Vox

Leitura de 25 min

Refactor Your Labeling Techniques

Summary

As we approach the end of 2020, Jonathan Cutrell introduces a series of resolution-worthy topics for developers, beginning with the importance of examining the labels we use. He explains that labels are not just descriptive but actively shape our perception and understanding of the world, and by extension, influence our actions. The episode focuses on how reductive or incomplete labels can be harmful, especially when they compress critical information or present contextual situations as absolute truths.

Cutrell provides several examples of potentially reductive labels. He discusses how listing “excellent presenter” on a resume might be accurate but fails to capture the underlying portable skills like communication or teaching. He examines catch-all labels like declaring “2020 was a bad year,” which compresses the variability of experiences into a single judgment. He also highlights labels that frame contextual situations as factual, such as “I don’t have enough money/time for X,” which often hide implicit choices and priorities.

The core of the episode presents three techniques to “refactor” our labeling. First, invert the label: imagine the exact opposite statement to uncover the implicit context that makes the original label seem true. Second, zoom out: ask “what else must be true?” to find more abstract, portable concepts (e.g., “communication skills” instead of just “good presenter”). Third, zoom in: decompose a broad label into its constituent parts to allow for richer, more granular understanding, such as describing specific difficult parts of a year rather than labeling the entire year as “terrible.”

Cutrell emphasizes that while our brains naturally use labels and heuristics for efficiency, it’s crucial to consciously inspect them. By challenging our labels, we can improve our perception, make better decisions, and communicate more effectively. The goal is not to eliminate labeling but to ensure our labels serve us well, helping us achieve clarity and purpose in our careers and lives.

Recommendations

Tools

  • New Relic — An observability platform praised for consolidating monitoring tools. It offers a telemetry data platform, full-stack observability, and applied intelligence with AI/ML features. A free tier is available with 100GB per month and one full user.

Topic Timeline

  • 00:00:00 — Introduction to New Year’s resolutions and the power of labels — Jonathan Cutrell introduces the end-of-year theme and the importance of New Year’s resolutions. He explains that while the calendar change isn’t magical, our brains mark transitions, making it a good time for reflection. He introduces the episode’s focus: how the labels we use intentionally shape our perception, communication, and ultimately our actions, making this a resolution-worthy topic for personal and professional growth.
  • 00:03:18 — Examples of reductive and potentially harmful labels — Cutrell outlines three types of labels worth inspecting. First, resume skills like “excellent presenter,” which are reductive and don’t capture underlying portable abilities. Second, catch-all labels like “2020 was a bad year,” which compress complex, variable experiences into a single judgment, losing important nuance. Third, labels that present a contextual situation as factual truth, such as “I don’t have enough money/time,” which often hide implicit choices and priorities rather than stating absolute constraints.
  • 00:10:54 — Sponsor message from New Relic — A sponsored segment for New Relic, framed within the episode’s theme of seeing things from multiple perspectives. New Relic is presented as an observability platform that consolidates monitoring tools into three core products: a telemetry data platform, full-stack observability, and applied intelligence. The offer includes free access with 100GB per month and one full user to simplify application monitoring.
  • 00:12:35 — First technique: Inverting labels to expose implicit context — Cutrell presents the first refactoring technique: imagine the exact opposite of your label. For example, change “I don’t have enough money for this book” to “I do have enough money for this book.” This exercise forces you to identify the implicit context (e.g., budget allocations) that made the original statement seem true. It reveals that many constraints are actually choices, shifting the focus from impossibility to decision-making and priority inspection.
  • 00:17:05 — Second technique: Zooming out to find portable, abstract skills — The second technique is to ask “what else must be true?” and zoom out to a more abstract level. Using the “excellent presenter” example, this leads to identifying portable skills like communication, design, and teaching. These abstract skills are more valuable because they apply across many situations, not just presentations. This helps in creating mental models and making skills on a resume more versatile and impactful.
  • 00:20:06 — Third technique: Zooming in for richer, granular understanding — The final technique is to zoom in and decompose a broad label into its components. Instead of saying “2020 was a terrible year,” describe the specific parts that were difficult (e.g., “the pandemic was challenging”). This avoids compressing out positive experiences and allows for a more nuanced, accurate understanding. Similarly, instead of labeling someone a “good developer,” describe the specific qualities and actions that constitute good work.
  • 00:23:27 — Conclusion on the importance of inspecting our labels — Cutrell concludes by emphasizing that while our brains use reductive labels for efficiency, we must consciously request deeper thinking when those labels hinder our goals. Inspecting and refactoring our labels is a powerful way to improve perception, decision-making, and communication. He encourages listeners to share the episode to help the show grow and reach more engineers.

Episode Info

  • Podcast: Developer Tea
  • Author: Jonathan Cutrell
  • Category: Technology Business Careers Society & Culture
  • Published: 2020-12-14T10:00:00Z
  • Duration: 00:24:43

References

Podcast Info

Transcript

[00:00:00] we’re coming up on the end of 2020 it’s been a long year for pretty much everybody that i’ve

[00:00:12] been in touch with and i imagine it’s been a long year for you the listeners of this show

[00:00:16] and as you move from one year into another as we talked about basically every year the show has

[00:00:23] been around we start thinking back on the year and we think forward on the year to come

[00:00:29] now there’s nothing particularly special about going into a new year there’s no

[00:00:35] switch that we can flip that suddenly makes 2021 better or even worse than 2020

[00:00:42] but our brains do have some understanding of transition of marking time we do care about this

[00:00:52] as human beings and so as much as we’d like to imagine that

[00:00:57] january 1st

[00:00:59] is just another day just like december 31st was or some random day in july

[00:01:04] our brains actually do mark that time and we do care and so for that reason we talk about

[00:01:12] resolutions on this show every year and i am hoping to inspire you with a couple of episodes

[00:01:21] about things that may make it onto your resolution list all right these are things that could

[00:01:29] very likely help uh engineers who are listening to this show and can help people in general even

[00:01:36] if you’re not an engineer my name is jonathan cutrell you’re listening to developer t

[00:01:39] my goal on the show is to help driven developers like you find clarity perspective and purpose in

[00:01:44] their careers and on today’s episode this series i guess uh we can kind of start a series of

[00:01:52] episodes here about resolution worthy topics resolution worthy topics

[00:01:59] today’s episode we’re going to talk about labels why is it so important the labels that we use

[00:02:07] are intentionally chosen why is that so important well first of all it should be obvious to anyone

[00:02:16] who listens to the show but language matters language matters quite a bit in fact this

[00:02:22] podcast wouldn’t exist if it didn’t matter because the way that we use language the way that we form

[00:02:29] we use to understand a topic is the perception that we shape for ourselves language matters and

[00:02:37] labels are all about language and so when we use a label not only are we expressing our own

[00:02:46] perception but we’re also creating our perception we’re shaping our own understanding by putting a

[00:02:56] label on something and of course if that’s the case then we’re not going to be able to

[00:02:59] label is something that we’ve shared in some way we’ve communicated it to other people

[00:03:03] we are somehow shaping their perception as well and so what happens when our labels shape

[00:03:12] our perception well what can often happen is we end up using labels that are reductive

[00:03:18] that are not complete that are taking shortcuts in some way in today’s episode i want to challenge

[00:03:26] you to reimagine your

[00:03:29] labels this is a resolution worthy topic because if you challenge the way you think about labeling

[00:03:38] if you challenge the way you think about defining those labels you are fundamentally working on your

[00:03:46] perception this is important because our perception shapes so much of our actions

[00:03:53] right this is why one of the main things we want to do on the show

[00:03:59] is to help you gain perspective perspective and perception are very tightly interlinked

[00:04:05] so what i want to do on this episode is give you a couple of examples of labels

[00:04:11] that might be reductive and reductive isn’t always a bad thing necessarily in fact our brains require

[00:04:20] reducing right a dimensionality reduction is an important topic it’s an important

[00:04:26] skill to be able to master and to be capable of doing that and to be able to do that in a way that

[00:04:29] because we can’t always think about every single facet of every single topic that we ever encounter.

[00:04:35] But when we need to think about those things, we should be able to recognize when the reduction is

[00:04:42] harmful. So we’re going to talk about a few labels that are likely reductive. And then later on in

[00:04:49] the episode, we’ll talk about ways that you can get out of labeling things in potentially harmful

[00:04:57] ways. Or perhaps a better way to put it is ways that you can put new labels or find new labels

[00:05:03] for what you’re talking about. But first, I want to talk about the kinds of labels that we

[00:05:11] want to inspect a little bit more. So we’ll start out with something that’s very much

[00:05:16] not obvious. And if you’re following along at home, if you want to do a little bit of homework,

[00:05:21] I encourage you to pause this and pull out your resume. If you don’t have one, then pause it for

[00:05:27] a little bit longer and go and make one. All right. And most resumes, they have a section

[00:05:33] called skills. And what ends up happening is we put into our skills, our own self-perception

[00:05:42] of what our skills are. And very often a skill might be phrased something like this.

[00:05:49] Jonathan is an excellent presenter. This is something that you might see in a testimonial,

[00:05:57] maybe from someone. And so you might put presenting or public presentation on your

[00:06:05] skill list, on your resume. Now, there’s nothing necessarily wrong or inaccurate about this,

[00:06:11] right? This is why it’s so sneaky and why it’s worth talking about, that our skills are necessarily

[00:06:18] in some cases, specific and reductive. But let’s think for a moment about how this matters

[00:06:26] to someone who’s listening or reading, who’s listening to that person talk about us and

[00:06:33] provide our credentials or who’s reading our resume. This label of excellent presenter,

[00:06:40] what else can we pull out of this? Right? This is reductive because it doesn’t talk about

[00:06:46] what it means to be an excellent presenter.

[00:06:48] How is that valuable outside of the context of presenting? We’re going to talk about how you

[00:06:57] might change that label later on in the episode. Another label is something like a catch-all.

[00:07:05] All right. This is trying to jump around the real complexity of an issue and compress it

[00:07:13] into a singular concept. So for example, we might say,

[00:07:18] 2020 was not a good year because of the pandemic.

[00:07:24] This presupposes a bunch of things. And here’s all the problems with what I just said, okay?

[00:07:29] This presupposes a lot of things. One, that a given year should be or can be evaluated

[00:07:37] on a scale from terrible to wonderful. Certainly, a year is a long thing to judge. And

[00:07:47] each individual is a different thing. And so, I’m going to talk about that.

[00:07:48] circumstance might change the answer for this. By reducing an entire year down to a couple of

[00:07:58] words or trying to put it on some kind of linear scale from bad to good, we often remove or

[00:08:06] compress a lot of information out of that experience. We compress a lot of information

[00:08:12] out of the label. And this can be harmful, especially if that information is particularly

[00:08:18] relevant. Let’s imagine that you had three wonderful days and four terrible days. Well,

[00:08:26] if we average those out, then the week that you had on those seven days was bad. But that totally

[00:08:33] compresses out the variability in that week. Now, it wouldn’t be the right thing to turn around and

[00:08:41] say, oh, well, anytime there’s a good day in the year or anytime there’s a good day in the week,

[00:08:46] we should say it’s a good week.

[00:08:48] Same kind of problem in the opposite direction. You’re still compressing out a lot of information.

[00:08:53] And so, it makes sense to pay attention to when we are compressing things, compressing

[00:08:59] critical information, especially by applying a label. The next kind of label that we should pay

[00:09:06] a lot of attention to is when we present a contextual situation as a factual truth.

[00:09:15] A contextual situation is a factual truth.

[00:09:18] So, let’s imagine that there’s a book that you want to buy. And this book is going to help you

[00:09:25] in your career. But you think to yourself, well, I just don’t have the money to buy this book.

[00:09:35] Now, very often, this is the kind of labeling that we would put on the situation. It’s not a

[00:09:39] single word. It’s not an adjective. Instead, it is an understanding or a frame, a lens that we’re

[00:09:47] putting on the situation.

[00:09:48] This is just as much of a label as anything else would be. And so, we’re taking this label of not

[00:09:54] having enough money to buy the book and applying it to the situation. But that may not necessarily

[00:09:59] be true. We may be preloading that lens or preloading that kind of framing with a lot of

[00:10:09] implicit context. For example, do you actually not have that much money or is that money allocated

[00:10:17] towards something else?

[00:10:18] In your predetermined budget. These are two different sets of context, but they both have

[00:10:25] completely different implications. This one most often shows up in our lives as not having or

[00:10:32] imagining labeling as if we don’t have enough time to do X, Y, or Z.

[00:10:39] Now, we’re going to talk about all of these labeling techniques and how we can kind of exit

[00:10:45] or inspect at the very least.

[00:10:48] Our labeling and hopefully improve our labeling right after we talk about today’s sponsor,

[00:10:54] New Relic. And speaking of labeling and seeing things from multiple perspectives,

[00:11:00] New Relic wants to help you see your application from all of the relevant angles. New Relic is

[00:11:07] observability made simple, after all. And monitoring a complex digital architecture

[00:11:12] typically takes a bunch of different tools. It’s not just down to one tool.

[00:11:17] It takes a lot of different things coming together to make it all work.

[00:11:22] And this is meant troubleshooting by jumping between a bunch of different products,

[00:11:27] a bunch of different dashboards and data. You might have to remember 17 different logins to

[00:11:32] know what’s going on with your application. And New Relic wants to change that. They’ve

[00:11:37] designed everything you need in three products, not 17. First is a telemetry data platform,

[00:11:44] which creates a fully manageable,

[00:11:46] schemeless,

[00:11:47] time series database of all of your data from any source the second is full stack observability for

[00:11:54] analyzing visualizing and troubleshooting your whole stack and finally applied intelligence with

[00:12:00] all of that data of course we can use ai and ml to automate anomaly detection and incidence

[00:12:09] intelligence correlation and best of all you can get access to these things these three services

[00:12:15] for free there’s no host-based pricing no constant upsell for more functionality you get 100 gigabytes

[00:12:23] a month to one full access user no strings attached head over to new relic.com to check it

[00:12:30] out today new relic is observability made simple so i want to talk about a few techniques that you

[00:12:35] can use that will help you create better labels all right better labels and you can substitute

[00:12:44] perception

[00:12:45] or lenses for labels here but labels tend to be a very important key to understanding this because

[00:12:54] typically we think about labels or we think about you know concepts subjects in terms of almost a

[00:13:01] key value in our minds when you think about a particular subject there’s likely a label that

[00:13:08] jumps to mind right in our previous example of not having enough money for a book you might

[00:13:15] immediately think about a label that jumps to mind right in our previous example of not having

[00:13:15] enough money for a book you might immediately think about a label that jumps to mind right in our previous example of not having

[00:13:16] of a label like unreachable or not available right these are labels that we use so i want you to think

[00:13:25] about your labels more in the upcoming year and one way to do this is to inspect or interrogate

[00:13:33] our own labels i’m going to give you a couple of ways you can do that on this episode the first way

[00:13:39] is to imagine the exact opposite imagine the exact opposite of what labels are and what you’re

[00:13:45] whatever your label suggests so in our book example let’s imagine that the exact opposite

[00:13:53] is i do have enough money to buy this book now is either one of these falsifiable in other words

[00:14:03] can we prove without any kind of flexibility or context shifting that either of these statements

[00:14:11] is true or false and once we can recognize

[00:14:15] that for example in this case if the context was implicit right if the context was implicit

[00:14:23] then we can change the context to make the opposite true in other words if you came into

[00:14:29] believing that you don’t have enough money for a book because you have an implied context

[00:14:34] that says that your money has to go towards some bill for example or that you’re already

[00:14:39] spending your money on some other thing if you change that statement and you say okay

[00:14:45] how can i make it true that i do have enough money for this book i can change the context i can

[00:14:54] change my budget in some way right i can instead of buying that other thing that i don’t need

[00:14:59] i can buy the book and so i do have enough money for the book and the outcome here should be

[00:15:06] i do have enough money for the book but but i have chosen to spend it in another place

[00:15:14] and what ends up happening

[00:15:15] when we do these opposites right when we when we kind of invert our statements is we recognize

[00:15:23] what things were implied in our original statements so what this requires you to do is

[00:15:31] become incredibly aware of the things that you otherwise were ignoring and instead of making

[00:15:38] everything implied it makes certain that you are having that you’re creating a habit of making

[00:15:44] things more explicit

[00:15:45] and so what this what this will help you do in addition to making things more explicit is

[00:15:52] identify areas that require change right or that might uh require some shift in context or shift

[00:16:02] in understanding and so you may want to buy the book but you’ve chosen to spend your money elsewhere

[00:16:09] in order to buy the book you need to change your decision about how to spend the money

[00:16:12] if we apply this to spending our time

[00:16:15] you can say okay well i didn’t have enough time to do that thing maybe you did have enough time but

[00:16:21] instead you chose to spend that time i don’t know listening to this podcast right you chose to spend

[00:16:28] that time in some other way and if you know that you’ve chosen to spend your time in some other way

[00:16:33] then you can start to inspect your choices a little more clearly and this is why you know

[00:16:39] we’ve we’ve kind of expounded out now in this concept of adding labels this is why inspecting

[00:16:45] our labels matters because if you don’t have enough time to spend your time in some other way then you can

[00:16:45] say well i didn’t have time to do it then we allow ourselves to believe that that’s unequivocally true

[00:16:53] because the label says it’s true rather than interrogating that label is it actually true

[00:17:00] that i didn’t have the time i’m going to share with you two more techniques

[00:17:05] to refactor your labels right the second technique is to imagine what else must be true

[00:17:15] this is a abstraction technique or kind of you can think about this as a zoom up technique

[00:17:22] in our very first example of outlining our skills on our resume if we are a good presenter

[00:17:31] right if we have the skill of being a good presenter then what else must be true well we

[00:17:37] probably need to have some understanding for communication skills we probably need to have

[00:17:45] some understanding for communication skills we probably need to have some understanding for

[00:17:45] how a good presentation should be structured and so maybe we have some design skill and we probably

[00:17:55] need to be able to educate other people so perhaps we even have some teaching skills

[00:18:00] so there’s some abstract skills that all are necessary to be a good presenter and these

[00:18:07] abstract skills are likely to be more portable and this is the key that i want you to focus on

[00:18:13] when you’re zooming out of a

[00:18:15] label all right zooming out of a label in order to make whatever it is that you’re talking about

[00:18:21] make sense in more than one place make it portable make the the information that you have

[00:18:27] understand it in a way that you can compare it to another situation where you can port it into

[00:18:34] another situation why is this helpful well if you can compare your situation to something else

[00:18:41] right then you’re creating a model and we talk about this a lot in our videos but if you’re

[00:18:45] doing this on the show you’ve talked you’ve probably heard other podcasts that talk about

[00:18:49] mental models all of the time but this is what happens this is this is how we extract models out

[00:18:55] of our daily lives we take these labels and we go up above the label and we ask the question

[00:19:03] what else must be true and when we ask this question and we zoom up out now we can say okay

[00:19:10] i have communication skills what other things can communicate communication skills apply to

[00:19:15] this is particularly important for example for your resume because if you can talk about portable

[00:19:22] skills rather than atomic skills like being a presenter being having a portable skill like being

[00:19:28] a good communicator now applies in many other situations rather than just when there’s a

[00:19:36] presentation to give so we’ve talked about two kind of techniques the first is to think about

[00:19:42] the exact opposite invert whatever it is that you

[00:19:45] whatever that label was the second technique is to zoom up out or ask the question what else

[00:19:52] must be true but the goal of finding something that is portable from one situation one domain

[00:19:59] to another domain the final uh the final kind of way of thinking or refactoring technique

[00:20:06] for your labels is to zoom in to zoom in what do we mean by this well a lot of our labels

[00:20:15] are fundamentally about collecting a lot of complex information and reducing it down

[00:20:24] to much less complex information this is kind of the definition of heuristics heuristics are

[00:20:32] a way of thinking that takes out a lot of the complexity of what you’re thinking about but

[00:20:40] sometimes the things that get compressed out are really important like we talked about in our

[00:20:44] example about kind of walking around the world and you’re thinking about what’s going on in your

[00:20:45] world and you’re thinking about what’s going on in your world and you’re thinking about what’s going on in your

[00:20:46] entire year of 2020 as bad this compresses a lot of information and there are a lot of other more

[00:20:54] harmful kind of labels that we can apply that are equally damaging and compressing uh very

[00:21:02] important information as well for example what is a good developer well this is very reductive

[00:21:08] right it requires us to create on a single scale or singular scale what it means to be a good

[00:21:15] developer now this can be instructive in some ways because there are some things that most good

[00:21:22] developers share but if we were to imagine that there’s only one prescription for being a good

[00:21:28] developer then we’ve created this very restrictive environment where multiple skill sets and multiple

[00:21:36] perspectives are kind of eradicated in favor of a more reductive one so the technique here is to

[00:21:44] imagine that you couldn’t

[00:21:45] have that broad label imagine that you had to describe a software engineer without using the

[00:21:54] term software engineer now you have to go down to a more component level right if you can think

[00:22:02] about composition in software engineering you can probably think about composition as it relates to

[00:22:06] your labels think about the things that make up this particular person what is it that we appreciate

[00:22:15] about the work that they’re doing and what is it that they’re doing that they’re doing that they’re

[00:22:15] doing rather than trying to say that the work that they do is what makes a good engineer so in this

[00:22:24] example about washing the entire year as good or bad we might instead say there are parts of this

[00:22:31] year that were difficult this is a much richer statement right there are parts of this year the

[00:22:39] pandemic was by and large difficult for me that might be a more

[00:22:45] a richer statement than saying well 2020 was a terrible year and this also allows for the good

[00:22:53] to come back through right to be able to say oh but there was a wonderful experience that i had

[00:23:00] in some month of this year despite the fact that it was happening in parallel to this other

[00:23:06] not so great situation so it’s important to be able to do this because what you’re doing is you’re

[00:23:12] you’re allowing the granularity of the work that you’re doing to make sure that you’re doing the

[00:23:15] of a given situation to shine through rather than trying to collapse it all into a singular

[00:23:22] statement a singular label hopefully this is a challenging way for you to think about

[00:23:27] the words that you use the concepts that you use to wrap up your life because we do this

[00:23:34] in all aspects of our lives we imagine things through as reductive of a lens as possible

[00:23:41] not because we’re malicious or because we’re short-sighted

[00:23:44] but because the work that we’re doing is so important to us and that’s why we’re doing this

[00:23:45] this is what makes our brains work efficiently this is important for our brains to be able to do

[00:23:51] but it’s also necessary for us as thinking human beings to request from our brain to think a little

[00:23:59] bit more sometimes to find out in what situations are those labels actually not helping us achieve

[00:24:06] what we care about achieving thanks so much for listening to this episode thank you to today’s

[00:24:10] sponsor new relic we wouldn’t be able to do this show without two things one you listening

[00:24:15] to our sponsors new relic is observability made simple head over to new relic.com to get started

[00:24:22] with one full access user 100 gigabytes a month all for free if you enjoyed this episode and you

[00:24:29] think someone else you know might enjoy it as well please share the episode with that person

[00:24:34] as it turns out this is the best way to help this show continue to grow and reach new engineers

[00:24:41] like yourself

Visão de gráfico