Basic guidelines

The following steps lay out some basic guidelines for planning a processing project.

1. Assign a Collection Number and Provisional Title (see Description tab on this page for information on collection titles)
2. Set Up Documentation
   1. JIRA ticket to track progress for collaborative projects
   2. G: drive collection folder to store documents
3. Conduct Initial Research
4. Conduct a Collection Survey
5. Create a Processing Plan

Collection survey

During a collection survey, the processor makes a pass through the entire collection to become familiar with the materials. This will give you an idea of what kinds of materials are present and help you determine how you to arrange the collection. Take good notes as you survey, but don’t spend too much time on each box since you will go through the materials in more detail later. Make sure your notes will allow you to fill out the processing plan when the survey is complete. Track this information by creating a table, which lists each box and its contents. Once you have the contents listed for each box, use a third column to categorize each box according to potential series; some boxes may contain materials that fall under more than one series. It is not necessary to list every folder or item contained in a box! You merely want a general idea of each box’s contents so you can plan for processing the collection as a whole. See below to get an idea of the level of detail necessary during surveying:

Some things to think about as you survey the collection:

• Is there an existing arrangement or order to the collection? Does the arrangement fall into natural groupings or main categories? Remember that the principle of original order says that it is best to maintain the original arrangement of the records. However, if no order exists then you will need to devise an organizational scheme.
• What types of materials are there—photographs, oversized materials, scrapbooks?
• What events, activities, or circumstances are documented? What is missing?
• What date span does the collection cover? Is there a date span during which the bulk of the materials were created?
• Are there records that should be restricted due to privacy concerns?
• Are there preservation issues that will need to be addressed?

Remember to think about the collection from the perspective of the researcher. Your arrangement should be simple enough for a wide range of researchers to follow. You will find it useful to consult completed finding aids to get an idea of conventional tactics for arranging and describing collections similar to yours.

Create a Processing Plan

Download a copy of the Johns Hopkins Processing Plan Template from G:\archives-and-manuscripts\forms-and-templates.

Your processing plan will set the course of action for your project; a well thought out plan will help processing go smoothly.

Use the collection file or existing inventories to fill in initial information in the template, then complete the rest of the plan based on the initial collection survey. See the accession folder on the G: drive and the collection file in the filing cabinets for existing inventories and other documentation. Save your processing plan to the collection folder on the G: drive.

Archival Hierarchy

In order to best reflect the way that records within a collection relate to one another and to make collections easier for researchers to navigate, archivists usually arrange and describe material within collections hierarchically. An item is usually contained within a folder, which can be part of a series, which is a subset of the entire collection. Common levels of arrangement in descending hierarchical order are as follows:

Collection > Series > Subseries > Folder > Item

Not all collections are arranged and described at each of these levels.  For example, if a curator purchases a single manuscript item that is not being added to an existing collection, no arrangement is necessary and the collection can be described at a single level.  When proposing subgroups for a larger collection, such as series and subseries, consider the logical groupings that already exist. For example, if there is a significant amount of correspondence in a collection, that may be a logical series, and if the creator divided their correspondence into business and personal correspondence, those may be logical subseries. If the creator of a collection divided up their material by subject matter, those are potential series as well. The archivist may also create a series from material that a person, organization, or office generated in the course of a specific action or a specific function, such as an administrative series.

There is no “one size fits all” way to arrange and describe a collection! Let the current arrangement of the collection and the needs of researchers guide you to the appropriate level of arrangement and description, and consult with your fellow archivists if you are unsure whether the arrangement you are considering for a collection is appropriate.

Maintaining original order during arrangement

(from the George Mason University processing manual 4, 17, revised)

Archival processors try to maintain the original order of how the materials in a collection were initially arranged, to the extent that it is possible and usable.  Maintaining original order does not mean that a processor has to leave materials that a donor randomly placed into a box as they are, but they always needs to be aware of how a creator kept his or her records, and if a creator organized papers in a particular way, the processor should keep the materials in that order (See Meghan Lyon, Rubenstein Library Processing Manual, 3). 

Some collections will need no additional physical re‐arrangement,, while others may require significant re‐arrangement in order to be useful to researchers This can be one of the most challenging aspects of archival work, but you will have already laid the groundwork for the final arrangement in your processing plan for the collection. Trust yourself! – archival judgment is an important thing to develop. Making decisions can be scary, particularly during the arrangement phase, but usually, if you can explain why you made a particular choice and you document that choice clearly, you are completely justified.

In general, wherever possible, work with the existing arrangement of folders within the collection, particularly if the creator of the material clearly organized the material a certain way. If you have identified series within the collection and need to shift some of the folders in order to logically fit into the different series, you can do so, but make sure that you mention this fact in the finding aid’s Arrangement note and Processing note. In a collection where there is little or no discernable order, you are not obligated to leave folders in the order that you find them (or the order in which you foldered the material) if it will not facilitate use by researchers. In these cases, you can organize the material in the collection in whatever way will be most useful. For financial and administrative records, reports, and similar types of material, chronological order will probably be most helpful to researchers. Alternatively, alphabetical order may be appropriate for correspondence (arranged by the last name of the correspondent), or for unpublished but titled papers. Ideally, you will already have made some of these decisions during the processing plan phase, but sometimes the arrangement of the collection, or of specific series and subseries, evolves during the physical processing stage. Regardless of the way that you end up arranging a collection, document what you did so that it is clear to researchers what organizational choices were the creator’s and what organizational choices were yours.

Foldering/refoldering/labelling

Depending on the physical state of a collection when it arrives in Special Collections, you may folder/refolder materials and label those folders at different times in the arrangement process.  For example, if there are boxes in a collection that contain loose papers that have no clear organization, do not physically arrange or rearrange that material until you have put the material into folders. Unless you have a clear, very good reason for doing so, you do NOT need to sort individual loose papers into different folders in an arrangement that seems logical to you. You may group them in a way that seems logical, but you should not spend time analyzing and organizing them page by page. This will both save you time and make it less likely that you will accidentally sever the archival bond, the relationship between materials that provides information about the context in which they were created.  

In contrast, if the material in the collection that you are processing is already in folders and the folders have titles that accurately describe their contents, but there are folders in another box that clearly belong in the same series (such as correspondence), you can arrange those folders together in one or more boxes before you refolder (if necessary) and relabel them. You should always remove hanging folders from a collection (the metal can damage the material around it). Generally, the folders that records are housed in when they arrive in Special Collections are not acid free, and unless they are in very good shape and already have neat titles written or typed on then, you will probably want to refolder the records using archival-quality folders. Again, this is a situation where an archivist's professional judgement plays a role - there is not always a right answer to when you should refolder or relabel material during arrangement, but the guidelines above should help you decide what will be best for the collection you're working on.  

In the process of foldering material in the collection or going through existing folders, you might find small amounts of material that should be weeded out of the collection. This might include multiple copies of a document or publication (keep two copies, but discard beyond that), receipts, bank statements, etc. For additional guidance on what you should weed out while processing a collection, see the appraisal section of this manual.

See the "Description" tab in the Processing section of this manual for information on devising folder titles. 

Here is an example of the layout for a properly labelled folder from the Willie Lee Rose papers (MS-0905):

While you will probably write collection numbers and titles on folders at some point while you are still physically arranging materials, it is best to wait to write box and folder numbers on your folders until you have finished physically arranging everything in the collection. This will (hopefully) save you from having to renumber a large amount of folders if something needs to be moved.

Once you are confident in the arrangement of folders within a collection (or within a series or subseries of a collection), you can enter the information from the folders into a container list spreadsheet template which you will use to upload the folder inventory into ArchivesSpace. See the ArchivesSpace Import tab on this page for instructions on this process.

Intellectual vs. Physical Order

The physical order of a collection refers to the way that the physical items, folders, and boxes are arranged, while the intellectual order refers to the way that these physical items are grouped in the finding aid that describes the collection.  The physical and intellectual order of a collection are usually the same (and it is easier for both researchers and reference archivists when this is the case), but occasionally, it may make sense to group certain folders or boxes together in a series intellectually (i.e., in the finding aid), but not rearrange them physically.  

To offer one extreme example of why the intellectual and physical arrangements may differ for a collection, at 1093 boxes and 55,350 folders/items, the Barbara A. Mikulski papers are such an enormous collection that implementing a physical arrangement was not feasible. It would have been too time-consuming to physically rearrange all of the materials after they had been processed. Instead, the materials in that collection are generally physically arranged in boxes according to their arrangement in the original boxes. While this mostly conforms to the intellectual arrangement imposed on the collection, there are many exceptions, especially in cases where the original boxes had no order to them. We chose to prioritize processing speed with this collection and therefore make the collection available for research much sooner, although it will result in more work for special collections staff as they support researcher requests because restricted materials are often housed alongside unrestricted materials in the same box. Had this been a smaller collection, the various types of access restrictions associated with the Mikulski papers would have been an excellent reason to ensure that the intellectual and physical arrangements match; instead, the sheer size of this collection trumped all other considerations.

As an example of a situation where the intellectual and physical arrangements of a collection should have been the same but were not, the Keyser-Wyman family papers (originally processed in the 1980s) were (for the most part) initially arranged intellectually by individual creator within the family, but physically arranged mostly by format (ledgers and other bound materials were housed together, photographs together, etc.)  The collection (excluding oversize material) is only 13.66 cubic feet, so it is not particularly large. The intellectual arrangement by individual was problematic on a few different levels - the boundaries between who is creating what material and for what purpose within a family is not always clear enough to categorize material at the individual level accurately and usefully, and assigning material to a particular family member sometimes obscures the presence of other family members in the record. Additionally, the mismatch between physical and intellectual arrangement created a great deal of confusion for reference archivists and researchers, to the point that the former University Archivist decided to limit access to the papers. For this and other reasons, the papers were reprocessed in 2023 with a simpler arrangement (no series for individual family members) where both the physical and intellectual organization are the same. 

When determining whether it makes sense to implement a physical arrangement that is different from the intellectual arrangement, you should consider the following:

• Collection size
• Format, size, or shape of items in the collection
• Access restrictions
• Ease of use for staff and researchers

Sources

Beckman, Liz.  Processing Manual, George Mason University, Special Collections Research Center. Spring 2018 (revised March 2019).

Lyon, Meghan. Rubenstein Library Processing Manual. Duke University, 2016.
https://library.duke.edu/rubenstein/sites/default/files/rubenstein/pdf/faculty/Processing%20Zine%20B
ooklet.pdf

Describing Archival Collections and Creating Finding Aids

This section of the manual provides an overview of local best practices for describing archival collections and writing finding aids, guided by the elements defined in Describing Archives: A Content Standard (DACS) and the Reparative Description tab in this section of the guide. Archivists should consult DACS for additional guidance and examples when needed.

This section covers the primary sections of a resource record, or finding aid, in ArchivesSpace. There is overlap between fields utilized in finding aids and in other types of ASpace records, so this section includes links to other pages of the guide when appropriate. This section does not describe every field available in an ASpace finding aid, only fields that are required or frequently used. Keep in mind that some finding aid information can be imported in bulk using specific ingest templates. See the ArchivesSpace Import tab in this section of the guide for information on performing a bulk import.

To create a new resource record in ASpace, select Create>Resource in the menu bar on the Staff Interface. Resource records can also be spawned from Accession records. To do so, select Spawn>Resource in the menu bar of an accession record. Spawning a resource record automatically creates Related Accession and Related Resource links between the two records.

Basic Information

A. Titles (required)

Rules on capitalization in archival titles: titles should be capitalized in the same way as ordinary sentences, i.e., capitalize the first letter of a title, as well as proper nouns, but everything else should be lower case.

Collection titles

Most titles of archival collections contain two (or occasionally three) elements - the name of the person or corporate body responsible for creating or collecting the material, the type of material being described, and (sometimes) the subject of the material (see DACS 2.3.3 and 2.3.4). Usually, the person accessioning a collection will devise the title, but a processor might change the title if they find additional or contradictory information while working on the collection.

Examples: Office of the President records, James Stimpert papers, Ida B. Wells pamphlet on voting rights

Series titles

Series titles are generally concise descriptors of what is contained in the series, whether it is a format or a subject - you do not need to repeat creator information in a series title unless it differs from the information at the collection level. Some series titles commonly used in archival collections, particularly collections in University Special Collections and Archives, are format based and include things like Correspondence, Working papers, Reports, Writings, Administrative files, Photographs, Audiovisual material, etc. In the title field, you will also include the series number.

Examples: Series 1: Correspondence, Series 5: Photographs, Series 3: Book projects, Series 4: Garland Hall sit-in

Folder titles

Devised folder titles usually consist of a concise summary of the folder's contents, including the three basic elements described above in the collection title section (creator/type[s] of material/subject). New processors sometimes make the mistake of writing overly long folder titles because they try to include a description of everything contained in a folder in the title. YOU DO NOT NEED TO DESCRIBE EVERYTHING IN THE FOLDER IN THE FOLDER TITLE. Feel free to consult with your fellow processors if you are having trouble figuring out how to title a folder where the contents seem hard to summarize.

Examples: Basil Gildersleeve correspondence with Alexander Graham Bell, Black Student Union meeting space articles and demands

Titles of Works

Titles of formal works should be italicized no matter their location in the archival description (collection title, archival object title, notes fields, etc.). Italicization of the titles of works is accomplished by wrapping the title in the following markup: <title render="italic”>Proper Title</title>. 

Some titles warrant quotation marks instead of italics, such as short stories, newsletters, and articles. In these cases, quotation marks should be keyed directly into the title field rather than using <title render> markup. 

Consult Chicago Manual of Style guidelines for proper use of italics and quotation marks for works.

B. Identifier a.k.a. Collection Number (required)

Collection numbers are assigned sequentially, and most frequently begin with either an “RG” for Archives or “MS” for manuscripts. To assign a manuscript collection number, determine the last number used and assign the next number. For records groups, determine the record group to which the materials belong. Use the RG associated with the department making the transfer. You may look the office up in ASpace or on the list of RG numbers stored in G:\archives-and-manuscripts\records-management. If you are unsure about a number or if the transferring department does not have an assigned RG number, please contact the University Archivist for clarification. Collection numbers are added to Resource records in the Identifier field. Each segment of the identifier must be added to a separate subfield without any punctuation. The following format applies depending on the type of material: 

C. Level of Description (required at all levels)

A resource record “Level of Description” is required by ArchivesSpace at the collection, series, file, and item levels in a finding aid. For manuscript collections, the top-most resource record level should be “Collection.” For institutional records, the top-most resource record level should be “Record Group.”

D. Resource Type (not required)

E. Publish? (required)

Indicate whether or not this collection should be available to the public. Collections should be left as Unpublished while they are being processed, and set to publish upon completion of the work. 

BEWARE: the “Publish All” button at the top of any resource record, which can be used to publish ALL information in a resource, including all notes. Consider internal notes before using this function. 

F. Restrictions? (required)

Indicate whether or not this collection contains restricted material at any level. Please note that our legacy collections did not require this note, verify presence of restrictions when handling legacy collections.  

G. Repository Processing Note (not required)

This note functions as an internal note; it does not publish or export to public views. This note can and should be used to record information about a collection that is relevant to future or current archivists, such as “Provenance research conducted in 2015, no creator found.”

H. Bib Number (required)

This refers to the Alma MMS ID number for the collection (not the bib or MMS ID number for the accession record). If you do not have one when creating the resource record, please remember to add it when you do. Please see the Alma Guide below for instructions. Paste the MMS ID in the General Notes field.

Languages (required at collection-level)

See the Languages section of this guide for information on assigning collection language and script.

Dates (required at all levels)

See the Dates section of this guide for information on assigning collection dates.

Extents (required at collection-level)

See the Extent section of this guide for information on calculating and formatting collection extent statements.

Finding Aid Data (select fields required locally as outlined below) 

The Finding Aid Data section of a resource record includes general top-level information about a completed finding aid. This section outlines formatting information for only the required fields in the Finding Aid Data section.

A. EAD ID

The collection number, rendered as MS.0001 or RG.01.001. 

B. EAD Location

The URL of the finding aid for the resource being described. This field is important to include for representation in union catalogs such as ArchiveGrid and WorldCat. You may temporarily publish a finding aid to generate and copy this link. 

C. Finding Aid Title

The full collection title with “Guide to the” appended to the beginning.  For example, “Guide to the Hanke family papers.” 

D. Finding Aid Date

Record the current year at least. You may use a more precise date if you wish, ex. 2023 October 10. 

Note: The date a finding aid is significantly altered should be included in a Processing Information note. See Collection Notes > Processing Information for more information. 

E. Finding Aid Author

Multiple names can be recorded in this field. This data does duplicate the content of the Process Info note, but that is intentional. See Collection Notes > Processing Information for more information. 

F. Description Rules

Set to Describing Archives: A Content Standard.

G. Language of Description

Set to English.

H. Script of Description

Set to Latin.

Related Accessions

This section is used to link related ArchivesSpace accession records to the relevant collection record. Begin typing in the Accession search bar, or select Browse from the drop-down list to search additional records. Linking a related accession also creates a Related Resource link in the linked accession record.

Agents (required at collection-level)

To link an Agent record to a finding aid, begin to type in the term in the “agent” field and see if the term you are looking for pops up as already in the system. Before creating a new record, check carefully to be certain that one (either the term you are looking for or a close corollary) does not already exist. It may sometimes be helpful to consult similar accession and/or resource records already in ArchivesSpace for guidance, and/or browse the existing agent records already in ArchivesSpace.

See the Agents section of this guide for guidance on creating new agent records in ArchivesSpace.

Level of description: Agents can be assigned at any level of description within a resource record, though local practice is to typically only assign at the collection level. However, For large collections, such as the Barbara Mikulski papers, or for artificially assembled collections with multiple creators, assigning agents at the series or file-levels can enhance usability.

Subjects (required at collection-level)

To link a Subject record to a finding aid, begin to type in the term in the “subject” field and see if the term you are looking for pops up as already in the system. Before creating a new record, check carefully to be certain that one (either the term you are looking for or a close corollary) does not already exist. It may sometimes be helpful to consult similar accession and/or resource records already in ArchivesSpace for guidance, and/or browse the existing subject terms already in ArchivesSpace.

See the Subjects section of this guide for guidance on creating new subject headings in ArchivesSpace.

Level of description: Subjects can be assigned at any level of description within a resource record, though local practice is to typically only assign at the collection level. However, for large collections, assigning subjects at the series or file-levels can enhance usability.

Collection Notes (select fields required locally as outlined below) 

A. Abstract (locally required at the collection-level) 

A brief summary combining a single sentence from the Biographical or Historical note plus one or two sentences from the Scope and Content note summarizing who or what the collection is about, what materials it includes, and when those materials date from. 

Examples:

• Professional and personal papers of Willie Lee Rose, a historian of the Civil War and the Reconstruction era and faculty member in the history department at Johns Hopkins University. (MS.0905)
• John W. Baldwin (1929-2015) was a historian of medieval France and Charles Homer Haskins Professor of History at Johns Hopkins University; his papers date from 1929 to 2015. The collection primarily documents Baldwin’s career as a student, a teacher, and a historian through his research files and scholarly works. (MS.0871)

B. Arrangement (optional) 

The arrangement note explains the current organization of the collection. Whenever series or subseries are present in a collection, an arrangement note can be provided at the collection level. An arrangement note can also be used to explain the organization of files or items at the series and sub-series level. See DACS 3.2 for more information. 

Examples:

• This collection is arranged into five series by media type.
• This collection is arranged into three series: Correspondence, Photographs, and Scrapbooks.
• This collection retains the original arrangement of its creator. 
• This series is arranged chronologically.

C. Biographical/Historical (locally required at the collection-level, may occur at lower levels) 

The biographical/historical note (colloquially referred to as bioghist) provides information about the creators and/or the subject matter of the materials described. See DACS 2.7 for further information. Information about the content of the collection itself should be provided in a Scope and Contents note, as described below. If more than one entity was responsible for the collection, separate biographical or historical notes should be provided for each (DACS 2.7.8). 

Content:  All description provided in biographical/historical notes should be relevant to the collection being described. For example, while a Johns Hopkins alum may have gone on to be a United States President, the biographical note for his papers should not overly emphasize his presidency if the collection only pertains to his student years at Johns Hopkins. His political career may be noted in the biography, but more emphasis should be placed on vital statistics (birth, death, graduation date, residency, etc.) and student activities. 

If there’s no information available: No biographical or historical information is available at this time.

Differences between biographical and historical notes: A biographical note describes an individual, while a historical note relates the history of a corporate body, organization, family, or other non-person. ArchivesSpace has a label field, which can be used to differentiate between biographical and historical notes.

Formatting note: Biographical/historical notes are typically entered as prose, but may take any format, such as ordered lists.

Citations: Chicago Manual of Style is our default style guide for citing outside sources. Citations should be provided in the bioghist note itself.

D. Conditions Governing Access (required at the collection-level and at the lowest level restrictions apply)

All finding aids require a collection-level Conditions Governing Access note, even when the collection is open for use. See the Restrictions section of this guide for more information on restriction types and formatting requirements.

E. Conditions Governing Use (required at the collection-level) 

This note is meant to record any restrictions on the intellectual use of the information contained within a collection. Consult accession records including the deed of gift, records transfer form, and donor correspondence for possible restrictions on publication, reproduction, or other usage of materials.

Standard Conditions Governing Use statement: Single copies may be made for research purposes. Researchers are responsible for determining any copyright questions. It is not necessary to seek our permission as the owner of the physical work to publish or otherwise use public domain materials that we have made available for use, unless Johns Hopkins University holds the copyright.

Examples of exceptions: 

MS.GAR.019: Permission to publish from this collection must be requested in writing from the Evergreen Foundation at: The Evergreen House Foundation, Inc. 4545 North Charles Street Baltimore, MD 21210 410-516-0341. 

MS.0056: Single copies may be made for research purposes. Researchers are responsible for determining any copyright questions. Permission to publish from this collection must be obtained from the estate of Ethel White: The literary rights to Edward Lucas White's unpublished papers were retained by White's sister Ethel White (1868-1955). In her will (1949) Miss White left any royalties owed to her from the publication of Edward Lucas White's manuscripts to the Woodstock Theological Center in Washington, D.C. A copy of Miss White's will was deposited in the Woodstock College Archives located at Georgetown University. As of March 31, 1999, The Ethel White Estate was turned over to the Jesuit Seminary/Mission Bureau, which was later called the Ignatian Apostolic Partnerships Office located in Baltimore, MD. 

F. Existence and Location of Copies (optional) 

This optional field may be used, according to DACS 6.2, to record the, “existence, location, and availability of copies or other reproductions of the materials being described when they are available for use in an institution, or for loan or purchase, or available electronically.”  As such, it may be used most often to point to online digitized versions of collection materials, at any level of description. 

Example: The entire collection of News-letter issues has been digitized, but only some issues are available online. Other digitized copies are available offline and can be made available by request.

G. Immediate Source of Acquisition (required at the collection-level and at the appropriate level for accruals) 

This note provides an area to record information about the source of the acquisition. Generally, the note includes the name of the source, as well as the date of the transfer/gift/purchase. 

Examples:

• Donated by William L. Boyle, Jr., July 2013. 
• There is no known acquisition information for this collection. 
• Purchased from Ian Brabner in December 2022.

Artificially assembled collections may use the following format for the collection-level note: Information about specific accruals is found in the Immediate Source of Acquisition notes at the relevant file levels.

For items adopted by donors in Adopt a Book campaigns, include the adoption information in this note. Use the language specified by Development for this donor information.

Examples:

• Purchased from Ruby Lane in August 2022. Adopted by Dr. Linda M. Odgen-Wolgemuth and Dr. Eric Ogden-Wolgemuth in honor of Emerys Ogden-Wolgemuth; FY2025.
• Purchased from eBay in February 2024. Adopted by Ms. Clare C. Miller and Hon. John P. Miller; FY2025.

H. Other Finding Aids (optional) 

Used when it is known that the full inventory is not presented in the online finding aid, or, when there is a supplementary published version of a finding aid or a guide to a collection. 

Example: “An offline inventory is available. Please contact Special Collections for more information.” 

I. Physical Characteristics and Technical Requirements (required if applicable at the collection-level and relevant lower levels)—reference/reconcile with the digital archives section

According to DACS 4.3., this note is used to provide, “information about access restrictions due to any technical requirements that restrict or otherwise affect access to the materials being described, such as equipment or specific hardware/software required for use.”

Example: The items contained in this collection are stored in [XYZ format / a variety of multimedia formats including X, Y, and Z]. Therefore, access to these items may be limited depending on the physical condition of each item, as well as the department's ability to support the playback of recordings stored on outdated media. Please contact Special Collections for more information. 

Level of description: Phystech notes may be added at any level of description, but should be added at the highest relevant level. 

J. Physical Description (optional at the collection-level and relevant lower levels)

Used to record information about the physical condition of the material described. In Special Collections practice, the types of information most often recorded include: 

• Physical dimensions of the material, particularly oversized or oddly shaped items (use the imperial measurement system)
• Condition concerns (possible mold, brittleness, handling instructions) 
• Extent statements that don’t fit neatly into the ArchivesSpace Extent section 

A Physical Description note should not be confused with the Container Summary field, which is a free text field describing how the collection is housed (ex. 1 record center carton). 

Level of description: May be added at any level of description, but, given the nature of the note, should be added at the lowest relevant level.

K. Physical Location (optional if applicable, two notes usually required) 

Used to record information about the physical location of an item or items for items with locations that are not managed in either ArchivesSpace or Horizon. 

Two notes are generated: one published (public view) and unpublished (staff view). Note that the use of the note will be a depreciated practice once the location module is fully implemented. 

 In the following example, the first note is unpublished: 

Prior to March 2019, location notes were used most frequently to describe oversized materials housed in the map cases. Some legacy notes may persist, but items stored in map cases are now managed in the Location module. See Housing > Oversized Materials for more information on the Location module.

L. Preferred Citation (required locally at collection level) 

This note informs patrons how to properly cite material from the collection. The note always takes the form:

[Name of folder or item], [Date], [Box number], [Folder number], [Collection title], [Collection number], Special Collections, The Johns Hopkins University.

M. Processing Information (required locally) 

This note is used to record information about the processing or re-processing of a collection at any level. Recording any other change to a collection or finding aid, like a minor revision, is optional. Do not confuse this note with the Accruals note, which is intended to record accessions. While this note does duplicate data found in the Finding Aid Data fields, it has been decided to record it in both places for consistency between processed and unprocessed collections. 

Formatted per the following examples: 

• This collection was processed in [date] by [name] (with additional description by [name] in [date]). 
• There is no known processing information for this collection.  
• This collection has not been processed. Please contact Special Collections for more information. 
• Processed by Elizabeth Beckman in July 2022. Two folders in Series I: Correspondence, one for Sylvia D. Hoffert and John E. Scott Jr., were empty and were removed from the papers.

Level of description: Processing information notes are required at the collection-level and at the relevant level for any accrual to a pre-existing collection.

N. Related Materials (optional)

According to DACS 6.3., this note is used to record, “[...] the existence and location of archival materials that are closely related to the materials being described by provenance, sphere of activity, or subject matter, either in the same repository, in other repositories, or elsewhere.” 

As these notes are “see also” notes, all efforts should be made to record a complete set of notes in each collection if all the collections being referenced are Special Collections materials. So if one collection is related to two others, ideally all three collections would have notes referencing each other.

Formatting notes pointing to other Special Collections materials:

The citation provided should follow the format outlined in Preferred Citation above. However, while the minimum is a citation to another collection, some context to the note can be relevant, as in the following examples:

• Hans Eugster's consulting work for the Maryland firm Harry T. Campbell Sons' Corporation was begun by Ernst Cloos; these related materials can be found in the Ernst Cloos papers, MS.0124, Special Collections.
• A series of letters written jointly to Alice and Elisabeth Gilman by their parents is filed with Alice's papers in the Daniel Coit Gilman papers, MS.0001, Special Collections.

Formatting notes pointing to non-Special Collections and non-JHU materials:

If known, use whatever citation style is provided by the repository. However, in most cases the following examples will suffice:

• The Roland Park Company Records, MS.504, also contains records related to the history of land development in Baltimore.
• The Jorge Guillen papers are located at Harvard University Houghton Library.

O. Scope and Content (locally required at collection level; optional at lower levels when appropriate) 

The Scope and Content note is a brief, narrative description of the contents of the collection as a whole. The scope note should generally include the types of documents (i.e. correspondence, diaries, financial records, etc.) and general topics covered. It should also include dates covered, notable gaps in the material, and notes on unexpected or particularly unique material.  

Differences between scope and contents notes and biographical and historical notes: A scope note should directly describe the content of a collection; whereas, a biographical and historical note describes the collection’s creator.

Formatting note: Scope and contents notes are typically entered as prose (but may take any format, including ordered lists).  Paragraphs are separated by line breaks in the ArchivesSpace Content field for the note.  There is no reason to enter <p>’s in the body of the note. 

Generally, the Scope and Content note should begin as follows: “The [full name of the collection], [dates of collection], contains…” 

•  Ex. “The Roland Park Company records, 1896 to 1954, contain…” 

Artificially assembled collections may use the following format for the collection-level note: "This is an artificially assembled collection with manuscript material chosen by the curators of Special Collections, related to or created by Maurice Sand. "

Citations: Since the content of a scope and contents note is purely a description of the collection, no citation should be necessary. Information requiring a citation may be better placed in a biog/hist note.

Containers and Digital Objects

A. Container Instances

To manually create and link boxes and folders to a finding aid:

• Select Add Container Instance in the Instances section.
• Select a material type from the drop-down list in the Type field. The default material type is "Mixed Materials."
• To create a new container, click on the arrow next to Top Container and select Create from the drown-down list. See the "Creating top containers in ASpace" instructions in the Housing > Top Containers and Container Profiles section of the guide.

• To link an existing Top Container, begin typing container information (box number or barcode) in the field to search for available records. You can also select Browse from the Top Container drop-down list to search for boxes already in the system.

To add folders to a record select Child Type: Folder, then manually type the folder number into the Child Indicator field. If you are assigning item numbers, select Item from the Child Type menu. The Child Barcode field can be left empty. The "Grandchild" fields are optional, used to provide more detailed container information if necessary.

To add disk information to a record select Child Type: Disk, then manually type the disk number into the Child indicator field. For guidance on assigning disk numbers to digital records, see the Digital Preservation section of this guide.

B. Digital Objects

Digital Objects are used to provide access to digitized or born-digital archival collections. They include file paths to access copies saved on local drives, as well as links to online content. Digital Objects are linked to collections, series, files, etc. via the Instances section of a record. For guidance on creating Digital Objects, see the Digital Records tab in this section of the guide.

Digital Processing Overview

Processing digital records involves the same principles as processing analog records – arrangement of the files, preservation actions to stabilize files and protect them from degradation, review for sensitive information, description and documentation, and preparation for long-term storage and access. How these principles are applied and what their workflows end up looking like differ due to the digital environment the files live in as well as unique properties of digital records. Digital processing work follows the best practices outlined in the NDSA Levels of Digital Preservation and OAIS Reference Model.

Typically the digital archivist will do the preservation work, while the description work for the finding aid will fall into the queue of a processing archivist.

Digital Arrangement

Born digital records typically receive no re-arrangement by archivists for two main reasons – to maintain original order and the ability of researchers to sort records themselves by different file characteristics (e.g. name, file extension, date last modified) in a file explorer window. Any file arrangement must happen within the disk number assigned to files when they are accessioned. See Accessioning > Digital Records for more details about disk numbering.

There is more arrangement leeway with digitized records as this is an artificially created original order. Disk number assignment at the point of accessioning is likewise flexible, but the digital archivist must be consulted.

Contact the digital archivist if there is any discussion about arranging digital records as this could impact the integrity of files.

Digital Preservation

Bag structure and transfer

At the end of the preservation process, all disks in a collection will be placed in bags following the Library of Congress’ BagIt structure. BagIt, a python-based command-line tool, wraps files in a "bag" data folder and generates administrative, technical, and preservation metadata. It has a graphical user interface version, Bagger.

To prepare the bag structure, all disks will be placed in a parent folder, under the applicable set of folders that house the original files, the metadata, and any derivative files.

• Collection folder
  • Access
    • Disk folder 001
    • Disk folder 002
  • Metadata
    • Disk folder 001
    • Disk folder 002
  • Originals
    • Disk folder 001
    • Disk folder 002
  • Preserv (depending on the originals’ file format)
    • Disk folder 001
    • Disk folder 002

Files are securely transferred to the computer station, typically through TeraCopy or as a disk image via FTK Imager. TeraCopy is a file transfer tool that will check the checksums after copying files to a new location. FTK Imager is a disk imaging tool that saves the information of a disk in one or more files that can be recreated later and also generates an internal checksum.

The transferred files are saved in a disk folder within the Originals folder with either an .md5 file from TeraCopy or bagged with BagIt, which will create a manifest and list of checksums.

The files from the Originals folder are to be left alone – no action should be taken unless there is a virus.

Collecting metadata

Before any action is taken, metadata on the original files will be collected using the FITS tool. FITS, the File Information Tool Set, pulls technical metadata from each file and generates an xml file. The xml file will be saved in the metadata folder within the disk number folder.

Stabilizing and reviewing files

The digital archivist will next work to make the files stable by working on a copy of the original files in a Sandbox folder. This includes converting the access copies into more user-friendly file formats as well as renaming files to remove any certain special characters and to replace spaces with underscores. Renaming the files will make them more computer-reader-friendly. After all preservation steps, these files will become the preservation and/or access copies. Preservation copies are created in cases where the original files are in obsolete formats and a high-quality copy can be made. A lower quality copy will become the Access copy.

Before preservation or access copies are moved into the Preserv or Access folders, PII screening may be run. Any PII report documentation will be added to the Metadata folder within the disk number folder.

Preparing for storage and access

Once the files are organized in the bag structure, have metadata files, and are stabilized, it's time to package this all up with BagIt. BagIt will create a final manifest that includes file paths, as well as a list of checksums. This bag will then be zipped up as a .tar file. Contact LAG through a Jira ticket to move the .tar file to VSM storage. The person moving the .tar file to VSM should reply back with the checksums and file paths, which will be added to the Preservation DOs (see below).

Move a copy of the Access folder to the U drive for internal access.

Wrapping Up

Now that the files are digitally processed and in their storage and U drive access locations, update the project Jira ticket. This acts as the handoff to the processing archivist and goes through this checklist:

1. The files are in VSM storage and on the U drive
2. The Preservation DO has been created (see below)
3. Extent information is added to the Accession and Resource Records
4. List any issues related to the files that researchers should know (e.g. a file is partially corrupted)
5. Confirm the location of the physical box (if present) and next steps for housing

Digital description: Preservation-level

The digital preservation process creates two sets of records: preservation copies and access copies.

After the preservation process is completed, the preservation copies are put in long-term storage and will have a corresponding preservation-level Digital Object (DO) in ArchivesSpace linked at the collection-level. This Preservation DO contains metadata necessary for long-term record management conducted by the digital archivist. Metadata fields used in ArchivesSpace include Title, Identifier, Type, File Version, Date created, Extent, and Additional Notes. The Events field is also used to track the progress of the files through the preservation process. See Digital-Object-Preservation-Guide.docx on the G drive for more details (G:/archives-and-manuscripts/electronic-records/Guides).

 
The Preservation DO is assigned to both the Accession Record and the Resource Record. The Preservation DO is not published and is created by the digital archivist.

Digital description: Access-level

Access copies of collection digital files are meant for researchers or internal archival use. Access copies are in more user-friendly formats and tend to be smaller in size and lower quality. If available, these files will be located on the U drive under the collections folder, but they may also be located online (for instance on JScholarship).

Access copies of files will have a corresponding access-level Digital Object in Archives-Space, usually connected to Archival Objects. If there are physical media along with digital files, add instances for both the box that the media is in and the access files.

The Access DO contains metadata relevant for researcher needs and retrieval by archivists, including:

• Title: A descriptive name of the digital object being described.
• Identifier: The disk number.
  • If there needs to be multiple DOs associated with one disk number, add -do01, -do02, -do03, etc. The "do" stands for Digital Object.
    • E.g. RG-04-290_005-do01 and RG-04-290_005-do02
    • In this instance, there are two interviews of different individuals on one disk. Each interview has its own Archival Object and therefore needs separate DOs assigned to each Archival Object. ArchivesSpace will not allow duplicative identifier IDs, so the -do01 and -do02 will keep the disk number connection.
• Type: This describes the kind of digital record, such as Moving Image or Still Image. If there are more than one types, choose Mixed Materials
• File Version: There will be two file versions for each Digital Object. It does not matter which order they appear in ArchivesSpace.
  • The first File Version creates a clickable link for the researcher to look at an online copy (if available) or to a form they can fill out to request access to offline copies. If the file version goes to an online copy, add a Caption to say "This digital material is online. Click here to access it." If the file version is for the form, add a Caption to say "This digital material is not online. Click here to request access." This should be published.
    • This is the form link: https://forms.office.com/Pages/ResponsePage.aspx?id=OPSkn-axO0eAP4b4rt8N7D_G7sFMP-9Hr_2ux3YeXRpUMFpEQjBEN05IMUo0TENDT0NaTVpFTTI4MiQlQCN0PWcu.
  • The second File Version is meant for internal tracking to be able to find where the access copies are stored on the U drive. Use the file path to the disk number on the U drive for the File URI. Do not publish.
• Date of record: This lists the date of creation of the record.
• Extent: This documents the size in gigabytes of the access files. Use the Container Summary to list the disk number ("Disk RG-14-290_001"). Use the Physical Details field to describe the type, such as "3 videos" or "1 PDF".
• Use a Note if applicable. This could be to indicate that a file was partially corrupted and to connect the digital object to the collection as a whole "The digital material is part of the xyz collection, collection identifier 123."
• Add Conditions Governing Access and Use as necessary.

The Access DO is published and is usually created by the processing archivist. Since the metadata fields rely on disk numbers and file locations, contact the digital archivist about creating Access DOs if the records are not already ingested and on the U drive.

Web Archiving

Websites are described in two places - ArchivesSpace and Archive-It - and these two locations have differing description goals. Websites are not accessioned.

ArchivesSpace

The approach to describing websites in ArchivesSpace Archival Objects aligns with local best practices for other archival material, such as a description of the content and context for it as it relates to the collection. See the Dates and Extent tabs in the Basic Metadata section for specific details. Each website seed should get its own AO.

Two Digital Object modules are connected with each website seed - one associated with the specific sites on Archive-It and another connected to the broader collection on Archive-It. For the specific sites on Archive-It, create new DO instances by following the directions for creating Access DOs. The identifier will be the link to the Archive-It seed capture review page (e.g. https://wayback.archive-it.org/org-626/*/http://www.jhukranti.com/), and the caption should say "Click here to access archived versions of the XYZ website: XYZ.com". The other Digital Object for the larger collection may already exist, in which case browse and link the appropriate DO to the AO. If necessary, create a new Archive-It collection by following the directions for creating Access DOs. The identifier will be the link to the collection (e.g. https://archive-it.org/collections/5126) and the caption should say "Click here to see the main XYZ Collection Archive-It page". Make sure all Access DOs are published.

Archive-It

The approach to describing websites in Archive-It differs slightly from the descriptions in ArchivesSpace. Here, it is assumed that a researcher might not have come directly from the link in the finding aid and would therefore need more description about the group, individual, or organization who created the site. The collection ID is entered in the Identifier field, as well as the link to the ArchivesSpace website Archival Object. An Archive-It metadata template can be found G:\archives-and-manuscripts\electronic-records\web-archiving.

Importing Container Lists via Spreadsheet

The primary way to import collection container lists (series, sub-series, file, and item-level description) is by using the ArchivesSpace Import Plugin available in the staff interface. In order to use this feature, your container list data must be formatted to the relevant bulk import template, which are available to download on the ArchivesSpace staff interface: Import Templates (requires staff login).

The templates require a basic understanding of archival hierarchy and note labels, but are easy to grasp with practice. If you have questions about the templates, ask a colleague for assistance. The ArchivesSpace Help Center has useful instructions for using the import templates (Help Center requires staff log in). All imports should be tested in ArchivesSpace Dev prior to importing to production.

To begin, download and save a copy of the Bulk Import AO spreadsheet from the list of templates, either the Excel or the CSV version depending on your preference:

It is recommended that processors utilize these templates at the outset of processing projects to avoid needing to migrate data later on. If you've been working in a different template, you will need to migrate your data to fit the ArchivesSpace import template prior to ingest.

Note: The ArchivesSpace container list ("archival object") templates cannot be used to import collection-level description, only data which is nested below the collection-level, such as series, sub-series, and file-level data. You must create a collection-level record in ArchivesSpace prior to using the bulk import feature.

Once you've created a collection-level record, perform a bulk import by selecting Load via Spreadsheet in the Edit toolbar of a finding aid:

In the next window, upload your CSV or Excel spreadsheet and select Import from Spreadsheet:

Once you select Import, a background job will begin with a status read-out as the data imports. If the import fails, ArchivesSpace will give an error message. The bulk import generates a log file which includes any error messages, which you can download and review. It can be difficult to determine the specific import issue, so if you have a large container list, trying importing data in smaller sections so it is easier to identify and troubleshoot any errors that occur. Per the ArchivesSpace Help center: "If you receive notification that your background job has failed due to errors in some rows of your spreadsheet, any valid archival object records within that spreadsheet will still be created. This may result in partial upload of your spreadsheet and unexpected data clean up. We highly recommend validating your spreadsheet using the “only validate” option before importing the spreadsheet to avoid this outcome."

Importing Records via the ArchivesSpace API

General information and tools

ArchivesSpace API

• To learn about the ArchivesSpace API, check out Valerie Addonizio’s 2021 ArchivesSpace Online Forum workshop: Introduction to the ArchivesSpace AP and API Playbook accompaniment.

• The ArchivesSpace API documentation is available online.

• There are two primary ways to access the ASpace API: using a scripting language such as Python, or installing an API Client such as Postman, which Valerie describes in her workshop (linked above). There are pros and cons to each.

• To use the ASpace API with Python, you need two simple files saved in the same directory as your script(s) - one authenticates you to the ASpace development API, the other authenticates you to the ASpace production API. The scripts written by Michelle Janowiecki and Eric Hanson for the Sheridan Libraries reference these files. [add g: drive examples for folks to use]

Python

• To view and edit python scripts, you can download a text editor to your work station. ex. Kate.

• To run python scripts, you need to use a terminal with Python 3 installed. ex. Anaconda.

• Michelle Janowiecki’s course Python lessons for librarians hosted in GitLab is an immensely helpful starting place. Her explanation of setting up Python, how the different applications work together, and the lessons she provides to get you started are worth a read. Her instructions include set up tips for Windows and Mac users.

Project planning

• Test API scripts in ASpace development prior to ASpace production. Ask Mark Cyzyk to refresh development with production data if necessary.

• Run scripts in ASpace production after work hours to avoid interrupting colleagues using the staff interface during the day. Notify Mark Cyzyk of your plans so he can back up production data and refresh the database if any errors occur.

• If your project results in the creation of csvs which might be useful in future projects - agent lists, subject lists, container lists - save those files to the relevant collection folder on the G: drive. It takes work to generate these csvs, so be mindful of what you want to save versus what is project-specific and can be discarded after use.

Common Data Cleaning Workflows (Google doc)

ArchivesSpace Collection Ingest Documentation aka "The Mikulski Method" (GitHub repository)

Additional ASpace API Scripts (GitHub repository)

General Metadata Editing Scripts (GitHub repository)

Checklist

1. All boxes must be properly labeled and barcoded. See the Housing page for instructions on labeling and barcoding.
2. A bib record for the collection and item records for all boxes must be created in Alma before they are sent to LSC. For instructions on adding or updating item records in Alma, see Housing > Barcoding and prepping boxes for LSC. Those instructions cover both newly processed and reprocessed materials. 
3. The ArchivesSpace finding aid must contain the following:
   1. Collection-level description as specified on the Description - ArchivesSpace Notes tab, as well as description at lower levels where relevant.
   2. Arrangement information for the collection. This includes series and any subseries, as well as an inventory of the collection (i.e., archival objects) arranged according to series and subseries at whatever level of arrangement was used for this collection (i.e., item, folder, box).
   3. Top containers must be added to ArchivesSpace and connected to the collection. See Housing > Top Containers and Container Profiles for more information.
   4. Bib number from Alma copy and pasted into Repository Processing Note.
4. Delete empty containers from Alma and the collection's ArchivesSpace accession record, since those old containers no longer exist. See Housing > Barcoding and prepping boxes for LSC for details on this step.
5. Add a Processed event to the collection's ArchivesSpace accession record. DO NOT use the "Processing Complete" event, only the "Processed" event. In the Processed event, under Record Links, add an Outcome that links to the collection's resource record.

Culturally sensitive language and reparative description

Archival description should strive to reflect the full humanity of the people who created collections as well as those whose lives are represented in them, but both historically and in the present day, archivists have fallen short of this ideal. To remedy this, processors in Special Collections at JHU practice reparative description as an integral part of our work to amend inadequate or offensive representation of people who have experienced oppression. For existing collections, this might involve rephrasing or contextualizing problematic language and/or adding new description to include people represented in the records who were not previously described. For new collections, this requires the processor to be thoughtful and sensitive in their approach to describing people and groups.

As you evaluate existing description or create new description, ask yourself whether it meets these criteria:

1. Does it describe everyone represented in the records, and are they all be described respectfully and accurately?
2. Does it include all relevant contextual information in a description, even if that history is uncomfortable? People and groups may leave behind complicated or uncomfortable legacies, but eliding this context on their lives and their impact through omission or euphemism would be unjust to those harmed by them. Our job as archivists is not to bring glory to the collection's creator, but to accurately describe the collection and its context so that researchers may understand its contents.
3. Is it specific as possible? Name people, groups, and places. Consider your default assumptions around aspects such as place (i.e., US-centric), race, gender, sexuality, etc. as you determine which details to include and how they are phrased.

Our values

We want to be mindful of how we describe both marginalized and dominant groups represented in our collections. Biased description includes both derogatory and aggrandizing terms and can even be found in seemingly neutral language when that “neutral” language omits information that might be important to understanding who the people represented in those records are and what they did during their lifetime.

We want to describe people using the terms that they and/or their communities have chosen. Be mindful that there is often considerable debate within marginalized communities about how to describe their identities in ways that best communicate their personhood and heritage, and consensus over preferred language can shift over time for various reasons. The best way to show care and respect for the people represented in the records we hold is to let them speak for themselves using the language they chose rather than language that others have chosen for them.

We want to be continually responsive to community feedback. Editing legacy description is rarely a simple “find and replace” task; it usually requires thoughtful review of the language and materials at hand. It is also a task that we cannot fully complete after a single audit of our collections. Rather, it involves ongoing repair work as communities update their preferred language and community members identify additional cases of problematic language and content in our collections that require some form of reparative description (e.g., content warning, contextualization, rewording). We echo the Yale Library Reparative Description Working Group's decision to “embrace slow archives by pursuing flexible practices that allow for changing course, and for prioritizing collaborative relationships with community stakeholders.”

Vocabulary and description guidelines

Refer to our Reparative Description Style Guide for the complete, most up-to-date version of our descriptive guidelines. The style guide includes specific guidance on language to use or avoid, rules for how to rephrase or contextualize problematic language, templates for content warnings, and additional trusted resources such as community-created vocabularies and advice for archivists carrying out this important work.

In general, processors should follow these rules for describing marginalized identities:

1. If the processor knows that an individual or group prefers to be described in a particular way, then use that language to describe them. Be mindful that there is often considerable debate within marginalized communities about how to describe their identities in ways that best communicate their personhood and heritage, and consensus over preferred language can shift over time for various reasons. The best way to show care and respect for the people represented in the records we hold is to let them speak for themselves using the language they chose rather than language that others have chosen for them. For example, if you know that a person self-identifies as an autistic person and rejects person-first language (i.e., person with autism), then use "autistic person" rather than "person with autism" to describe them.
2. If the processor knows that an individual or group described themselves in a particular way, but that language is outdated, use that self-description but also provide context. We want to respect both the people represented in the records and the contemporary members of those communities. Using quotes and specifying that the language is self-description ensures that the people in the records are represented as they labeled themselves, but providing additional historical context and using updated language elsewhere in the description ensures that researchers understand where this language is coming from and signals respect for contemporary members of the community who may identify differently.
3. If the processor does not have specific information about an individual or group's preferred terminology, refer to the guidelines in the Reparative Description Style Guide and the resources it links to.

Processors should familiarize themselves with the following resources before working on collections:

• Archives for Black Lives in Philadelphia (A4BLiP) anti-racist description guidelines
• Protocols for Native American Archival Materials
• National Center on Disability and Journalism disability language style guide
• ADA Knowledge Translation Center disability language style guide
• Homosaurus LGBTQ+ linked data vocabulary
• GLAAD media reference guide
• For a full list of vocabularies and other resources, see the Reparative Description Style Guide.

Collections may need to be combined or separated due to a variety of reasons, such as discovery of unrelated materials in a collection, or discovery of closely related materials housed in two separate collections. It should be noted that these combining or separating collections should only take place under direction of the University Archivist. 

Combining collections

Nominate one record to be retained as the “new” record and the other record to become inactive. This same decision will be reflected in both AS and Alma. In addition to moving all notes and updating the extent in the new combined record, the following should apply to the old records: 

ArchivesSpace

• Prepend old AS record title with [TOMBSTONE].
• Change extent to 0 cubic feet and update the second resource record with the additional extent.
• Record fate of collection in Repository Processing Note in old record and Custodial History note in new or continuing record. Include the date this happened.
• Unpublish (do not Suppress) the old record in ArchivesSpace.
• Save any relevant justification (such as email correspondence) to the new record’s folder in the G: drive. 

Alma and physical boxes

• Work with Technical Services to move relevant information from the old bib record to the new bib record.
• Call back boxes and physically change the old call number to the new call number.
• Move item records from the old record to the new record and change box numbers as appropriate.

Separating material to form a new collection

Follow all guidelines in the creation of a new resource record being sure to address the following: 

• Create two or more Custodial History notes (one for every record affected) documenting the separation of materials, the reasoning, and the date. These notes should appear in the original record and any new records created as a result of the separation.
• Update the Extent statements and container links for all affected records.
• Save any relevant justification (such as email correspondence) to the new record’s folder in the G: drive.

Encountering classified records

In the rare event that a collection contains material with classified markings or a classified cover sheet, those records must be deaccessioned and returned to the proper authorities. If you believe these records are no longer classified, then the Center for Legislative Archives at NARA can perform a classification review. If, in consultation with the originating agency, they determine that the record is not classified or may be declassified, then we may be permitted to retain it. We may not keep anything that is deemed classified.

The most likely place to encounter such materials is in a congressional collection. Classified records are supposed to be weeded from congressional collections before the member leaves office, but some items may slip through the cracks.

Who to contact

In most cases, the appropriate agency to contact is NARA's Information Security Oversight Office (ISOO). If the record is in a U.S. senator's collection, contact the Office of Senate Security and the Senate Archivist instead.

ISOO or the Office of Senate Security will come pick up the classified record. If it is being sent for review, they will help coordinate the classification review process. NARA's Center for Legislative Archives can perform classification reviews, and the originating agency and other relevant agencies will also participate in the review process.