The Los Altos Hills Digital Scrapbook Project -
the "why" and the "how".

Objectives

The primary objective is of course to "preserve" the content of the town's historical archives (scrapbooks, newspaper clippings, related ephemera) so that they will be avaiable in the extended future - at least in "virtual" form as digital files that are easily accessible to interested readers.

This simply stated objective implies a number of indirect objectives which must also be met if the project is to succeed.

Adaptability and mutation over time to reflect changing technology.

Any given digital format (either the physical recording media or the bit patterns recorded on that media) is likely to become obsolete within a time span measured in years - let alone decades or centuries. Therefore, the archives must be (insofar as we can understand the issues of the future) amenable to be copied and re-recorded onto whatver new technology comes down the pike.

It was just a few years ago that critical archives about technology developments and related patents were recorded on 8-inch floppy disks using Word★Star software - today that data is already lost forever in practical terms. Early photo images taken of the back of the moon by pioneer space probes, radioed back to Earth and archived on massive reels of 1 inch tape and recorded in a helical format were nearly lost forever and only recently recovered at huge expense because both the equipment to read such tapes was no longer available, and the magnetic "imprint" on the tapes had deteriorated over time. Luckily a retired engineer from Ampex could be found as well as an old tape drive which he managed to refurbish and hook to a modern computer with home-brewed hardware and software so the images could be read and re-recorded on more modern media and thus preserved for posterity - but only just in the nick of time.

Whatever technology is used must be widely available and widely understood

This project is now (and likely ever will be) sustained by volunteers from the community and on a shoe-string budget. This implies that any tools used must be widely available to those volunteers (ideally already personally owned by them) and how to use those tools must widely understood within the community to assure a wide population of qualified people from which volunteers can be recruited.

This means no "special" or tailor-made museum software should be used, and file formats must adhere to some standard to maximize the probability that files can be processed or interchanged over time as technolgical advances occur. To this end, the assets used so far include

  • Computer Hardware - Consumer grade home computer, running current version of popular operating system, DVD disks and/or USB Memory sticks

  • Computer Software - Microsoft Excel spreadsheet, Image manipulation and processing (Photoshop or similar), Web browser.

  • Camera - Consumer grade Single Lens Reflex with appropriate lens capable of producing high-resolution images in "Jpeg" format. (A full page of newspaper text readable in a single frame) with "flip out" preview screen (so that the operator can monitor image position and zoom level while the camera is mounted on the copy stand) and cable release

  • Camera "copy stand" with built-in lighting for consistent exposure values and camera-to-document distance adjustment capability

  • Magnetic Bulletin Board large enough to hold a full sheet of newspaper, spray-painted matte black - together with a selection of small black magnets. This is used as a background on which to place pages holding clippings that have been folded (or loose clippings that have been folded). The magnets can be judiciously positioned to hold the document flat without obscuring critical text or harming the document. (A large sheet of perspex was also suggested and tried, but this brought with it lighting reflections from the oblique cross-lights on the copy stand as well as other abient light sources.)

Method

The process used to create a digital scrapbook involves five major sub-processes

  • Photograph each page of a scrapbook or "collection"

  • Process each page image (Align, crop, adjust hue/saturation and contrast) Photoshop or any similar image processing software should be able to do this

  • Create "index" spreadsheet (one per book or collection) with one entry per image, each entry has link to the image file and textual digest of page contents

  • Generate web-format HTML-format index file from spreadsheet by doing appropriate "save as..."

  • Place all processed image files and related Excel and HTM files in a single appropriately named folder, create backup and publish
Each of these processes is described in greater detail below.

Photograph Each Page

The camera is mounted on the copy stand at a height such that the mid-rang of the zoom on the lens (the sweet spot) takes in approximately a page of the typical scrapbook size. (11 by 14 or so). This allows fine adjustment during the photo session via the zoom capabilities of the lens to cause the target image to fill the frame.

The small viewing screen on the camera is flipped out and turned so that the operator can view the image prior to and immediately after shutter releaase.

Even if there is enough ambient light, the two electric lamps help to provide an even light spread over the target. Relying on ambient light coming laterally from one side or another may cause difficulties in achieving an even exposure of the image.

The camera operating mode is set to "Aperture Priority" and the aperture itself is also set to a mid-range value (sweet spot again, say f9 or f11) to optimize image sharpness - shutter speed and ISO sensitivity can then be chosen automatically by the camera system as being of lower priority than image sharpness/resolution.

Autofocus is used specifying the center spot of the image. On some pages, the camera may have difficulty focussing if this center spot happens to co-incide with a blank space on the document being photographed. You will hear the lens "hunting" and no focus-confirming beep will be emitted. In this case you can either shift the document slightly to make sure a contrasty "edge" falls into the focal target area, or place a small target with contrasty lines on the image at the focal target area,

After releasing the shutter with the cable release (to avoid jogging the camera), check via the camera screen that all edges of the image fell within the frame boundaries and proceed to the next page.

A magnetic bulletin board that has been spray-painted matte black can be used as a background surface on which to place loose documents. Small magnets can then be used to hold documents flat that are badly creased or have a tendency to curl.

 

Process images

In most cases, the work flow for a given image consists of the following main steps

  • Rotate the image 180° to make it "right side up"
    Original ImageRight Side Up

  • If necessary, further rotate the image 1 or 2 degrees clockwise or counter-clockwise to fine tune the document to appear "straight" within the image frame.
    Rotated 180°Fine tuned vertical by another 2 °

  • Crop the image to remove extraneous material around the periphery of the image
    Cropped

  • Drive the overall yellow/orange hue in the image (caused by aging of the newsprint and further emphasized by the narrow-band spectrum of the fluorescent light source) towards black/white by reducing saturation - usually around 50% is a good value. (In Photoshop, use Ctrl-U to bring up the control do this this)
    Yellow Hue driven to Gray/WhiteSaturation Control

  • Enhance the contrast in the image by anchoring (or even darkening by pushing down slightly) any black/grey tones in the image, while boosting the brightness of the "white space". (In Photoshop, use Ctrl-M to bring up the control to do this. Anchor the dark part of the spectrum with an inflexion point about 1/5 of the way across the spectrum, and then push the line up ("to taste") with a second inflexion point in the center of the remaining part of the spectrum.)
    Boost ContrastHold darker shades while brightening lighter shades

  • "Save" the image with jpeg compression level 7 - this minimizes the size in storage of the image while speeding up access.The fidelity loss caused by such compression is not noticeable with this kind of image.
Create Index

The index page is initialy created (and sbsequently updated) using Microsoft Excel. It is suggested that the index page for a new book be started by cloning an existing one from another book, both because it saves effort and it provides some consistency in the format and layout of the page across books. To clone a spreadsheet, make a copy of an existing one and delete the individual rows that are linked to images while retaining headings, column widths, cell formats etc.

The location for this file must be in the same "folder" as the prepared images for this book. (Each book is contained within a separate folder)

After making any necessary adjustments to the headings, you can start to create the individual index entries. You start a new row by creating the link to the image referenced by that row - select the cell where the link is to be created (click on it) and then initiate the "Insert Hyperlink" dialog. (On the main menu choose Insert --> Hyperlink or more quickly just hit Ctrl-K). This will bring up a list of files that reside in the same folder as the spreadsheet - select the desired file and click "OK".

Once the link has been created and it is clearly established which image file you are dealing with, you can now proceed to enter key phrases or words (in the wide column of that row) that will provide a useful digest to give a subsequent reader a good indication of the type of information to be found on that page. A useful start may be just to transcribe the "headline" above each article - but this is not an inviolable rule - feel free to edit, re-word it for clarity or add some key words (names of people, street names, etc) that might be helpfull on subsequent "search by keyword" operations.

Because there are (usualy) multiple articles on a page, so you will be creating multiple strings of phrases or keywords within a given index entry. Normaly such "sentences" would each be ended with a period but it is suggested you use a hyphen (with a space either side of it) to delimit such snippets as that makes it much easier for a speed reader or somebody just scanning the page to pick out the "chunks".

An example of such an index entry might look like this:

LAH residents to vote January 10, 7 candidates for 5 posts on council - Sketch map of town showing areas opted out of incorporation (dotted lines) - Los Altos Hills expected to stay as name of new town - Candidates Bledsoe Fowle Rothwell Wilson Dungan Frame Treat

Generate web-format HTML

When you have completed the spreadsheet, save it in its native Microsoft Excel format (i.e. an .xsl or .xslx file) as this file represents the "master copy" which can be updated and revised.

Next we will use Microsoft Excel to generate a another file derived from the "Master" - but in this case a file formatted as an HTML document suitable for processing by any standard web-browser.

We do this by saving the file again, but this time doing a "save as Web Page *htm" and specifying the file name as "index.htm" (Use of this name simplifies specifying URL links later as it is the default file name a web-browser will look for when directed to a named folder). This will result in a file called "index.htm" being stored in the folder with the spreadsheet and the images as well as (caution!) a nested folder called "index_files". It is this complex of the "index.htm" file together with its sibling folder "index_files" that constitute the mechanism to make the information available via web-brower. Both of these entities need to be deleted and re-generated from the "master" spreadsheet when the latter is updated.

Publish/Distribute

Eventualy, the folder containing this book will be copied and added to the "collection" of such folders (each representing a "book") and from which copies can be spawned for interested parties as well as to assure redundant back-up in case the master copy gets corrupted, lost, or becomes unreadable for any reason. At this point it seems reasonable to propagate copies of the collection on either a DVD disk or USB memory stick.

Note to Potential Volunteers

Of all the steps described above, the one that takes the most time (and is the least boring...) is the process of creating the index. And this is where the town's History Committee could use some real help. If you have basic computer skills and are passingly familiar with spreadsheets, and would like to get involved, please contact the committee.