that was found by one of the group’s members in a Los Angeles thrift store. The front cover of the album has the word “Snaps” written on it, which gave us the title for the collection. The album contains more than 70 pages with black and white photographs pasted onto each. Decisions about all aspects of this project were made via e-mail conversations, D2L discussions, and live online meetings. We created shared documentation using documents on Google Drive to collaborate on project documentation, metadata development, and this summative paper.Planning for DigitizationSnaps, the online collection for Group 2, consists of scans of pages from an album of black and white amateur photographs, and can be viewed online at . The album contains over 300 pages, including more than 70 pages with photographic prints, as well as blank pages. Group member ‘Student B’ purchased the album in 2010 at a Los Angeles thrift store, and the provenance of the photographs and the album are unknown. Most of the images depict people in locations that appear to be in Southern California during the early 1900s.We considered other ideas for our collection, including family photos, postcards, vinyl record covers, and compact disc cover art. We rejected the family photo idea because it wasn’t very original. Vinyl records were rejected because these would require an oversized scanner bed, a piece of equipment we didn’t have. And postcards didn’t seem like the best idea because some of us didn’t own very many postcards. We seriously considered compact disc cover art, but after much discussion, we could not decide how these would make a coherent collection, who our user groups would be, or why users would be interested in that collection. Some of us liked the idea of sharing our favorite musical artists with the rest of the class, while others thought our audience ought to be collectors or music historians interested in unique album cover art. Thus, we couldn’t arrive at consensus with any of these invented collections.Student B mentioned a photo album that he’d purchased in a Los Angeles thrift store, and this was the one collection we could all see ourselves working on. We were able to easily identify potential users of the images in this album. As Student B stated on September 9, this mystery photo album “is a more straightforward collection, which we don't need to impose a unifying theme upon. The fact that the album is bound automatically defines it as a self-contained collection.” That we could more easily arrive at consensus with this album brings to mind an observation one of our team members has heard. A director she worked with communicated to Student C that, where possible, students in the SLIS program should be encouraged to write about real collections or real repositories so that their effort is focused on the topic of the class, and not on creating a collection or imagining a fictitious repository (personal communication, November 16, 2013). Working with this real collection made it easier for our group to answer the questions of why to digitize this collection, how to select a subset of the album pages, and how to present the images. These questions are all asked in the Collaborative Digitization Program’s "BCR-DPS CDP Getting Started Questionnaire." (Collaborative Digitization Project, n.d.).We anticipated that the primary users of this collection would be anyone interested in amateur photography, Southern California history, early twentieth-century history, or found art objects. The type of audience that reads Found magazine, which collects and publishes found items that give insight into peoples’ lives, would be very interested in this collection (). We also considered the possibility that we could connect the album with descendants of the people in the photographs. We discussed waging an online effort to put the album in front of people whose families lived in Los Angeles in the early 1900s. For this purpose, a digitized, online surrogate of the photo album permits “multiple, simultaneous access” to the album (Scott, 2013). ?What are the needs of these audiences? We discussed several possible needs. Audiences may want to know when and where the photographs were taken with as much specificity as possible. They may want to know where the photo album was purchased. They may be interested in the physical characteristics of the photographs, and their placement in the original photo album. They may want to browse through the photographs and to compare photographs taken at or near the same location. Finally, researchers might look for photographs that depict specific locations such as the Ocean Park Bath House, or items, such as clothing from a certain era. By providing searchable fields with ample subject identification, we hope that we are permitting this type of search. ?CopyrightThe Snaps photograph album that is the source of our collection came from a thrift shop in Los Angeles. The album has no identifying marks or writing on it, so it would be difficult to determine ownership. Because the album is very old, it would be equally difficult to identify the people depicted in the photographs. Although specific copyright ownership could not be determined, we decided it would still be valuable to reproduce this collection digitally. The album is also not dated, but based on the clothing and scenes depicted, we estimated a reasonable time frame of creation to be somewhere between 1899 and 1920. We were even able to identify a few buildings depicted in the photographs, and researched the styles of clothing that the people in the photographs are wearing. Since we estimate that the photographs were created between 1899 and1920, this album may still be under copyright protection. Under U.S. Copyright law, an unpublished, anonymous work may be under copyright for 120 years after date of creation. Thus, only works created before 1893 are free of copyright (Hirtle, 2013). With this in mind, we did some research on the process of publishing “orphan works” since we do not know the owner of the work, the subjects, or the creator. Nor do we have any way of contacting anyone to ask for permission to use the works. According to the Society of American Archivists (SAA) (2009), it is necessary to make every reasonable effort to “find and make contact with” the copyright owners before publishing the works. Even in this instance, proving that a reasonable effort to contact the rightful owners was made may only reduce the penalties of infringement on the infringer. Therefore, SAA has set forth a group of principles to consider when making the decision to publish an orphan work. These include considering the likelihood of finding the rights holder, the age of the work, an anonymous or non-professional creator, and the scope of distribution of the work by the archivists. We believe that we have a low likelihood of finding the rightful owner, whom we believe to be an amateur photographer. Furthermore, we are only digitizing a portion of the collection for a very small audience, so we decided to proceed with digitization.We drafted the following copyright statement and included it with each image in our collection. We believe that the statement below sufficiently informs potential users that the collection contains orphan works, limits our liability in the event of any possible infringement by another party, and includes a means of contacting our group in case a viewer can offer us more information, or requests that we remove the content.Objects in this collection are considered orphan works because the identity of the rights owner is unknown; they may still be under copyright. The collection is being provided for personal, research, and non-commercial use. Some materials in this collection may be protected by the U.S. Copyright Law (Title 17, U.S.C.). For more information about seeking permission to publish or reproduce, contact the Group 2 Archival Project at Dr. Martin Luther King, Jr. Library, One Washington Square, San José, CA 95192-0028, (408) 808-2000.Since this digitization project was for a class, and we did not have time or adequate resources available, our process of attempting to actually locate a rightful owner was minimal. However, we have enabled the Tags and Comments portions on our collection’s website as a means of allowing our users to contact us, as well as any potential copyright owners.SelectionAs stated above, the Snaps album contains over 70 pages, each of which holds single or multiple prints, and some of which are blank. In order to best reflect the original artifact, we decided to scan and catalog images of full album pages, rather than cropping out individual prints. We also determined that the time allotted for this project did not allow us to scan every page of the album to be included in the online collection. Thus, we had to develop some selection criteria in order to decide which pages we would scan, and which would not be digitized. First, we decided not to scan blank pages of the album, and that any page with multiple photographs would be treated as one page. We decided not to scan other pages from the album based on the condition of the photographs and of the album paper. Some of the pages are deteriorating, and the scanning process may have caused further fading, ripping, tearing, or other physical damage. We also eliminated pages that had scenes or subjects that were depicted in multiple other pages, and chose the best quality images of those scenes and subjects. Some of the photographs were too faded to provide enough discernable detail for us to describe and assign subject terms. If these pages had been scanned, the digital image would have required a lot of post-processing. Post-processing would not only be time-consuming, but any digital editing would also change the image in a way that was not representative of the original. After developing these criteria, and making the decisions about which pages would be best to scan, we were left with 43 pages that were suitable for digitization.Technical Production InformationBecause Student B was the only project member with access to the photo album, he digitized the album pages using standards agreed upon by the group. Scanning was conducted in three stages: preliminary scans, resolution test scans, and the final master image scans.Preliminary ScansA preliminary scan was conducted for the purpose of planning the project so that all members had a sense of the scope of the album, and so that we had a single frame of reference for our discussions. The scan consisted of all two-page spreads in the album that contained at least one print. Images were captured using a Ricoh Aficio MP 6001 multifunction digital copier, in black and white, 8-bit, 300 dpi resolution, and delivered to the group as a single PDF file.Resolution Test ScansOur options for digital imaging were dictated largely by the equipment available to us. Through his employer, Student B had access to a Hewlett Packard Scanjet G4010 flatbed photo scanner capable of high quality scans. According to vendor documentation, this model is capable of capturing images at a resolution of up to 19200 dpi and color at depths of?8, 24, or 96 bits per pixel (Hewlett Packard Company, 2011). We decided early in the project that capturing the images in color was important for several reasons:The collection represents the album, not just the prints contained within.The yellowing of the pages, the color of the leather cover and binding, and the color of the various print processes would all be of interest to our audience.The provenance of the album is unknown, so our audience of researchers is more interested in photography in this era, less so the specific histories of the people depicted.Color can help determine the document's age, physical condition, or previous use. Capturing at a color-capable bit depth (at least 24-bit) would also help capture the high dynamic range of the black and white prints in the album.To determine a scanning resolution, Student B took measurements of the album and Susan Edwards (Susan E.) used those measurements to calculate resolution, pixel dimensions, and file sizes at various quality indices (QI) using the University of Illinois Library’s Image Quality Calculator (). The album pages measure 220mm x 140mm and the smallest discernable detail of a sample print was between 0.5mm and 0.25mm. Table 1 records results of these calculations for images scanned at 24-bit color.Table 1 – Measurements for the Snaps collection, at 24-bit color, for capturing various levels of detail and QISmallest detail visible.5mm .5mm .33mm .33mm .25mm .25mm QI(QI=8)(QI=5)(QI=8)(QI=5)(QI=8)(QI=5)Recommended resolution615 dpi384 dpi932 dpi582 dpi1230 dpi769 dpiPixel dimensions5326 x 21163325 x 21168072 x 51375040 x 3,20710653 x 67796660 x 4238Est. file size52 MB20 MB119 MB46 MB207 MB80 MBBased on these results, it seemed that capturing details as small as .25mm would require very large file sizes. We wondered if the images scanned at a higher resolution would be appreciably better. File size was not the only consideration in determining the specifications for our scans; speed of the scanning process was also a concern, just as it would be for any institution undertaking a digitization project. In order to compare the quality of these various resolutions visually, Student B conducted 24- and 96-bit test scans of a page that featured matte and glossy images, at 300, 600, and 1200 dpi. The resulting TIFF files were shared on Google Drive for the group to compare, and to discuss the trade offs between quality, file size, and scan time. Table 2 presents the results of those tests.Table 2 - Measuring scanning time at various resolutions and bit depthsResolutionBit DepthScan TimePixel DimensionsFile Size300 dpi2400:17.522574?×?175513.6 MB300 dpi9600:20.292574?×?175527.1 MB600 dpi2401:27.665148?×?351054 MB600 dpi9601:15.795148?×?3510108.4 MB1200 dpi2405:19.6410296?×?7020216.8 MB1200 dpi9605:35.6410296?×?7020433.7 MBBased on the results of these tests, the group decided that the 96-bit scans did not provide sufficient additional detail to warrant the much larger file sizes and longer scanning times, and decided to scan at 24-bit. When we visually examined the three 24-bit images, it was clear that the 300 dpi image did not capture sufficient detail—pixelation was apparent when zooming in on the image. While the image scanned at 1200 dpi did contain much more information (the texture of the paper pulp on the album page was readily visible), it didn’t seem to capture any more detail in the images than the scan created at 600 dpi. We decided that an additional test comparing images captured at 600 dpi and 750 dpi should be conducted to see if there was a discernable difference in quality. Student B posted the results in Table 3 to the group.Table 3 - Comparing 600 dpi and 700 dpiResolutionBit DepthScan timePixel DimensionsFile size600 dpi240:55.945136 x 350251.4 MB750 dpi244:53.176419 x 437780.3 MB-62865908050Figure 2 - Comparison of 24-bit scans at 600 dpi and 750dpi. At the same physical dimensions, there is no visible improvement in detail captured in the 750 dpi image, compared to the 600 dpi image.Figure 2 - Comparison of 24-bit scans at 600 dpi and 750dpi. At the same physical dimensions, there is no visible improvement in detail captured in the 750 dpi image, compared to the 600 dpi image.No discernable difference was visible to the group between the 600 and 750 dpi images (See Figures 2 and 3). While it was possible to zoom in further on the 750 dpi scans than on the 600 dpi scans, there was no appreciable increase in the amount of detail visible. As demonstrated in Figure 2, we observed that the 600 dpi scan, viewed at 100% was sufficient to capture the smallest details in a sample image. Figure 3 shows the level of pixilation visible in each of these sample scans at 500% zoom. Again, although the image pixelates at a shallower zoom point in the 600 dpi image, we decided that the amount of detail discernable is sufficient. The 700 dpi image does not provide more detail than the 600 dpi image, when viewed with the naked eye. Through these tests, and discussions, the group came to a consensus that the 24-bit, 600 dpi images, captured as TIFF files, provided the best balance of image quality, file size, and speed.514351440180Figure 3 - Comparison of details from 24-bit scans at 600 dpi and 750 dpi, and at 500% zoom.0Figure SEQ Figure \* ARABIC 3 - Comparison of details from 24-bit scans at 600 dpi and 750 dpi, and at 500% zoom.-62865-1188720003023235-116078000 Student B proceeded to scan the album pages from the HP ScanJet G4010 using “HP Scan” software version 2.4.4 running on Mac OS 10.7.5. The biggest challenge to capturing the album pages was the construction and condition of the album. Due to its age, the album needed to be handled with care, but the stitched binding produced large, heavy shadows and distortion in the page gutters if the album was just placed atop the scanner platen. Additionally, the album was larger than the scanner and needed to be supported. To overcome this, a weight was placed on top of the album, close to the binding, and improvised supports were placed at each side of the scanner to support overhanging pages (see Figure 4). 6229352903220Figure 4 - Scanning station setupFigure 4 - Scanning station setup622935254000Using the HP software, a few preferences were required to ensure consistent image capture. First, all images were captured at the same size by locking the scan selection area to 5148 pixels wide by 3510 pixels high. Second, many of the built-in features of the scanning software were disabled, such as those that restore faded color and remove scratches. Enabling those features may make sense for home users that wish to enhance their own photos, but one of the goals of our project is to present an accurate representation of the original artifact. Third, the software was set to produce sequentially numbered file names (Snaps_IMG##.tif) for easier management of the files.Image ProductionBecause we chose to use full pages, and the scan selection was locked to the page size, the images produced did not need to be cropped. The TIFF files that were produced during the scanning process were uploaded to the CONTENTdm servers using the Project Client software, which automatically produced JPEG2000 compressed files for end-user access. This initial upload was actually done in error—the group had planned on providing uncompressed TIFF files to users—but we found that the resulting JPEG2000 files were of very high quality without any noticeable loss of image detail, so the group decided to keep those as derivative access files. By converting to JPEG2000, files were reduced from an average size of 54.2MB to 2.1MB.MetadataOur metadata development started with assessing the Dublin Core fields that were already available in CONTENTdm for their potential value for our audience. Overarching considerations included how each field could be useful (or not) for search, or for our users’ research interests, and how we would connect our individual records for the album pages together so it would be clear they were part of a larger collection. We set up a Google document with the Dublin Core fields laid out in a table, and divided them among ourselves. We researched how each field should be utilized according to the Dublin Core Metadata Initiative (DCMI); looked for examples of how archives have utilized the field in similar collections; thought about whether it made sense to keep the field, rename it, or delete it; and assessed whether a vocabulary authority would make sense for the field. In the end, we used many Dublin Core fields as they were configured in CONTENTdm, deleted a few fields, changed the name and configuration of a few, and added a few new fields. Our records entail primarily descriptive metadata, but we also have some structural, administrative, and technical fields. In every case, we used an existing authority or standard for the field, if possible. In some cases a standard vocabulary was used, and in other cases, we use formats from authorities like the International Organization for Standards (ISO).Descriptive MetadataWe spent a lot of time discussing the descriptive metadata. Because our album has an unknown creator, and we have very little information about it, we realized right away that several of the Dublin Core fields that are usually key in image records, such as the Title, Creator, Geographic location, and Date, would not be of use for accessing this collection. Instead, we would have to focus on describing the images visually as best we could in the Title, Description, and Subject fields in order to create intellectual access points into these images. Coverage, Language, Publisher, and Source. We deleted the Creator and Contributor fields because we simply don’t have the information. Although we aren’t certain of the date, or the specific geographic location of the images in the album, we were able to make an educated guess based on clues in the images themselves. We amended the DCMI Coverage field to include this information. We created a Coverage-Spatial field to indicate the approximate geographic location, renamed it “Location Depicted,” and used the Thesaurus of Geographic Names (TGN) vocabulary. We are pretty confident that most of the images are from Los Angeles County generally, and a few images depict a building or landmark that we could use to tie it to a specific location that was named in this field. We also created a Coverage-Temporal field and renamed it “Date of Photograph” to include a very general timeframe of “Early 20th century” for all the images. We used the Library of Congress Name Authority as a standard for formatting the language of this phrase. Although the album has very little text, it is in English where present, so we used the Language field to indicate on all the album page records that this is an English-language album, following the ISO standard code, “eng.” DCMI Publisher and Source fields were used in the standard way to describe the digitized document. All records have the same publisher—our group—and source, which describes where the original album was obtained—by our group member, Student B. Title, Description, and Subject. Our most intensive discussions involved the Title, Description, and Subject fields. None of the images or pages in the album contained any text that we could use for our titles. So, the Title and Description fields would both have to be descriptive, narrative text. We decided to use the Thesaurus of Graphic Materials (TGM) thesaurus in the Subject field to provide some standardized access. Our discussions about these fields generally concerned how much detail to describe, and what aspects of the album pages and the photographs we should include in our descriptions. We discussed how to balance descriptive terms that would be the same for each image in the collection (e.g., Photographic print) with terms that are unique to the individual album pages. Terms that are the same for each item would not be very useful for researchers searching for images within this collection. But such terms are valuable descriptors, and would be useful if these images are viewed within a larger context, for example within an archive of many collections. Description. The Description field was used to describe the album page itself, the scenes in the photographic images on the pages, and any remarkable characteristics of the photographic prints. Because we scanned full album pages, and consider the entire album (not just the photographs) to be of interest to our audience, we decided that the description should describe the album pages as a whole. We decided that the most useful information for our audience would be to know how many prints were on each page; so, we include language about the number of photographs on the page in our descriptions (“Album page with two photographs…”). Narrative description of the images in the photographs consists of the bulk of the Description field. For each image, we described the scene generally and then provided more detail about the specific people depicted, and any significant aspects. Finally, because our audience includes photography historians and amateur photography enthusiasts, we also include some information about the photographic print format. Such characteristics include whether the print was trimmed, if it is a reprint (there are several reprinted photographs in the album), whether the full negative was printed, etc.Title. The Title field followed on the Description field, as in most cases a sentence from the Description was used or modified for the Title. We debated whether this should include reference to the photographs on an album page. We decided to not include this in order to keep the titles as short as possible, and to avoid repeating the word “photograph” in every title. Subject. In our discussions of the Subject field, we debated whether to include as many terms as possible, or to focus on aspects of the images that are unique and that represent the image as a whole. This came up when the team took a first pass at adding metadata for all of the objects. One team member used a small number of general subject terms that described the images as a whole (e.g. Landscape, Portrait), while another team member used subject terms to describe as many of the details in the images as possible (e.g., Hats). We were curious what the standard approach was and surveyed some online collections to find out. We found a range of approaches, and determined that there may be no standard. We did note that detailed subject terms describing many aspects of an image can be a useful aid to browsing, especially in collections where the audience is not likely to search by a creator or title. We also noted that in CONTENTdm, the full text of the description field is searchable, so any terms mentioned in the Description may also provide additional access. With this in mind, we decided to include subject terms for significant details in the photographs, such as hats, even though the photograph may not be about hats, but not to describe every single element in a photograph.We chose several subject terms to use globally for every item in our collection. Two of these describe the material nature of the album itself and the photographs: ‘Photographic albums’ and ‘Photographic prints.’ The other heading, ‘Amateur works,’ acknowledges the unknown origin of the photographs, and the fact that they are not fine art photographs. In addition, over a third of the images in the collection are portraits of groups of people. We had a discussion about what characteristics of these images constitutes “portraits” per se, and decided that a group of people arranged intentionally and facing or posing for the camera constituted portraiture. We noted that, in the scope note for ‘Group portraits,’ that the term should also be double indexed with a term describing the medium, so we decided to include the term ‘Portrait photographs’ along with that term. Finally, because there were four of us entering the subject terms for different items in the collection, we inevitably had discrepancies in the subjects assigned. In order to reconcile these differences, Student C created a map of all the terms used in the collection, and listed each item that used that term. This helped the team see where terms had been applied inconsistently, and where there were opportunities to utilize some terms more systematically. It turned out to be a great advantage to have all four of us entering these terms, because we each viewed the images slightly differently, and some of us discovered valuable terms that the others did not find or think about. Table 4 summarizes all of the values and rules we established for our descriptive metadata fields.Table 4 - Descriptive metadata for the Snaps collectionField nameDCMI mapField data, or rulesTitleTitleDescribe the scene in the photograph(s) generally. Don’t use the terms ‘images’ or ‘photographs’ in the title. You might use a sentence from the Description field as a title. SubjectSubjectUse Thesaurus of Graphic Materials (TGM). List as many subject headings as applicable to describe the general scene, and details.Use these headings for every image: Photograph albums; Amateur works; Photographic printsFor all items with the subject ‘Group portraits,’ also include the heading ‘Portrait photographs.’DescriptionDescription1. ?Describe the number of prints on the album page.2. ?Describe the content of the photo. Use 1-3 sentences to describe the images on the page. Begin by describing the general types of scenes (this first sentence can be used also as the title in the Title field), then use additional sentences to provide more detail.3. ?Describe any remarkable characteristics of the print format.Example (3rd image in the album):Album page with one photograph of five people standing in a landscape. An open field is in the foreground, with rolling hills and trees in the distance. One woman and two(?) men stand in the right foreground, and two women walk away from the camera in the middle distance. The photo appears to be a reprint.PublisherPublisherLIBR284-15 Group 2Location DepictedCoverage-SpatialUse Thesaurus of Geographic Names (TGN). Los Angeles (California, United States : county) will be used for most items, unless a more specific location can be positively identified in the image.Date of PhotographCoverage-TemporalEarly 20th century (uses term from LoC Name Authority - )LanguageLanguageeng (follows ISO 639-2)Structural MetadataWe created two structural metadata elements in order to link all the album pages together, and to indicate the order of their arrangement in the original object, the photo album. These fields both use the DCMI Relation element. We created a “Collection” field, mapped to the DCMI Relation-IsPartOf element, in order to indicate that the album pages are part of a larger collection, an album called “Snaps.” We also created a “Page Number” field, mapped to the DCMI Relation-IsPartOf element,?in order to indicate which page each image refers to in the larger album. This field will help the audience reconstruct the order of pages in the original album. Additionally, because we did not scan every page in the album, this field allowed us to indicate the exact location in the album of each image and accounts for the missing pages. Table 5 summarizes the values and rules we set for our structural metadata. Table 5 - Structural metadata for the Snaps collectionField nameDCMI mapField data, or rulesCollectionRelation-IsPartOfSnaps: NumberRelation-IsPartOfRecord album page number as it appears in original album sequence. Example: “Page 2”Administrative MetadataTwo of our administrative fields use the standard DCMI elements and are populated with the same data for each record—Type and Rights. Two other administrative fields use amended DCMI elements. We renamed the DCMI Date element “Digitized Date” in order to record the date that the album pages were scanned. Because we decided to use the filenames of the images for our collection identifier, it made sense to rename the DCMI Identifier field “File Name” to be more specific, so our audience would understand exactly what that number represents. The files were named with the name of the collection, followed by the image number indicating the order in which the images were scanned (e.g., “Snaps_IMG02.tif”). Although the images were scanned in the order that they appear in the album, from beginning to end, the filename numbers do not correspond to the actual page numbers in the album. This is because the digitized collection does not include every page, since a selection was made from the full album to be digitized. Thus, we sorted the collection display on the Filename field in CONTENTdm so that the display would be sequenced as the pages appear in the original album. However, because this field does not accurately represent the location in the album, we added the Page Number field to help users understand where the original image appeared in the album.Table 6 - Administrative metadata for the Snaps collectionField nameDCMI mapField data, or rulesTypeTypeImage (uses DCMI vocabulary)RightsRightsObjects in this collection are considered orphan works because the identity of the rights owner is unknown; they may still be under copyright. The collection is being provided for personal, research, and non-commercial use. Some materials in this collection may be protected by the U.S. Copyright Law (Title 17, U.S.C.). For more information about seeking permission to publish or reproduce, contact the Group 2 Archival Project at Dr. Martin Luther King, Jr. Library, One Washington Square, San José, CA 95192-0028, (408) 808-2000. Digitized DateDate2013-10-27 (follows ISO 8601)FilenameIdentifierFile name of digitized image, without file extension. Example: “Snaps_IMG02”Technical MetadataTechnical metadata in our collection is recorded in two fields, which contain the same values for all of the images. We used the DCMI Format field to indicate the file format for all of the images. We also added a technical field called “Digital Reproduction Information” to record the specifications and process of the actual digitization process for these images. Content for these fields is provided in Table 7.Table 7 - Technical metadata for the Snaps collectionField nameDCMI mapField data, or rulesFormatFormatjp2 from archival tiff (follows MIME type standard)Digital Reproduction InformationNoneArchival TIFF masters scanned from a Hewlett Packard Scanjet G4010 at 600dpi/24-bit resolution using HP Scan version 2.4.4 software, converted to JPEG2000 by CONTENTdm Project Client.ConclusionDigitizing the Snaps?photo album in order to create an online collection gave Group 2 the opportunity to practice and reflect on the planning, selection, production, and metadata development aspects of digitizing an archival collection. In addition to the work documented in this paper, we also had discussions about work that could be done on this collection, which was not possible due to the scope of this project within the bounds of this course. If time were not an issue, we would have liked to scan the entire album, and performed post-processing on the very faded images in order to bring them to life again. We also wanted to spend more time researching the photographs in order to determine the physical locations where they were taken. We also would have liked to attempt to identify the descendants of the people depicted in this photo album. We speculated that incorporating user-generated content into our online collection, such as inviting comments and tags, could be a way to gain more information about this collection, and possibly locate the descendants of the people depicted. We anticipate that going through this process on a real project, as a team, will improve the quality of our future digitization endeavors. ReferencesCollaborative Digitization Program. (n.d.). “BCR-DPS CDP Getting Started Questionnaire” Provided in Week 2 course readings.Hewlett Packard Company. (2011). HP Scanjet G4010 photo scanner datasheet. Retrieved from , P.?(2013). “When Works Pass Into Public Domain in the United States: Copyright Term for Archivists and Librarians.” Cornell Copyright Information Center. Retrieved from , A. (2013). “Project Planning and Management.” Slides from Week 2 lecture. Society of American Archivists (SAA). (2009). “Orphan Works: Statement of Best Practices.” Retrieved from ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download