mDesktop in Depth - Main File Layout

Summary

The best way to configure a Main File with all the fields required for processing your data file is to use the Setup Wizard.  If, for some reason this is not possible or not an appropriate option, this section explains the fields your Main File must have so that mDesktop can do what is required.

There are basically two kinds of fields that the system uses.  First, there are Input Fields.  These are the fields that you, the user, will supply for mDesktop's processing.  If they are name and address fields, you must give them standard mDesktop field names, because otherwise mDesktop will not recognize that it must apply appropriate processing to these fields.  For example, it must know if a field is a freeform field containing a person's name, or if it contains just the surname, or if it is an address line etc.  Although mDesktop can search through name fields for job titles, through address lines for names or zip codes etc., it will work much more effectively if you can label your data with appropriate mDesktop standard field names.

Additionally, mDesktop uses Generated Fields.  These are fields that probably weren't in your input data, but mDesktop will generate them from your data. They are critical to the main processing steps (deduplication etc).  It is normal to have the input fields at the start of your Main File layout, in one block, and to have the generated fields in another block at the end of the layout. We recommend that you keep to this unless you have a special reason not to, as this allows mDesktop to work out which fields are Input Fields and which are Generated.

However, as shown in mDesktop Field Names and Usage, some fields can be Input in one file but Generated in another.  Also, some Input fields can be replaced by Generated fields.

mDesktop Field Names and Usage

The table in Appendix A - Field Names & Usage  lists field names and usage in alphabetical order.  

Quality Assurance Field

mDesktop requires a FLAG field to prepare the Data Summary and for View Records by Category to allow you to examine certain categories of a record.  Entries in the Flag field are explained in Appendix B.

mDesktop Field names by Type

These are the main data fields that mDesktop recognizes specifically for import or matching:

Personal Name

• ADDRESSEE, CONTACT, FORENAMES, FORENAMES2, INITIALS, INITIALS2, NAME, NAME1, NAME2, NAME3, NAME2FOUND, NAME_KEY, PREFIX, PREFIX2, QUAL, SALUTATION, SEX, SUFFIX, SURNAME, SURNAME2

Company Name

• COMPANY, COY_KEY, COY_NAME, NAME1, NAME2, NAME3

Job

• DEPARTMENT, JOB_TITLE

Address

• ADDRESS1 to 9, ADD_KEY, COUNTY, COUNTRY, DPS, POSTCODE, POST_IN, POST_OUT, PREMISE, STATE, TOWN, ZIP

Telephone

• FAX, REST_PHONE, STD_CODE, TELEPHONE

Campaign

• MAILING_ID, MAIL_DATE

Match Set Grouping

• MATCH_REF, SET_DUPS, OVERLAPREF, OVERLAPCNT, MATCHSCORE

Deletion Priority

• DEL_PRI

Quality Assurance

• FLAG

Unique Reference

• UNIQUE_REF

Using Your Own Field Names

You may use your own field names within mDesktop.  If you do, mDesktop will not recognize them for special processing (e.g. if you name a field as CUSTOMER instead of ADDRESSEE, mDesktop will not apply its personal name processing to that field).  This is because mDesktop uses the field name as the main indicator to tell it what kind of data is likely to be in that field.

Although mDesktop won't apply any special processing to your own field names, it will apply its standard string handling routines.  This means that you can use any field you like as either a matching key or for matching weights. We don't recommend doing this until you are confident that you completely understand how mDesktop works, as using a non standard field can have a drastic effect on your matching results.

To allow you to use your own field names safely, we have set down a few guidelines on what to call your fields.  If you follow these rules, you should have no problems using your own fields in mDesktop:

• Each field name in a Main File should be unique. If you have fields that do a similar thing, number them – for example, "CUSTOMER1", "CUSTOMER2" etc.
• Whenever possible, select the field name "OTHER" from the suggested list in the Setup Wizard for fields that you know mDesktop won't need to know about.  We appreciate that in many situations, these names would not be sufficiently meaningful for processing, so in these cases, follow the naming rules below.
• Field names must not be greater than 10 (ten) characters long
• Field names must ONLY contain the letters A-Z or the numbers 0-9, or the underscore character ("_").  Field names must NOT contain spaces or any other kind of punctuation.
• Field names must start with a letter (A-Z) NOT a number or an underscore.
• Field names are not case sensitive, so 'CUSTNAME' is the same as 'CustName' or 'custname'
• Field names must be AT LEAST three characters long.
• Although you should follow these rules for your field NAMES, you can use different TYPES of fields (character, numeric, logical, date etc.).  However, mDesktop can only match on character fields.
• Do not use any field names specified in the reserved list, as use of these names may conflict with mDesktop's operation.

Words Not To Use As Field Names

Words Not To Use As Field Names

Words Not To Use As Field Names

There follows a provisional list of "reserved names" in mDesktop.  It is not guaranteed to be complete, but the majority of names to avoid should be included.