I index books for a living. Usually when I tell people this they express surprise that this job is actually done by human beings. I guess people assume it’s done by some sort of computer program.
Anyway, I just finished up a couple of deadlines and I’m bored this weekend, so I thought maybe I’d do a thread on this in case anyone is curious about it.
How do you choose the categories you’ll include?
How do you make sure you don’t miss bits that should be included in the categories?
Ever read Cat’s Cradle, where one character (an older woman) is a book indexer? Can’t be too many books featuring indexers as a character- do you know of anymore?
She says that writers should never index their own books as it reveals too much personal info. In fact she reveals that one character is gay, based on how he indexed his travel guide!
People have literally written entire books on this topic. 
But in a nutshell, every topic in the book that gets more than a passing mention should be covered in the index. If the author has done a good job of organization his information, it makes it easier to decide how to set up the structure of the index, i.e. which topics are going to be cross-referenced to other ones; which ones will be double-posted; which ones will only exist as subheadings of other entries, etc. If the author has not done a good job of this, it’s a bit more difficult, but you can still arrange the information into grouped categories. I’ve found that the more I do this, the faster I get.
By reading every word and paying careful attention.
I have not, but now I’m going to have to read it! (To be honest, I’m not a huge Vonnegut fan, but I’ll make an exception for this.) And no, I am unaware of indexing being featured as a profession or a plot point in other works of fiction. Which is really shocking, because it’s such a fascinating job, what with all the sitting around staring at a computer screen and drinking coffee.
I appreciate good indexes. I’ve seen a few cookbooks that didn’t have indexes and it drove me crazy; I think that really reduces the usability of a cookbook. What types of books have you indexed?
Oh, here’s one that I encountered today - Mark Bittman’s “How to Cook Everything” has under the “Squash” section of the index the entry - “-- winter - see Winter squash.” (Winter squash are butternut squash, acorn squash, that sort of thing, rather than summer squash. So that index sends you to “W” for this category of vegetable.) Right call, or dumb?
I realize I’m biased on this topic, but in my opinion a cookbook without an index is an unusable cookbook. It’s like having a textbook without an index. Why would you do this?! (I know that the answer is “to save money” but as a consumer I will not buy a cookbook that lacks an index, so I don’t think the publisher is saving that much money in the long run.)
I spent the first several years of my career solely doing computer books for a small publishing company based in Seattle. (Not Microsoft, though we did some production for them.) So, a lot of books on Office products and that sort of thing. In the last couple of years I’ve started securing a lot more freelancing clients. I’m still doing a lot of tech/computer stuff, but I’ve also done a book on corporate entrepreneurship, one on advanced oil crop biorefineries, and a recent one on Canadian privacy laws, to name a few. (The oil crop biorefinery thing was surprisingly interesting.)
So, to be clear here, they’ve got all the summer squash recipes under “squash” but then for winter squash you have to go to “winter squash”? Yeah, no, I disapprove. I’d double-post in that case. So I’d go ahead and put the winter squash under “winter squash” but I’d also have them listed under the main “squash” heading. The goal with an index is to get the reader out of it as quickly as possible, so you want to avoid sending people chasing across the index needlessly.
Another alternative, if space is an issue, is to have “squash. See summer squash; winter squash.” So then the reader just needs to decide what type of squash she wants to use, and go to the appropriate location. But I don’t really like this, because what if you don’t know the difference between summer and winter squash? I think it’s better to just have all squashes under S for squash, and then double-post where appropriate to give the reader multiple points of access.
I read it in high school, and it was a throw away line that really stuck with me. Her being an indexer had no other plot contribution, and what revealed didn’t either. It was such a quirky and interesting line that it stuck with me.
I also appreciate a well indexed book- so thanks to you and carry on!
Did you have any other jobs/pursuits before your indexing career? If so, do you feel that diverse experiences have helped make you a better indexer?
How many years do you think you’ll be able to index before you start banging your head against the wall?
It’s worse than that, really:
Squash
- filling for fresh pasta
- summer
-- about
-- pancakes
-- pickled, quick
-- rounds cooked in butter
-- sauteed
- winter. See Winter squash
[...]
Summer squash
- about
- pancakes
- pickled, quick
- rounds cooked in butter
- sauteed
[...]
Winter squash
- about
- flavorings for
(etc.)
So the only unique entry under “Squash” is for pasta filling, then it duplicates the entries that are also listed under “Summer squash”, but directs the reader to a separate section for “Winter squash.” It’d be much more sensible to put “See Squash” notes under the “Summer squash” and “Winter squash” entries.
I love the cookbook, and this is the only place I’ve noticed this, but it’s likely there are other places in the index where this happens.
Any topics that seem harder to index than others, for whatever reason?
Why do indexes sometimes use sub-entries (if that is the right terminology)? To give a made up (and unrealistic) example, a large index might contain an entry for Fish, where you would expect it alphabetically, and then list all different kinds of fish - cod, perch, haddock, roach, etc., and maybe things like fish, reproduction, or fish, nutritional value of - under the main entry for fish.
If I want to know about roach, or the nutritional value of fish, my first thought is to look under R and N respectively, and sometimes it can be hard to guess what the indexer has assigned the thing you are looking for to be a subcategory of. (Also, I have seen indexes where these category-type entries can run over more than a page, which makes it hard to figure out where in the index you are. You are looking for porridge, but open the index and see entries for roach and other fish-related things beginning with R, and think you at R, so you page back looking for the Ps, but you were really in the Fs for fish, and so find yourself in the Es.)
So, what is the rationale for using categories like this? Is it still considered good indexing practice to use them? (My impression it is that this is more common in older books, but then, older books often seem to be more thoroughly indexed anyway.)
¬¬¬¬¬¬
Also, how is the increasing availability of electronically searchable text files affecting the indexing profession? Is it a dying trade, or do you think indexers will nevertheless remain indispensable? Are you (indexers in general) having to modify how you do things in order to continue providing added value?
I actually rather enjoy what I do so I don’t really anticipate burning out on it. Particularly since I started branching out and getting different sorts of books besides just Office product ones. 
My first job out of college was doing data entry/proofreading, and then my next job was as a proofreader/copyeditor. That was the job where someone parked me in front of a book and said, “This needs an index. You do it.” (I shudder to think what that index looks like. Not good, I suspect.) But I realized that this was something I liked, and since nobody else at the company would touch an index with a ten-foot pole, I became the de facto in-house indexer. Then I found out that I could make a lot more money by offering my services as a freelancer, so I’m doing that now. (That’s the extremely truncated version of the story, but it’s basically accurate.)
Yeah, that’s no good. If you’re going to list summer squash under “squash” then you should list winter squash there also. And what’s with that random “filling for pasta” entry?
Although you must have the first edition of HTCE - I just pulled out my copy, which is the revised tenth anniversary edition, and they’ve fixed it! Under squash it simply says “see summer squash; winter squash” to direct you to either of those locations for the squash recipes. It’s not my first preference, but it’s not a bad way to handle it, either, and I suspect that for this index, space was certainly an issue.
I’m going to assume that you really only mark words, then a computer program does the actual grunt work of storing the data, sorting and formatting it as you have instructed it.
If so, then I’ve done that, too, as part of writing or editing books.
Genealogical books are time consuming, but easy to do. You need to tag every person’s name and every place. Some places will have to be cross-indexed if their name has changed over time, and some people’s names should be cross-referenced after marriages as well.
Philosophy is a stone-cold bitch. With something like a computer textbook, all of the topics are pretty clearly outlined and fall into simple categories, easy to organize. Something like philosophy, or political theory, or anything like that is going to require a lot more depth of analysis in order to properly organize the index.
The idea is to get the reader as close to the information he is searching for without making the index unwieldy or disorganized. So, for example, let’s say that in your hypothetical book, there is an entire chapter on fish. Without subentries, the “fish” entry might simply have a giant 40-page range after it: “fish, 40-80”. That is helpful in that it lets you know that this is where the main discussion of fish is, but it is not terribly helpful in getting you to the specific information you need. So instead the indexer will break it down: “fish, 40-80: cod, 42-46; haddock, 47-50; turbot, 51-53” and so forth.
However, each of those subentries should also be double-posted as its own main entry, so that you can go to C and find “cod, 42-46”, and so forth. That way you pick up the reader who specifically knows he’s looking for cod, as well as the reader who isn’t quite sure what type of fish he’s interested in, but knows that he wants to read something about fish. Or maybe he just wants to know what sorts of fish are covered in the book.
Having a giant list of subentries that covers multiple pages is indeed confusing, for the reason you mention, and is to be avoided when possible. I don’t really like having a list of subentries that goes for more than a column, although occasionally it is unavoidable. (Also, this is sometimes out of the indexer’s hands; the index goes through page layout after I’ve submitted it, and I have no control over how many columns or what point size the font is in or whatever.)
Well, it’s a problem, because publishers are trying to get away with providing e-books that don’t have indexes, on the grounds that you can just do a full-text search on “fish” in the book or whatever. But my response to that is that, again taking the example we just used, doing a full-text search on “fish” does not provide usable information to the reader. An index tells the reader where the main discussion occurs, and breaks that discussion down into parts, letting you know what types of fish are discussed, and letting you know where to find that information. Doing Ctrl-F on “fish” and getting 452 results seems less helpful to me. There’s a push in the indexing industry to start working with publishers to include indexes that are clickable and take you to the point in the book where the discussion occurs, and my hope is that something ultimately comes of this. I don’t think this is a dying trade per se, although I do think we are at a point where we have to convince the publishing industry of our continuing relevance.
I’m not sure exactly what you mean by “mark words.” Do you mean the embedded indexing function in Microsoft Word?
Do you have clients/companies that give you a consistent amount of work, or do you bop around here and there? Or perhaps a bit of both?
Do indexers have conferences or get-togethers of any sort?
Any inside jokes or entries that you insert just for funzies?
I do get regular work from certain publishers, yes. It is always a nice feeling when I get that second call from a publisher or author that I’ve done an index for in the past. Like, confirmation that they didn’t get my index and think, “well, this is total shit.” 
The American Society of Indexers has a national conference every year, and the regional chapters tend to have weekend workshops and the like, at regular intervals. I’ve been to the Chicago Great Lakes chapter’s fall workshop a few times. They’re usually pretty good, and give me an excuse to visit Chicago also.
No inside jokes per se, although recently I was somewhat amused by including an entry for “exploding pie charts” in a book on Excel.
I haven’t used MS Word, so I don’t know their terminology. I’m more familiar with InDesign, and even more, PageMaker.
If you mark/tag a word somehow, and tell the program what level and other tag info like cross-references apply to that word, then the program should associate the page number with that tag and when asked, sort and format your data. So you do the stuff that needs intelligence and the computer does the rest, right?
Right, yeah. The indexing functionality is basically the same in all of those programs (Word, PageMaker, InDesign) with superficial differences. So if you’re doing an embedded index, yes, you insert tags that have the index entries in them at the appropriate locations and then you generate the index, and it spits it out with correct page locators and so forth. I find that embedded indexes usually take quite a bit of editing at the end to make sure that everything looks nice and polished, but a lot of publishers like to use them for various reasons.
The other method of creating an index, and the one I use more frequently, is to use a stand-alone indexing program, e.g. Sky Index (my program of choice), Cindex, or Macrex. The publisher or author sends you PDFs of the page proofs, and you create a record in your program for each index entry, with associated page number or page range. At the end, you generate the index, and it compiles it as an RTF file.
I know that there will be variation from book to book but approximately how long will it take you to index a 500 page book? I am just looking for a ballpark figure.