Cleaning Up Old Personal Knowbase Data Files (Part 2)

If your old Personal Knowbase data files are getting cluttered and you feel like you're no longer familiar with what they're storing, it might be time for a little file maintenance.

In Part 1, we talked about the cleaning up your file's keywords. Here are more ideas for working with your articles and the files themselves.

5) Check Duplicate Articles

To locate articles which are exact duplicates of each other, run the Report command from the File menu and select the Check for Duplicate Articles option from the cascading submenu. It reports duplicates found and allows you to select ones to delete.

Note that articles are considered duplicates if both the titles and contents are identical. Their dates, keywords, and attachments may be different, so delete articles with caution.

6) Locate Articles with No Keywords

Another problem to check is articles with no keywords assigned to them, so they would never appear in the Index Window when keywords are selected. To locate them, see our previous tip on Finding Articles Which Have No Keywords.

Sometimes when I import articles from text files, I forget to add keywords. So I want to track those down and fix them.

7) Remove Obsolete Notes

Browse your full titles list looking for articles that you no longer need. I sort the Titles column by date for this, with the oldest articles on top, since the oldest articles are most likely to be obsolete. Obsolete keywords (see item (3) in Part 1) can tip you off to outdated topics.

For example, I found articles with tips and trouble-shooting information for software and services that I no longer use. In some cases, the sites or software no longer existed. Even if I still had an account on a site, these tips would be unlikely to work now.

I also found articles with URL references to web pages that no longer exist. Oops. I should have copied the whole web page into the article. Storing dead links is not useful.

In cases like these, you can delete the articles entirely. Or add a keyword such as obsolete or archive and use the Build Query command to omit them from results. If you want to split these old articles out into a separate archival file, see item (9) below.

8) Watch for Entries That Belong Elsewhere

Sometimes you'll run across an article that should have been stored somewhere else.

I found a few quotations with the keyword quote. They should have been in my quotations data file. So I moved them from the reference file to the appropriate file. For details on moving articles between files, see our previous post on Copying Articles from One Knowbase File to Another.

I deleted the keyword quote from the articles, since it was unnecessary in the quotations file. Then I went back to the reference database and also deleted quote there to discourage future additions of quotations.

9) Consider Splitting or Combining Files

Over time, we may change what data we collect and in what categories.

Although Personal Knowbase has no set limits on file size, you may find that a large file becomes difficult to manage. Or perhaps a minor project or issue becomes a major one, requiring more data. In that case, consider splitting your Knowbase file into multiple files. For example, several years ago, I split my quotations database out into its own file.

On the flip side, you may have a project that originally warranted its own file, but became less important later. If a file shrinks or the data no longer needs its own file, consider combining the data file into another one.

10) Watch for Other Issues

While I was browsing through my files, I watched for other items that I could tidy up. For example:

• I capitalized proper noun keywords like "facebook", "apache", and "wordpress" using the Edit Keyword command (Keyword menu).
• I made the titles of related articles consistent, like changing "File - " to "File: ".
• I fixed misspellings that pre-dated PK's spell-checker.
• I used Edit Keyword to rename keywords to be parallel. For example, I had Win10, but Windows 7.

11) Run Compact

PK files don't use much hard drive space relative to modern hard drive sizes, so compacting your data file is less necessary for saving space. If you mostly enter new notes, but rarely edit or delete, you won't benefit much from running the Compact command (File menu). But if you delete a lot, your file may become fragmented, with longer articles distributed in blocks throughout your file, which can slow access. Compact will defragment your file.

PK's Compact command also detects file errors and unused keywords.

When a data file gets old and cranky, give it some love.