This was a fine system, except for one problem: the recipes were hand-written. Some were easier to read than others, and those which were faxed from one lodge to another were especially difficult to deal with. The executive pastry chef apologized at the beginning of the ski season for the hand-written recipes, and promised that solutions were being looked at. Many of you are thinking, what's the big deal? Why not just type up and print out the recipes like normal people?
Yes, they could do that. It would be a lot of work, probably taking up a week or so of a single baker's time, and when you're only open for 4 1/2 months out of the year, you tend to focus your energy during that time on managing the rest of the bakery. When the season is over, the majority of bakers leave town, often to work at another resort across the country or world until the next ski season. The skeleton crew that is left focuses their resources on other events for the next several months. And yet, that wasn't really the problem.
The problem had to do with how the recipes were laid out. Let's take my basic cheesecake for example. Sometimes, the bakery only needs to produce one cheesecake. This isn't often, but it happens. More often, something like eight cheesecakes would need to be made. Sometimes only four. Sometimes as many as twelve. With experience, these numbers were somewhat predictable.
We need the recipe for a single cheesecake. Even if we never make just one, this is the basis for which all other amounts will be calculated. We have eight spring form pans, but we rarely need to bake more than five cheesecakes at once. Except in the busy seasons, anything more than five would result in a shelf life longer than a week, which would result in a loss of quality. During the busy weeks, we may need to make eight cheesecakes at a time, and we may need to do so every other day. So we need, at the very least, ingredient lists printed up for 1, 5 and 8 cheesecakes. It would be nice to have them all printed up on the same page, mostly for manageability. And it would also be nice to have at least one or two extra columns on that page, just in case a new amount needs to be calculated (for instance, we buy another 8 spring form pans).
Going from memory (and likely leaving out some subtle details), the form (complete with the aforementioned cheesecake recipe) looked something like this:
|white, granulated sugar
+ 1 tsp
+ 1 tsp
1. Preheat oven to 275F.
2. Allow the cream cheese to warm up to room temperature.
3. In a mixer, cream together the cream cheese and the sugar with the mixer's paddle attachment until light and fluffy.
4. Add the vanilla.
5. Add one egg and mix on low speed. When it is fully integrated, stop the mixer and scrape down the sides and the paddle with a rubber spatula.
6. Continue adding eggs the same way, one by one.
7. When the mixture is smooth and there are no lumps, pour it into a prepared 9-inch round cake pan.
8. Move to a 275F oven and set the timer for 1 hour.
9. After an hour, jiggle the pan a little. If the center still seems liquid, close the door and wait a few minutes before checking again.
10. When the center is only a little wiggly (but no longer liquid), turn the oven off.
11. Leave the oven door open for 1 minute, and then close it again.
12. Allow the cheesecake to sit in the oven for 1 more hour as both the cheesecake and the oven cool.
13. Move the cheesecake to the refrigerator and chill for at least 6 hours (overnight is better) before attempting to cut and serve.
You'll notice that I've abandoned the format suggested in my post about standardizing recipes. I hope that I made it clear in that post that that particular format was not applicable to every kitchen. In the case of this bakery, it would only serve to add confusion and leave no room for multiple scalings.
Take note of the multiple scalings. First of all, this bakery would never have stuck with volume measurements. They would have been converted to weight almost immediately. You'll notice that each column is simply laid out, and easy to follow. Some bakers, to make sure that they don't accidentally look at the wrong column, would cover up the others with masking tape until they were done with the recipe. If they needed a temporary scaling, they would cover the last (blank) column with masking tape and write the new amounts on it. If they used it enough, they would pull out the actual paper and fill in a new column.
I added a couple of other things that I have not yet mentioned. The oven temp is clearly noted at the top of the page, so that the baker can set the oven without having to search the instructions for the temp, and then start to gather ingredients. There's also a place where the baker can circle whether they set the oven fan to high or low. Many professional bakeries make use of convection ovens which have only two fan settings: low and high (but not off). In retrospect, I think the bakery would rather have had a temperature setting for low, and another one for high. Maybe even a third one for "no fan", for their older ovens.
I do not know of any recipe software on the planet that takes all of these things into account. And from talking to other bakers, cooks and chefs, this feature is sorely needed. The corporation that owned that ski resort had spent years trying to find suitable software, and had come up empty-handed. The last I heard, they were working on building something in-house, and it did not look promising. The bakery was not interested in investing any time in typing recipes until the software was made available. Any significant difference in how they typed them and how the new software would handle data entry could represent days worth of lost time and duplicated efforts.
Let's say we got software that supported printing out multiple scalings one a single page. There's another problem here: saving the multiple scalings. Some of you are thinking, "why not just save the amounts for a single cheesecake, and then choose other amounts just before printing?" Let me ask you: have you ever made rice before?
I think Alton Brown summed this up nicely in the landmark Good Eats episode, Power to the Pilaf. Most rice packages instruct you to use one part rice to two parts water, by volume, regardless of how much rice you're actually making. But as he points out, you can actually cook one cup of rice in a cup and a half of water. Two cups of rice will cook nicely in 2 3/4 cups of water. And Three cups of rice will cook properly in 3 1/2 cups of water.
As much as we (especially programmers) would like to think that food fits neatly into our carefully-calibrated mathematical world, it does not. Food is organic. It is grown, largely by a set of rules that we still don't completely understand. No cup of rice is the same, no pound of flour. All that we can do, even in the best of times, is approximate. Doubling recipes will often work, to a certain degree. But there will be a point in which our ingredients will rebel against us, and only the hand of an experienced cook or baker will be able to control the chaos. And when that cook or baker discovers that their approximation has changed because of the sheer amount of food that they are making, they make a note of it. And if you think they want to figure it out every time, or let the next guy figure it out from scratch, you're out of your mind.
Recipe software needs to have the ability to not only scale properly, but also record the (user-adjusted) changes based on amount, and optionally print it out on the same sheet, with easy to reference columns side-by-side for the cook or baker's convenience. We don't currently have software that supports this. We, as home cooks, may not need it. But our professional counterparts do.