Loading...
HomeMy WebLinkAboutMetadata - White PaperManage Your Metadata White Paper November 2008 The information contained in this document represents the current view of Compulink Management Center, Inc on the issues discussed as of the date of publication. Because Compulink must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Compulink, and Compulink cannot guarantee the accuracy of any information presented after the date of publication. This chapter is for informational purposes only. COMPULINK MAKES NO WARRANTIES, EXPRESS OR IMPLIED, AS TO THE INFORMATION IN THIS DOCUMENT. Table of Contents What Is Metadata? ................................................................................................. 3 What Is Metadata For? .......................................................................................... 3 Types of Metadata ................................................................................................. 3 Fields and Templates ........................................................................................ 3 Tags ................................................................................................................... 6 Document Relationships ................................................................................... 7 Versions ............................................................................................................. 7 General Principles of Metadata Use ..................................................................... 9 Plan Metadata Use ............................................................................................ 9 Streamline Metadata Types .............................................................................. 9 Use Appropriate Metadata Types ....................................................................1 0 2 What Is Metadata? In technical terms, metadata is information about information. In that sense, many elements of the Laserfiche repository can be referred to as metadata: the names of documents, the folder structure, the annotations, the security, and so on. However, when we speak of Metadata in the context of a Laserfiche repository, we mean a specific set of information. Metadata in that sense refers to user-defined information that is applied to the entry as a whole. By that definition, Laserfiche has five types of metadata: templates, fields, document relationships (also called links), tags, and versions. For the rest of the paper, the term `metadata' will be used to mean those five items as a group. What Is Metadata For? Metadata stores any information you want to have connected with an entry. You can use metadata to help you organize and locate entries, to make entries more easily searchable, to provide additional information that would be useful to have alongside an entry, or to jump directly to related entries. You should use metadata to store information that will be useful to you in the future, either to find the entry you're adding it to, to find another document, or to provide additional information when you have a document open. Types of Metadata There are five types of metadata -fields, templates, tags, document relationships, and versions. As they are closely related concepts, fields and templates will be discussed together. Fields and Templates What are fields? Fields store text information about an entry. That information can be general text (such as a name, address, set of notes, and so on), a list, a number, a date or time, or a chunk of data to use with integrations. The type of the field will determine what can be stored in it -for example, a Character field can hold any kind of plain text, a Number field can only hold numbers (but can contain negative or decimal numbers), and a Date field can only contain a date in the local computer's date format. While a Character field is more versatile ~d~~r~~~°~~ F'~'~~... because it can contain any type of ,a ~ ~~,~h~ ~ text, more specific character types like Date and Number are better- 1: A character field named "Author," containing the value "John Smith." 3 suited for those specific types of data, as they handle sorting and displaying dates and numbers in a more useful way. TlVhile most field types are self-explanatory (Number fields store numbers, Time fields store times, etcetera), List fields merit more explanation. A List field is a list of values determined by the person who creates the field. when a user views a document containing a List field, the Metadata dialog will display the pre-determined values and allow Add;'Remave Fields... them to select one. They must choose a value Language ~ from the list; the cannot add another value. List y fields are useful for fields that will contain one of En fish `~ a limited number of options. For example, you French Partu~~e~e might use a list to store the state a document is relevant to -the user can simply select a state 2: This °°Language°' list field allows , the user to select from a from the fifty available, without needing to type predefined list of languages® it in or worry about typos. One of the optional settings possible with fields is allowing the field to store multiple values. Amultiple-value field can hold more than one value within the same field. All of the values in the field will be viewable and searchable, and once a field has been designated as multiple-value it can contain as many values as you need. What Are Templates? A template is a collection of fields, gathered together to allow users to easily apply related fields to relevant documents. You can choose which fields to include, and what order they will be displayed in. You can also create new fields from within the template creation dialog. Tenlp~~ate; P~cec~unts Payable +~ Campan~t Plame Dae Company Ine Quote Number 101 Quote Date 07J31?200 l_'J Amount Due X508.99 ~, L'J Transaction Plumper 9954 AP Clerk ~ppro5~al v Camptraller r~ppraval v 3: An Accounts Payable template, For information about fields and templates, and the options available during their creation, see the Laserfiche Administration Reference. 4 How Should I Use Fields and Templates? Fields are often used to make documents more searchable. For example, you might have a repository with many documents by different authors, and users who want to find documents by a particular author. You could use an "Author" field to store the names of document authors. A user who wanted to find everything by John Smith could simply search for documents with that name in the "Author" field. LSear~h Name ....................................................................Templake Name..................................Pages........In~ex ~ ~~-~0~~ Notes ..................................... .........General.............................................. ....................1......Yes ~~~2-~Q~~ Repark,xls Reporks Yes ~~03-2Q08 Notes General 1 Yes ®~~4-~0~~ ~epork,xls Reporks Yes "~Mana~e Your Meka~aka Library Dotumenks ~ Yes ~~Mana~e Your Mekadaka ~~~ Library Dotumenks ~ Yes 4: This search for the author "constance anderson" located results in three different templates, because the same Author field was used in all three templates, Fields can also be used to keep information relevant to a document stored with that document. For example, you might have a file pertaining to a particular customer, and want to have contact information for that customer on hand in case you need to follow up with something. You could create PhoneNumher 555-986-6543 ~ ~( template fields to contain the phone 555-1~3-456, ~ number and e-mail address for the user. - - ~ You could even use amultiple-value field 5: This Phone Number field can accept to store more than one phone number or multiple values to store more than one e-mail address in the same field, number. You can also use fields with integrations. For example, you might want to pull relevant information about a document out of another database and store it with the document. Fields are a good way to store that information directly as part of the document. Furthermore, there is a field type -binary -intended solely for storing binary data that is needed for an integration. (This also means that, if you are not writing an integration or customization using Toolkit, you can usually ignore the Binary field type.) The most important use for templates is to quickly and consistently apply the same set of fields to many related documents. For example, you might scan invoices into your system that should always contain certain fields. You can 5 add the necessary fields to invoices quickly and easily by creating an Invoices template and then applying that template to your invoices. You can also use templates to categorize your documents. The template gives you a quick way of determining which documents belong to which groups. Tags What Are Tags? A tag is a marker that can be used to indicate the . , , . , . k type or status of the document. There are two types Fields T~g~;1 Links Versions of tags: informational tags, which are used simply Tag Desaripkian Cammenks to provide additional data about the entry they are ^ ~ Canfidenkial ^ ~ High Prioriky applied to, and security tags, which restrict access ^ ~ Praaessing 0 ~ Ta Be Filed to the entry. If a document has been assigned a security tag, users will only be able to see the s: This document has been document in the repository if they have also been assigned the "To Be Filed" assi ned that to . g g gag' For information about tags, see the Laserfiche Administration Reference and the Laserfiche User Guide. How Should I Use Tags? Informational tags are useful for classifying documents in cases where the classification is a simple `yes' or `no.' A tag has only two states: it either has been applied to an entry, or it has not. You might therefore use tags to mark a yes-or-no status, such as "To Be Filed" or "High Priority." If a document has not yet been filed, it should have the "To Be Filed" tag; if it has been filed, the tag should be removed. Another use for informational fields is as part of workflow routing. For example, you could apply the "Processing" tag to all documents currently being managed by UVorkflow. Users browsing the repository would then know that that document might be altered or moved by a Uljorkflow process. Security tags allow you to apply security directly to an entry, without affecting Entry Access rights or inheritance. Security tags are useful for documents that should only be visible to certain users and groups, and that will be moved around in the repository but that should always remain restricted wherever they are. For instance, some documents might be classified, and should only be viewed by the "Authorized" group. You could create a "Classified" tag and apply it to the relevant documents, and then grant it to that group. No matter where your workflow routed the document in the repository, it would remain secure, available only to the "Authorized" group. 6 Document Relationships What Are Document Relationships? Document relationships connect one document to another. A document relationship is a link (sometimes they are referred to as "links," in fact) that indicate that two documents refer to one another in some way. The link type for the relationship indicates how they are connected: "E-Mail" and "Attachment," for example, or "Agenda" and "Packet." In the Metadata dialog, you can directly open the document on the other end of the link. For information about document relationships, see the Laserfiche Administration Reference and the Laserfiche User Guide. How Should I Use Document Relationships? Document relationships are useful whenever you want to be able to keep two documents associated with one another - especially if they are not going to end up in the same part of the repository from one another. For example, you might file your e-mail messages in a "Communication" folder by date or sender, but put the attachments to those messages in other folders depending on their content. If you connected e-mails with their attachments using a document relationship, you could quickly find the report that goes with a particular e-mail message even if the report was in a subfolder of the "Reports" folder and the e-mail was in a subfolder of the "Communications" folder. ~,,,.,,,,.,,,,,,v,,,.,ti„v , v,,.Y, .~r..r~.,.,,, ~,vY,~"„ , „~~ vf~r",vY., ~ please revie~ ~uk~aek ,,, i~~11~~ooa i~;~~;i~ aM 151 i~~ii~~a~~ 1 ~ a M ~ Fields Tags Links; 1 Versiens Name Link Type Reverse Link Pakh ~~~008 Repork,~ls Akkathmenk Message Tesk 7: An e-mail message with its attachment linked to it. It is a good idea to use document relationships to connect an item with its source (as in the e-mail and attachment example), or to mark related documents (such as a meeting agenda and the minutes for that meeting), or to indicate a status (such as the "Supersedes" and "Superseded by" relationship used with Records Management). Versions What Are Versions? A version is a modification of a document that has been saved as a separate, but related, entry. For example, if your repository hold scanned copies of health insurance forms for your employees, and you need to scan in renewals 7 to those forms every year, you could scan ~~~-~ ~ ~ - - - ~ k ~ - those renewals in as new versions of the ~ Thefallawingdacumenthasbeenmadified; . ~ MyRepasikary~Gavernment~Ciky Clerk~2008 Repark,xls original forms. This allows you to quickly Vt+hak would you like to da with the changes? access any year's renewal from the original ~}~5aveasanewversianafthedacument --,~ 5ave as a new acumenk form, or to go back and view the original ~~':~Overwritekheexiskingdacumenk form from any of the later renewals. If you `_-iDiscardkhechanges are working with electronic documents, ^aan~kaskmeagain OK Cancel you can also save edited versions of the files as new versions. You might open a 8. You can save a modified electronic UVord file and add a new paragraph to document as a new version, reflect additional information. You could save the edited document as a new version rather than overwriting the original document, and therefore maintain the original document for comparison with the new version. For information about versions, see the Laserfiche Administration Reference and the Laserfiche User Guide. How Should I Use Versions? Versions are useful for making changes to documents "safely." If you scan or save modified documents as new versions, the old version of the document remains untouched, eliminating the chances of accidentally losing information by overwriting. This allows you to go back to older versions for reference. It also allows you to keep track of a document through its lifespan, and monitor it for changes over time. Fields Tags Links Versions; 3 Name J Nauka, Pieker -Healkhcare Renewal 2008 j Nauka, Pieker -Healkhcare Renewal 2007 ~] Nauka, Pieker -Healkhcare Cerkificakian Version DakejTime 3 9f1f2008 12;34;08 2 9~1~200712;33;53 1 10f 14f 2006 7;51;21 ~~~ OK Cancel Help 9: Storing an application and its successive renewals as versions of the same document makes finding the original and its versions straightforward. Keeping many versions of a document can result in a large number of closely- related documents in your repository. Uljhile this keeps a wealth of data at your fingertips, it can make it harder to find the information you need. You 8 can reduce this problem by setting your Client to display only the most recent version of documents with more than one version, or to return only the most recent version in the search results. This allows you to keep the older versions, but only present users with the newest version to reduce confusion when browsing or searching for documents. The older versions can still be found by opening them from the Version tab of the Metadata dialog, or by temporarily turning off this option. You can set these options in the Client Options dialog, in the Browser and Search sections respectively. General Principles of Metadata Use Plan Metadata Use It is a good idea to plan your use of metadata as early as possible. Deliberately designing your metadata allows you to have a more structured, useful and intuitive system than adding metadata elements on the fly. For example, if you plan your fields ahead, you can create a single "Customer" field that can be used for all templates, which reduces the total number of fields a user must contend with and streamlines search and retrieval. However, if users are creating fields as they go, one user might create an "Invoice Customer" field and another a "Customer Report" field, both containing client names, In that case, if you need to search for all documents pertaining to John Smith, you'd need to search each of those fields separately, which is both less intuitive and less efficient. Planning ahead avoids these problems. Streamline Metadata Types In general, it is a good idea to use as few metadata types as you can while still storing all the information you need. The reason for this is twofold. First, the more metadata elements you have in your repository, the more performance will be impacted. UVhen you open a list of available fields, tags, or document relationships, the list will open much more quickly if it contains thirty entries than if it contains three hundred. Second, and perhaps more important, users generally find it easier to select the appropriate element from a shorter list than a longer one. If a user is presented with ten informational tags, they can scan the list quickly and choose the appropriate tag within seconds. If a user is presented with hundreds of informational tags, however, the process is more difficult. If the user knows what tag they want, they must still scroll through hundreds of irrelevant tags to get to it, which is unwieldy and often frustrating. Ullorse, if the user doesn't know what tag to apply, they must skim over the options, and then decide which of hundreds of tags seems most appropriate. In addition, if a user wants to search for items by tag, they will have the same decision-making process when confronting the list of 9 hundreds, whereas a list of five or ten presents much less of a hurdle to searc ing. In many cases, a single metadata element can serve more than one purpose, simplifying the process of choosing and applying them. For example, it's not necessary to have five different "Customer" fields just because you have five templates. If all of those fields will contain a client name, you can create a single "Customer" field and use it in all of the templates. In another case, you might have three different document relationships to indicate that one document is a supplement to another -for example, "Report" and "Supplement," "Agenda" and "Packet," and "Presentation" and "Handout." In some cases, it may be important for the way your organization handles searching to keep these separate, but in many cases they can be combined into a single document relationship: "Document" and "Supplement." This allows users to choose the relationship more easily, as they are confronted with fewer choices, and also makes searching simpler. Use Appropriate Metadata Types Certain kinds of information can be stored in more than one way, by more than one metadata type. In some cases, more than one metadata type is equally appropriate, but in most cases one type isbetter-suited to the task. In general, you should consider how the information will be set and used, and choose the type most suited to the task. For example, when creating a field, consider carefully which field type would be most suited to what you are trying to do. For example, you might want to find a way to indicate a document's status. If your company has a workflow process with several different possible statuses - "To be filed," "Pending approval," "Approved," and "Legacy," for example - a list field with each of those pre-defined options would be the most efficient way to allow users to select and view the current status of the document. Users could select the relevant status from the list, or search for all documents with a particular status. If you only need to mark whether a document has been filed, though, an informational tag might be better suited to the task. The document could be tagged "To Be Filed" and then users could remove the tag once the document has been filed, or search for that tag to find all documents that need to be put in the appropriate folder. 10 Manage Your Metadata December 2008 Author: Constance Anderson Technical Editor: Justin Pava Compulink Management Center, Inc. Global Headquarters 3545 Long Beach Blvd. Long Beach, CA 90807 U.S.A Phone: +1.562.988.1688 www.laserfiche.com Laserfiche is a trademark of Compulink Management Center, Inc. Various product and service names references herein may be trademarks of Compulink Management Center, Inc. All other products and service names mentioned may be trademarks of their respective owners. Copyright ©2008 Compulink Management Center, Inc. All rights reserved 11