COMMUNITY MUSEUM PROJECT
SPECIFICATIONS FOR DIGITIZATION
Completed October, 2006
Prepared by Theodore Gerontakos for the Olympic Peninsula Community Museum Project
Preface. Areas of Description in these Specifications
Introduction: brief explanation of these specifications
1. Media Type 1: Photographic Prints
2. Media Type 2: Negatives
3. Media Type 3: Slides (and other positive transparencies)
4. Media Type 4: Three Dimensional Objects (and generally anything digitized using a digital camera)
5. Media Type 5: Audio
6. Media Type 6: Video
7. Media Type 7: Text (page images)
8. Summary Table
[##].1. Master Files
[##].1.1. Device(s)
[##].1.2. Input Resolution
[##].1.3. Bit Depth
[##].1.4. File Type
[##].1.5. File Name
[##].1.6. Use of Capture Software
[##].1.7. Storage Media
[##].1.8. Storage Media Name
[##].1.9. Backup
[##].1.10. Copy for University of Washington Archive
[##].1.11. Targets
[##].1.12. Quality Control
[##].1.13. Required Metadata at Time of Capture
[##].1.14. Miscellaneous notes on digitizing [media type]
[##].2. Derivative Files / Screen Presentation
[##].2.1. File Type
[##].2.2. Compression
[##].2.3. Use of Photo Editing Software
[##].2.4. Thumbnail surrogate
[##].2.5. Storage
[##].2.6. Uploading
[##].2.7. Band / Watermark / Brand
[##].3. Print Presentation
[##].3.1. Printing Master File
[##].3.2. Printing Derivative File
These specifications were written to guide the digitization of all media types for the Olympic Peninsula Community Museum Project. This was necessary due to the distributed nature of the digitization process and the need to follow standards and best practices. Thus this was written as a working document, a practical document, not as a theoretical document or work of scholarship.
As the digitization work for this project was "distributed" -- that is, project planners designed this project so that many different people in many different places would create digital objects for the project -- the specifications were not intended to be read by all digitizers; in fact, the specifications were most often used to draft "how to" sheets custom-written to meet specific needs. In all cases, whether guided by these specifications or by instruction sheets based on these specifications, the recommendations of these specifications were only loosely followed. Enforcement was not practical.
Given the marginal importance of these specifications in the project, they were written on-the-fly and a bibliography of sources was not practical to maintain, let alone annotations for specific passages and assertions. We publish these specifications without citations, annotations or explanations/justifications for our decisions primarily as an indicator of our project's direction rather than to give direction to other projects. Anything more ambitious was outside the scope of our efforts.
1.1. Master Files
1.1.1. Device(s)
1.1.1.1. Flatbed Scanner.
1.1.1.2. Formats too large for a flatbed scanner:
1.1.1.2.1. Digital camera.
1.1.1.2.2. Analog camera (scan from negative).
1.1.2. Scan Resolution (this refers to the actual resolution at which the scanner scans): recommendations:
1.1.2.1. Images less than 2 inches in the long dimension: scan at 800 ppi.
1.1.2.2. Images between 2 and 7 inches in the long dimension: scan at 600 ppi.
1.1.2.3. Images between 7 and 12 inches in the long dimension: scan at 400 ppi.
1.1.2.4. Images larger than 12 inches in the long dimension: scan at 300 ppi.
1.1.2.5. Notes:
1.1.2.5.1. Required: minimum resolution is 300 ppi, unless an organization arranges a smaller minimum.
1.1.2.5.2. Required: in no case should interpolated resolution be used.
1.1.2.5.3. Do not create an input resolution that differs from the scan resolution; that is, do not "scale" the image at the scanner for printing.
1.1.2.5.4. Recommended: input resolution number should be an integer divisor of the maximum resolution of the scanner being used. If the input resolution recommended above is not an integer divisor of the maximum resolution, use the nearest resolution that is an integer divisor.
1.1.2.5.5. Recommended: If the input resolution recommended above is not available using your scanner software, use the nearest available resolution
1.1.3. Bit Depth: recommendations:
1.1.3.1. Most black and white prints: 8 bit grayscale.
1.1.3.2. Black and white prints with color tones (for example, a sepia tone) worth preserving: 24 bit RGB.
1.1.3.3. Color prints: 24 bit RGB.
1.1.4. File Type
1.1.4.1. Required: TIFF 6.0.
1.1.5. File Name: file names given to master images will be preserved by all derivatives. This is a central organizing principle for the project; project staff should review all incoming master files and edit file names whenever necessary; files should be named carefully and in accordance with the following recommendations:
1.1.5.1. Each file name should have 4 components:
1.1.5.1.1. An organization code prefix, preferably 2 characters only.
1.1.5.1.1.1. Examples: Hoh Tribe = HT; Clallam County Historical Society = CC.
1.1.5.1.2. A four digit numeral directly following the organization code, representing the sequence of digital images (regardless of original format) in the organization's entire digital collection for this project.
1.1.5.1.2.1. Numerical sequence: 0001, 0002, 0003, ... 0045, 0046, ... 0101, 0102, ... 4672, 4673, ....
1.1.5.1.3. The organization's identifier for the print, separated from the four digit numeral with an underscore.
1.1.5.1.3.1. Identifiers include accession numbers, negative numbers, etc.
1.1.5.1.3.2. Replace all punctuation in the identifier with dashes.
1.1.5.1.3.2.1. Example: accession number 26.32/1 would become _26-32-1.
1.1.5.1.3.2.2. Example: negative number 2003.A-7/2002: 14AB would become 2003-A-7-2002-14AB.
1.1.5.1.4. File type extension. File extension ".tif" will be appended to all master files.
1.1.5.1.5. Examples:
1.1.5.1.5.1. The Timber Museum's 212th image scanned for this project; the print is Timber Museum accession number 43.2.2: "TM0212_43-2-2.tif."
1.1.5.1.5.2. Clallam County Historical Society's 2,032nd scanned image for this project; the print is CCHS negative number 13/4-2003, 6.1: "CC2032_13-4-2003-6-1.tif."
1.1.5.1.6. When organizations scan their own images, we recommend file names that are identical to the organization's identifier, with punctuation intact. CMP staff can edit the file names later to conform with the project-wide file-naming recommendations.
1.1.6. Use of Capture Software
1.1.6.1. Set scanner to scan the media type you are scanning.
1.1.6.1.1. Microtek, for example, has an option "Scan Material" on their "Preferences" menu. For photographic prints select "Reflective."
1.1.6.2. Your scanner may have other overall settings that need to be adjusted from your main menus; for example, your preview window may need to be configured, your view of scanner controls may need to be configured, you may have a choice between scanning for quality or for speed (choose quality), filters may have to be disabled, etc.
1.1.6.3. Overview and/or preview image.
1.1.6.4. Crop image (does not need to be exact, although an exact crop is preferred).
1.1.6.5. Set dynamic range to "automatic."
1.1.6.6. Set resolution and bit depth as recommended above.
1.1.6.7. Scan the image.
1.1.6.8. Store the file on the appropriate storage device.
1.1.6.9. Required: do not compress the data.
1.1.6.10. Recommended: do not use filters (sharpening, blurring, descreening, etc.).
1.1.6.11. Recommended: do not clip the endtones off the image's actual dynamic range (although clipping of very sparse and "insignificant" tone is encouraged).
1.1.6.12. Recommended: do not create an input/output differential (that is, do not scale).
1.1.6.13. Recommended: review image after it is scanned for verisimilitude to original.
1.1.6.14. Note: we do not want to "improve" the print in any way; the master should capture the print with the highest fidelity possible (without using filters). Photo editing software can be used at a later date, when creating derivative files, to improve, touch-up, enhance or sharpen the derivative image.
1.1.7. Final Storage Media
1.1.7.1. Original master file on one optical disc and a copy of the original master file on a different optical disc should be stored on any of the following: CD-R, DVD+R, DVD-R, DVD+RW. All files will be stored at University of Washington. Copies can be prepared for any collection owner upon request.
1.1.7.2. Workstation hard drives. As space permits, images can be stored temporarily on local hard drives. Ultimately they will be transferred to optical discs and deleted from these hard drives.
1.1.7.3. Other Alternatives: owners requiring other options can talk to CMP staff to discuss borrowing hardware, using FTP, etc.
1.1.8. Final Storage Media Name
1.1.8.1. The name of each optical disc should have three components:
1.1.8.1.1. A three-character project code: "CMP."
1.1.8.1.2. The optical disc type -- CD (indicated by the lack of any code) or DVD (indicated by "DVD"). Examples
1.1.8.1.2.1. CD = CMP[rest of disc number]
1.1.8.1.2.2. DVD = CMPDVD[rest of disc number]
1.1.8.1.3. The disc number in the project's complete sequence of discs.
1.1.8.1.4. Examples:
1.1.8.1.4.1. CDs: CMP1, CMP2, CMP3, ... CMP27, CMP28,... CMP96, CMP97...
1.1.8.1.4.2. DVDs: CMPDVD1, CMPDVD2, ... CMPDVD27, CMPDVD28...
1.1.8.2. The name should be digitally assigned to the disc and written on the disc case, along with other descriptive information about the disc con tents.
1.1.8.3. There are two options to identify the name of the actual optical disc, to pair it with its case and contents:
1.1.8.3.1. On the case, write the two numbers printed and/or engraved around the center hole of the actual CD. This is the preferred method.
1.1.8.3.2. Use a marker to write the CD name for this project on the innermost cylinder of the disc, around the center hole. Try not to cover any numbers printed or engraved on the CD. DO NOT write on the parts of the CD under which data is stored.
1.1.9. Backup
1.1.9.1. Local backup system at the owning organization. This will be done on a case-by-case basis.
1.1.9.2. Images on CDs: One run of optical discs (CDs and DVDs) will accommodate all original masters and copies of masters (backups). Ultimately all derivatives will also be stored on these discs.
1.1.10. Copy for University of Washington Archive
1.1.10.1. The run of optical discs will contain all project files. These will be kept in a fire-protection file cabinet until the files can be moved to a digital repository, probably D-Space (which may never happen, as it will happen outside the boundaries of this project).
1.1.11. Targets
1.1.11.1. Targets are not required.
1.1.11.2. Organizations can choose to use targets.
1.1.11.2.1. Targets may be scanned separately, before a batch is scanned, or with each individual image.
1.1.11.2.1.1. When scanned separately, targets should be distinguished from image files.
1.1.11.2.1.1.1. Targets should have file names that identify them as targets and that identify the images in the batch assessed by the target, ultimately to be stored on the final optical disc. In the event that a target assesses a batch not represented by a consecutive series of image file names, the target file name should refer to an additional document that serves as a concordance of target to batch.
1.1.11.2.1.2. When scanned with individual prints, resolution should be determined by the total dimensions of the print without the target.
1.1.12. Quality Control
1.1.12.1. During scanning:
1.1.12.1.1. Calibrate monitors and devices, according to the operator's judgment. Project staff can provide a calibration worksheet upon request.
1.1.12.1.2. Check the fidelity of the digital image to the original image before transferring the file. In almost all cases, this is the only time the digital image will be compared to the original.
1.1.12.2. After scanning, before derivatives are produced:
1.1.12.2.1. Project staff will review each TIFF while producing derivatives.
1.1.12.2.1.1. The original print will not be available during this review.
1.1.12.3. The "Reviewing Digital Images for Quality" worksheet will help guide the quality review process.
1.1.13. Required metadata at time of capture
1.1.13.1. The following aspects of every item should be described during scanning; if possible, record this information on a spreadsheet:
1.1.13.1.1. File name for digital image;
1.1.13.1.2. Organization's identification number for the print (accession number, negative number, etc);
1.1.13.1.3. Describe what the item scanned is (a negative, not a print, not a slide, etc.); please be as specific as you can;
1.1.13.1.4. Subject headings assigned by organization;
1.1.13.1.5. Notes accompanying the print, written on the print, etc.; try to indicate how and where anything is written;
1.1.13.1.6. Dimensions of the print scanned, rounding up to the nearest 1/8 inch if possible;
1.1.13.1.7. Resolution at which the item was scanned;
1.1.13.1.8. The precise wording on how somebody can order the item, image, or other facsimile;
1.1.13.1.9. The precise word-by-word property statement that will appear beneath the image, in the form "Property of....," "Image courtesy of..." etc. This must be included and must be stated with pinpoint accuracy. It cannot be changed later;
1.1.13.1.10. Anything else worth noting
1.1.13.1.11. Name of the person who scanned
1.1.13.2. For all information not known, enter "unknown" in the spreadsheet.
1.1.13.3. A metadata form can be obtained online for use during scanning; visit the Community Museum Project Cataloging Web Page to see a selection of metadata forms from which to choose.
1.1.13.3.1. During scanning, only minimal metadata is required; however the forms at the Community Museum Project Cataloging Web Page are to assist full description of each item. Use the forms as you see fit at any time in the process of building the digital collection.
1.1.14. [No miscellaneous notes on scanning photographic prints]
1.2. Derivative Files / Screen Presentation
1.2.1. File Type: JFIF.
1.2.2. Compression: JPEG.
1.2.2.1. The compression ratio should be about 150:1, or even 200:1.
1.2.2.1.1. When using Adobe Photoshop to create JPEGs, use JPEG quality level 3.
1.2.3. Use of Photo Editing Software
1.2.3.1. Rotate image if it is skewed.
1.2.3.2. Crop image, as necessary.
1.2.3.2.1. Each organization should create their own cropping criteria; for example, should informative borders be scanned, should image space only be scanned, etc.
1.2.3.3. Resize image.
1.2.3.3.1. When height is longer than width: resize height to 600 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc. -- the word used depends on the software used).
1.2.3.3.2. When width is longer than height: resize width to 700 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc., depending on the software used).
1.2.3.4. Adjust luminance and/or color using histogram ("levels"), curve tool, or tool of your choice.
1.2.3.4.1. White and black points should be adjusted when the dynamic range of the scanner was set larger than the actual dynamic range of the print.
1.2.3.4.2. Do not clip the endtones off the image's actual dynamic range (although clipping of very sparse and "insignificant" tone is encouraged)
1.2.3.5. Unsharp Mask. Most derivatives will require use of this filter.
1.2.3.5.1. Recommended settings: amount 100%, radius 1.0, threshold 0.
1.2.3.5.1.2. If the recommended setting is too extreme, try less amount, smaller radius or a higher threshold.
1.2.3.6. Save As. Make sure you "Save As" (rather than save) so you do not alter the master image! (Recommended: store master images in a read-only directory.)
1.2.3.6.1. Name of master file should be used (with the new file type extension).
1.2.3.6.2. Format: JPEG.
1.2.3.6.3.JPEG options: Adobe Photoshop Quality level 3, Baseline ("Standard") Format Option.
1.2.4. Thumbnail surrogate.
1.2.4.1. Use one of two types of thumbnails:
1.2.4.1.1. CONTENTdm generated. CONTENTdm will produce a thumbnail for every print automatically when the JPEG is uploaded.
1.2.4.1.2. Custom thumbnail. If desired, we can use the thumbnail of our choice. These are easily imported into CONTENTdm. Their maximum size is 160 W x 120 H pixels.
1.2.4.1.2.1. Custom thumbnails will probably be unnecessary for all photographic prints.
1.2.5. Storage
1.2.5.1. Derivative images will be stored on the same optical discs as the master files. The web images will also be saved on the web site servers.
1.2.6. Uploading
1.2.6.1. Derivatives will be loaded into the CONTENTdm Acquisition Station first.
1.2.6.1.1. Images can be imported by CONTENTdm individually or in batches.
1.2.6.1.1.1. Upon request, project staff will produce an instruction sheet on importing into the CONTENTdm Acquisition Station.
1.2.6.1.2. Any metadata created prior to the import should be imported simultaneously with the images.
1.2.6.1.3. Bands must be added at the time images are imported into CONTENTdm.
1.2.6.1.3.1. "Property of" statements must be obtained before importing into CONTENTdm.
1.2.6.1.3.2. All these "Property of..." statements will be displayed using the CONTENTdm band option, in Verdana type, 8 point, normal script, pure white background (FFFFFF) and black type.
1.2.6.2. Derivatives will be uploaded from the CONTENTdm Acquisition Station to a development server at UW. The built-in file transfer feature in CONTENTdm will be used to make a transfer from the development server to the live server. Once loaded into the development server, the images and metadata can be edited and viewed using the web-based CONTENTdm Administration Station.
1.2.6.3. Derivatives will be loaded onto the University of Washington public server from the development server using the automated system configured in the University Libraries Digital Initiatives department. This will be done for each digital collection when it is completed.
1.2.7. Band / Watermark / Brand
1.2.7.1. Bands will be used in CONTENTdm, unless the organization negotiates different arrangements.
1.2.7.2. Bands will be added during uploading to CONTENTdm Acquisition Station (as outlined above in section 1.2.6.1.3.2).
1.3. Print Presentation
1.3.1. Printing Master File: will be determined by the printing parties.
1.3.2. Printing Derivative File
1.3.2.1. JPEG will not print professionally, no matter how it is done. The JPEG does not re-size or scale well. However the images can be printed well enough to serve as reference images.
1.3.2.2. Printing from a CONTENTdm environment (the following was not conclusively tested; it was tested on one machine only running a Windows XP Professional operating system):
1.3.2.2.1. Images appear to print scaled at 100 dpi.
1.3.2.2.1.1. 800 pixel wide images print 8 inches wide.
1.3.2.2.1.2. 700 pixel high images print 7 inches high.
1.3.2.2.1.3. All images print with the property band displayed.
1.3.2.3. Printing outside a CONTENTdm environment
1.3.2.3.1. Images can be saved to the local hard drive. They are saved with the property band attached.
1.3.2.3.2. When opened in Photoshop, image appears to re-scale to 1 dpi (this was not conclusively tested).
1.3.2.3.2.1. To print, image must be edited.
2.1. Master Files
2.1.1. Device(s)
2.1.1.1. Flatbed scanner.
2.1.1.1.1. Two settings may prevent use of a flatbed scanner for negatives:
2.1.1.1.1.1. Resolution: resolution must set high enough to meet our resolution recommendation (see section 2.1.2).
2.1.1.1.1.2. Dynamic range: the scanner must be able to capture the larger dynamic range of negatives.
2.1.1.1.1.2.1. Quick check: scan the negative with the dynamic range setting at "Full." If the endpoints on the histogram show significant clipping, consider scanning with a different device.
2.1.1.1.1.2.2. Another check: look for random noise (speckles) in the highlight areas of the digitized negative. If there is a lot of noise, consider scanning with a different device.
2.1.1.1.1.2.2.1. Multi-scanning to eliminate noise will be the exception, not the rule, for the Community Museum Project.
2.1.1.2. Film scanner, if available.
2.1.2. Scan Resolution (this refers to the actual resolution at which the scanner scans)
2.1.2.1. Set the resolution to produce an image with spatial resolution of 3000 pixels across the long dimension. (This does not have to be exact, but exactly 3000 pixels is recommended.)
2.1.3. Bit Depth
2.1.3.1. Black and white negatives: 8 bit grayscale.
2.1.3.1.1. If there is a reason to scan black and white negatives using RGB, see if the software can disable any orange mask compensation.
2.1.3.2. Color negatives: 24 bit RGB.
2.1.3.2.1. The scanner software may allow you to customize according to film types and their idiosyncrasies; results will vary.
2.1.4. File Type
2.1.4.1. Required: TIFF 6.0.
2.1.5. File Name: file names given to master images will be preserved by all derivatives. This is a central organizing principle for the project; project staff should review all incoming master files and edit file names whenever necessary; files should be named carefully and in accordance with the following recommendations:
2.1.5.1. Each file name should have 4 components:
2.1.5.1.1. An organization code prefix, preferably 2 characters only.
2.1.5.1.1.1. Examples: Hoh Tribe = HT; Clallam County Historical Society = CC.
2.1.5.1.2. A four digit numeral directly following the organization code, representing the sequence of digital images (regardless of original format) in the organization's entire digital collection for this project.
2.1.5.1.2.1. Numerical sequence: 0001, 0002, 0003, ... 0045, 0046, ... 0101, 0102, ... 4672, 4673, ....
2.1.5.1.3. The organization's identifier for the print, separated from the four digit numeral with an underscore.
2.1.5.1.3.1. Identifiers include accession numbers, negative numbers, etc.
2.1.5.1.3.2. Replace all punctuation in the identifier with dashes.
2.1.5.1.3.2.1. Example: accession number 26.32/1 would become _26-32-1.
2.1.5.1.3.2.2. Example: negative number 2003.A-7/2002: 14AB would become 2003-A-7-2002-14AB.
2.1.5.1.4. File type extension. File extension ".tif" will be appended to all master files.
2.1.5.1.5. Examples:
2.1.5.1.5.1. The Timber Museum's 212th image scanned for this project; the print is Timber Museum accession number 43.2.2: "TM0212_43-2-2.tif."
2.1.5.1.5.2. Clallam County Historical Society's 2,032nd scanned image for this project; the print is CCHS negative number 13/4-2003, 6.1: "CC2032_13-4-2003-6-1.tif."
2.1.5.1.6. When organizations scan their own images, we recommend file names that are identical to the organization's identifier, with punctuation intact. CMP staff can edit the file names later to conform with the project-wide file-naming recommendations.
2.1.6. Use of Capture Software
2.1.6.1. Set scanner to scan the media type you are scanning.
2.1.6.1.1. Microtek, for example, has an option "Scan Material" on their "Preferences" menu. For negatives select "Negative Film."
2.1.6.2. Your scanner may have other overall settings that need to be adjusted from your main menus; for example, your preview window may need to be configured, your view of scanner settings may need to be configured, you may have a choice between scanning for quality or for speed (choose quality), filters may be automatically set, etc.
2.1.6.3.Overview and/or preview image.
2.1.6.4. Crop image (does not need to be exact although an exact crop is preferred).
2.1.6.5. Set dynamic range to "automatic."
2.1.6.6. Set resolution and bit depth as recommended above.
2.1.6.7. If you are using an infrared channel (for example, DigitalICE on a film scanner), it is recommended that you disable this feature; however you may choose to configure this channel to remove dust and scratches: please use this option with caution, as it can blur the image.
2.1.6.7.1. Infrared channels will not work with silver-based black and white film (silver blocks infrared).
2.1.6.8. Scan the image.
2.1.6.9. Store the file on the appropriate storage device.
2.1.6.10. Required: do not compress the data.
2.1.6.11. Recommended: do not use filters (sharpening, blurring, descreening, etc.).
2.1.6.11.1. The use of a filter is recommended for removing the orange mask while scanning color negatives.
2.1.6.12. Recommended: do not clip the endtones off the image's actual dynamic range (although clipping of very sparse and "insignificant" tone is encouraged).
2.1.6.13. Recommended: do not create an input/output differential (that is, do not scale).
2.1.6.14. Recommended: review image to ensure quality.
2.1.6.15. Note: we do not want to "improve" the negative in any way; the master should capture the negative with the highest fidelity possible (without using filters). Photo editing software can be used at a later date, when creating derivative files, to improve, touch-up, enhance or sharpen the derivative image.
2.1.7. Storage Media
2.1.7.1. Original master file on one optical disc and a copy of the original master file on a different optical disc should be stored on any of the following: CD-R, DVD+R, DVD-R, DVD+RW. All files will be stored at University of Washington. Copies can be prepared for any collection owner upon request.
2.1.7.2. Workstation hard drives. As space permits, images can be stored temporarily on local hard drives. Ultimately they will be transferred to optical discs and deleted from these hard drives.
2.1.7.3. Other Alternatives: owners requiring other options can talk to CMP staff to discuss borrowing hardware, using FTP, etc.
2.1.8. Final Storage Media Name
2.1.8.1. The name of each optical disc should have three components:
2.1.8.1.1. A three-character project code: "CMP."
2.1.8.1.2. The optical disc type -- CD (indicated by the lack of any code) or DVD (indicated by "DVD"). Examples
2.1.8.1.2.1. CD = CMP[rest of disc number]
2.1.8.1.2.2. DVD = CMPDVD[rest of disc number]
2.1.8.1.3. The disc number in the project's complete sequence of discs.
2.1.8.1.4. Examples:
2.1.8.1.4.1. CDs: CMP1, CMP2, CMP3, ... CMP27, CMP28,... CMP96, CMP97...
2.1.8.1.4.2. DVDs: CMPDVD1, CMPDVD2, ... CMPDVD27, CMPDVD28...
2.1.8.2. The name should be digitally assigned to the disc and written on the disc case, along with other descriptive information about the disc contents.
2.1.8.3. There are two options to identify the name of the actual optical disc, to pair it with its case and contents:
2.1.8.3.1. On the case, write the two numbers printed and/or engraved around the center hole of the actual CD. This is the preferred method.
2.1.8.3.2. Use a marker to write the CD name for this project on the innermost cylinder of the disc, around the center hole. Try not to cover any numbers printed or engraved on the CD. DO NOT write on the parts of the CD under which data is stored.
2.1.9. Backup
2.1.9.1. Local backup system at the owning organization. This will be done on a case-by-case basis.
2.1.9.2. Images on CDs: One run of optical discs (CDs and DVDs) will accommodate all original masters and copies of masters (backups). Ultimately all derivatives will also be stored on these discs.
2.1.10. Copy for University of Washington Archive
2.1.10.1. The run of optical discs will contain all project files. These will be kept in a fire-protection file cabinet until the files can be moved to a digital repository, probably D-Space (which may never happen, as it will happen outside the boundaries of this project).
2.1.11. Targets
2.1.11.1. Targets are not required.
2.1.11.2. Organizations can choose to use targets.
2.1.11.2.1. Targets may be scanned separately, before a batch is scanned, or with each individual negative or group of negatives.
2.1.11.2.1.1. When scanned separately, targets should be distinguished from image files.
2.1.11.2.1.1.1. Targets should have file names that identify them as targets and that identify the images in the batch assessed by the target, ultimately to be stored on the final optical disc. In the event that a target assesses a batch not represented by a consecutive series of image file names, the target file name should refer to an additional document that serves as a concordance of target to batch.
2.1.11.2.1.2. When scanned with individual negatives or groups of negatives, resolution should be determined by the long dimension of the negative only.
2.1.12. Quality Control
2.1.12.1. During scanning:
2.1.12.1.1. Calibrate monitors and devices, according to the operator's judgment. Project staff can provide a calibration worksheet upon request.
2.1.12.1.2. Check the quality of the digital image before transferring the file.
2.1.12.2. After scanning, before derivatives are produced:
2.1.12.2.1. Project staff will review each TIFF while producing derivatives.
2.1.12.2.1.1. The original image will not be available during this review.
2.1.12.3. The "Reviewing Digital Images for Quality" worksheet will help guide the quality review process.
2.1.13. Required metadata at time of capture
2.1.13.1. The following aspects of every item must be described during scanning; if possible, record this information on a spreadsheet:
2.1.13.1.1. File name for digital image;
2.1.13.1.2. Organization's identification number for the negative (accession number, negative number, etc);
2.1.13.1.3. Describe what the item scanned is (a negative, not a print, not a slide, etc.); please be as specific as you can;
2.1.13.1.4. Subject headings assigned by organization;
2.1.13.1.5. Notes accompanying the negative, written on the sleeve, etc.; try to indicate how and where anything is written;
2.1.13.1.6. Dimensions of the negative scanned, using the common unit for measuring the negative (4 x 5, 35 mm, etc.);
2.1.13.1.7. Resolution at which the item was scanned;
2.1.13.1.8. The precise wording on how somebody can order the item, image or other facsimile;
2.1.13.1.9. The precise word-by-word property statement that will appear beneath the image, in the form "Property of...." This must be included and must be stated with pinpoint accuracy. It cannot be changed later;
2.1.13.1.10. Anything else worth noting;
2.1.13.1.11. Name of the person who scanned.
2.1.13.2. For all information not known, enter "unknown" in the spreadsheet.
2.1.13.3. A metadata form can be obtained online for use during scanning; visit the Community Museum Project Cataloging Web Page to see a selection of metadata forms from which to choose.
2.1.13.3.1. During scanning, only minimal metadata is required; however the forms are to assist full description of each item. Use the forms as you see fit at any time in the process of building the digital collection.
2.1.14. Miscellaneous notes on scanning negatives
2.1.14.1. Orienting the film can be difficult. Here are some tips:
2.1.14.1.1. Scan the shiny side (not the emulsion side, which has a duller, rougher surface). This is difficult to detect for color negatives however.
2.1.14.1.1.1. Sometimes scanning the emulsion side produces better images. If the emulsion side is scanned, be sure to flip the image using the scanning software.
2.1.14.1.2. Look at the negative for any written words in the image; scan the side where the words are read normally, left to right.
2.1.14.1.3. Usually any writing along the edge of the film is on the side of the negative we scan (not on the emulsion side).
2.1.14.2. Cleaning dust. Non professional recommendations (for best results, consult a professional!):
2.1.14.2.1. If necessary, use a very clean, natural soft-hair brush (like camel hair). After brushing, an anti-static, lint-free cloth can be used. Also, a liquid film cleaner can be applied and removed with a lint-free cloth.
2.1.14.2.1.1. Glass negatives: DO NOT use a brush on the emulsion side of the negative!
2.1.14.2.2. DO NOT USE: water, cloth that leaves lint or static, cans of compressed air.
2.1.14.3. Some flatbed scanners require a "Transparent Media Adapter" to replace the cover of the bed. Without the adapter, the software probably won't allow you to scan film.
2.2. Derivative Files / Screen Presentation
2.2.1. File Type: JFIF.
2.2.2. Compression: JPEG.
2.2.2.1. The compression ratio should be about 150:1, or even 200:1.
2.2.2.1.1. When using Adobe Photoshop to create JPEGs, use JPEG quality level 3.
2.2.3. Use of Photo Editing Software
2.2.3.1. Rotate image if it is skewed.
2.2.3.2. Crop image, as necessary.
2.2.3.2.1. Each organization should create their own cropping criteria; for example, should informative borders be scanned, should image space only be scanned, etc.
2.2.3.3. Resize image.
2.2.3.3.1. When height is longer than width: resize height to 600 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc. -- the word used depends on the software used).
2.2.3.3.2. When width is longer than height: resize width to 700 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc., depending on the software used).
2.2.3.4. Adjust luminance and/or color using histogram ("levels"), curve tool, or tool of your choice.
2.2.3.4.1. White and black points should be adjusted when the dynamic range of the scanner was set larger than the actual dynamic range of the print.
2.2.3.4.2. Do not clip the endtones off the image's actual dynamic range (although clipping of very sparse and "insignificant" tone is encouraged).
2.2.3.4.3. Adjust color of the color negative, especially if the orange mask was not accounted for during scanning.
2.2.3.5. Unsharp Mask. Most derivatives will require use of this filter.
2.2.3.5.1. Recommended settings: amount 100%, radius 1.0, threshold 0.
2.2.3.5.1.1. If the recommended setting are too extreme, try less amount, smaller radius or a higher threshold.
2.2.3.6. Save As. Make sure you "Save As" (rather than save) so you do not alter the master image! (Recommended: store master images in a read-only directory.)
2.2.3.6.1. Name of master file should be used (with the new file type extension).
2.2.3.6.2. Format: JPEG.
2.2.3.6.3.JPEG options: Adobe Photoshop Quality level 3, Baseline ("Standard") Format Option.
2.2.4. Thumbnail surrogate.
2.2.4.1. Use one of two types of thumbnails:
2.2.4.1.1. CONTENTdm generated. CONTENTdm will produce a thumbnail for every image automatically when the JPEG is uploaded.
2.2.4.1.2. Custom thumbnail. If desired, we can use the thumbnail of our choice. These are easily imported into CONTENTdm. Their maximum size is 160 W x 120 H pixels.
2.2.4.1.2.1. Custom thumbnails will probably be unnecessary for all negatives.
2.2.5. Storage
2.2.5.1. Derivative images will be stored on the same optical discs as the master files. The web images will also be saved on the web site servers.
2.2.6. Uploading
2.2.6.1. Derivatives will be loaded into the CONTENTdm Acquisition Station first.
2.2.6.1.1. Images can be imported by CONTENTdm individually or by batches.
2.2.6.1.1.1. Upon request, project staff will produce an instruction sheet on importing into the CONTENTdm Acquisition Station.
2.2.6.1.2. Any metadata created prior to the import should be imported simultaneously with the images.
2.2.6.1.3. Bands must be added at the time images are imported into CONTENTdm.
2.2.6.1.3.1. "Property of" statements must be obtained before importing into CONTENTdm.
2.2.6.1.3.2. All these "Property of..." statements will be displayed using the CONTENTdm band option, in Verdana type, 8 point, normal script, pure white background (FFFFFF) and black type.
2.2.6.2. Derivatives will be uploaded from the CONTENTdm Acquisition Station to the development server at UW. The built-in file transfer feature in CONTENTdm will be used to make this transfer. Once loaded into the development server, the images and metadata can be edited and viewed using the web-based CONTENTdm Administration Station.
2.2.6.3. Derivatives will be loaded onto the University of Washington public server from the development server using the automated system configured in the University Libraries Digital Initiatives department. This will be done for each digital collection when it is completed.
2.2.7. Band / Watermark / Brand
2.2.7.1. Bands will be used in CONTENTdm, unless the organization negotiates different arrangements.
2.2.7.2. Bands will be added during uploading to CONTENTdm Acquisition Station (as outlined above in section 2.2.6.1.3).
2.3. Print Presentation
2.3.1. Printing Master File: will be determined by the printing parties.
2.3.2. Printing Derivative File
2.3.2.1. JPEG will not print professionally, no matter how it is done. The JPEG does not re-size or scale well. However the images can be printed well enough to serve as reference images.
2.3.2.2. Printing from a CONTENTdm environment (the following was not conclusively tested; it was tested on one machine only running a Windows XP Professional operating system):
2.3.2.2.1. Images appear to print scaled at 100 dpi.
2.3.2.2.1.1. 800 pixel wide images print 8 inches wide.
2.3.2.2.1.2. 700 pixel high images print 7 inches high.
2.3.2.2.1.3. All images print with the property band displayed.
2.3.2.3. Printing outside a CONTENTdm environment
2.3.2.3.1. Images can be saved to the local hard drive. They are saved with the property band attached.
2.3.2.3.2. When opened in Photoshop, image appears to re-scale to 1 dpi (this was not conclusively tested).
2.3.2.3.2.1. To print, image must be edited.
3.1. Master Files
3.1.1. Device(s)
3.1.1.1. Flatbed scanner.
3.1.1.1.1. Two settings may prevent use of a flatbed scanner for slides:
3.1.1.1.1.1. Resolution: resolution must set high enough to meet our resolution recommendation (see section 3.1.2).
3.1.1.1.1.2. Dynamic range: the scanner must be able to capture the much larger dynamic range of slides.
3.1.1.1.1.2.1. Quick check: scan the slide with the dynamic range setting at "Full." If the endpoints on the histogram show significant clipping, consider scanning with a different device.
3.1.1.1.1.2.2. Another check: look for random noise (speckles) in the shadow areas of the digitized slide. If there is a lot of noise, consider scanning with a different device.
3.1.1.1.1.2.2.1. Multi-scanning to eliminate noise will be the exception, not the rule, for the Community Museum Project.
3.1.1.2. Film scanner, if available.
3.1.2. Scan Resolution (this refers to the actual resolution at which the scanner scans)
3.1.2.1. Set the resolution to produce an image with spatial resolution of 3000 pixels across the long dimension. (This does not have to be exact, but exactly 3000 pixels is recommended.)
3.1.3. Bit Depth
3.1.3.1. Black and white slides: 8 bit grayscale.
3.1.3.2. Color slides: 24 bit RGB.
3.1.4. File Type
3.1.4.1. Required: TIFF 6.0.
3.1.5. File Name: file names given to master images will be preserved by all derivatives. This is a central organizing principle for the project; project staff should review all incoming master files and edit file names whenever necessary; files should be named carefully and in accordance with the following recommendations:
3.1.5.1. Each file name should have 4 components:
3.1.5.1.1. An organization code prefix, preferably 2 characters only.
3.1.5.1.1.1. Examples: Hoh Tribe = HT; Clallam County Historical Society = CC.
3.1.5.1.2. A four digit numeral directly following the organization code, representing the sequence of digital images (regardless of original format) in the organization's entire digital collection for this project.
3.1.5.1.2.1. Numerical sequence: 0001, 0002, 0003, ... 0045, 0046, ... 0101, 0102, ... 4672, 4673, ....
3.1.5.1.3. The organization's identifier for the print, separated from the four digit numeral with an underscore.
3.1.5.1.3.1. Identifiers include accession numbers, negative numbers, etc.
3.1.5.1.3.2. Replace all punctuation in the identifier with dashes.
3.1.5.1.3.2.1. Example: accession number 26.32/1 would become _26-32-1.
3.1.5.1.3.2.2. Example: negative number 2003.A-7/2002: 14AB would become 2003-A-7-2002-14AB.
3.1.5.1.4. File type extension. File extension ".tif" will be appended to all master files.
3.1.5.1.5. Examples:
3.1.5.1.5.1. The Timber Museum's 212th image scanned for this project; the print is Timber Museum accession number 43.2.2: "TM0212_43-2-2.tif."
3.1.5.1.5.2. Clallam County Historical Society's 2,032nd scanned image for this project; the print is CCHS negative number 13/4-2003, 6.1: "CC2032_13-4-2003-6-1.tif."
3.1.5.1.6. When organizations scan their own images, we recommend file names that are identical to the organization's identifier, with punctuation intact. CMP staff can edit the file names later to conform with the project-wide file-naming recommendations.
3.1.6. Use of Capture Software
3.1.6.1. Set scanner to scan the media type you are scanning.
3.1.6.1.1. Microtek, for example, has an option "Scan Material" on their "Preferences" menu. For slides select "Positive Film."
3.1.6.2. Your scanner may have other overall settings that need to be adjusted from your main menus; for example, your preview window may need to be configured, your view of scanner settings may need to be configured, you may have a choice between scanning for quality or for speed (choose quality), filters may may need to be disabled, etc.
3.1.6.3.Overview and/or preview image.
3.1.6.4. Crop image (does not need to be exact although an exact crop is preferred).
3.1.6.5. Set dynamic range to "automatic."
3.1.6.6. Set resolution and bit depth as recommended above.
3.1.6.7. If you are using an infrared channel (for example, DigitalICE on a film scanner), it is recommended that you disable this feature; however you may choose to configure this channel to remove dust and scratches: please use this option with caution, as it can blur the image.
3.1.6.8. Scan the image.
3.1.6.9. Store the file on the appropriate storage device.
3.1.6.10. Required: do not compress the data.
3.1.6.11. Recommended: do not use filters (sharpening, blurring, descreening, etc.).
3.1.6.12. Recommended: do not clip the endtones off the image's actual dynamic range (although clipping of very sparse and "insignificant" tone is encouraged).
3.1.6.13. Recommended: do not create an input/output differential (that is, do not scale).
3.1.6.14. Recommended: review image to ensure quality.
3.1.6.15. Note: we do not want to "improve" the slide in any way; the master should capture the slide with the highest fidelity possible (without using filters). Photo editing software can be used at a later date, when creating derivative files, to improve, touch-up, enhance or sharpen the derivative image.
3.1.7. Storage Media
3.1.7.1. Original master file on one optical disc and a copy of the original master file on a different optical disc should be stored on any of the following: CD-R, DVD+R, DVD-R, DVD+RW. All files will be stored at University of Washington. Copies can be prepared for any collection owner upon request.
3.1.7.2. Workstation hard drives. As space permits, images can be stored temporarily on local hard drives. Ultimately they will be transferred to optical discs and deleted from these hard drives.
3.1.7.3. Other Alternatives: owners requiring other options can talk to CMP staff to discuss borrowing hardware, using FTP, etc.
3.1.8. Final Storage Media Name
3.1.8.1. The name of each optical disc should have three components:
3.1.8.1.1. A three-character project code: "CMP."
3.1.8.1.2. The optical disc type -- CD (indicated by the lack of any code) or DVD (indicated by "DVD"). Examples
3.1.8.1.2.1. CD = CMP[rest of disc number]
3.1.8.1.2.2. DVD = CMPDVD[rest of disc number]
3.1.8.1.3. The disc number in the project's complete sequence of discs.
3.1.8.1.4. Examples:
3.1.8.1.4.1. CDs: CMP1, CMP2, CMP3, ... CMP27, CMP28,... CMP96, CMP97...
3.1.8.1.4.2. DVDs: CMPDVD1, CMPDVD2, ... CMPDVD27, CMPDVD28...
3.1.8.2. The name should be digitally assigned to the disc and written on the disc case, along with other descriptive information about the disc contents.
3.1.8.3. There are two options to identify the name of the actual optical disc, to pair it with its case and contents:
3.1.8.3.1. On the case, write the two numbers printed and/or engraved around the center hole of the actual CD. This is the preferred method.
3.1.8.3.2. Use a marker to write the CD name for this project on the innermost cylinder of the disc, around the center hole. Try not to cover any numbers printed or engraved on the CD. DO NOT write on the parts of the CD under which data is stored.
3.1.9. Backup
3.1.9.1. Local backup system at the owning organization. This will be done on a case-by-case basis.
3.1.9.2. Images on CDs: One run of optical discs (CDs and DVDs) will accommodate all original masters and copies of masters (backups). Ultimately all derivatives will also be stored on these discs.
3.1.10. Copy for University of Washington Archive
3.1.10.1. The run of optical discs will contain all project files. These will be kept in a fire-protection file cabinet until the files can be moved to a digital repository, probably D-Space (which may never happen, as it will happen outside the boundaries of this project).
3.1.11. Targets
3.1.11.1. Targets are not required.
3.1.11.2. Organizations can choose to use targets.
3.1.11.2.1. Targets may be scanned separately, before a batch is scanned, or with each individual negative or group of slides.
3.1.11.2.1.1. When scanned separately, targets should be distinguished from image files.
3.1.11.2.1.1.1. Targets should have file names that identify them as targets and that identify the images in the batch assessed by the target, ultimately to be stored on the final optical disc. In the event that a target assesses a batch not represented by a consecutive series of image file names, the target file name should refer to an additional document that serves as a concordance of target to batch.
3.1.11.2.1.2. When scanned with individual slides or groups of slides, resolution should be determined by the long dimension of the slide only.
3.1.12. Quality Control
3.1.12.1. During scanning:
3.1.12.1.1. Calibrate monitors and devices, according to the operator's judgment. Project staff can provide a calibration worksheet upon request.
3.1.12.1.2. Check the quality of the digital image before transferring the file.
3.1.12.2. After scanning, before derivatives are produced:
3.1.12.2.1. Project staff will review each TIFF while producing derivatives.
3.1.12.2.1.1. The original image will not be available during this review.
3.1.12.3. The "Reviewing Digital Images for Quality" worksheet will help guide the quality review process.
3.1.13. Required metadata at time of capture
3.1.13.1. The following aspects of every item must be described during scanning; if possible, record this information on a spreadsheet:
3.1.13.1.1. File name for digital image;
3.1.13.1.2. Organization's identification number for the slide (accession number, negative number, etc);
3.1.13.1.3. Describe what the item scanned is (a slide, not a print, not a negative, etc.); please be as specific as you can;
3.1.13.1.4. Subject headings assigned by organization;
3.1.13.1.5. Notes accompanying the slide, written on the mount, etc.; try to indicate how and where anything is written;
3.1.13.1.6. Dimensions of the slide scanned, using the common unit for measuring the negative (for example, 35 mm);
3.1.13.1.7. Resolution at which the item was scanned;
3.1.13.1.8. The precise wording on how somebody can order the item, image or other facsimile;
3.1.13.1.9. The precise word-by-word property statement that will appear beneath the image, in the form "Property of...." This must be included and must be stated with pinpoint accuracy. It cannot be changed later;
3.1.13.1.10. Anything else worth noting;
3.1.13.1.11. Name of the person who scanned.
3.1.13.2. For all information not known, enter "unknown" in the spreadsheet.
3.1.13.3. A metadata form can be obtained online for use during scanning; visit the Community Museum Project Cataloging Web Page to see a selection of metadata forms from which to choose.
3.1.13.3.1. During scanning, only minimal metadata is required; however the forms on the Community Museum Project Cataloging Web Page are to assist full description of each item. Use the forms as you see fit at any time in the process of building the digital collection.
3.1.14. Miscellaneous notes on scanning slides
3.1.14.1. Orienting the slide can be difficult. Here are some tips:
3.1.14.1.1. Scan the shiny side (not the emulsion side, which has a duller, rougher surface). This is difficult to detect for color slides however.
3.1.14.1.2. The mount may have a logo or writing; if so, it is probably on the emulsion side.
3.1.14.1.3. Look at the slide itself for anything that can help you orient the film.
3.1.14.2. Cleaning dust. Non professional recommendations (for best results, consult a professional!):
3.1.14.2.1. If necessary, use a very clean, natural soft-hair brush (like camel hair). After brushing, an anti-static, lint-free cloth can be used.
3.1.14.2.2. DO NOT USE: water, cloth that leaves lint or static, cans of compressed air.
3.1.14.3. Some flatbed scanners require a "Transparent Media Adapter" to replace the cover of the bed. Without the adapter, the software probably won't allow you to scan film.
3.2. Derivative Files / Screen Presentation
3.2.1. File Type: JFIF.
3.2.2. Compression: JPEG.
3.2.2.1. The compression ratio should be about 150:1, or even 200:1.
3.2.2.1.1. When using Adobe Photoshop to create JPEGs, use JPEG quality level 3.
3.2.3. Use of Photo Editing Software
3.2.3.1. Rotate image if it is skewed.
3.2.3.2. Crop image, as necessary.
3.2.3.2.1. Each organization should create their own cropping criteria; for example, should informative borders be scanned, should image space only be scanned, etc.
3.2.3.3. Resize image.
3.2.3.3.1. When height is longer than width: resize height to 600 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc. -- the word used depends on the software used).
3.2.3.3.2. When width is longer than height: resize width to 700 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc., depending on the software used).
3.2.3.4. Adjust luminance and/or color using histogram ("levels"), curve tool, or tool of your choice.
3.2.3.4.1. White and black points should be adjusted when the dynamic range of the scanner was set larger than the actual dynamic range of the print.
3.2.3.4.2. Do not clip the endtones off the image's actual dynamic range (although clipping of very sparse and "insignificant" tone is encouraged).
3.2.3.5. Unsharp Mask. Most derivatives will require use of this filter.
3.2.3.5.1. Recommended settings: amount 100%, radius 1.0, threshold 0.
3.2.3.5.1.1. If the recommended setting are too extreme, try less amount, smaller radius or a higher threshold.
3.2.3.6. Save As. Make sure you "Save As" (rather than save) so you do not alter the master image! (Recommended: store master images in a read-only directory.)
3.2.3.6.1. Name of master file should be used (with the new file type extension).
3.2.3.6.2. Format: JPEG.
3.2.3.6.3.JPEG options: Adobe Photoshop Quality level 3, Baseline ("Standard") Format Option.
3.2.4. Thumbnail surrogate.
3.2.4.1. Use one of two types of thumbnails:
3.2.4.1.1. CONTENTdm generated. CONTENTdm will produce a thumbnail for every image automatically when the JPEG is uploaded.
3.2.4.1.2. Custom thumbnail. If desired, we can use the thumbnail of our choice. These are easily imported into CONTENTdm. Their maximum size is 160 W x 120 H pixels.
3.2.4.1.2.1. Custom thumbnails will probably be unnecessary for all slides.
3.2.5. Storage
3.2.5.1. Derivative images will be stored on the same optical discs as the master files. The web images will also be saved on the web site servers.
3.2.6. Uploading
3.2.6.1. Derivatives will be loaded into the CONTENTdm Acquisition Station first.
3.2.6.1.1. Images can be imported by CONTENTdm individually or by batches.
3.2.6.1.1.1. Upon request, project staff will produce an instruction sheet on importing into the CONTENTdm Acquisition Station.
3.2.6.1.2. Any metadata created prior to the import should be imported simultaneously with the images.
3.2.6.1.3. Bands must be added at the time images are imported into CONTENTdm.
3.2.6.1.3.1. "Property of" statements must be obtained before importing into CONTENTdm.
3.2.6.1.3.2. All these "Property of..." statements will be displayed using the CONTENTdm band option, in Verdana type, 8 point, normal script, pure white background (FFFFFF) and black type.
3.2.6.2. Derivatives will be uploaded from the CONTENTdm Acquisition Station to the development server at UW. The built-in file transfer feature in CONTENTdm will be used to make this transfer. Once loaded into the development server, the images and metadata can be edited and viewed using the web-based CONTENTdm Administration Station.
3.2.6.3. Derivatives will be loaded onto the University of Washington public server from the development server using the automated system configured in the University Libraries Digital Initiatives department. This will be done for each digital collection when it is completed.
3.2.7. Band / Watermark / Brand
3.2.7.1. Bands will be used in CONTENTdm, unless the organization negotiates different arrangements.
3.2.7.2. Bands will be added during uploading to CONTENTdm Acquisition Station (as outlined above in section 3.2.6.1.3).
3.3. Print Presentation
3.3.1. Printing Master File: will be determined by the printing parties.
3.3.2. Printing Derivative File
3.3.2.1. JPEG will not print professionally, no matter how it is done. The JPEG does not re-size or scale well. However the images can be printed well enough to serve as reference images.
3.3.2.2. Printing from a CONTENTdm environment (the following was not conclusively tested; it was tested on one machine only running a Windows XP Professional operating system):
3.3.2.2.1. Images appear to print scaled at 100 dpi.
3.3.2.2.1.1. 800 pixel wide images print 8 inches wide.
3.3.2.2.1.2. 700 pixel high images print 7 inches high.
3.3.2.2.1.3. All images print with the property band displayed.
3.3.2.3. Printing outside a CONTENTdm environment
3.3.2.3.1. Images can be saved to the local hard drive. They are saved with the property band attached.
3.3.2.3.2. When opened in Photoshop, image appears to re-scale to 1 dpi (this was not conclusively tested).
3.3.2.3.2.1. To print, image must be edited.
4.1. Master Files
4.1.1. Device(s)
4.1.1.1. Digital camera
4.1.1.2. Analog camera: will be used only by special request. This section (section 4) describes the use of a digital camera only.
4.1.2. Camera resolution and related issues:
4.1.2.1. Resolution settings that should be addressed: megapixels, focal length, compression. This is the bare minimum of what needs to be addressed; an advanced camera operator may like to be more exact in setting resolution than is recommended here.
4.1.2.1.1. Megapixels: if the camera claims to use 6.5 megapixels or less, use the largest uninterpolated pixel dimensions available. If the camera claims to use more than 6.5 megapixels, create an image with pixels dimensions 3072 x 2048 or the nearest equivalent.
4.1.2.1.1.1. "Interpolation" here refers to constructing additional pixels that add to the width and height of the digital image based on an algorithmic prediction. This does not include other types of interpolation, like those performed to simplify color filtering, which may be unavoidable.
4.1.2.1.2. Focal Length: use of optical zoom is fine; there should be no use of digital zoom. Macro mode is fine; close up filters should not be used.
4.1.2.1.3. Compression: use the least amount of compression the camera allows; if there is a no-compression setting, use it.
4.1.3. Bit Depth
4.1.3.1. If a setting is available, use at least 24-bit or, preferably, higher for color; at least 8-bit, preferably 16, for black and white.
4.1.3.1.1. Many cameras will not have a bit depth setting that can be changed; in these cases you will have to accept the default value.
4.1.4. File Type
4.1.4.1. TIFF, when available
4.1.4.2. JFIF/JPEG, when TIFF is not available.
4.1.5. File Name: file names given to master images will be preserved by all derivatives. This is a central organizing principle for the project; project staff should review all incoming master files and edit file names whenever necessary; files should be named carefully and in accordance with the following recommendations:
4.1.5.1. Each file name should have 4 components:
4.1.5.1.1. An organization code prefix, preferably 2 characters only.
4.1.5.1.1.1. Examples: Hoh Tribe = HT; Clallam County Historical Society = CC.
4.1.5.1.2. A four digit numeral directly following the organization code, representing the sequence of digital images (regardless of original format) in the organization's entire digital collection for this project.
4.1.5.1.2.1. Numerical sequence: 0001, 0002, 0003, ... 0045, 0046, ... 0101, 0102, ... 4672, 4673, ....
4.1.5.1.3. The organization's identifier for the print, separated from the four digit numeral with an underscore.
4.1.5.1.3.1. Identifiers include accession numbers, negative numbers, etc.
4.1.5.1.3.2. Replace all punctuation in the identifier with dashes.
4.1.5.1.3.2.1. Example: accession number 26.32/1 would become _26-32-1.
4.1.5.1.3.2.2. Example: negative number 2003.A-7/2002: 14AB would become 2003-A-7-2002-14AB.
4.1.5.1.4. The word "digicam" separated from the identifier with an underscore, to distinguish this image as the original.
4.1.5.1.5. File type extension. File extension ".tif" will be appended to all master files, although some digicams will only produce JPEGs and the master image will have to be a JPEG with ".jpg" file extension.
4.1.5.1.6. Examples:
4.1.5.1.6.1. The Timber Museum's 256th digital object for this project, including all scans, audio files, etc., which is a photo of a farm implement with the accession number FI1996.17.3: "TM0256_FI1996-17-3_digicam.tif."
4.1.5.1.6.2. Clallam County Historical Society's 2,172nd digital object for this project, a photograph of an old pair of shoes with the classification number Shoes/17/1987/43a: "CC2172_Shoes-17-1987-43a_digicam.tif."
4.1.5.1.7. Most digicams will assign an automated file name to each image. When the images are moved to a computer workstation, we recommend that file names get changed to accord with the above recommendations. In all cases, CMP staff should review all images from digicams and assign the appropriate file names.
4.1.6. Use of Capture Software. Capture software is on the camera itself.
4.1.6.1. Set resolution and bit depth as outlined above.
4.1.6.2. ISO setting: automatic recommended. Advanced operators should set as needed.
4.1.6.3. Focus: automatic recommended, sometimes using multiple frames of reference, as available. Advanced operators may want to control this manually.
4.1.6.4. Exposure: automatic recommended, usually using multiple frames of reference, as available. When on automatic, some cameras will have an exposure compensation or EV option that can beset manually. At some point all operators are expected to benefit from having manual exposure controls, and they should be used at the camera operator's judgment.
4.1.6.5. Shutter speed: most cameras offer a "long" setting for evening shooting. In almost all cases, this should be disabled.
4.1.6.6. Formatting: this option should be available when using CompactFlash memory cards. Sometimes it is necessary to format the card before using the camera.
4.1.6.7. White balance: automatic recommended. Advanced operators may want to set this manually.
4.1.6.8. Flash: automatic recommended, or disable the flash. Many operators will want to set this manually. We recommend avoiding the flash as often as possible, as inexpert use of flash usually yields poor results.
4.1.6.9. Effects: please, no effects. This includes "vivid," "blur" or "low sharpening," "sepia," etc. However "black and white" is often considered an effect; although black and white photos are not anticipated from digital camera usage in this project, it is a justifiable option.
4.1.6.10. Use macro mode when using auto focus. Close up filters should not be used.
4.1.6.11. Other options: there are many more options, such as AE Lock, FE Lock, AF Assist Beam, most which are either on or, by default, off. Many of these options help make better photographs; operators should use these features at their own discretion.
4.1.7. Storage Media
4.1.7.1. Original master file on one optical disc and a copy of the original master file on a different optical disc should be stored on any of the following: CD-R, DVD+R, DVD-R, DVD+RW. All files will be stored at University of Washington. Copies can be prepared for any collection owner upon request.
4.1.7.2. Workstation hard drives. As space permits, images can be stored temporarily on local hard drives. Ultimately they will be transferred to optical discs and deleted from these hard drives.
4.1.7.3. Other Alternatives: owners requiring other options can talk to CMP staff to discuss borrowing hardware, using FTP, etc.
4.1.7.4. Camera's memory card. This is where the images will get stored initially. They need to be transferred to the next storage device, usually a workstation.
4.1.8. Final Storage Media Name
4.1.8.1. The name of each optical disc should have three components:
4.1.8.1.1. A three-character project code: "CMP."
4.1.8.1.2. The optical disc type -- CD (indicated by the lack of any code) or DVD (indicated by "DVD"). Examples
4.1.8.1.2.1. CD = CMP[rest of disc number]
4.1.8.1.2.2. DVD = CMPDVD[rest of disc number]
4.1.8.1.3. The disc number in the project's complete sequence of discs.
4.1.8.1.4. Examples:
4.1.8.1.4.1. CDs: CMP1, CMP2, CMP3, ... CMP27, CMP28,... CMP96, CMP97...
4.1.8.1.4.2. DVDs: CMPDVD1, CMPDVD2, ... CMPDVD27, CMPDVD28...
4.1.8.2. The name should be digitally assigned to the disc and written on the disc case, along with other descriptive information about the disc contents.
4.1.8.3. There are two options to identify the name of the actual optical disc, to pair it with its case and contents:
4.1.8.3.1. On the case, write the two numbers printed and/or engraved around the center hole of the actual CD. This is the preferred method.
4.1.8.3.2. Use a marker to write the CD name for this project on the innermost cylinder of the disc, around the center hole. Try not to cover any numbers printed or engraved on the CD. DO NOT write on the parts of the CD under which data is stored.
4.1.9. Backup
4.1.9.1. Local backup system at the owning organization. This will be done on a case-by-case basis.
4.1.9.2. Images on CDs: One run of optical discs (CDs and DVDs) will accommodate all original masters and copies of masters (backups). Ultimately all derivatives will also be stored on these discs.
4.1.9.3. Images on the camera's memory card: these cannot be backed up and should be transferred to a workstation and backed up at the end of each photographing session.
4.1.10. Copy for University of Washington Archive
4.1.10.1. The run of optical discs will contain all project files. These will be kept in a fire-protection file cabinet until the files can be moved to a digital repository, probably D-Space (which may never happen, as it will happen outside the boundaries of this project).
4.1.11. Targets
4.1.11.1. Targets are not expected with original digital photographs for this project.
4.1.11.2. If targets or resolving tests are used, files should be created showing the result of the test as a digital image. These files should be distinguished from other image files.
4.1.11.2.1. Targets should have file names that identify them as targets and that identify the images in the batch assessed by the target, ultimately to be stored on the final optical disc. In the event that a target assesses a batch not represented by a consecutive series of image file names, the target file name should refer to an additional text document that serves as a concordance of target to batch.
4.1.11.2.2. If an assessment of camera performance based on the target is created as some sort of text file, it should be given a file name and stored on the final optical disc. The file name of the target-as-a-digital-image should refer to this object.
4.1.12. Quality Control
4.1.12.1. During a photographing session:
4.1.12.1.1. Review images on the camera's LCD monitor or digital viewfinder.
4.1.12.1.2. Check the quality of the digital image on a workstation before transferring the file to Community Museum Project staff.
4.1.12.2. After loading to a computer workstation (and possibly burning to an optical disc), but before derivatives are produced:
4.1.12.2.1. Project staff will review each master image while producing derivatives.
4.1.12.3. The "Reviewing Digital Images for Quality" worksheet may help guide the quality review process (even though it is primarily intended to assess scanned images).
4.1.13. Required metadata at time of capture (or at time of upload to a computer workstation):
4.1.13.1. The following aspects of every item must be described before transferring the file to Community Museum Project staff; if possible, record this information on a spreadsheet:
4.1.13.1.1. File name for digital image;
4.1.13.1.2. Organization's identification number for the item photographed (accession number, classification number, etc);
4.1.13.1.3. Describe what the item scanned is (a basket, a re-created room, etc.); please be as specific as you can;
4.1.13.1.4. Subject headings assigned by organization;
4.1.13.1.5. Notes accompanying the object, whether written on the object, typed on a paper, , etc.; try to indicate how and where anything is written;
4.1.13.1.6. Dimensions of the item photographed, rounding up to the nearest 1/8 inch;
4.1.13.1.7. Camera used;
4.1.13.1.8. Camera settings used;
4.1.13.1.9. The precise wording on how somebody can order the item, image or other facsimile;
4.1.13.1.10. The precise word-by-word property statement that will appear beneath the image, in the form "Property of...." This must be included and must be stated with pinpoint accuracy. It cannot be changed later;
4.1.13.1.11. Anything else worth noting;
4.1.13.1.12. Name of the person who photographed.
4.1.13.2. For all information not known, enter "unknown" in the spreadsheet.
4.1.13.3. A metadata form can be obtained online for use while photographing; visit the Community Museum Project Cataloging Web Page to see a selection of metadata forms from which to choose.
4.1.13.3.1. While photographing, only minimal metadata is required; however the forms available at the Community Museum Project Cataloging Web Page are to assist full description of each item. Use the forms as you see fit at any time in the process of building the digital collection.
4.1.14. Miscellaneous notes on using digital cameras
4.1.14.1. Use of lighting, backdrops, external meters, and so on, that are used while photographing are entirely the judgment of the camera operator. If possible, briefly describe the use these accoutrements in your metadata.
4.2. Derivative Files / Screen Presentation
4.2.1. File Type: JFIF.
4.2.2. Compression: JPEG.
4.2.3. Use of Photo Editing Software
4.2.3.1. If necessary, change bit depth to 24-bit and color "mode" to RGB.
4.2.3.2. Rotate image if it is skewed.
4.2.3.3. Crop image, as necessary.
4.2.3.3.1. Each organization should create their own cropping criteria; for example, should objects be tightly framed, etc.
4.2.3.4. Resize image.
4.2.3.4.1. When height is longer than width: resize height to 600 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc. -- the word used depends on the software used).
4.2.3.4.2. When width is longer than height: resize width to 700 pixels. Make sure the ratio of dimension proportions remains constant (that is "keep" proportions, "constrain" proportions, etc., depending on the software used).
4.2.3.5. Adjust luminance and/or color using histogram ("levels"), curve tool, or tool of your choice.
4.2.3.5.1. White and black points should be adjusted when the dynamic range shows unused pixels at the extreme white and black points.
4.2.3.5.2. Do not clip the endtones off the image's actual dynamic range (although clipping of very sparse and "insignificant" tone is encouraged).
4.2.3.6. Unsharp Mask. Most derivatives will require use of this filter.
4.2.3.6.1. Recommended settings: amount 100%, radius 1.0, threshold 0.
4.2.3.6.1.1. If the recommended setting are too extreme, try less amount, smaller radius or a higher threshold.
4.2.3.7. Save As. Make sure you "Save As" (rather than save) so you do not alter the master image! (Recommended: store master images in a read-only directory.)
4.2.3.7.1. Name of master file should be used (with the new file type extension).
4.2.3.7.2. Format: JPEG.
4.2.3.7.3.JPEG options: Adobe Photoshop Quality level 3, Baseline ("Standard") Format Option.
4.2.4. Thumbnail surrogate
4.2.4.1. Use one of two types of thumbnails:
4.2.4.1.1. CONTENTdm generated. CONTENTdm can produce a thumbnail for every image automatically when the JPEG is uploaded.
4.2.4.1.2. Custom thumbnail. If desired, we can use the thumbnail of our choice. These are easily imported into CONTENTdm. Their maximum size is 160 W x 120 H pixels.
4.2.4.1.2.1. Custom thumbnails will be unnecessary for most original digital photographs; however if multiple photographs are to be treated as one object -- for example multiple views of the same artifact -- we may want to import these into CONTENTdm as a compound object and may want to create a custom thumbnail.
4.2.5. Storage
4.2.5.1. Derivative images will be stored on the same optical discs as the master files. The web images will also be saved on the web site servers.
4.2.6. Uploading
4.2.6.1. Derivatives will be loaded into the CONTENTdm Acquisition Station first.
4.2.6.1.1. Images can be imported by CONTENTdm individually or by batches.
4.2.6.1.1.1. Multiple images can be imported as a single "compound object." These must be imported from a common directory without additional images that are not part of the compound object.
4.2.6.1.1.2. Upon request, project staff will produce an instruction sheet on importing into the CONTENTdm Acquisition Station.
4.2.6.1.2. Any metadata created prior to the import should be imported simultaneously with the images.
4.2.6.1.3. Bands must be added at the time images are imported into CONTENTdm.
4.2.6.1.3.1. "Property of" statements must be obtained before importing into CONTENTdm.
4.2.6.1.3.2. All these "Property of..." statements will be displayed using the CONTENTdm band option, in Verdana type, 8 point, normal script, pure white background (FFFFFF) and black type.
4.2.6.2. Derivatives will be uploaded from the CONTENTdm Acquisition Station to the development server at UW. The built-in file transfer feature in CONTENTdm will be used to make this transfer. Once loaded into the development server, the images and metadata can be edited and viewed using the web-based CONTENTdm Administration Station.
4.2.6.3. Derivatives will be loaded onto the University of Washington public server from the development server using the automated system configured in the University Libraries Digital Initiatives department. This will be done for each digital collection when it is completed.
4.2.7. Band / Watermark / Brand
4.2.7.1. Bands will be used in CONTENTdm, unless the organization negotiates different arrangements.
4.2.7.2. Bands will be added during uploading to CONTENTdm Acquisition Station (as outlined above in section 4.2.6.1.3).
4.3. Print Presentation
4.3.1. Printing Master File: will be determined by the printing parties.
4.3.2. Printing Derivative File
4.3.2.1. JPEG will not print professionally, no matter how it is done. The JPEG does not re-size or scale well. However the images can be printed well enough to serve as reference images.
4.3.2.2. Printing from a CONTENTdm environment (the following was not conclusively tested; it was tested on one machine only running a Windows XP Professional operating system):
4.3.2.2.1. Images appear to print scaled at 100 dpi.
4.3.2.2.1.1. 800 pixel wide images print 8 inches wide.
4.3.2.2.1.2. 700 pixel high images print 7 inches high.
4.3.2.2.1.3. All images print with the property band displayed.
4.3.2.3. Printing outside a CONTENTdm environment
4.3.2.3.1. Images can be saved to the local hard drive. They are saved with the property band attached.
4.3.2.3.2. When opened in Photoshop, image appears to re-scale to 1 dpi (this was not conclusively tested).
4.3.2.3.2.1. To print, image must be edited.
5.1. Master Files
5.1.1. Device(s): at present we have no devices we can rely on. If planning the creation of audio files, please be advised that CMP staff will help you use any devices at your disposal; you may want to consider the following situations at the planning stage:
5.1.1.1. Analog recording to digital; equipment may include:
5.1.1.1.1. Analog playback device, preferably a stand alone device (for example, a cassette player, a turntable, etc.). Required in all cases.
5.1.1.1.2. Mixing board: optional intermediary device (or software) between a playback device and the CD burner.
5.1.1.1.3. CD Recorder or other stand alone digital recording device: use with playback device and optional mixer. Playback device should connect directly to the recorder using standard stereo cables.
5.1.1.1.4. Analog to digital converter: "card" or "board" that can be inserted into slot on a workstation's system board. Many good "soundcards" will do-the-trick.
5.1.1.1.5. Computer workstation. Minimal features of this computer:
5.1.1.1.5.1. 256 MB or more of memory.
5.1.1.1.5.2. A recent microprocessor, rated at a minimum of approximately 800 MHz, like a Pentium III (or higher).
5.1.1.1.5.3. Sound editing software (for example, CoolEdit/now, Adobe Audition, Sound Forge, etc.).
5.1.1.1.5.4. Optical disc burning software (many operating systems will suffice).
5.1.1.1.5.5. Internal or external optical disc burning hardware.
5.1.1.2. Live speech to digital
5.1.1.2.1. Digital audio recording device; could be a powerful, stand alone device, or it could be a very small MiniDisc player/recorder. Possibilities include: handheld computers, desktop computers, portable computers, consumer player/recorders, MiniDisc players, voice recorders, professional solid state recorders, CD Recorders.
5.1.1.2.1.1. Small digital audio playback/recording devices are best when able to support an external microphone and have an onboard level indicator to set the microphone -- provided there is a microphone available that will produce a better recording than the device's built-in microphone.
5.1.1.2.1.1.1. Most PocketPCs and Palm Devices will not support this recommendation.
5.1.1.2.1.2. Most MiniDisc players require real-time re-recording of their contents (not uploading), making these devices less practical than others.
5.1.1.2.2. Computer workstation: it is possible to record directly to a workstation with an adequate sound card and equally adequate audio input device. It may not support an external microphone and sound editing software may be required in this case.
5.1.1.3. Digital to digital
5.1.1.3.1. Computer workstation with the features listed under 5.1.1.1.5. above.
5.1.1.3.1.1. Especially useful in this case: sound editing software.
5.1.1.3.2. File transfer hardware (wires, CD burning drive, etc.) and/or software (FTP, local network, CD burning software, etc.)
5.1.2. Sampling resolutions
5.1.2.1. Sample rate: 44.1 kHz for highest quality; a sample rate of 22.05 kHz is acceptable for recording voice only.
5.1.2.2. Channels: one (mono) is fine for voice; when required, two channels should be used for stereo. This decision will have to be made on a recording-by-recording basis.
5.1.2.3. Amplitude resolution: see 5.1.3. below, "Bit Depth."
5.1.3. Bit Depth: 16-bit, if available.
5.1.4. File Type
5.1.4.1. Preferred: Waveform Audio File Format (WAV, WAVE, .wav).
5.1.4.2. Acceptable: Audio Interchange File Format (AIFF, .aif, .ief)
5.1.4.2.1. NOT AIFF-Compressed, AIFF-C or AIFC!
5.1.5. File Name
5.1.5.1. Each audio file name should have 3 or 4 components:
5.1.5.1.1. A two character organization code prefix.
5.1.5.1.1.1. Examples: Hoh Tribe = HT; Clallam County Historical Society = CC.
5.1.5.1.2. A four digit numeral directly following the organization code, representing the sequence of digital objects (regardless of original format) in the organization's entire digital collection for this project.
5.1.5.1.2.1. Numerical sequence: 0001, 0002, 0003, ... 0045, 0046, ... 0101, 0102, ... 4672, 4673, ....
5.1.5.1.3. When applicable: the organization's identifier for the original recording -- when the original is not the digital recording itself -- separated from the four digit numeral with an underscore.
5.1.5.1.3.1. Identifiers include accession numbers, tape numbers, etc.
5.1.5.1.3.2. Replace all punctuation in the identifier with dashes.
5.1.5.1.3.2.1. Example: accession number 26.32/1 would become 26-32-1.
5.1.5.1.4. File type extension. File extension ".wav" for WAV files; file extension ".aif" for AIFF files. Other extensions as needed for service files that do not follow this recommendation.
5.1.5.1.5. Examples:
5.1.5.1.5.1. The Timber Museum's 44th digital object for this project, including all scans, audio files, etc., which is a recording of an interview on a cassette tape with cassette number 1999: 13/7: "TM0044_1999-13-7.wav."
5.1.5.1.5.2. Clallam County Historical Society's 2,032nd digital object for this project, a recording of songs on reel to reel tape with the classification number 17/1987/43a: "CC2032_17-1987-43a.aif."
5.1.6. Use of Capture Software
5.1.6.1. At the discretion of the device operator. In general the device should try to capture the original with the highest fidelity possible. Any "improvements" can be made with sound editing software.
5.1.6.2. Do not record in "Long Play" modes.
5.1.6.3. Set resolution, bit depth and file type as outlined above in 5.1.2 through 5.1.4.
5.1.6.4. Channel settings -- mono or stereo -- should be selected.
5.1.6.5. Microphone levels can be set manually to minimize distortion.
5.1.6.6. Some features are obviously beneficial (e.g. Motor Noise Reduction Structure when using a microphone) while others are a judgment call (Gain Control, MIC sensitivity, Track Marking, Equalization) based on the knowledge of the operator and the requirements of the situation.
5.1.7. Storage Media
5.1.7.1. Original master file on one optical disc and a copy of the original master file on a different optical disc should be stored on any of the following: CD-R, DVD+R, DVD-R, DVD+RW. All files will be stored at University of Washington. Copies can be prepared for any collection owner upon request.
5.1.7.2. Workstation hard drives. As space permits, images can be stored temporarily on local hard drives. Ultimately they will be transferred to optical discs and deleted from these hard drives.
5.1.7.3. Other Alternatives: owners requiring other options can talk to CMP staff to discuss borrowing hardware, using FTP, etc.
5.1.8. Final Storage Media Name
5.1.8.1. The name of each optical disc should have three components:
5.1.8.1.1. A three-character project code: "CMP."
5.1.8.1.2. The optical disc type -- CD (indicated by the lack of any code) or DVD (indicated by "DVD"). Examples
5.1.8.1.2.1. CD = CMP[rest of disc number]
5.1.8.1.2.2. DVD = CMPDVD[rest of disc number]
5.1.8.1.3. The disc number in the project's complete sequence of discs.
5.1.8.1.4. Examples:
5.1.8.1.4.1. CDs: CMP1, CMP2, CMP3, ... CMP27, CMP28,... CMP96, CMP97...
5.1.8.1.4.2. DVDs: CMPDVD1, CMPDVD2, ... CMPDVD27, CMPDVD28...
5.1.8.2. The name should be digitally assigned to the disc and written on the disc case, along with other descriptive information about the disc contents.
5.1.8.3. There are two options to identify the name of the actual optical disc, to pair it with its case and contents:
5.1.8.3.1. On the case, write the two numbers printed and/or engraved around the center hole of the actual CD. This is the preferred method.
5.1.8.3.2. Use a marker to write the CD name for this project on the innermost cylinder of the disc, around the center hole. Try not to cover any numbers printed or engraved on the CD. DO NOT write on the parts of the CD under which data is stored.
5.1.9. Backup
5.1.9.1. Local backup system at the owning organization. This will be done on a case-by-case basis.
5.1.9.2. Images on CDs: One run of optical discs (CDs and DVDs) will accommodate all original masters and copies of masters (backups). Ultimately all derivatives will also be stored on these discs.
5.1.10. Copy for University of Washington Archive
5.1.10.1. The run of optical discs will contain all project files. These will be kept in a fire-protection file cabinet until the files can be moved to a digital repository, probably D-Space (which may never happen, as it will happen outside the boundaries of this project).
5.1.11. Targets
5.1.11.1. Targets, resolving tests, or other similar assessment tools are not anticipated for sound files.
5.1.11.2. If a target of some sort is used, it should be saved (or some type of surrogate should be saved) on the project optical discs. The file(s) should be named in such a way to distinguish it as a target file. The name of the file should also contain the range of files assessed by the target. If a range of audio files are not easily summarized in a file name, create another file -- a text file -- that lists the audio files assessed by the target. In this case, when naming the target file, point to the text file. Additionally if there are test results that are not clear from the graphic representation of the target, create a text file that explains the test results; the target file name should, in the case, point to this file as well.
5.1.12. Quality Control
5.1.12.1. Analog recording or digital recording to digital recorder
5.1.12.1.1. Check sections of the recording for quality assurance. Any objective measurement that can be applied using sound editing software would be helpful.
5.1.12.1.2. If re-recording is necessary when loading the files on a workstation (as with MiniDisks), be sure to check the quality of the re-recorded files.
5.1.12.1.3. Community Museum Project staff will only provide spot checks of audio files -- both masters and derivatives -- when producing derivatives.
5.1.12.1.4. If the quality is not adequate, re-do the digital recording. Be careful to distinguish what is poor about the original and do not attribute that poor quality to the new digital recording.
5.1.12.2. Live recording to digital recorder
5.1.12.2.1. Prevention of mishaps is really the best way to ensure quality in this case:
5.1.12.2.1.1. Make sure your device works well before going into the field.
5.1.12.2.1.2. Make sure you know how to use your device.
5.1.12.2.1.3. Make sure you set up and configure all peripherals so that they work well.
5.1.12.2.1.4. Run a quick check before you begin a recording session.
5.1.12.2.2. Spot check your final product. If you believe repairs are necessary and you can make them, do so, using sound editing software. Be sure you preserve a copy of the original recording in case something goes very wrong. You will not be able to simply reproduce these recordings if you no longer have the original!
5.1.12.2.3. If re-recording is necessary when loading the files on a workstation (as with MiniDisks), be sure to check the quality of the re-recorded files.
5.1.12.2.4. Community Museum Project staff will only provide spot checks of audio files -- both masters and derivatives -- when producing derivatives.
5.1.12.2.5. If sound quality is not adequate, the recording will have to be discarded.
5.1.13. Required metadata at time of capture
5.1.13.1. The following aspects of every item must be described before transferring the file to Community Museum Project staff; if possible, record this information on a spreadsheet:
5.1.13.1.1. File name for audio file;
5.1.13.1.2. If there is an original from which the audio file was derived, list the organization's identification number for the item digitized (accession number, classification number, cassette number, etc);
5.1.13.1.3. Describe what the item digitized is (a sound disc, a sound tape reel, a piano roll, etc.); please be as specific as you can;
5.1.13.1.4. Subject headings assigned by organization;
5.1.13.1.5. Notes accompanying the object, whether written on the object, typed on a paper, , etc.; try to indicate how and where anything is written;
5.1.13.1.6. Dimensions of the item recorded, rounding up to the nearest 1/16 inch;
5.1.13.1.7. recording devices used;
5.1.13.1.8. settings used on each device;
5.1.13.1.9. The precise wording on how somebody can order a copy of the item;;
5.1.13.1.10. A precise word-by-word property statement that will be associated with the audio file; it may be appended to a surrogate image for the sound file and it may be entered in the public metadata display; in the former case the property statement must be included and must be stated with pinpoint accuracy. It cannot be changed later;
5.1.13.1.11. Anything else worth noting;
5.1.13.1.12. Name of the person who did the recording.
5.1.13.2. For all information not known, enter "unknown" in the spreadsheet.
5.1.13.3. A metadata form can be obtained online; visit the Community Museum Project Cataloging Web Page to see a selection of metadata forms from which to choose.
5.1.13.3.1. While recording, only minimal metadata is required; however the forms are to assist full description of each item. Use the forms as you see fit at any time in the process of building the digital collection.
5.1.14. [No miscellaneous notes on sampling audio]
5.2. Derivative Files / Screen Presentation
5.2.1. File Type
5.2.1.1. Preferred: MPEG I - audio layer 3 compression (MP3, .mp3).
5.2.1.2. Accepted: Real Audio file (.ra, .rm, .ram) -- only when MP3 is not available.
5.2.2. Compression
5.2.2.1. Preferred: "MP3" compression.
5.2.2.2. Accepted: "Real Audio" compression.
5.2.3. Use of Sound Editing Software
5.2.3.1. Presently entirely at the discretion of the creator of the derivative.
5.2.4. Thumbnail surrogate
5.2.4.1. Custom thumbnail imported into CONTENTdm. Choose an appropriate image, size to 160W x 120 H and import into CONTENTdm as a custom thumbnail.
5.2.5. Storage
5.2.5.1. Derivative files will be stored on the same optical discs as the master files. The thumbnail images will also be saved on the web site servers.
5.2.6. Uploading
5.2.6.1. Derivatives will be loaded into the CONTENTdm Acquisition Station first.
5.2.6.1.1. Audio files can be imported by CONTENTdm individually or in batches.
5.2.6.1.1.1. Multiple files can be imported as a single "compound object." These must be imported from a common directory containing only the files for the compound object.
5.2.6.1.1.2. Upon request, project staff will produce an instruction sheet on importing into the CONTENTdm Acquisition Station.
5.2.6.1.2. Any metadata created prior to the import should be imported simultaneously with the audio files.
5.2.6.1.3. Bands/brands/watermarks will probably not be used. Possibilities here include creating property statements in the file header (using sound editing software) and, of course, in the public metadata display.
5.2.6.2. Derivatives will be uploaded from the CONTENTdm Acquisition Station to the development server at UW. The built-in file transfer feature in CONTENTdm will be used to make this transfer. Once loaded into the development server, the files can be heard and the metadata can be edited using the web-based CONTENTdm Administration Station.
5.2.6.3. Derivatives will be loaded onto the University of Washington public server from the development server using the automated system configured in the University Libraries Digital Initiatives department. This will be done for each digital collection when it is completed.
5.2.7. Band / Watermark / Brand
5.2.7.1 Undecided, but no band, brand or watermark is anticipated.
5.3. Print Presentation: not applicable.
6.1. Master Files. Note: master files will be produced for presentation only, not preservation. We may want to think of the masters as "service files" rather than as "master files;" compared to derivatives, the service copy will be less compressed than a derivative. In some cases, the service file will be equivalent to the derivative. These service files will be stored on the project optical discs with all other digital objects for the project.
6.1.1. Device(s): at present we have no devices we can rely on. In planning the creation of video files however the following recommendations suggest devices and settings to create these files:
6.1.1.1. Analog video tape to digital
6.1.1.1.1. Analog playback device, preferably a stand alone device (for example, a VHS tape player, a beta tape player, etc.).
6.1.1.1.1.1. Note: most videotape players transmit an analog signal, either "S-video" or "composite." Very high quality recordings will transmit "component" signals. In all cases care should be taken that the adapter card on the computer system board accepts the type of signal transmitted.
6.1.1.1.1.2. Note: the physical connector for the video transmission will probably be a BNC connector or an S-Video connector. RCA audio connectors may also be needed.
6.1.1.1.2. Analog videotape to digital converter: a "card" or "board" that can be inserted into a slot on a workstation system board; a home video unit that performs the conversion and that can be connected to a computer workstation will also work well.
6.1.1.1.2.1. This converter must support the input signal from the analog tape player. All other features are added features; supporting analog input is the necessary feature for conversion to a digital file.
6.1.1.1.3. Camcorder: some cameras have a port for video/audio input and convert the analog signal in the camera, using the camera's file storage method (DV tape, etc.). The resulting digital video can then be sent to a workstation for processing and storage.
6.1.1.1.4. Computer workstation.
6.1.1.1.4.1. 256 MB memory -- preferably much more.
6.1.1.1.4.2. A new microprocessor, rated at a minimum of approximately 1 GHz, like a Pentium 3 -- this is low power however for processing video. The video card used may have an additional processor on board to help process video.
6.1.1.1.4.3. Video editing software (for example, Adobe Premiere, etc.). If you plan to drive the video card using this software, the card must support the software specifically.
6.1.1.1.4.4. DVD disc burning hardware and software. (CDs can be used for short video clips.)
6.1.1.1.4.5. An port appropriate to the needs of input devices used; most often, a "firewire" port is required (IEEE 1394); sometimes USB2 will be okay.
6.1.1.2. Live video recording to digital
6.1.1.2.1. Digital video recording device
6.1.1.2.1.1. Camcorder. DV/Mini-DV camcorders are generally good quality, reasonably priced and readily available. Digital8, DVD and memory card camcorders are also fine.
6.1.1.2.1.2. The camcorder should create a digital file. The camcorder should also have an IEEE 1394 ("Firewire") port in which a cable can be inserted and connected to a workstation to transfer the video file.
6.1.1.2.1.3. The camcorder should support external microphones.
6.1.1.2.1.4. The digitization should be performed by the camera; however, if the card on the workstation system board is expected to process the file further, it should be compatible with the camcorder data type. Areas of compatibility include: file type, type of compression, audio type.
6.1.1.2.2. Analog video recording device (also called a camcorder)
6.1.1.2.2.1. If an analog camera is all that's available, create a video and use the steps outlined in section 6.1.1.1 to create the digital file.
6.1.1.2.3. Microphones: the camcorder can probably support two channels of sound and, by extension, two microphones. Selection of microphone types is the judgment of the camera operator.
6.1.1.2.4. Card on system board
6.1.1.2.4.1. When the camcorder creates the digital file, only a straight-through IEEE-1394 port is needed on the system board (this is a feature often included on new workstations, without adding a card to the system board).
6.1.1.3. Film to digital video: procedures yet to be determined.
6.1.1.4. Digital video to digital video
6.1.1.4.1. Use a straight through IEEE 1394 data transfer (that is, firewire cable between two devices with firewire ports). Change file formats, if necessary, using video editing software.
6.1.2. Resolution
6.1.2.1. Video resolution will generally be determined by the file type. For service files in MPEG-2 format, 720 x 486 pixel resolution is expected.
6.1.2.2. Audio should be captured at 44.1 kHz and 16 bits for highest quality; for conversation/ interviews only, a sample rate 32 kHz and 12 bits is okay (actually voices can be adequately sampled at 22.05 kHz). Use of channels is at the operator's discretion.
6.1.3. Bit Depth
6.1.3.1. At least 24 bits per pixel for color reproduction, if available.
6.1.3.2. 16 bits per pixel for black and white (grayscale) preferred, if available.
6.1.4. File Type
6.1.4.1. Preferred: MPEG-2 for service files.
6.1.4.2. Acceptable: MPEG-1, especially when service files are identical to derivatives.
6.1.4.3. Note: D1 files (uncompressed digital video) are excellent but impractical and will not be used. DV and MJPG files should be converted to the more standard formats listed above. Most other file formats are suitable only for derivatives: Quick Time, AVI, Real Video, Windows Media, etc.
6.1.5. File Name
6.1.5.1. Each video file name should have at least 3, and as many as 5, components:
6.1.5.1.1. A two character organization code prefix.
6.1.5.1.1.1. Examples: Hoh Tribe = HT; Clallam County Historical Society = CC.
6.1.5.1.2. A four digit numeral directly following the organization code, representing the sequence of digital objects (regardless of original format) in the organization's entire digital collection for this project.
6.1.5.1.2.1. Numerical sequence: 0001, 0002, 0003, ... 0045, 0046, ... 0101, 0102, ... 4672, 4673, ....6.1.5.1.3. When applicable: the organization's identifier for the original video -- when the original is not the digital video itself -- separated from the four digit numeral with an underscore.
6.1.5.1.3.1. Identifiers include accession numbers, tape numbers, etc.
6.1.5.1.3.2. Replace all punctuation in the identifier with dashes.
6.1.5.1.3.2.1. Example: accession number 26.32/1 would become 26-32-1.
6.1.5.1.4. If the complete video is broken into separate files for practical reasons, distinguish the separate files only by adding a part number to a common file name. The part number should come after the organization's identifier for the original recording, separated from that identifier with an underscore. The form of the part number should be "_part1," "_part7," "_part14," etc.
6.1.5.1.5. File type extension. File extension ".mpg" for MPEG-1 and MPEG-2 files. Others as needed.
6.1.5.1.6. Examples:
6.1.5.1.6.1. The Quileute Tribe's 144th digital object for this project, which is a collection of 3 digital video files of the VHS tape with VHS number 1999: 41: "QT0144_1999-41_part1.mpg," "QT0144_1999-41_part2.mpg," "QT0144_1999-41_part3.mpg."
6.1.5.1.6.2. Clallam County Historical Society's 6,032nd digital object for this project, a digital video file of a VHS tape featuring song performances, with the classification number 17/1982/143a: "CC6032_17-1982-143a.mpg."
6.1.6. Use of Capture Software
6.1.6.1. The camcorder, the microphones and the video card will be used at the discretion of the operator. In general the device should try to capture the original with the highest fidelity possible. Any "improvements" can be made with video editing software. The main required settings are detailed above: namely, file type and audio resolution. "Quality" (not objectively measured however) is also required.
6.1.7. Storage Media
6.1.7.1. Original service file on one optical disc and a copy of the original master file on a different optical disc, using any of the following: CD-R, DVD+R, DVD-R, DVD+RW. Alternatively, video may be stored permanently on camcorder storage, like DV tape. All files will be stored at University of Washington. Copies can be prepared for any collection owner upon request.
6.1.7.2. Workstation hard drives. As space permits, images can be stored temporarily on local hard drives. Ultimately they will be transferred to optical discs and deleted from these hard drives.
6.1.7.3. Other Alternatives: owners requiring other options can talk to CMP staff to discuss borrowing hardware, using FTP, etc.
6.1.8. Final Storage Media Name
6.1.8.1. The name of each optical disc should have three components:
6.1.8.1.1. A three-character project code: "CMP."6.1.8.1.2. The optical disc type -- CD (indicated by the lack of any code) or DVD (indicated by "DVD"). Examples
6.1.8.1.2.1. CD = CMP[rest of disc number]
6.1.8.1.2.2. DVD = CMPDVD[rest of disc number]
6.1.8.1.3. The disc number in the project's complete sequence of discs.
6.1.8.1.4. Examples:
6.1.8.1.4.1. CDs: CMP1, CMP2, CMP3, ... CMP27, CMP28,... CMP96, CMP97...
6.1.8.1.4.2. DVDs: CMPDVD1, CMPDVD2, ... CMPDVD27, CMPDVD28...
6.1.8.2. The name should be digitally assigned to the disc and written on the disc case, along with other descriptive information about the disc contents.
6.1.8.3. There are two options to identify the name of the actual optical disc, to pair it with its case and contents:
6.1.8.3.1. On the case, write the two numbers printed and/or engraved around the center hole of the actual CD. This is the preferred method.
6.1.8.3.2. Use a marker to write the CD name for this project on the innermost cylinder of the disc, around the center hole. Try not to cover any numbers printed or engraved on the CD. DO NOT write on the parts of the CD under which data is stored
6.1.8.4 If the storage for the service files is not an optical disc but a DV tape or some other camcorder storage, try to name the storage unti in accordance with these recommendations; if that is impractical, simply do-the-best-you-can to produce service file storage with names meaningful in their context.
6.1.9. Backup
6.1.9.1. Video on the camcorder: these probably cannot be backed up and should be transferred to a workstation and backed up at the end of every recording session.
6.1.9.2. Video on hard drives: local backup system.
6.1.9.2.1. If there is no local backup system, or if the local backup system cannot accommodate the files, the files should be transferred after each recording session.
6.1.9.3. Images on CDs: One run of optical discs (CDs and DVDs) will accommodate all original masters and copies of masters (backups). Ultimately all derivatives will also be stored on these discs.
6.1.10. Copy for University of Washington Archive
6.1.10.1. The run of optical discs (and perhaps some other storage media, like DV tape) will contain all project files. These will be kept in a fire-protection file cabinet until the files can be moved to a digital repository, probably D-Space (which may never happen, as it will happen outside the boundaries of this project).
6.1.11. Targets
6.1.11.1. Targets, resolving tests, or other similar assessment tools will not be used for video, unless the operator or organization requires them.
6.1.11.2. If targets of some sort are used, they should be represented similarly to targets for other media types.
6.1.12. Quality Control
6.1.12.1. Producers of a video file should review the entire file before submitting to Community Museum Project staff. Accepted imperfections should be noted and Project staff should be informed.
6.1.12.2. Project staff will review some or all contents of a digital file while producing derivatives. Any service files that are poor quality will be a candidate for removal from the collection. This decision should be made collectively.
6.1.13. Required metadata at time of capture
6.1.13.1. The following aspects of every item must be described before transferring the file to Community Museum Project staff; if possible, record this information on a spreadsheet:
6.1.13.1.1. File name for video file;
6.1.13.1.2. If there is an original from which the video file was derived, list the organization's identification number for the item digitized (accession number, classification number, cassette number, etc);
6.1.13.1.3. Describe what the item digitized is (a VHS tape, a 16mm film, etc.); please be as specific as you can;
6.1.13.1.4. Subject headings assigned by organization;
6.1.13.1.5. Notes accompanying the digitized item, whether written on the item, typed on a paper, etc.; try to indicate how and where anything is written;
6.1.13.1.6. Dimensions of the item digitized, rounding up to the nearest 1/16 inch;
6.1.13.1.7. Digitizing devices used;
6.1.13.1.8. Settings used on each device;
6.1.13.1.9. The precise wording on how somebody can order the item, image or other facsimile;
6.1.13.1.10. The precise word-by-word property statement that may accompany the image, in the form "Property of...." This must be included and must be stated with pinpoint accuracy. It may not be changeable later;
6.1.13.1.11. Anything else worth noting;
6.1.13.1.12. Name of the person who did the digitizing.
6.1.13.2. For all information not known, enter "unknown" in the spreadsheet.
6.1.13.3. A metadata form can be obtained online; visit the Community Museum Project Cataloging Web Page to see a selection of metadata forms from which to choose.
6.1.13.3.1. While digitizing, only minimal metadata is required; however the forms are to assist full description of each item. Use the forms as you see fit at any time in the process of building the digital collection.
6.1.14. Miscellaneous notes on creating digital video files
6.1.14.1. None at present
6.2. Derivative Files / Screen Presentation
6.2.1. File Type
6.2.1.1. For streaming files use formats supported by DigitalWell at present: Windows Media, Quicktime, RealMedia, MPEG-2, MPEG-4.
6.2.1.2. For downloadable files (to be used only if space is not available on the server from which files are streamed):
6.2.1.2.1. Preferred: MPEG-1.
6.2.1.2.2. Other formats can be accepted when MPEG-1s cannot be produced; this should be arranged on a case-by-case basis. (VOB/IFO/BUP files may not be useable; check with CMP staff before you produce or submit these types of video files.)
6.2.2. Compression. Determined by file type.
6.2.3. Use of Video Editing Software
6.2.3.1. Convert the file to the proper format. A typical process for a streaming video file would include:
6.2.3.1.1. Open the file.
6.2.3.1.2. Choose a function equivalent to exporting the file.
6.2.3.1.3. Choose an export mode such as "Save For Web."
6.2.3.1.4. Select a format for the exported file.
6.2.3.1.4.1. Advanced settings have not yet been determined; for example, options such as "Video for Web Servers (56 Kbps)," "Video for single channel ISDN (64 Kbps)," etc., have not been discussed. Whenever possible however, a multi-rate stream should be created.
6.2.3.1.4.2. As DigitalWell is a stream management environment, files for http streaming should not be created (for example, "progressive download" formats).
6.2.3.1.5. Save the file, which will be uploaded to DigitalWell; a hyperlink will be created in CONTENTdm metadata probably using the file name you create at this point.
6.2.3.2. Besides converting the file format as necessary, the use of video editing software will be at the discretion of operators.
6.2.4. Thumbnail surrogate
6.2.4.1. Custom thumbnail imported into CONTENTdm. Choose an appropriate image, size to
160W x 120 H and import into CONTENTdm as a custom thumbnail.6.2.5. Storage
6.2.5.1. Derivative video files intended to stream will be stored on DigitalWell servers.
6.2.5.1.1. Arrange an upload with CMP staff. The file will be stored on a DigitalWell server and accessed through a hyperlink uploaded to CONTENTdm.
6.2.5.2. Derivative video files intended for download will be stored on UW servers.
6.2.5.1.1. Files will be uploaded to a development server.
6.2.5.1.1. When the database is ready to be served to the public, the video files will be
duplicated on the public server. Video files will remain on the development server also, where
edits can be made. The public server is currently updated on a regular basis with changes from
the development server.6.2.5.3. Derivative video should always be copied the project's optical discs just like the service files, and these discs will be stored at University of Washington Libraries.
6.2.6. Uploading
6.2.6.1. Streaming files will be uploaded to DigitalWell by CMP staff only. Three files, (1) a custom thumbnail image, (2) an HTML file containing the link to the actual video file, and (3) a text file containing descriptive metadata will be loaded into the CONTENTdm Administration Station.
6.2.6.2. Downloadable files
6.2.6.2.1. Load into the CONTENTdm Acquisition Station first.
6.2.6.2.1.1. Video files can be imported by CONTENTdm individually or in batches.
6.2.6.2.1.1.1. Multiple files can be imported as a single "compound object." These must be imported from a common directory without additional files that are not part of the compound object.
6.2.6.2.1.2. Upon request, project staff will produce an instruction sheet on importing into the CONTENTdm Acquisition Station.
6.2.6.2.2. Any metadata created prior to the import should be imported simultaneously with the video files.
6.2.6.2.3. Bands/brands/watermarks will not be used. Possibilities here include creating property statements in the file header (using video editing software) and, of course, in the public display of metadata.
6.2.6.3. Upload files in the CONTENTdm Acquisition Station to the development server at UW. The built-in file transfer feature in CONTENTdm will be used to make this transfer. Once loaded into the development server, the files can be viewed and the metadata can be edited using the web-based CONTENTdm Administration Station.
6.2.6.4. Derivatives will be loaded onto the University of Washington public server from the development server using the automated system configured in the University Libraries Digital Initiatives department. This will be done for each digital collection when it is completed.
6.2.7. Band / Watermark / Brand
6.2.7.1 Undecided. However CONTENTdm will probably not be useful for this.
6.3. Print Presentation : not applicable.
7.1. Master Files
7.1.1. Device(s)
7.1.1.1. Flatbed scanner
7.1.2. Scan Resolution (this refers to the actual resolution at which the scanner scans): recommendations:
7.1.2.1. Text only
7.1.2.1.1. Images less than 2 inches in the long dimension: scan at 800 ppi.
7.1.2.1.2. Images between 2 and 7 inches in the long dimension: scan at 600 ppi.
7.1.2.1.3. Images between 7 and 12 inches in the long dimension: scan at 400 ppi.
7.1.2.1.4. Images larger than 12 inches in the long dimension: scan at 300 ppi.
7.1.2.2. Text with images printed using a halftone process (newspapers, magazines, most
books).7.1.2.2.1. Text is considered more important than images: use the resolution guidelines in 7.1.2.1, selectively decreasing resolution to improve image quality, if desired.
7.1.2.2.2. Images are considered more important than text: there are several options:
7.1.2.2.2.1. Use the descreening filter on your scanner, if it works satisfactorily. Be sure to describe the use of this filter in the preliminary meta (see 7.1.13 below).
7.1.2.2.2.2. The scanner operator should determine an adequate decrease to the resolution
guidelines in 7.1.2.1, thereby decreasing the detail at which the halftone patterns are captured.7.1.2.2.2.3. Equivalent or increased resolution to that recommended in 7.1.2.1 above; this will require intensive use of photo editing software; specifically, it should be treated by a software editing program's blur (first), resize (second), and sharpen (last) tools, at the operator's judgment.
7.1.2.2.3. Text and images are equally important: this will be at the operator's judgment; generally slightly lower resolutions can be used than those outlined above in 7.1.2.1; also use of photo editing software will be highly advised. The scanner's descreening tool may also be helpful.
7.1.3. Bit Depth
7.1.3.1. 1-bit bitonal: black and white text only, line art, or anything that can be sufficiently presented in two tones only.
7.1.3.2. 8-bit grayscale: text with black and white images, text where shades of gray should be preserved; newspaper print is generally improved with grayscale also. However grayscale is often not a good choice for text, as anti-aliasing blurs the letters.
7.1.3.3. 24-bit RGB: when color is required, even if only to preserve tones in black and white printing.
7.1.4. File Type
7.1.4.1. Required: TIFF 6.0.
7.1.5. File Name
7.1.5.1. Each text image file name should have at least 3, and as many as 5, components:
7.1.5.1.1. A two character organization code prefix (required).
7.1.5.1.1.1. Examples: Hoh Tribe = HT; Clallam County Historical Society = CC.
7.1.5.1.2. A four digit numeral directly following the organization code, representing the sequence of digital objects (regardless of original format) in the organization's entire digital collection for this project (required).
7.1.5.1.2.1. Numerical sequence: 0001, 0002, 0003, ... 0045, 0046, ... 0101, 0102, ... 4672, 4673, ....
7.1.5.1.3. When applicable: the organization's identifier for the item scanned, separated from the four digit numeral with an underscore.
7.1.5.1.3.1. identifiers include accession numbers, call numbers, etc.
7.1.5.1.3.2. Replace all punctuation in the identifier with dashes.
7.1.5.1.3.2.1. Example: accession number 26.32/1 would become 26-32-1.
7.1.5.1.4. If there is more than one scanned page from a common item broken into separate files, distinguish the separate files only by adding a page number to a common file name. The page number should come after the organization's identifier for the original item, separated from that identifier with an underscore. The form of the page number should be "_page1," "_page7," "_page14," etc.
7.1.5.1.5. File type extension. File extension ".tif" at the end of the file name.
7.1.5.1.6. Examples:
7.1.5.1.6.1. The Timber Museum's 6,562nd digital object for this project, which is a collection of 3 files of 3 scanned pages from a book with the call number 75TT47XyTJLM: "TM06562_75TT47XyTJLM_page1.tif," "TM06562_75TT47XyTJLM_page2.tif," "TM06562_75TT47XyTJLM_page3.tif."
7.1.5.1.6.2. Clallam County Historical Society's 12,000th digital object for this project, the 722nd page from a book with the classification number Book:Series: 45/76.3:QL45 7171-A39627/ Book:Series-Series2: 2001June/input47.a1: "CC12000_Book-Series-45-76-3-QL45-7171-A39627-Book-Series-Series2-2001June-input47-a1_page722.tif."
7.1.6. Use of Capture Software
7.1.6.1. Set scanner to scan the media type you are scanning.
7.1.6.1.1. Microtek, for example, has an option "Scan Material" on their "Preferences" menu. For printed items select "Reflective."
7.1.6.2. The scanner may have other overall settings that need to be adjusted from the main menus; for example, the preview window may need to be configured, there may be a choice between scanning for quality or for speed (choose quality), filters may be manually disabled, etc.
7.1.6.3. Overview and/or preview the page; when scanning batches, it may be necessary to preview only the first page of a batch.
7.1.6.4. Crop pages when necessary. Much cropping can be done later, using photo editing software, if necessary.
7.1.6.5. Levels, use of a histogram tool, etc.
7.1.6.5.1. Grayscale and RGB: set dynamic range to "automatic;" however manual adjustments using a histogram tool that clarifies anti-aliased text will be accepted (be sure to record these adjustments ion the preliminary metadata -- see section 7.1.13 below). Do not use indexed color; only use 24-bit or higher RGB to scan color materials.
7.1.6.5.2. Bitonal: leave the threshold level at 128 (the midpoint); however, pages that have automatically or manually set threshold levels will be accepted.
7.1.6.6. Set resolution and bit depth as recommended above.
7.1.6.7. Scan the image.
7.1.6.8. Store the file on the appropriate storage device. In most cases the files will be saved in a directory on your computer's hard drive.
7.1.6.9. Required: do not compress the data.
7.1.6.10. Although the use of filters is generally discouraged, while scanning text and text with images, some filtering will be acceptable (such as descreening). Please record all filters used and their settings in the preliminary metadata.
7.1.6.11. Recommended: do not create an input/output differential (that is, do not "scale" the image).
7.1.6.12. Recommended: review pages after scanned to check for either (1) verisimilitude to original, or (2) legibility.
7.1.6.13. Although "improvements" to the pages will be accepted, it may still be best to create a master image with minimal touch-up; photo editing software can be used at a later date, when creating derivative files, to improve, touch-up, enhance or sharpen the derivative image.
7.1.7. Final Storage Media
7.1.7.1. Original master file on one optical disc and a copy of the original master file on a different optical disc should be stored on any of the following: CD-R, DVD+R, DVD-R, DVD+RW. All files will be stored at University of Washington. Copies can be prepared for any collection owner upon request.
7.1.7.2. Workstation hard drives. As space permits, images can be stored temporarily on local hard drives. Ultimately they will be transferred to optical discs and deleted from these hard drives.
7.1.7.3. Other Alternatives: owners requiring other options can talk to CMP staff to discuss borrowing hardware, using FTP, etc.
7.1.8. Final Storage Media Name
7.1.8.1. The name of each optical disc should have three components:
7.1.8.1.1. A three-character project code: "CMP."
7.1.8.1.2. The optical disc type -- CD (indicated by the lack of any code) or DVD (indicated by "DVD"). Examples
7.1.8.1.2.1. CD = CMP[rest of disc number]
7.1.8.1.2.2. DVD = CMPDVD[rest of disc number]
7.1.8.1.3. The disc number in the project's complete sequence of discs.
7.1.8.1.4. Examples:
7.1.8.1.4.1. CDs: CMP1, CMP2, CMP3, ... CMP27, CMP28,... CMP96, CMP97...
7.1.8.1.4.2. DVDs: CMPDVD1, CMPDVD2, ... CMPDVD27, CMPDVD28...
7.1.8.2. The name should be digitally assigned to the disc and written on the disc case, along with other descriptive information about the disc contents.
7.1.8.3. There are two options to identify the name of the actual optical disc, to pair it with its case and contents:
7.1.8.3.1. On the case, write the two numbers printed and/or engraved around the center hole of the actual CD. This is the preferred method.
7.1.8.3.2. Use a marker to write the CD name for this project on the innermost cylinder of the disc, around the center hole. Try not to cover any numbers printed or engraved on the CD. DO NOT write on the parts of the CD under which data is stored.
7.1.9. Backup
7.1.9.1. Local backup system at the owning organization. This will be done on a case-by-case basis.
7.1.9.2. Images on CDs: One run of optical discs (CDs and DVDs) will accommodate all original masters and copies of masters (backups). Ultimately all derivatives will also be stored on these discs.
7.1.10. Copy for University of Washington Archive
7.1.10.1. The run of optical discs will contain all project files. These will be kept in a fire-protection file cabinet until the files can be moved to a digital repository, probably D-Space (which may never happen, as it will happen outside the boundaries of this project).
7.1.11. Targets
7.1.11.1. Targets are not required.
7.1.11.2. Organizations can choose to use targets.
7.1.11.2.1. Targets may be scanned separately, before a batch is scanned, or with each individual page scan.
7.1.11.2.1.1. When scanned separately, targets should be distinguished from other image files.
7.1.11.2.1.1.1. Targets should have file names that identify them as targets and that identify the images in the batch assessed by the target, ultimately to be stored on the final optical disc. In the event that a target assesses a batch not represented by a consecutive series of image file names, the target file name should refer to an additional document that serves as a concordance of target to batch.
7.1.11.2.1.2. When scanned with individual pages, resolution should be determined by the dimensions of the page, as if the target was not present.
7.1.12. Quality Control
7.1.12.1. During scanning:
7.1.12.1.1. Calibrate monitors and devices, according to the operator's judgment. Project staff can provide a calibration worksheet upon request.
7.1.12.1.2. Check the fidelity of the digital page to the original page before transferring the file. In almost all cases, this is the only time the digital page will be compared to the original. However, when making page scans, fidelity to the original may not be as important as legibility and/or clarity of either text or images. A document-by-document strategy should be decided before scanning, so that quality can be measured accurately. If possible, record any unusual strategies in the preliminary metadata (see section 7.1.13 below).
7.1.12.2. After scanning, before derivatives are produced:
7.1.12.2.1. Project staff will review TIFFs while producing derivatives; in some cases all TIFFS will be checked, in others selected TIFFS only will be checked.
7.1.12.3. The "Reviewing Digital Images for Quality" worksheet will help guide the quality review process and is available upon request.
7.1.13. Required metadata at time of capture
7.1.13.1. The following aspects of every item must be described during scanning; if possible, record this information on a spreadsheet:
7.1.13.1.1. File name for each digital image (that is, for each page scan);
7.1.13.1.2. Organization's identification number for the original item (accession number, negative number, etc);
7.1.13.1.3. Describe what the item scanned is (a book, not a slide, not a slide, etc.); please be as specific as you can;
7.1.13.1.4. Subject headings assigned by organization;
7.1.13.1.5. Notes accompanying the original item, written on the item, etc.; try to indicate how and where anything is written;
7.1.13.1.6. Dimensions of the item scanned, rounding up to the nearest 1/8 inch if possible;
7.1.13.1.7. Resolution at which the item was scanned;
7.1.13.1.8. The precise wording on how somebody can order the original item, a scan of the original item or some other facsimile;
7.1.13.1.9. The precise word-by-word property statement that will appear beneath the image, in the form "Property of...." This must be included and must be stated with pinpoint accuracy. It cannot be changed later;
7.1.13.1.10. The use of any filters (like descreening) during scanning and the setting used;
7.1.13.1.11. Any luminance of color adjustments -- using, for example the histogram, adjusting the threshold -- and the settings used.
7.1.13.1.12 Anything else worth noting
7.1.13.1.13. Name of the person who scanned
7.1.13.2. For all information not known, enter "unknown" in the spreadsheet.
7.1.13.3. A metadata form can be obtained online for use during scanning; visit the Community Museum Project Cataloging Web Page to see a selection of metadata forms from which to choose.
7.1.13.3.1. During scanning, only minimal metadata is required; however the forms are to assist full description of each item. Use the forms as you see fit at any time in the process of building the digital collection.
7.2. Derivative Files / Screen Presentation
7.2.1. File Type: JPEG (JFIF) or PDF.
7.2.2. Compression: JPEG (RGB and grayscale), or ZIP (for bitonal), or any of the compressions used by Adobe Acrobat while creating a pdf file.
7.2.2.1. Select the lowest quality that is adequate. Page images may not have a common quality level.
7.2.3. Use of Photo Editing Software
7.2.3.1. Rotate the page if it is skewed.
7.2.3.2. Crop the page, as necessary.
7.2.3.2.1. Each organization should create their own cropping criteria; for example, should informative borders be scanned, should text space only display, should full pages display, etc.
7.2.3.3. Resize the page scan. The project standard is no more than 600 pixels high or than 700 pixels wide, whichever comes first; however text will usually have to deviate from this standard, providing larger image areas if it is required for legibility. The height/width proportion should always remain constant.
7.2.3.4. Adjust luminance and/or color using histogram ("levels"), curve tool, threshold tool or another tool of your choice.
7.2.3.4.1. White and black points should be adjusted when the dynamic range of the scanner was set larger than the actual dynamic range of the print, or when the it is required for legibility or image clarity.
7.2.3.4.2. Clipping of endtones will be at the discretion of the operator. This can be a useful tool in sharpening edges that have been dulled by anti-aliasing.
7.2.3.5. Blur tools. They have various names: "Blur," "Gaussian Blur," etc. This will often be required to correct moire patterns (usually seen in scans of images produced using the halftone printing process). Use as needed.
7.2.3.6. Unsharp Mask. Most derivatives will require use of this filter.
7.2.3.6.1. Most commonly used setting for images in this project: amount 100%, radius 1.0, threshold 0. When applied to an image that has been blurred, the settings will vary image to image. When applied to text, customized sharpening will be required. The operator should exercise judgment in creating settings.
7.2.3.7. Save As. Make sure you "Save As" (rather than save) so you do not alter the master image!
7.2.3.7.1. Name of master file should be re-used (with the new file type extension).
7.2.3.7.2. Format: JPEG or PDF. Settings for compression are at the operator's discretion.
7.2.4. Thumbnail surrogate.
7.2.4.1. Use one of two types of thumbnails:
7.2.4.1.1. CONTENTdm generated. CONTENTdm will produce a thumbnail for every page scan automatically when the JPEG is uploaded.
7.2.4.1.2. Custom thumbnail. Use the image of your choice to represent the document. These are easily imported into CONTENTdm. Their maximum size is 160 W x 120 H pixels.
7.2.5. Storage
7.2.5.1. Derivative images will be stored on the same optical discs as the master files. The web images will also be saved on the web site servers.
7.2.6. Uploading
7.2.6.1. Derivatives will be loaded into the CONTENTdm Acquisition Station first.
7.2.6.1.1. Images can be imported by CONTENTdm individually or in batches.
7.2.6.1.1.1. Upon request, project staff will produce an instruction sheet on importing into the CONTENTdm Acquisition Station.
7.2.6.1.2. Any metadata created prior to the import should be imported simultaneously with the images.
7.2.6.1.3. Bands must be added at the time images are imported into CONTENTdm.
7.2.6.1.3.1. "Property of" statements must be obtained before importing into CONTENTdm.
7.2.6.1.3.2. All these "Property of..." statements will be displayed using the CONTENTdm band option, in Verdana type, 8 point, normal script, pure white background (FFFFFF) and black type.
7.2.6.2. Derivatives will be uploaded from the CONTENTdm Acquisition Station to a development server at UW. The built-in file transfer feature in CONTENTdm will be used to make a transfer from the development server to the live server. Once loaded into the development server, the images and metadata can be edited and viewed using the web-based CONTENTdm Administration Station.
7.2.6.3. Derivatives will be loaded onto the University of Washington public server from the development server using the automated system configured in the University Libraries Digital Initiatives department. This will be done for each digital collection when it is completed.
7.2.7. Band / Watermark / Brand
7.2.7.1. Bands will be used in CONTENTdm, unless the organization negotiates different arrangements.
7.2.7.2. Bands will be added during uploading to CONTENTdm Acquisition Station (as outlined above in section 7.2.6.1.3.2).
7.3. Print Presentation: text can be prepared for screen viewing only.
Master Files --------------------------------- |
Derivative Files ---------------- |
|||||
Media Type | Input Resolution | Bit Depth | File Type | File Type / Compression | Size | Required |
1 Photographic Prints | 300-800 ppi, depending on size of the original | 8-bit grayscale, 24-bit RGB | TIFF | JPEG | Either 600 pixels high or 700 pixels wide, whichever comes first | Text of property statement in the band beneath the image. |
2 Negatives | Set to produce an image 3000 pixels along the long dimension | 8-bit grayscale, 24-bit RGB | TIFF | JPEG | Either 600 pixels high or 700 pixels wide, whichever comes first | Text of property statement in the band beneath the image. |
3 Slides | Set to produce an image 3000 pixels along the long dimension | 8-bit grayscale, 24-bit RGB | TIFF | JPEG | Either 600 pixels high or 700 pixels wide, whichever comes first | Text of property statement in the band beneath the image. |
4 Dimensional Artifacts | Use maximum uninterpolated resolution for cameras under 6.5 megapixels; use least amount of compression available | 8-bit grayscale, 24-bit RGB (minimums)
|
TIFF | JPEG | Either 600 pixels high or 700 pixels wide, whichever comes first | Text of property statement in the band beneath the image. |
5 Audio | 44.1 kHz, but recorded voice only can go as low as 20.05 kHz. | 16-bit (also known as "amplitude resolution"), if available | WAVE, with AIFF acceptable | MP3s, with Real Audio files acceptable | Varies | No "watermarking," banding, etc. |
6 Video (all files are presentation files only; anything unique in the "master file" are properties of a "service" file, which is not archival quality). | 720x486 pixel resolution expected; sound at 44.1 kHz, with recorded voice okay at 20.05 kHz | 24-bit color, if available; 8-bit grayscale okay | MPEG-2 is recommended for service files; MPEG-1 is acceptable; streaming formats (Windows Media, Quicktime, RealMedia) will be accepted with reluctance. | WMV, MOV, RM; others okay upon request | Varies | No "watermarking," banding, etc. |
7 Text | 300-800 ppi recommended (depending on the needs of the document and of the typeface), with "legibility" the primary determinant | 1-bit bitonal, 8-bit grayscale, 24-bit RGB | TIFF | JPEG, PDF | Recommended: either 600 pixels high or 700 pixels wide, whichever comes first; however this will vary according to the needs of legibility and/or clarity | Text of property statement in the band beneath the image. |