ODIN stands for the Open Document Image Network developed at the University of Washington. Based on the World Wide Web, ODIN is an electronic storage and retrieval system for images of documents. It is modeled on a simple filing system so it can be used in a variety of offices.
ODIN is designed to be as user-friendly as possible. Instructions are included directly on the screens to reduce the need for printed materials. In addition, the entire contents of this manual are available online by clicking the `Help' link from any ODIN page. The purpose of this manual is to document the functions and features of ODIN for reference. It is not a substitute for training.
The three main functions of ODIN are scanning, indexing and retrieving document images. The scanning is done on a PC using custom-developed software called ODINScan. The indexing and retrieval are done from a PC or an X terminal using a web browser such as Netscape Navigator.
ODINScan makes electronic copies of documents and stores them in a location where they can be indexed and retrieved by ODIN. The images are like pictures, not like text, so they cannot be edited.
Indexing is the process of assigning key data to an image. This information is used to file, search for and identify the image in the future. Without indexing, the scanned image is not associated with any file and so it can not be retrieved in ODIN.
Once an image is indexed, it can be retrieved and displayed by any authorized user any number of times. Images are organized into files, and a table of contents can be viewed for any file before selecting a specific document image for display.
On a Nebula PC, the web browser can be found on the desktop. The current web browser is Netscape.
Several ODIN screens require choices from a List Box. These boxes look like text fields with an arrow pointing down on the right side. Click anywhere in the box or on the arrow to see the list of options, use the scroll bar if necessary, then click on an option to choose it.
It is possible to copy text between ODIN and other applications using the Windows clipboard. Highlight text and choose Edit, Copy (or press Ctrl-C) to copy it to the clipboard. Position the cursor in the destination field and select Edit, Paste (or use Ctrl-V or Shift-Insert) to paste the text from the clipboard.
Netscape and other Microsoft Windows applications have been known to cause General Protection Faults and other errors. Sometimes Netscape or even Windows must be restarted after an error. These errors do not affect the images in ODIN. Although it is unlikely that they can be prevented in all cases, please report errors to your computer support person if they are a problem.
On an X terminal, the Netscape browser can be started by holding down the left mouse button in the background and highlighting `Reference', then highlighting `Netscape' and releasing the button. Alternatively, it can be added to the xdock so it can be started with a single click on a button.
Several ODIN screens require choices from a List Box. These boxes look like buttons with a shaded dash on the right side. Click on the arrow with the mouse to see the list of options, then click on an option to choose it.
It is possible to copy text between ODIN and other applications using the X terminal Copy and Paste Feature. Mark the text to be copied by highlighting it with the left mouse button. Insert the highlighted text by clicking the middle mouse button in the destination field.
Printing ODIN images from Netscape requires a PostScript printer. The PostScript printing can be very, very slow.
Authorization is required for anyone to scan images into ODIN. If an unauthorized user starts ODINScan, access to the directories will be denied and images cannot be scanned.
ODINScan images are stored in a special shared network directory. Authorized users must log on to the Nebula PC scanning station in order to gain access to that directory. If a user is not logged on, then access to the directories will be denied and images cannot be scanned.
Never pull a page out of the scanner while it is being scanned. Instead, please remove excess documents and wait for the page to finish or choose `Cancel' from the Scan menu.
In the event of a jam in the paper feeder, lift up on the tab that says `UP' before removing the paper.
Set the paper guides on the feeder to match the paper width of the batch by gently squeezing the sides of both guides at the same time and moving them both in or out. Note that the guides always center the paper in the feeder.
Documents can be placed in the feeder in batches or they can be scanned from the flatbed one sheet at a time.
The current scanner setup options can be displayed and modified by choosing 'Scanner Settings' from the Scan menu. Changes apply only to the current session unless they are saved. When ODINScan is started, it uses the most recently saved settings. The initial defaults can be reloaded and saved if desired.
The contrast setting affects how sharp the image appears. The recommended setting is Automatic.
The brightness setting affects how light or dark the overall image appears. The automatic option is not available on all scanners. The initial default is fairly dark to capture faint images such as dot matrix printing and pencil.
This is the resolution of the image. Higher resolution results in more detailed images, but they take more storage space and they take longer to display. The recommended setting is 200 dpi.
The paper size should be set to match the actual size of the input documents. The paper must always be fed into the scanner in the lengthwise (portrait) direction. Landscape images should be captured by selecting `Scan Rotated Batch' (see below), not by feeding them into the scanner sideways.
Changes in the scanner settings apply only to the current session unless they are saved. Choose 'Save Scanner Settings' to make ODINScan remember them in future sessions.
This option will change all the scanner settings to the original defaults. As with other changes, these apply only to the current session unless they are saved.
Selecting this menu item turns the `Show Scanned Image' feature on or off, as indicated by a check mark. When on, the image is displayed after it is scanned and the batch number and page are shown in the ODINScan title bar. When off, the image is not displayed but the batch number and page are shown inside the ODINScan window. Scanning is somewhat faster when this is turned off, but quality can be checked immediately when it is on.
ODINScan has been preconfigured to work with your scanner. Contact your support person to change scanners.
ODINScan allows documents to be scanned in batches (of course, a batch may contain only a single page). All documents in a batch will be scanned with the same scanner settings and in the same orientation, so they should be divided into batches according to size and orientation. Pages can be fed in through the document feeder on top of the scanner or they can be placed one at a time on the glass flatbed. Either way, choosing `Scan Portrait Batch' on the Scan menu begins the process.
If there are pages in the document feeder, ODINScan begins scanning them immediately and ignores anything on the glass. Most documents can be scanned in this way. The pages should be placed face down with the top to the left (toward the output tray). The paper guides should be set to match the width of the paper.
Pages with staples or paper clips cannot be fed through the document feeder, so they should be scanned one at a time on the flatbed portion of the scanner. Badly torn documents, very small documents, and some fax paper are also best handled in this way. Documents should be placed face down with the top to the left (toward the output tray), and they should be aligned with the back left corner of the glass (at the triangular mark). If there are no documents in the feeder, ODINScan will present a dialog box with Scanning Options. Change the paper source to Flatbed and select `Start Scanning'.
When ODINScan finishes scanning all documents in the feeder or any document on the flatbed, it displays a dialog box to ask whether to stop or continue the batch. Placing additional documents in the feeder or on the glass and selecting `Continue Scanning' will add them to the current batch. Selecting `Stop Scanning' will close the batch and make it available for indexing. Selecting `Options' makes it possible to switch between the feeder and the flatbed in the middle of a batch.
This option automatically rotates images 90 degrees before displaying or saving them so that landscape documents can be scanned correctly. The pages should be fed lengthwise into the scanner, face down with the top of the page toward the back (where the cables connect). In all other respects, this works the same as `Scan Portrait Batch'.
The data and images in ODIN are confidential, so access is limited to authorized users. When someone attempts to start ODIN or access any page within the system, they will be prompted for their user name and password. If they do not have authorization, access will be denied.
For additional security, access is only available on machines at the University of Washington.
There are many kinds of computers and many kinds of web browsers, each of which offers slightly different features. Where possible, ODIN is designed to conform to standards so it does not depend on a particular browser. However, the development and testing was done using Netscape Navigator on X terminals and Nebula PCs. Some features may not be available on other platforms.
ODIN can be accessed with world wide web browsers. See the appropriate Special Notes section above for instructions on how to start a web browser on a PC or an X terminal.
Regular users should create an entry in their bookmarks to simplify access to ODIN. The location (URL) for ODIN is
http://www.washington.edu/odin/system/
ODIN requires a valid user name and password that matches an authorized account on a C&C host computer (for example, bank, red, carbon, or green).
Files in ODIN are divided into groups. Each group has a list of users who can see all the files in the group. The group code for the Student Loan files, for example, is `slo'.
It is possible to bypass the group code entry screen by using a location (URL) that specifies the group. A valid user name and password will still be required. For group slo, the URL is
http://www.washington.edu/odin/system/slo.html
ODIN uses links to move from one page to another. These links work the same as links on any other web page. In the default browser configuration, links appear underlined. Click once on any underlined link to go to that page. Use the `Search' link to search for files and the `Index' link to enter key index information for recently scanned documents. Other links appear when applicable.
Standard browser features, such as `Back', can also be used to move from page to page. The `Back' button returns the last web page that was displayed before the current one, essentially going back one page in history. Caution: it is possible to go back to ODIN pages which are no longer valid, and errors may result. For example, after confirming the index entry for a page, it is possible to use `Back' to redisplay the confirmation screen. However, clicking on the link to confirm the index again will result in an error, since a page can only be indexed once.
The look and feel of ODIN can be customized by changing the browser preferences, such as point size and color.
The browser scroll bars can be used to move around on a single page in three ways: click on the arrows at the ends of the scroll bars to move a small amount, click in the scroll bar itself to move one screen at a time, or drag the box inside the scroll bar to move by any amount.
ODIN is designed to display images using the full width of your browser window. To change the size of the displayed page, resize the browser window. Wider windows display larger images but fit less on the screen and take more time. More narrow windows display more of the image on the screen by making it smaller.
Click anywhere in the image to zoom in and display it at full resolution, regardless of window size. The resulting page is usually larger than a computer screen, so scrolling can be used to move around and see the whole document. Use the `Back' button of your browser to return to the standard page display.
ODIN is designed to eliminate or at least radically reduce the need for paper copies of filed business documents. Once an image is filed in ODIN, it can always be retrieved by any authorized user. It can even be saved and emailed (although ODIN cannot prevent tampering with any copies of the image saved outside of ODIN), so printing should not be necessary. Printing copies of scanned images reduces the benefits of an imaging system and is strongly discouraged. It is possible, however, by using the browser `Print' button or choosing File -> Print.
When images are scanned using ODINScan, they are placed into batches for indexing. At this point, ODIN does not contain any information about the pages, so they are not filed until the indexing is complete. Indexing is similar to filing a paper document. A document must be indexed, or placed in the file, before it can be used.
Choose 'Index' in ODIN to go to the Batch Index Selection form, which displays a list of all unindexed batches for a group. The batch numbers correspond to the numbers assigned by ODINScan. Click on a batch number to display the first page of the batch for indexing. It may be helpful to have the batches of paper copies handy. The batches can be indexed in any order. It is a good practice to keep the list of unindexed batches short.
For each batch, ODIN displays the number of pages which remain to be indexed, the total number of pages that were scanned in the batch, and the date they were scanned.
When all pages in a batch have been indexed, that batch number disappears from the Batch Index Selection form. It will reappear if a page is removed from a file for reindexing, but the batch information changes. The date field becomes the date it was removed from the file instead of the date scanned and the page counters are based on the number of pages removed from the file instead of the number of pages originally scanned.
After selecting a batch number, the first page of the batch is displayed for indexing in the Page Index form. The fields on this form are used to identify the document so that it can be retrieved in the future. All fields must be filled in.
The pages in a batch do not have to be indexed in the order they are scanned. Use the `Prev Page' and `Next Page' links to move forward and backward through a batch without indexing or discarding pages. Once a page is indexed or discarded, it is removed from the batch and it will be skipped by these links, which only appear when there is a previous or next page in the current batch.
Electronic images of documents are divided into files in ODIN, just as paper documents are stored in file folders. There may be multiple different ways to identify a file, such as by name or by number. Only one of these keys needs to be entered to identify the file for indexing, but all methods will be available to search for the file after it is indexed. Select one method from the list box.
Type the identifying information in the space to the right of the list box. If you selected name from the list box, enter the file name. If you selected number from the list box, enter the file number. If the information does not uniquely determine a file, then a list of matching files will be displayed.
There can be documents of various types within a file. Select the correct document title from the list of options. This is the title of the document which will appear in the electronic table of contents for the file.
The page number defaults to one (1), but it should be changed to the correct page for multiple-page documents.
Enter the date that will be used to identify the document. This should be the date of the document, not the date it is added to ODIN or indexed. The month, date, and year should be entered as numbers separated by slashes, as in `9/15/96'. All pages with the same document title and date are displayed on a single line in the table of contents when searching for documents.
All the index fields are initialized when entering a batch. The data entry fields are set to blank except for page which has a default value of one (1). The file identifying information, document title, and document date are carried forward from one page to the next within a batch in order to save data entry time. They can still be changed as necessary, but this reduces the need for typing when the documents in a batch are related.
The `Reset Fields' button initializes all the index fields just as they are initialized when first entering a batch.
If a page was scanned by mistake or if it should not be moved into any file, it can be discarded. The discard option is permanent. The only way to recover a discarded page is to scan it again. To discard a page, click on the `Discard this Page' button. No other information is required on the form. ODIN will ask you to confirm your request to discard the page. Select `Okay to DISCARD' to confirm the deletion. Select `Cancel Discard and Return to Index' to undo the discard action and to index the page.
Once the index fields are all filled in, click on the `Store Index Entry' button to add this page and the entered information to the ODIN files.
If the information entered does not uniquely identify an ODIN file, then a list of all matching files is displayed. All the key fields are shown for each file to help identify the correct one. Click on a file to select it or click `Back' to change the entered information.
If the index information matches exactly one ODIN file, the `ODIN Confirm Index' page redisplays the image and the information that was entered, along with the rest of the key information for the file. If the display is correct, click on `Confirm Index Entry' to file the page in ODIN so that it can be retrieved.
If the information displayed on the ODIN Confirm Index page is not all correct, click on `Cancel and Return to Index Page' to change it.
ODIN checks all entries to make sure they are valid. If a required field is left blank or if a field contains invalid data, an error is displayed. After reading the error, use the `Back' button to return to the Page Index screen and correct it.
Electronic images of documents are divided into files in ODIN, just as paper documents are stored in file folders. The ODIN search form is used to search for files in ODIN. Once the correct file is located, individual documents can be selected from the list of the file contents.
There may be multiple different ways to identify a file, such as by name or by number. Select one method from the `Search by' box.
Type the identifier for the desired file in the `Search for' space to the right of the list box. For example, if you selected name from the list box, enter the file name.
To erase the `Search for' field and reset the `Search by' box to the default search method, click on `Reset Form'.
To execute the search, click on the `Begin Search' button or simply press return. Pressing return works on this form because there is only one text entry field. Forms with more than one text entry field cannot be submitted by pressing enter.
If the search submitted does not uniquely determine a file, then a list of matching files will be displayed. All the key fields are shown for each file to help identify the correct one. Click on a file to select it or click `Back' to change the entered information.
If the search submitted does not match any files in ODIN, then a message is displayed which says, ``No Match/Not Found''. Use the `Back' button to change your entry and try again. If, on the other hand, the search does locate a file but the file contains no documents, then a message is displayed which says, ``EMPTY FILE - NO DOCUMENTS FOUND.''
Every file has a table of contents which allows any page of any document in the file to be selected. To display a page, click on the page number to the right of the document title (don't click on the document title itself). If there is only one document in the file, the document is displayed automatically instead of the file contents.
At the top of the ODIN Page Display screen there are several navigation links which allow easy movement through a file.
`TOC' stands for `Table of Contents'. Clicking on `TOC' displays the list of documents in the current file.
The `Prev Page' link moves backward one page in the current document, if there is a previous page. Otherwise it does not display.
The `Next Page' link moves forward one page in the current document, if there is a next page. Otherwise it does not display.
The `Prev Doc' link moves to the first page of the previous document in the current file, if there is a previous document. Otherwise, it does not display.
The `Next Doc' link moves to the first page of the next document in the current file, if there is a next document. Otherwise, it does not display.
To remove a page from a file, search for the file and select the page for display. Then click on `Remove from File'. ODIN will redisplay the page and all the index information for confirmation. Click on `Remove this Page from File' to confirm the action or `Cancel and Return to Page Display' to undo it. Removing a page from a file does not delete the page from ODIN. Instead, it places it back in the list of unindexed documents. The page will appear under its original batch number and can be re-indexed or discarded through the indexing functions.
Until the administrative functions of ODIN are enhanced, access to a group of ODIN files can only be granted or revoked by sending email to odin@cac.washington.edu. Requests such as this must be sent from the designated ODIN contact for the group.
Until the administrative functions of ODIN are enhanced, document titles can only be added to or removed from the list of available titles by sending email to odin@cac.washington.edu. When a document title is removed from the list of titles available in the indexing function, it will still show up on documents which were previously indexed with that title.
Until the administrative functions of ODIN are enhanced, a new file that is not included in the master database can only be added by sending email to odin@cac.washington.edu. For example, if a student loan file must be created in ODIN for a person who does not have a record in the Student Data Base, then the file can be requested by sending email including the name, social security number, and student number.
Open Document Image Network / ODIN / UW C&C © 1995-1998, University of Washington, All Rights Reserved