Mobile Monitoring Solutions

Search
Close this search box.

Presentation: Leading Technical Projects – and How to Get Them Done

MMS Founder
MMS Sarah Wells

Article originally posted on InfoQ. Visit InfoQ

Sarah Wells shares stories on how the Operations and Reliability team at the Financial Times built tools that are used by lots of their development teams: the challenges they faced, the things they tried and what worked for them.

By Sarah Wells

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Backcast a Time Series for COVID-19 Truths

MMS Founder
MMS RSS

Article originally posted on Data Science Central. Visit Data Science Central

A couple of months ago, Turkey’s Health Minister announced that the positive cases showing no signs of illness were not included in the statistics. This statement made an earthquake effect in Turkey, and unfortunately, the articles about covid-19 I have wrote before came to nothing.

The reason for this statement was the pressure of the Istanbul Metropolitan Mayor. He has said that according to data released by the cemetery administration, a municipal agency, the daily number of infected deaths were nearly that two times the daily number of the death tolls explained by the ministry.

So, I decided to check the mayor’s claims. To do that, I have to do some predictions; but, not for the future, for the past. Fortunately, there is a method for this that is called Backcasting. Let’s take a vector of time series {X_1, . . . ,X_n} and estimate X_{1-m}, m > 0.

  • One-step estimation for backcasting: X_0^t = phi_{t,1}X_1 + . . . + phi_{t,t}X_t = Phi^T X with X = (X_1, . . . ,X_t).
  • One- step estimation for forecasting, X_{t+1}^t = phi_{t,1}X_t + . . . + phi_{t,t}X_1 =Phi^TX with X = (X_t, . . . ,X_1)

As you can see above, the backcasting coefficients are the same as the forecasting coefficients(Phi). For instance, in this case, the model for new cases is ARIMA(0, 1, 2) with drift:

  • For forecasting: X_t = c + X_{t-1} +epsilon_t + theta_1epsilon_{t-1} + theta_2epsilon_{t-2}
  • For backcasting: X_t = c + X_{t-1} +epsilon_t + theta_2epsilon_{t-1} + theta_1epsilon_{t-2}

#Function to reverse the time series
reverse_ts <- function(y)
{
  y %>%
    rev() %>%
    ts(start=tsp(y)[1L], frequency=frequency(y))
}
 
#Function to reverse the forecast
reverse_forecast <- function(object)
{
  h <- object[["mean"]] %>% length()
   
  f <- object[["mean"]] %>% frequency()
   
  object[["x"]] <- object[["x"]] %>% reverse_ts()
   
  object[["mean"]] <- object[["mean"]] %>% rev() %>%
    ts(end=tsp(object[["x"]])[1L]-1/f, frequency=f)
   
   
  object[["lower"]] <- object[["lower"]][h:1L,]
  object[["upper"]] <- object[["upper"]][h:1L,]
  return(object)
}

We would first reverse the time series and then make predictions and again reverse the forecast results. The data that we are going to model is the number of daily new cases and daily new deaths, between the day the health minister’s explanation was held and the day the vaccine process in Turkey has begun. We will try to predict the ten days before the date 26-11-2020.

#Creating datasets
df <- read_excel("datasource/covid-19_dataset.xlsx")
df$date <- as.Date(df$date)
#The data after the date 25-11-2020:Train set
df_after<- df[df$date > "2020-11-25",]
#The data between 15-11-2020 and 26-11-2020:Test set
df_before <- df[ df$date > "2020-11-15" & df$date < "2020-11-26",]
#Creating dataframes for daily cases and deaths
df_cases <- bc_cases %>% data.frame()
df_deaths <- bc_deaths %>% data.frame()
 
#Converting the numeric row names to date object
options(digits = 9)
date <- df_cases %>%
  rownames() %>%
  as.numeric() %>%
  date_decimal() %>%
  as.Date()
 
#Adding date object created above to the data frames
df_cases <- date %>% cbind(df_cases) %>% as.data.frame()
colnames(df_cases)[1] <- "date"
 
df_deaths <- date %>% cbind(df_deaths) %>% as.data.frame()
colnames(df_deaths)[1] <- "date"
 
#Convert date to numeric to use in ts function
n <- as.numeric(as.Date("2020-11-26")-as.Date("2020-01-01")) + 1
 
#Creating time series variables
ts_cases <- df_after$new_cases %>%
  ts(start = c(2020, n),frequency = 365 )
 
ts_deaths <- df_after$new_deaths %>%
  ts(start = c(2020, n),frequency = 365 )
 
#Backcast variables
ts_cases %>%
  reverse_ts() %>%
  auto.arima() %>%
  forecast(h=10) %>%
  reverse_forecast() -> bc_cases
 
ts_deaths %>%
  reverse_ts() %>%
  auto.arima() %>%
  forecast(h=10) %>%
  reverse_forecast() -> bc_deaths

It might be very useful to make a function to plot the comparison for backcast values and observed data.

#Plot function for comparison
plot_fun <- function(data,column){
  ggplot(data = data,aes(x=date,y=Point.Forecast))+
    geom_line(aes(color="blue"))+
    geom_line(data = df_before,aes(x=date,y=.data[[column]],color="red"))+
    geom_line(data = df_after,aes(x=date,y=.data[[column]],color="black"))+
    geom_ribbon(aes(ymin=Lo.95, ymax=Hi.95), linetype=2,alpha=0.1,fill="blue")+
    geom_ribbon(aes(ymin=Lo.80, ymax=Hi.80), linetype=2, alpha=0.1,fill="blue")+
    scale_color_identity(name = "Lines",
                         breaks = c("black", "red", "blue"),
                         labels = c("After", "Real", "Backcast"),
                         guide = "legend")+
    ylab(str_replace(column,"_"," "))+
    theme_light()
}
plot_fun(df_cases, "new_cases")

plot_fun(df_deaths, "new_deaths")

Conclusion

When we examine the graph, the difference in death toll seems relatively close. However, the levels of daily cases are significantly different from each other. Although this estimate only covers ten days, it suggests that there is inconsistency in the numbers given.


References

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Potential Impact of COVID-19 on 3D Sensor

MMS Founder
MMS RSS

Article originally posted on Data Science Central. Visit Data Science Central

Charting a solid 7.2% CAGR (Compound Annual Growth Rate) over the forecast period of 2018 to 2026, the global 3D Sensor Market is set to see an improvement in market worth. As per a Transparency Market Research, the value by the end of the forecast period would be USD 2556.6 million. The market owes the growth to massive advancements in technology in industries like IT and Telecommunication, and Electronics. For instance, the 3D sensor that has been upgraded has transformed the wat delivery of services happened.

Another significant factor driving growth in the global 3D Sensor Market is that it is used in electronics to improve performance. And, thus, it is widely used in gaming devices, cameras, smartphones, and laptops. And, since it improves user experience, it is also used in gaming for augmented reality and virtual reality.

Thus, as demand for smartphones and other such consumer electronics increases, the global 3D Sensor Market will see an upward growth trajectory. Currently it being used in face recognition features in mobile phones, substituting for the use of PIN and fingerprint.

Experts in the industry feel that over the coming few years, it will be used as a authentication process in payment applications and mobile IDs, etc. as they are not only convenient but also more secure.

Some of the most significant industries that use 3D Imaging are defense, automotive and aerospace. These lead the global 3D sensor market on to a path of higher growth. Ad, since these markets are coextensive with 3D sensing, growth in former leads to growth in the latters.

North America will Dominate the Global 3D Sensor Market over the Forecast Period

Owing to a massive online gaming rage consuming the region of North America, significant CAGR will be charted by the region over the forecast period. The landscape is rife with virtual reality gaming and augmented reality games. Since 3D sensors improve a 360-degree experience by adding depth, they are massively used in gaming. Additionally, it is important to note here that owing to presence of leading manufacturers in the United States and Canada.

However, it is pertinent to note here that the fastest growing region would be Asia Pacific (APAC). The reason behind this significant growth is the extensive use of smartphones and a very large user base.

The Global 3D Sensor Market is Set to Witness Fragmented Competitive Landscape over the Forecast Period

The global 3D sensor market is fragmented and competitive and major players in the market include Infineon Technologies, Omnivision Technologies, Occipital, Inc., PMD Technologies AG, Microchip Technology, Cognex Corporation, Intel Corporation, Ifm electronic GMBH, LMI Technology, and Texas Instruments, among others. Acquisitions and partnerships form the core strategies of better market penetration and expansion of operations.

Get More Inforation about 3D sensor industry by TMR

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Can a Diploma from a Lower Ranking University Hurt your Data Science Career Prospects?

MMS Founder
MMS RSS

Article originally posted on Data Science Central. Visit Data Science Central

Here I specifically discuss the case of a PhD degree from a third-tier university, though to some extent, it also applies to master degrees. Many professionals joining companies such as Facebook, Microsoft, or Google in a role other than a programmer, typically have a PhD degree, although there are many exceptions. It is still possible to learn data science on the job, especially if you have a quantitative background (say in physics or engineering) and have experience working with serious data: see here. After all, learning Python is not that hard and can be done via data camps. What is more difficult to acquire is the analytical maturity. 

University of Namur

In my cased, I did my PhD at the University of Namur, a place that nobody has heard of. The topic of my research was computational statistics and image analysis. These were hot topics back then, and I was also lucky to work part-time in the corporate world for a state-of-the-art GIS (Geographic Information System) company, working with engineers on digital satellite images, as part of my PhD program, thanks to my mentor. Much of what I worked on is still very active these days, on a much bigger scale. It was the precursor of automated driving systems, and the math department in my alma mater was young and still very creative back then. This brings me to my first advice when choosing a PhD program.

Advice #1

  • If you come from a poor background, your options might be more limited (this was my case), and you need to leverage everything you can. My parents did not have the money to send me to expensive schools, and I ended up attending the closest one to avoid spending a lot of money on rent. On the plus side, I did not accumulate student loans.
  • Before deciding on a PhD program, carefully choose your mentor. Mine was not known for his research, but he was well connected to the industry, managed to get money to fund his projects, and was working on exciting, applied projects. 

A side effect on my last piece of advice is that if your goal is to stay in Academia, you may have to rely on yourself to make your research worthy of publications and susceptible to land you a tenured position. The way I did it is summarized in my next advice. You want ideally to leave all doors open, both Academia and other options.

Advice #2

  • Be proactive about reaching out to well respected professors in your field. Attend conferences and meet peers from around the world. Accept jobs such as reviewers. Start publishing in third-tier journals, move to second-tier, and then get a few ones in first-tier journals before completing your PhD. The one I published in Journal of Statistical Society, Series B, is what resulted in me being accepted as a postdoc at Cambridge University. Initially when it was accepted, it only had my name on it. 
  • It helps to be passionate about what you do. My very first paper was in Journal of Number Theory, during my first year as a PhD student. It happened because I had a passion for number theory that I developed during my middle-school and high-school years. I hated high-school math (repetitive, boring mechanical exercises) but loved the math that I discovered and self-learned myself during these years, mostly through reading. I was the only student to participate (and be a finalist) at the national Math Olympiads, in my school. When you are young, it’s something good to have on your resume. 

So to answer the original question – does it hurt coming from a low ranking school – at this point you know that you can still succeed despite the odds. But it requires patience, perseverance, and you must be very good at what you do. Perhaps the biggest drawback is the lack of great connections that top schools offer. You have to make up for that. Also great schools have state-of-the-art equipment and labs (so you can learn the most modern stuff), but somehow my little math department didn’t lack these, so I was not penalized for that. I also cultivated great relationships with the computer science department. At the end, my research was at the intersection of math, statistics and computer science.

My last piece of advice is about what happens next after completing your PhD. In my case, I started a postdoc at Cambridge then moved to the corporate world (after failing a job interview for a tenured position) and eventually became entrepreneur, VC-funded executive, and sold my last venture recently to a publicly traded company. I still do independent math research, even more so and of higher caliber than during my PhD years. 

Advice #3

  • Contact other successful professionals who came from a third-tier university to ask for their advice. In my math department, two other PhD students in my cohort ended up having a stellar career: Michel Bierlaire (postdoc MIT after Namur) is now full professor at EPFL; Didier Burton (also postdoc MIT after Namur) ended up as an executive at Yahoo. 
  • If you can, leverage the fact that you are very applied, don’t have student loans, and thus you can ask a lower salary, be more competitive, gain various horizontal experience in many places while developing world-class expertise in a few areas.  I eventually realized that working for myself (not as consultant, but entrepreneur) was what I liked best.

You may argue that you don’t need any diploma to create your own self-funded company, not even elementary school, but in the end I believe I got the best I could out of my PhD. In my case, it also implied relocating several times, from Belgium (due to lack of jobs) to UK to United States, and from the East Coast to the Bay Area and finally Seattle. I’ve been through various bubbles and market crashes; you may use your analytical skills to navigate them the best you can, selling and buying at the right time, understanding the markets, and emerge stronger each time. 

About the author:  Vincent Granville is a data science pioneer, mathematician, book author (Wiley), patent owner, former post-doc at Cambridge University, former VC-funded executive, with 20+ years of corporate experience including CNET, NBC, Visa, Wells Fargo, Microsoft, eBay. Vincent also founded and co-founded a few start-ups, including one with a successful exit (Data Science Central acquired by Tech Target). He is also the founder and investor in Paris Restaurant in Anacortes, WA. You can access Vincent’s articles and books, here.

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Building an Algorithm to Trade Items on the Steam Community Market

MMS Founder
MMS RSS

Article originally posted on Data Science Central. Visit Data Science Central

Machine learning is increasingly used to build trading algorithms: today’s machine learning algorithms can deal with incredibly complex problems, including portfolio choice. At the same time, there is now a vast amount of data freely available – alongside the computing power to crunch complex data science problems. 

Here at ELEKS, one of our colleagues – Volodymyr Mudryi – wanted to explore the capabilities of machine learning to train an algorithm to trade profitably. Let’s take a look. 

Building an agent that can trade for profit: an overview 

The stock market is a very complex environment, so we decided to experiment on a simpler market instead: the Steam Community Market. Steam is a popular online gaming platform. Gamers can use the Steam Community Market to buy and sell in-game items to fellow gamers. 

We proceeded to build what’s called a reinforcement learning agent. Reinforcement learning is a field of machine learning where the agent under training adjusts its behavior by observing the optimal way to behave in an environment in return for maximum rewards. 

The basic algorithm consisted of empirical rules, which we optimized with genetic algorithms. We then added a Deep Q-Learning algorithm which enables us to construct a rudimentary decision-making policy. 

As a final step, we used an Actor-Critic algorithm to optimize the decision-making policy at every step of the process, rather than right at the end of the learning exercise. After all, a real trader would evaluate their trades after every trade. 

Starting with a basic algorithm  

Our experiment used data from the Steam Community MarketThere are several thousand items in this market of which only 48 were selected, in line with three criteria that we set. We watched the price history of items for the day over 430 daysYou can read more about the full selection criteria and methodology here. 

Next, we set up a basic algorithm that accounts for the structure of the Steam Community Market and our observations. The algorithm followed these four steps: 

  1. Calculate the maximum price for the last three days of trading on the market. 
     
  1. If the current price is less than the maximum multiplied by 0.85 (given that 15% of the price is diverted to Steam market commission), then buy the item and consider it the new maximum price – the purchase price. Save the purchase price. 
  1. If the price for which the item was purchased is lower than the current price multiplied by 0.85, then put the item up for sale. 
     
  1. If the purchased item has not been sold within seven days, reduce the price by 5% and deduct three days from the number of days without sale. 

So, what were the results of this basic algorithm? Well, putting this algorithm to trade led to a profit of 179% across 16 months. 

Adding a genetic algorithm 

Next, we wanted to optimize the results of the basic algorithm above, so we added a genetic algorithm. At every step, we randomly generated 50 agents according to a set of statistical parameters. We then picked the best-performing agents based on the return they achieved in the market, adding a new mutation to each agent. We continued the iterations. 

In observing the resulting agents, we found that the returns fluctuated without consistently outperforming the basic algorithm by a large margin. There was clearly a learning problem involved.  

For example, in one iteration we found that the agent finished executing with a portfolio of items worth 49,630 monetary units, but with only 454 monetary units in the bank. Essentially, the agent anticipated high profits in the future and therefore never sold any items. 

Training the agent with a reinforcement learning algorithm 

We needed to add a machine learning algorithm to help the agent to maximize profits based on what it learns through interacting with its environment. Deep Q-Learning allowed us to find a relatively simple decision-making policy that’s not tied to a specific action at a specific point in time. 

For this stage of the experiment, we only selected one of the items – and compared the results to the basic algorithm applied to just one item.  

We won’t discuss the precise methodology of the Deep Q-Learning algorithm here – but the net result was that, after completing the training and running the model, the total portfolio value was 15,273 monetary units – up 21% on the base model, and 4% higher than the optimized basic model. 

Next, we tried an Actor-Critic algorithm running with the above data but using a reward function that included a range of features. For example, if the agent reaches the end of the data with a positive balance, the agent is rewarded with the amount of money earned. 

We also defined the trading environment, and models for both actor and critic. The training process lasted 5,000 epochs with a learning rate equal to 0.0003

 

Our agent simply didn’t trade – so we made some tweaks 

The result of the above Actor-Critic was disappointing: the agent decided that inaction is the best option – which meant the model had no value. We had to tweak a few elements of the model, but most of the work was done on the rewards function which we tweaked by, amongst other things, deleting the time elapsed parameter and changing the reward for purchase to zero. 

Unfortunately, these tweaks couldn’t put that agent back on the track to trade, so it shows that reward function for Actor-Critic algorithm requires more complex structure and extra domain knowledges. 

What did we learn? 

Well, it is clear that algorithms can be used to optimize decisionmaking in an uncertain market, and to do so achieving profit. Including the stock market, or indeed the Steam Community Market. Rules-based genetic algorithms work but require further input. 

A Deep Q-Learning algorithm can work for current data and furthermore, showed the best results. Also, it can always be combined with other approaches – or indeed, be expanded by further parameters. That said, our experiment did not involve a response to the environment because prices were predefined and historical. 

An Actor-Critic algorithm took a lot of time and efforts but with provided data it’s hard to achieve a good trading results, but we still made some investigations and assumption what could improve training process. Again, you can read full details of our Actor-Critic algorithm here. 

With a bit more time and experimentation, we may get to the point where these algorithms can be used to apply to real trading in the stock market – but without a doubt, there is much more to be done. 

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Article: Vue 3 Released with New APIs to Tackle Usage at Scale

MMS Founder
MMS Bruno Couriol

Article originally posted on InfoQ. Visit InfoQ

Vue 3 recently shipped with numerous new APIs that cater to using Vue at scale and in non-DOM environments. New Suspense and Teleport built-in components and new CSS scoping rules make for a more expressive template language.
Custom events and fragments allow Vue components to have a public API closer to that of regular DOM elements. The Vue ecosystem is finalizing its migration to Vue 3.

By Bruno Couriol

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


New Hybrid PCA-Based Facial Age Estimation Using Inter-Age Group Variation-Based Hierarchical Classifier

MMS Founder
MMS RSS

Article originally posted on Data Science Central. Visit Data Science Central

This article was written by Tapan Kumar Sahoo & Haider Banka

Abstract

In this paper, we propose hybrid principal component analysis (HPCA) to extract appearance feature of a face and inter-age group variation-based classifier (IAGVC) with regression to estimate age of a person. The proposed age estimation system is robust and less sensitive to outliers where nonuniform distribution of images at different age groups is existing. Under HPCA, we introduce two novel methods, extended SpPCA and extended SubXPCA. The issues, such as summarization of variance, variable component selection, computational complexity and classification accuracy of HPCA, have been addressed as well. The proposed HPCA operates on subpattern and whole pattern at a time and extracts appearance feature based on both local and global variation of faces. The IAGVC uses HPCA-based hybrid Eigen spaces of each training age group to estimate age group of a test image, and subsequently, support vector regressor estimates the specific age in the selective age group. The experimental results on FG-NET aging database show that the proposed HPCA-based IAGVC has better classification accuracy as compared to existing classical PCA, local SpPCA and SubXPCA over all age groups.

Table of contents 

1. Introduction

2. Background 

    2.1 Prior Work on Appearance Feature Extraction 

    2.2 Prior Work on Age Estimation 

    2.3 Overview of Proposed Work 

3 Database Description 

4 Methodology 

    4.1 Data Preprocessing 

    4.2 Appearance Feature Extraction Using Variation of PCAs 

          4.2.1 Hybrid Principal Component Analysis (HPCA) 

          4.2.2 Issues of HPCAs 

    4.3 Inter-Age Group Variation-Based Hierarchical Age Estimation 

5 Experimentation 

    5.1 Performance Evaluation 

    5.2 Experimental Setup 

    5.3 Results Discussion 

6 Conclusions and Future Works 

To read this article, click here.

 

DSC Ressources

Follow us: Twitter | Facebook  

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Deep Learning Reinvents the Hearing Aid

MMS Founder
MMS RSS

Article originally posted on Data Science Central. Visit Data Science Central

This article was written by DeLiang Wang.

My mother began to lose her hearing while I was away at college. I would return home to share what I’d learned, and she would lean in to hear. Soon it became difficult for her to hold a conversation if more than one person spoke at a time. Now, even with a hearing aid, she struggles to distinguish the sounds of each voice. When my family visits for dinner, she still pleads with us to speak in turn.

My mother’s hardship reflects a classic problem for hearing aid manufacturers. The human auditory system can naturally pick out a voice in a crowded room, but creating a hearing aid that mimics that ability has stumped signal processing specialists, artificial intelligence experts, and audiologists for decades. British cognitive scientist Colin Cherry first dubbed this the “cocktail party problem” in 1953.

More than six decades later, less than 25 percent of people who need a hearing aid actually use one. The greatest frustration among potential users is that a hearing aid cannot distinguish between, for example, a voice and the sound of a passing car if those sounds occur at the same time. The device cranks up the volume on both, creating an incoherent din.

It’s time we solve this problem. To produce a better experience for hearing aid wearers, my lab at Ohio State University, in Columbus, recently applied machine learning based on deep neural networks to the task of segregating sounds. We have tested multiple versions of a digital filter that not only amplifies sound but can also isolate speech from background noise and automatically adjust the volumes of each separately.

We believe this approach can ultimately restore a hearing-impaired person’s comprehension to match—or even exceed—that of someone with normal hearing. In fact, one of our early models boosted, from 10 to 90 percent, the ability of some subjects to understand spoken words obscured by noise. Because it’s not necessary for listeners to understand every word in a phrase to gather its meaning, this improvement frequently meant the difference between comprehending a sentence or not.

Without a better hearing aid, the world’s hearing will get worse. The World Health Organization estimates that 15 percent of adults, or roughly 766 million people, suffer from hearing loss. That number is rising as the population expands and the proportion of older adults becomes larger. And the potential market for an advanced hearing aid isn’t limited to people with hearing loss. Developers could use the technique to improve smartphone speech recognition. Employers could use it to help workers on noisy factory floors, and militaries could equip soldiers to hear one another through the noisy chaos of warfare.

It all adds up to a big potential market. The global US $6 billion hearing aid industry is expected to grow at 6 percent every year through 2020, according to the market research firm MarketsandMarkets, in Pune, India. Satisfying all those new customers, though, means finding a way to put the cocktail party problem behind us. At last, deep neural networks are pointing the way forward.

To read the rest of this article, click here.

 

DSC Ressources

Follow us: Twitter | Facebook  

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Article: Breaking the Taboo – What I Learned from Talking about Mental Health in the Workplace

MMS Founder
MMS Sophie Kuster

Article originally posted on InfoQ. Visit InfoQ

Key Takeaways

  • The topic of mental illness is often shrouded in taboo and stigma which hinders open communication and relationships.
  • Opening up about your struggles can help. It can make others feel less alone and encourage them to speak about theirs in turn. But it can also help by allowing yourself to ask for and accept help.  
  • Speaking openly about your experiences and opinions can be daunting. But rather than a character trait, it is more of a skill that can be learned and practiced.
  • Showing vulnerability, for instance by admitting to having a mental illness, is sometimes seen as a weakness. But contrarily, it can be an empowering experience – to yourself and others.  
  • In uncertain times like the ones we currently live through, many struggle emotionally and psychologically. By changing the way we talk about mental illness to a more open and unprejudiced conversation, we can make it easier to see the symptoms, understand them, and get help if needed.

Mental illness is a topic that does not get discussed openly very often. Many people concerned hide their own history for fear of being stigmatized, especially in the workplace. I was no exception to this, until one day I decided to speak openly, even with my boss and co-workers. Communication and relationships have changed a lot for me since then and – even though not all experiences have been positive – largely for the better. I recently told my story at the Agile Testing Days and afterwards many participants reached out to me, saying that they had had similar experiences with illness and self-imposed silence. This again solidified my belief that talking about mental illness helps yourself and others. So let me share with you what I have learned from breaking the taboo.

When I was in my early twenties and a math student at University, I fell ill. I did not know what was happening to me; I only knew things were going terribly wrong. University was not going well, my boyfriend and I were fighting incessantly and – without wanting to – I was pulling away from all of my friends. Even my own thoughts scared me; I felt sad and scared and overwhelmed and lonely most of the time, and sometimes I did not feel anything at all.

I felt so helpless that I desperately clung to the only aspect of my life of which I still felt in control: my weight. I started obsessively exercising while limiting my calorie intake, striving for a body that I hoped would satisfy my own perfectionism. But it never did. By the time I admitted to myself that I needed help, my body looked as sick and malnourished as my soul felt.

When I finally went to a doctor, I was diagnosed with depression, an anxiety disorder and eating disorder, and I spent the next years on antidepressants and in therapy. I got better, little by little, but it took time and effort.

However, this is not the story of my sickness or my recovery. This is the story of what happened when I decided to speak openly about mental health, even in the workplace. So we will have to fast forward a few years.

When I had finished my University degree and first applied for jobs, my mental health became a problem again. I felt much better but my illness had taken up so much mental capacity for so long that my studies had taken noticeably longer than they should have. So my CV looked bad and, fearing the stigma that comes with mental ill health, I did not want to disclose the reasons, least of all in a cover letter or a job interview. As a result, the job application process was challenging.

One day I decided to take a risk. I had been on an interview and had been asked – as always – why I had lost time at University and the interviewer was obviously not buying my evasive answers. I admitted to the eating disorder but nothing else, hoping that that was just enough to explain the gaps, but not so bad that he would think I would relapse under the slightest bit of pressure.

He gave me a chance and hired me, on a limited contract at first. I instantly liked the work; the atmosphere was great and I loved the colleagues. I was really happy there, and to this day I still am. But back in my first months, I was constantly worried that I would slip up and someone would find out my deep dark secret.

After my diagnosis, I had always talked rather openly about mental health (or lack thereof) with friends and family. But at work, I did not dare to. You simply do not talk about that in the workplace, I thought. I did not want to be seen as weak or vulnerable or unable to withstand stress, or even stupid, least of all by my boss. That of course meant I could not speak openly with my colleagues. The years I have spent being ill and getting better have had a big impact on who I am now as a person, and they have taught me a lot. And normally I like to share what I have learned, but you cannot share insights from therapy if you want to keep the fact that you have been in therapy a secret.

So there was always friction. On the one hand, I spent hours a day with my co-workers, working, having lunch, playing foosball or having after work beers, and I had become quite fond of them. But on the other hand, I always bit my tongue.

Then, two years ago, something changed. My colleagues and I went to the Agile Testing Days for the first time. I do not know what I expected exactly, but certainly not what I encountered: people talked about mental health. In a professional environment. At a tech conference, no less. After a talk about burnout, I asked the speaker how he had explained his absences to his boss and he answered, “The truth”. That got me thinking.

Commitment, focus, respect, courage and openness – we probably all know the Scrum values by heart. But trying to hide my own history certainly did not feel courageous and open. And could I not trust my boss and my colleagues to still respect me if they knew about my history?

As it happened, I had my end of year performance review on the very first day back at the office after the conference. My head still spinning from all the new impressions (and a serious lack of sleep), I was determined to rectify my own lack of openness. So I started the review by telling my boss the whole truth about my mental illness. And from that moment on, I talked openly about mental health whenever the topic came up, even with my colleagues.

Here is what I have learned since then:

1. You are never the only one.

When you mention illness or therapy in a casual conversation with a group of people, there is often enough someone who responds with their experiences. It is a simple matter of statistics. According to the State of Health in the EU Report of 2018, 17% of the population had a mental health problem during one year- that is more than one in six. So if the group is large enough, there is always somebody else who has had a similar experience.

It should not have come as a surprise that as soon as I started speaking about mental health issues with my colleagues, I learned I was not the only one who had been through hard times or had a loved one who had.

One day, a co-worker approached me. He said there was something that he had kept to himself and he did not want to hide it anymore, but did not know how to start talking about. With my openness, he felt safe to confide his story in me. And a short while later he talked about it at lunch in a larger group as well. Having helped someone become more comfortable with talking about themselves is one of the most rewarding experiences that I have had.

2. Openness is a habit, not a character trait

The second change I noticed was within myself. Anxiety and impostor syndrome make it hard to voice your opinions, especially if you have to contradict someone. After all, who would want to listen to silly me? Why should I be right?

In an agile team however, you have to be able to say, “Hey, I think we’re wrong here” or, “I think we should do it differently”. That is what the values of courage and openness are all about, aren’t they?

I used to think that being able to speak your mind openly and easily was a character trait that I just sadly did not have. But I have found that it becomes easier with practice. Once I had shared something so very personal, expressing my opinions about day-to-day business did not seem so overwhelming anymore. Over time I became more and more comfortable with being open. At lunch I would share a funny anecdote about therapy and during our sprint planning I would share my thoughts about the user story we were discussing, even if I thought that I might be wrong about it.

I will not pretend that it is now always easy. The anxiety and impostor syndrome did not magically disappear; it still takes a conscious effort to overcome them. But I have learned that openness, much like confidence, is like a muscle that can be exercised. When you start, it is hard and your muscles get sore. But the next time you try, it feels a little easier and the ache afterwards is not as bad.

3. There’s strength in showing vulnerability

Being a sensitive person plus having depression plus anxiety means I have never had thick skin. When I was younger, I was called a wimp and a cry baby for it. So naturally, I have always worried about conflicts at work. I never wanted to show vulnerability. But that too changed.

Even in the most harmonious, well-attuned teams, misunderstandings and conflicts occur, on a professional level as well as on an interpersonal one. And you cannot resolve a problem without accepting that there is one in the first place, and that sometimes means admitting you are hurt.

One day for example, I had a stupid misunderstanding with a teammate. He was rather new on the team, and during our daily stand-up meeting, I had mentioned a problem that I was facing. His response indicated that he thought I was not doing enough to solve it and he had to save me. I did not say anything and told myself he probably did not mean it, but it gnawed at me. I kept obsessing about it for hours. Old me would just have let it ruin my day and probably leave a dint in my confidence while putting on a happy face. But new and improved, open and talkative me did not want to. So I sent him a message saying I was bothered by what he had said and why. My therapist would have been so proud. All these hours of talking about healthy conflict resolution and here I was standing my ground, making I-statements.

We talked the problem through, sorted out the misunderstanding, and have much better communication ever since. Admitting I was hurt had not made me weak in his eyes. On the contrary, he appreciated me facing the issue.

4. Some people just don’t get it
 
If you break a leg, no one would ever suggest you should pull yourself together and stop having a broken leg. When you are physically ill, your body needs time to heal. Sometimes you need more than time. You need doctors and medicine and plaster casts and surgery and radiation and bandages. There is no shame in that.

But when it comes to mental illness, some people do not apply the same logic. Their reasoning seems to be that if it is your head that is suffering, it is “all in your head”. So you receive advice ranging from, “Look on the bright side”, “Just stay positive,” to “Antidepressants are bad. They mess with your brain” and, “You know you’re too skinny, so eat something. It’s not that hard.” It is for the most part probably well-meaning, but entirely unhelpful. Even worse than unhelpful, it can be dangerous. When you are constantly told that it is not a real disease, you might not see a doctor, you might not take your medicine. But it is a real disease, and it is potentially fatal. So do not laugh it off and do not make people feel bad about taking pills that could be saving their life.

So how do you deal with these people? Do you try to educate them or do you try to ignore them? I myself have over time accepted that some people just do not get it. And that has meant the end of more than one friendship. That does not mean that I have stopped caring. Not caring is not exactly an easy thing to do if caring too much is part of your diagnosis. I still try. I try to explain to people what it’s like and I try to be zen about it when that is fruitless.  
 
5. Some people get it really wrong
 
I talk quite openly about very personal things. And this openness at times gets misconstrued as a cry for attention, and in particular, male attention. I do not see how you can understand “My brain does not produce enough of the happy chemicals” as being flirty, but it has happened.

One day at a party, in a discussion about openness, a man told me, “I like your openness. And since we’re all being so open, here’s what I like in bed. Oh and by the way, I have been thinking about doing that to you while you were talking.” Maybe this man did not even mean any harm by what he said. Maybe he honestly thought he was cheering me up by complimenting me. But maybe he was exploiting my perceived weakness, my “confidence issues”, and that is appalling. It is predatory behavior and I am not easy prey.  

Thankfully, I have not had many of these encounters.

6. People want to help you. Let them

When my colleagues and I came to the Agile Testing Days for the first time, we were planning to fly in from Cologne Bonn Airport to Berlin Schönefeld, just over an hour in the air. The only problem was that I have a phobia of flying. When I am on an airplane, panic takes over me and I cry and yelp and hyperventilate. Obviously, I did not want my new colleagues to see me lose my cool like that, but I knew it was going to happen and there was nothing I could do to hide it.

So I decided to get ahead of the embarrassment and told one of them about my fear. But she did not laugh, she did not find it embarrassing, she did not tease. She asked, “What can I do to help?” I told her how to make a flight easier for me – distract me and remind me how to breathe in case I forget. And that is what she did during the entire flight. Not only that, but all my colleagues on that flight helped to keep my mind off of my fear, even the ones I had not briefed.

My fear of being laughed at for my silly phobia made me forget one very important thing: most humans are kind. They will want to help you, if you let them. Asking for help and accepting help can be hard, I know. But changing your perspective and looking at the effects on the person who is helping might make it easier; psychological research suggests that helping others also boosts your own well-being and lowers depression. We have probably all felt the warm glow that comes from helping others. So look at it this way: you get the help you need, they get the endorphins. Win – win! And if you are scared that people will like you less for being a burden, consider the Ben Franklin effect: a person who has helped another person before will be more likely to help them again than if they had been the one who was helped. Subconsciously they reason that if they have helped, they must like the person, and actually start to do so.

7. It makes things easier

One reason I began speaking openly about mental health was purely logistical. When I first thought about speaking publicly on this topic and wrote a proposal for a talk at the Agile Testing Days about it, I was going to say it has been years since I needed therapy or meds. But that was before the world got turned upside down by the pandemic. The threat of the virus, social distancing, and on top of it, having to battle another onset of ill health (this time physical) during the pandemic has been trying and I am considering going back to therapy. If I had never opened up about having struggled with my mental health before, would I be able to say that now? If I ever have to leave the office early to go to therapy, would I try to hide it? I do not know what I would have done, but I know that worrying about it would have taken up a lot of headspace. But now that the band-aid is ripped off, I do not have to think about it any longer.

8. Everybody should know the oxygen mask rule (although many don’t)

My favourite thing about not trying to hide my experience with illness and therapy is that I can now share what I have learned. Sometimes people ask me for the most helpful thing therapy has taught me, and I always tell them about the oxygen mask rule.

One day, my therapist suggested that I needed to stop thinking about how to make others happy and think about how I was treating myself. And it was true; I constantly worried about making my parents proud, being a great girlfriend and a good friend. But to myself I was really mean: constantly nagging, thinking of myself as a loser, starving myself and not taking care of my emotional well-being. I thought as long as I was nice to others that was okay; I even fancied myself rather selfless.

But neglecting your own happiness has nothing to do with selflessness. On a plane, the safety instructions tell you to “put on your own oxygen mask before assisting others,” because if you pass out while trying to put a mask on somebody else first, you are no use to anyone. Being kind to yourself and taking care of your emotional well-being is like putting on that oxygen mask. It is not only not selfish, it is your job. Because if you do not take care of yourself, somebody else has to. If you do not make an effort with your mental health, you will have to make an effort with your mental illness. In times of crisis, being kind is essential; to each other, yes, but also never forgetting to be kind to yourself. Give yourself air to breathe.

Talking about mental health is more important than ever.

It is about time we all get used to talking about mental health. The number of people who suffer from depression and other mental health issues has been high for years, and it will only get worse.  Long after this pandemic is over, we will still have to deal with the psychological aftermath of social distancing and not hugging people, not to mention living with existential dread for months on end. These are horrible, stressful times we live in, and they are taking a toll on even the strongest of characters.

Not many will come out of this with their psyche unscathed; some will need psychological help. But not everyone will be able to get it. Some will be too ashamed, and some will not see the symptoms as symptoms. So talking about the warning signs, about what you can do to look after your own mental health and where and how you can get professional help is more important than ever.

Thankfully, mental illness is much more openly discussed now than it was thirty years ago. But the way we talk about mental illness is still influenced by taboo, stigma and stereotypes. “Mentally ill” is something that is said about “lone wolf shooters,” while in truth, the vast majority of mentally ill people are people like you and me. Odds are, you know someone who is, but maybe you don’t even know that they are. How often do people say, “I’m sorry I haven’t called. I’ve been too depressed and anxious. I’m a bit better now. Let’s have coffee!”?

Talking is both the hardest and the easiest thing you can do to change the experience of patients who suffer from mental disorders. It takes some guts to speak up, but once you do, you can help other people just by talking and they can help you in turn.

I first wrote these notes for a talk at the Agile Testing Days. Days before the conference, I was incredibly nervous. What if everything I had to say was banal? What if nobody wanted to listen? But the conference taught me otherwise. The feedback I received had me alternately crying and beaming with happiness. People approached me, saying how important they thought my words were, how they had made them think and even cry and how they had resonated with them.

Many have had similar experiences with both mental illness and with fear of being stigmatized for it. But nobody should have to suffer alone and in silence. This is why I no longer hide my history; I was lucky enough to have been met with a lot of kindness, understanding and support. This puts me in the comfortable position to be able to pay it forward.

I now realize that with every person who speaks openly about mental illness, we can chisel away a tiny little bit of the walls of taboo and shame that surround the issue. Looking back at my younger self- lost, scared and ashamed- I know that it would have made a difference if these walls had not been there. So let us normalize talking about mental health and mental illness, bit by bit.

About the Author

Sophie Küster studied mathematics at university which did not go as planned. She finished with a diploma nonetheless. Now she works happily as an agile tester at cronn GmbH in Bonn. No stranger to the universe’s gut punches, she is passionate about mental health and raising awareness and communication. She recently celebrated her first conference speaking engagement at the Agile Testing Days.

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.


Meetings in a Time of Separation

MMS Founder
MMS Ben Linders

Article originally posted on InfoQ. Visit InfoQ

Having many people in virtual meetings can lead to people who only partly attend and become disengaged. We should question who should be attending the meeting and make information from the meeting available for those who decide not to attend to decrease meeting FOMO.

Michael Lopp spoke about meetings in a time of separation at Stretch Online 2020

These days people are doing a lot of video meetings. The primary issue in virtual meetings is one of engagement, as Lopp mentioned:

How do you show up in a meeting when we’re all virtual? You can hang back and half-listen, but I think that is inefficient. If you are going to a virtual meeting, your job is to engage. Video on. Active participant. If either of those two requirements aren’t achievable, I’d ask yourself, “Must I be in this meeting?”

If people feel disengaged, they could manage themselves out of a meeting as Lopp suggested. Meeting facilitators can support this by committing to making whatever happens in that meeting public knowledge, as Lopp explained:

Capture the notes, capture the decision, and make them publicly available in an obvious place. There are confidential meetings where this is not the right move, but the vast majority of meetings would benefit more from following this practice. It will decrease meeting FOMO, virtual or not.

InfoQ interviewed Michael Lopp about engaging people in virtual meetings.

InfoQ: Now that people are working more from home due to the Covid-19 pandemic, how has this impacted meetings?

Michael Lopp: The obvious impact is that we’re sitting in video conferences a lot more and that has a bunch of interesting implications. Once we collectively figured out the MUTE button, the next challenge is that meeting room size doesn’t really matter anymore so we can invite as many humans as we like.

This seems like a good idea from a transparency perspective, until you’re in a meeting that normally had 10 people and now has 50. There’s a bunch of folks with their camera on who are allegedly listening, but who are really half-listening. This means they are half-working. Is this a better set-up than this half-working person getting the meeting notes a little later? I don’t think so.

InfoQ: What difficulties do people have when doing virtual meetings?

Lopp: It’s an issue of engagement. Whoever is running the meeting has an additional responsibility to scan that video grid of humans and figure out who wants to speak. It’s easy in a real conference room because you can see Frank over there fidgeting when he wants to speak. Those visual cues are harder to find in a virtual meeting, but as important to discover.

InfoQ: How can we “read the room” in virtual meetings? And how can that help us to keep everyone involved?

Lopp: If I’m running a meeting, reading the room means starting by scanning the room, saying hello, noticing little things (new haircut, new background, etc.) and then kicking off the agenda. While the meeting is going on and again assuming that the majority of the folks leave the video on, I read the room by regularly scanning attendees for engagement. It’s so hard virtually, but you can see when someone is trying to get a word in edgewise. Ok, make a note and when the current talker stops talking, bring them in.

InfoQ: How can we spot that we’re losing people, and what can we do if we see that happening?

Lopp: Running a meeting is work. My practice to keep folks engaged is to keep a running inner dialog of what is happening in the meeting, how the narrative is evolving, and who I would expect to be jumping in to debate as that evolution occurs. When I find unexpected silence, I call on folks, “Juliet. We were debating this in our 1:1. What do you think?”

InfoQ: What alternative mediums besides meetings can we use to interact?

Lopp: I’m biased as I’m the former VP of engineer at Slack, but I think that tool can serve much of the purpose of a meeting. Get together humans, discuss a topic, make a decision. It’s often better than a meeting because it’s timeless and there’s a clear visible record. I’ve found it takes time for cultures to see this tool as a replacement for some meetings, but once they get it, it takes off. Also, it scales better than meetings.

InfoQ: What have you learned this past year when it comes to keeping in touch when working from home?

Lopp: I’ve been preaching consistent 1:1s for years and my core 1:1 principles are even more important as we all work from home:

  • 30 minutes (at least)
  • Every week
  • No matter what

My 1:1s are not check-ins on status. They are meant to be conversations. We discuss topics of substance. We riff on those topics and see where they will take us. Sure, we have work topics to discuss, but we’re also just connecting as humans and that’s more important than ever.

Subscribe for MMS Newsletter

By signing up, you will receive updates about our latest information.

  • This field is for validation purposes and should be left unchanged.