Hello,

I want to write today about a request I receive maybe a couple of times a year, which is to update a previously written index. The usual advice is to avoid updating and to rewrite the index from scratch. The rationale, as counterintuitive as it may seem, is that rewriting the index is often easier and faster than mucking about in the original index. And yet, is rewriting always the best option?

Like almost everything in indexing, it depends.

I updated an index last month and decided to go ahead and work from the original. While it proved to be more work than I anticipated, I think it was still the right decision.

To help me decide whether to update or rewrite, I’ve come up with four questions that I ask when assessing the request from the client. What’s at stake is my time and the client’s budget.

The four questions are:

1. Did I write the original index?
2. How much of the text has changed?
3. How complex is the index?
4. Is the original index any good?

​

Did I Write the Original Index?

This first question has to do with how much work it will be to understand and deconstruct the original index. Personally, I find it takes a lot more work to get inside someone else’s index. This does not necessarily mean that the original index is poorly written; it simply means that the original indexer had a different approach and style. In contrast, I know how I tend to index, and so even if I haven’t looked at the index in a few years, I find it much easier to get back into.

Does this mean that I always rewrite the index if I didn’t write the original? I am certainly far more inclined to rewrite, but it also depends on the answer to my second question.

​

How Much of the Text has Changed?

This second question is about the scope of changes which will need to be reflected in the index. Are the changed confined to a handful of paragraphs? Maybe a new introduction? Or have entire chapters been added, removed, or rearranged? Is updating the index primarily about adjusting locators for reflow (relatively straightforward) or will I need to add, remove, and revise main entries and subheadings? The more changes to the text, the more changes to the index. At a certain point, it is easier and quicker to rewrite the index from scratch.

This is also about the process of writing a new index compared to updating an index. For me, updating an index tends to be tedious and slow. Instead of reading the book and creating my own index entries, I scan and compare the proofs and index, page by page, looking for what needs to be changed. I find it to be slow, mechanical, and less creative.

I always ask for a PDF or manuscript with the changes marked, so that I can clearly see what has changed. Trying to identify the changes myself is too time-consuming and prone to error. As much as possible, I want to remove the possibility for me to overlook revisions. But even with the changes highlighted, the work can still be slow-going.

If just a few changes, then updating the index is quicker. If half the book has been changed, or maybe even a third or quarter, I’d rather rewrite the index. Rewriting is certainly more enjoyable, and will probably be quicker as well.

​

How Complex is the Index?

Related to the scope of changes is the complexity of the index. A short, simple index, for a short, simple book, is going to be easier to update than a long, complicated index for a long, complicated book. So in addition to the scope of changes, consider the index itself. How difficult will it be to deconstruct and update?

​

Is the Original Index Any Good?

Along with considering how much work is involved, it is also worth considering if the original index is even worth saving. One of the temptations when updating is to fix the index. Even for indexes I previously wrote, there is always something I question and want to do differently. If there are just a couple of issues to fix, that’s fine and shouldn’t add much time. But if the entire index is poorly written and I’m tempted to give it a complete overhaul—well, better to scrape the original and start anew.

Beginning again is also fairer and more transparent to the client, as I see fixing and updating as two different things. If I’ve agreed to update the index, I feel like I should stick within that scope.

​

My Recent Project

The index I recently updated was for a scholarly book I first indexed three years ago. The book is about NATO and gender. It is being reissued and the author contacted me about updating the index, since they had made a few revisions.

My first and fourth questions were easy to answer, as I had written the original index and on reviewing the index I thought it stood up well enough. To gauge the scope of work I asked to see a couple of sample chapters. Along with the proofs, the author also agreed to send me the manuscript in Word with track changes, so I could see both what had been added and what had been removed. With all that information, I decided to go ahead and update the index, anticipating that the work would take me six to eight hours.

In the end, I was both right and wrong. The changes to the text were more extensive than I anticipated from the samples I saw. While the book’s structure and chapters remained the same (a good sign), there were more paragraphs rewritten, added, or removed in each chapter than I realized. This also resulted in reflow, which proved to be more complicated to update, as the book is published by Oxford University Press and uses paragraph IDs.

The work took me twelve hours, which is more than I estimated. Thankfully the client was understanding and had the budget for the extra work. For my quote, I charged $105 per hour (I usually charge hourly when updating an index because the time needed can be difficult to estimate), with the final bill being $1260 plus sales tax. This was still quicker and cheaper, though, compared to a new index. A new index would have taken me at least twenty hours and I would have charged over $2000.

​

Did I Catch Everything?

Two final points about updating an index.

One fear I have is if I’ve missed something.

When writing an index, I am the one making decisions about which entries to include, and so I am much more aware about what is in the index. When updating an index, the entries already exist and I am less aware of all that are present. This is why it is important to get inside the index, to figure out how it is structured, how topics are divided and sorted, and how double-posts and cross-references are handled. A single piece of information may be indexed two or three different ways across an index, and if that information is changed in the text, all of those access points need to be updated.

When updating, I try my best to find every entry that needs to be updated, but I also fear that I have inevitably missed something. I’m not sure what to do about this beyond doing my best to be thorough. I don’t want to introduce errors, and the more extensive the changes to the book, the more likely, I feel, I’ll miss something. And yet if changes to the text are minor, then updating the index is likely the better option.

​

Utilize Your Software

When updating an index, I also rely heavily on Cindex to help me zoom in and identify what needs to be changed. I often place the index in page order sort so I can more easily compare the index to the proofs. I also search for entries within specific page spans to help me find entries for affected pages. I also search for specific terms, in case something is double-posted.

I assume other programs, such as Macrax, SKY, and Index-Manager, have similar capabilities. Whichever program you use, use it to your advantage.

​

Updating an index is not the most enjoyable work, in my opinion. I’d rather write a new index. But there are occasions when updating is the better choice. If you are trying to decide how best to proceed, try asking these four questions and take out some of the guesswork to find the quickest and easiest way forward.

Yours in indexing,

Stephen

​

PS. Thank you so much for responding to my brief survey about ways I might be able to help you grow and succeed as an indexer. It is really helpful to hear what you are interested in. I'll write more later about the results. For now, if you haven't responded yet and would like to, the link is here. Thanks again!