4

When needing to represent True / False or Yes / No which of the following examples is better from a geodatabase design point of view?

Our database architect hates the option on the right but when looking at the data using a non-esri tool using letters makes it easy to understand without needing to do a database join.

Store an integer or Store a string and use the first letter

0 - 'No' or 'N' - 'No'

1 - 'Yes' or 'Y' - 'Yes'

0 - 'False' or 'F' - 'False'

1 - 'True' or 'T' - 'True'

  • 2
    Numeric values allow you to use the fields in mathematical calculations/expressions more easily, which can be extremely useful. – nmpeterson Apr 28 '15 at 16:16
  • Just using a domain rather than open entry to the field is best practice. Whether you use an integer or string value to do that is up to you. Personally, I prefer the string type method because if the data ever leaves the database with the assigned domain, it will show the raw value. So 0's, and 1's may be confusing to those not knowing of the original domain code-description values. – evv_gis Apr 28 '15 at 16:28
  • I agree with @nmpeterson, but one consideration for integers is that if you end up going between excel or elsewhere, the values could get converted to 1.0 and 0.0 which is pretty annoying. Just a thought. – mr.adam Apr 28 '15 at 16:30
1

We have tested this on our enterprise database designs over a three year period. The theory was that an integer field with text domain values would have superior performance. Testing revealed no noticeable performance benefits.

We discovered that re coding numeric values to be a real processing problem when it came to changes. Also we found that users would export data to their scratch for analysis, and the numeric values were exported instead of the domains. While you can export the domain text instead of the numeric, this process is painful and sometimes not possible as the only option was to go dbf.

We concluded that text based domains were superior. Attribute context is not lost on the users, they are easier to admin, and the performance in million feature global extent data was pretty much the same as numeric domains.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.