Author Archives: Bialik Julia

A tech glance at the Oscars

Reading Time: 5 minutes

On Sunday night, the biggest film event took place in LA, albeit in a much different format than usual. Not only did the ceremony take place at a different venue (no Dolby Theater this time), but also it was a much smaller affair, with many attendees connecting through Zoom and with those inside the venue following social distancing protocols. While the ceremony might have looked different, it was still a celebration of the movie industry at its finest, which like virtually all other sectors took a hit in 2020. Let us take a look at the ceremony through a tech lens (which I hope is not the first thing which popped into your mind when thinking about the Oscars). 

Oscars 2014: Look back on the best and worst of years past - Los Angeles  Times

Source: LA Times

If you weren’t already aware, betting on which nominees are going to take the shiny statue home that night is a pretty big deal (and a lucrative business, might I add), with odds being taken as soon as the nominations come through. With lots of historical data readily available, it is no surprise that machine learning algorithms have started to be deployed to predict just who will be crowned the winner during the ceremony. 

One such attempt was made on a blog called BigML, where predictions were made for 8 main categories at the Oscars. For each separate category, different models were trained, which were fed historical data (previously nominated and victorious movies at the Oscars), but also information regarding the movies’ evaluation on IMDB, nominations and winners for 20 key industry awards (Golden Globes, BAFTAS, Critics Choice etc.), as well as basic information such as synopsis, duration, budget and genre. The whole dataset consisted of over 1k entries, with over a 100 features.

Overview of the dataset. Source: BigML blog

The 8 different models were trained with multiple approaches on BigML side, with a variety of their own products used in the process, which consisted of supervised learning algorithms solving classification and regression and optimization process for model selection and parameterization. Once the candidate models were created, Batch Predictions were made against the movies produced in 2020 that we had set aside in a separate dataset. As was the case last year, all approaches yielded more or less similar predictions supporting each other’s constructs. The individual Deepnets  models configured with the Automatic Network Search option were chosen in the end.

For reference, the field importance report below is that of the Best Picture Oscar and it shows that fields like user reviews, synopsis, votes, and wins in Critic’s Choice plus nominations for Online Film Critics Society, People’s Choice, Hollywood Film, and BAFTA awards all factored in strongly in the final scores for each Best Picture nominee.

Best Picture Deepnet summary report. Source: BigML blog

As was mentioned, BigML considered 8 categories, and the models created by them turned out to be right in 4 out of 8 cases – the models correctly predicted the victors in Best Picture, Best Director, Best Supporting Actress and Best Supporting Actor categories. While it definitely could be better, it correctly predicted the win for Nomadland in the Best Picture category, which is the hardest one of them all to get right (due to a different process of selecting the winner called preferential balloting). 

Speaking of the process of selecting the winner through preferential balloting, an article by Nicholas Parker written in 2020 is worth mentioning, where the method was clearly explained, and where the author created a bespoke algorithm mimicking the unique voting method. Using the example of 2020 Oscar nominees, Parker trained his own model utilising scraped data for each awards show’s (from the ‘award season’, such as Golden Globes or BAFTAS) nominees and winners from Wikipedia, and historical wins at the Oscars of the Best Picture category. He then merged them all into one dataset in Python using the Pandas and Beautiful Soup packages. 

Preferential balloting at the Oscars is a system where voters submit a ballot of all the nominees ranked from “best to worst”. Once the ballots are collected, each first choice on a ballot is tallied up as a vote for that particular movie. Then the least popular films are eliminated one by one, and ballots are once again ranked, until one film has greater than 50% of the #1 votes. After a film is eliminated from all ballots, the ballots which previously had the eliminated film at their #1 spot, now have their #2 move to the top spot, which increases the number of votes for the remaining films. This process continues until one film has greater than 50% of the #1 votes and then it is declared the winner. A simulation of this elimination process is shown below.

Simulation of preferential balloting elimination. Source: Towards Data Science

Parker then went on to use a Random Forest Classifier, where the ProbA values for each film were used in the test set, and then further used to create 1st – 9th place ranking of the film.

An overview of an Individual Decision Tree’s vote on the test. Source: Towards Data Science

In Parker’s own words: “Using my Preferential Balloting Random Forest I simulated this year’s Best Picture race. To de-correlate each Decision Tree, I varied which awards show each tree saw, similar to Random Forest’s max_featureshyperparameter. In this simulation, max_features represents what guild the voting academy member may be in, or how closely they follow the other awards shows that season. I also included a random noise feature for each Decision Tree to train on, representing each voter’s innate bias towards certain films. The Academy is made up of around 7,000 unique voters, so I fired up my Forest, which soon produced 7,000 ballots”. His final predictions looked like this, with 1917 being crowned the winner:

The final results after 6 rounds of preferential balloting elimination. Source: Towards Data Science

While Parker did not predict the winner correctly – the big prize went to Parasite that night, the movie was the second choice of the algorithm, meaning that Parker was definitely on to something! Anyways, the process itself is super fascinating to watch, and as we can see the results were super close to reality.

I hope this technological perspective at the Oscars was also as intriguing to you all as it was to me, and that they inspired you to look deeper for tech related issues in the most surprising of places. 


Tagged , , ,

How challenging is it to produce vegan cheese?

Reading Time: 5 minutes

The answer to the question is: pretty hard. With the rise of veganism over the last couple of years, individuals eating a primarily plant-based diet have been the target of multiple companies, who have taken it upon themselves to produce vegan alternatives to a variety of ‘traditional’ animal products. Unfortunately, the available alternatives are more often than not a flop – with a weird, chewy texture, lack of taste, and an expensive price tag. Seeing as plant-based diets, even if applied in a limited capacity (reducing the intake of animal products throughout the week), are and will for sure continue to gain popularity, the demand for animal-free alternatives will also rise. Companies will inevitably need to step up their game, and come up with solutions which will satisfy the taste buds of customers. Let’s take a look at one of the game-changers of the ‘vegan technology’ industry.

Example of a plant-based creamy cheese made by Prfect Day – look at that texture!

One of the products which (speaking from personal experience) is especially hard to replicate in a plant-based way is cheese, which has a distinctive and hard to copy taste and texture. To highlight just how big of a deal cheese is, it has been estimated that by 2030 the global vegan cheese market will triple in value to $7 billion. Rising to meet the demand and the challenge of replacing traditional cheese with a plant-based product is Ryan Pandya and his company Perfect Day. Pandya is a graduate chemical and biological engineering, so when in 2014 he was appalled at the existing alternatives to cheese, he decided to take action. He wanted to find what was the ‘magic ingredient’ in animal milk which couldn’t be found in plant-based alternatives. “Coming at it from a scientific perspective, you can’t help but realise that there’s no magic in cows or in milk,” he says. “It’s just chemistry and biology. So it stands to reason that if you had an almond milk that had whatever ingredients are in cow’s milk, and you make it act like cow’s milk, you would have a vegan version of milk.”

Pandya eventually identified the magic ingredients as casein and whey protein, which ensure that milk and milk-based products (for instance cheese, obviously) have a creamy texture and its distinctive tangy taste. Casein in particular (which has such a variable structure that there’s a lack of consensus over what it actually looks like) is the coagulating component that causes real cheese to stretch, melt and bubble so deliciously. 

Fermentation is the new alchemy (according to Perfect Day)

Taking a closer look at the technology behind Perfect Day’s plant-based products, at the heart of the process is fermentation – the company makes dairy proteins from plant sugars. Fermentation is, in a nutshell, a way microorganisms convert food into energy in an environment with no oxygen. The method of fermentation Perfect Day uses is known as precision fermentation, allowing the company to make super specific, highly pure ingredients. In the process, microflora convert sugar into whey and casein, dairy proteins that are useful to food makers for their top-notch nutrition, functionality, and versatility. 

Perfect Day utilises fermentation tanks (or bioreactors) to provide microflora with the adequate conditions to make lots of protein. The fermentation tank is filled with so-called ‘growth media’, a liquid containing plant-based sugars that microflora thrive on. Then, the growth media is inoculated with microflora. The environment inside the bioreactor is controlled by adjusting the temperature, pressure, pH, and stirring action to give the flora the perfect conditions to eat, grow, and multiply. As they consume sugar and increase exponentially, they also pump out copious amounts of our desired protein.  The fermentation process ends when the flora stop multiplying and the sugar is all used up. The protein is separated from the microflora and then dried, and results in an extreme pure protein powder ready for use by food makers. 

The fermentation process is actually the last step in Perfect Day’s process. Before microflora (called Trichoderma) undergoes fermentation, it receives the ‘recipe’ for making the specific whey protein which can be found in cow’s milk. The scientist at Perfect Day identified before starting the whole process which specific protein carries the properties we want, and utilised a widely accessible database (UniProt) of a cow’s sequenced genome to get the recipe for this specific protein. Then they “copied” the genetic information regarding the wanted protein and “pasted” it into the genome of microflora Trichoderma. The microflora understands the instructions and incorporates the DNA into its own genome. Then a process called homologous recombination takes place, which basically means that the flora searches the broken segments in its own DNA, and when it finds a gap with ends that match the connector’s ends, the flora fills the gap with that connector. Then the new gene is permanently incorporated into Trichoderma’s genome.

The process “visualized”

As you can see, the process of Perfect Day is fascinating and quite “simple” on paper. I especially appreciate the company giving such details of their process, so that such tech nerds as us can take a closer look at it.

Perfect Day is one of many companies currently innovating in the field of so-called vegan technology. Another fascinating example of a company in this field is Revo Foods, an Austrian startup who is producing 3D-printed smoked salmon, which was developed using new technology based on 3D food printing to recreate the texture and appearance of seafood. During the process, ingredients such as pea protein, algae extracts, and dietary fibers are combined to create a base that is high in protein, omega 3, and B12. Pretty cool, huh?

Revo’s printed smoked salmon

I look forward to following not only what next steps companies like Perfect Day and Revo Foods take (Perfect Foods recently announced they are working on replicating cow’s milk fat), but also to actually trying out the products they offer. I wonder if the products are as good as they look (take a look below at more of Perfect Day’s cream cheeses).

The texture looks fabulous + it has some interesting flavors (from the look of it)


Tagged , , , , ,

Sodinokibi: The Crown Prince of Ransomware

Reading Time: 2 minutes

Sodinokibi, also known as REvil (short for Ransomware Evil) is a ransomware threat group gaining more and more notoriety. Similar to some other ransomware families, REvil is what is called a Ransomware-as-a-Service (RaaS). Ransomware-as-a-Service is where a group of people maintain the code and another group, known as affiliates, spread the ransomware. Such RaaS models allow affiliates to distribute REvil ransomware in various ways, such as phishing campaigns or by uploading tools and scripts allowing them to execute the ransomware in the internal network of a victim. 

Sodinokibi hacks organizations by infecting them with a file blocking virus, which encrypts files after infection and discards a ransom request message. In the message, Sodinokibi explains that the victim needs to pay a ransom in bitcoins or else the files will be leaked. 

The group recently made headlines when they targeted Acer, a Taiwanese electronics company. On March 19th 2021, Acer was the subject of a hacker attack. The attackers, who are the REvil group, demanded the biggest known ransom to date in the history of cyber-attacks – $50 million. The hackers gave Acer until the 28th of March to pay the ransom, or all the stolen data will be released to the public. As of March 20th, Acer did not acknowledge that they were the victim of a security breach. 

Acer data leak on REvil ransomware site
Acer data leak on REvil ransomware site

The malware first surfaced in 2019, when it was discovered that in Oracle’s WebLogic server a serious flaw was noticed – a remote code execution bug which was remotely exploitable without authentication. This was an unusual attack from the side of the hackers, as it directly utilized the vulnerability of the server – and as researchers suggests, such attacks are typically executed with the involvement of user interactions, e.g., the act of opening an attachment to an email message or clicking on a malicious link. 

Sodinokibi has subsequently targeted organizations such as celebrity law firm Grubman Shire Meiselas & Sacks, foreigner currency exchange giant Travelex, Brown Forman Corp. (the owner of the Ritz Hotel in London) and as of recently Acer. 

REvil ransomware funcionalities

REvil is gaining momentum and notoriety, which is evident in the way the hacking group decided to target the tech giant Acer. This cyber security breach is worth following, as the repercussions for Acer may be substantial. This unfortunate event for Acer should also serve as a reminder to all internet users that cyber security attacks keep getting more refined and complex, and that substantial security measures should always be kept in place. 


Tagged , , , ,

Robots Taking Over Social Media

Reading Time: 3 minutes

If you’re active on social media and keep up to date with the fashion industry, you’ll probably have heard of internet sensation Lil Miquela (@lilmiquela on Instagram and Twitter) – an Instagram model, influencer and self-proclaimed activist. Lil Miquela first appeared on social media in 2016, quickly gaining thousands of followers and sparking interest all around the world, as people quickly started questioning whether Miquela was real. For 2 years Miquela’s identity was kept a secret, while the number of her followers skyrocketed and suspicions around her existence grew. Finally, in 2018, LA based company Brud came forward and officialy announced that Miquela was a digital art project created by them.

But what exactly is Brud? On their website they describe themselves as “transmedia studio that creates digital character driven story worlds, (…) story worlds that can create a more tolerant world by leveraging cultural understanding and technology”. Their team consists of artists, software specialist, storytellers and designers, who come together to create fictional characters and immerse them in real-life scenarios. While Lil Miquela is undoubtedly their most succesful project, Brud is also behing her fellow avatars BermudaisBae and Blawko22. They all interact together on social media, posting photos of each other having lunch, wearing the latest designer and streetwear clothes and meeting real-life celebrites. “The CGI celebrities” are causing quite the stir, and not only on the Internet. Miquela has currently over 1,5 mln followers on Instagram and the Instagram ‘model’ lives quite a busy life – she appeared in last year’s ad campaign for UGG, she took over Prada’s Instagram for Milan Fashion Week, she has been named by Path McGrath as her latest muse and the face of her make-up line, has released multiple songs on Spotify and other platforms, and even gave “interviews” to Vogue, the Guardian or the BBC. This year, Miquela is planning to create her own clothing brand and she is looking for help, you guessed it, on Instagram.

Lil Miquela pictured with Bermuda and Blawko – three robots created by Brud

But is not only Brud’s creations that are taking over social media – several other AI powered robots have been launched on social media and are becoming more and more in demand, when it comes to modeling campaigns (big names like Balmain or Fenty). It raises the question – are robots slowly taking over our world and are they integrating into our society? Miquela and other robots have been designed to ‘lead lives’ worthy of an Instagram post – their designers made a concious decision to make Miquela the latest ‘it girl’ and I believe their experiment turned out to be succesful – after all, for over 2 years Instagram users all over the world argued whether Miquela was just a mysterious influencer or a robot. People are actively observing Instagram robots, engaging with them in the comments, speculating with whom they will be photographed next and generally talking about them as if they were real.

Balmain’s campaign featuring robot models

I think this ‘experiment’ shows very acutely how social media platforms are often a facade, an illusion – after all, how much can we trust what we see on the Internet, when actual robots, who were designed by teams of specialists, are portrayed as if they were real? This phenomenon is much more than just another clever way of using AI and technology – it raises serious ethical questions, such as what is the role of AI in people’s lives, how people can eventually be  surpassed by robots and even do robots have any rights?

I’m very curious as to what you all think of the role of robots and AI in our lives, especially about artificially generated ‘internet personas’. Let me know in the comments.


World’s first foldable device

Reading Time: 2 minutes 


Chinese company Royale released world’s first foldable device – it’s called FlexPai and is a mix of a tablet and a smartphone. It has a 7.8 inch screen that you can bend (up to 180 degrees) and use it as a phone – in its “flat” form it looks like a tablet. It was officially launched on the 31st October 2018 in Beijing. The most impressive thing about it is that Royale beat other big companies like Samsung and created this device first – and just for around €1400 you can have it at home!



Apart from the obvious new breakthrough feature of the flexible screen, FlexPai is a mediocre device at best. Many reviewers claim that it’s very chunky, even though it may serve as a phone we cannot exactly put it in our back pockets. FlexPai system Water OS (custom made by Royale)  is also not the best – it supposedly keeps on crashing, opening up random apps, and even changes orientation during the folding process itself. It definitely needs some improvements!

Still, FlexPai is all about the screen – on the official site Royale claims “FlexPai’s screen is virtually unbreakable and extremely durable passing tests where the screen has been bent over 200 000 times”. While most of us probably wouldn’t purchase FlexPai due to its many flaws, it is undoubtedly a breakthrough and we can surely expect more companies coming out with foldable devices. One of the most notable ones is Samsung – it has been long speculated that the tech giant is going to release a foldable phone in 2019 (rumoured to be called Galaxy X or Galaxy F)




New Year, New Incredible Discovery

Reading Time: 2 minutes

2019 has so far been extremely successful when it comes to the discovery of space – on the 1st of January 2019, NASA’s New Horizons probe flew by Ultima Thule – now, the furthest place humankind has visited in history, as this lump of rock and ice is four billion miles away from Earth. What a way to start 2019, is it not?

Naturally, everyone is so excited about this achievement, that it had to be celebrated! That is why the mission team behind it threw a New Year’s Party, which was undoubtedly the most nerdy way to start 2019 ever.

Ultima Thule (MU69) looks a little bit like a snowman – in fact, this type of object is called a “contact binary”, which is basically two spheres touching. Ultima Thule is not big – in fact it measures only about 31 kilometers in length. It’s located in Kuiper Belt, a region that is beyond the orbit of major planets, although it is worth mentioning that it is home to Pluto.

Alas, NASA predicts that it will take up to 20 months to properly gather the data that has been recorded by New Horizons, as the new discovered body is so far away from Earth. Still, it will be worth it. It is widely believed that discoveries made on MU69 will help to explain how planets were formed four and a half billion years ago, both in our solar system and beyond.

It is not the end for New Horizons – scientists are looking to visit a third object in Kuiper’s Belt. And well, the sooner it happens the better, as somewhere in 2030s its generator (loads of hot plutonium) will no longer be able to hold the right temperature for New Horizons’ sensitive electronics. Still, even after its unfortunate end, New Horizons’ visit to Ultima Thule will have been recorded in history as an amazing achievement for humankind.





The next step in human evolution – designing babies?

Reading Time: 2 minutes

CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) is a method of editing genomes, such as human genomes, that can be done in a lab. But how does it work? In order to modify a genome, you damage the DNA and then instead of letting the natural repair processes do their work, we trick the DNA in order to make an edit that we want – we squeeze a specific piece of DNA we want to add in the centre and our genome is modified.

The most controversial aspect of CRISPR is the fact that we can use it to edit human genomes – in fact, some scientists have already done it, sparking outrage all around the globe. Some argue that CRISPR is the next ‘natural’ step in evolution – we can now design people not to die of heart attacks or not to suffer from Alzheimer’s, wouldn’t we want that? A healthy, perfectly happy society? Well, as every great tool, it can be misused. CRISPR allows scientist to design a genome – imagine picking your future child’s height, hair colour and the shape of their eyes. Can we really interrupt the natural process of reproduction and modify human genomes in a lab? Most of the concerns are ethical, after all how can we justify e.g. wiping out a whole species because we feel like it? But what a lot of people do not realise is that we cannot be sure whether using CRISPR will bring unwanted consequences in the future. Can we really risk it?

There have been reports that scientist in Shenzhen under He Jiankui have already started experimenting – supposedly, they modified human embryos to eliminate a gene called CCR5, which would make the offspring resistant to HIV, smallpox and cholera. A clinical document was found containing those claims, but we are not sure whether the experiment has produced a living birth. He Jiankui declined to comment when asked. But what we do know for sure is that the birth of the first clinically modified human will be a breakthrough point in not only the history of science, but in the history of the human species.




Have science fiction spaceships just become a reality?

Reading Time: 2 minutesRecently, MIT engineers have designed and built an electric airplane with no moving parts. The absence of turbines and propellers is possible thanks to “ionic wind”, which occurs when ions generated by the thruster on the aircraft collide with air molecules and propel the plane forward. This ground-breaking machine makes virtually no sound and is powered entirely by battery.

Source: MIT

The airplane made by MIT engineers has a 5 meter wingspan and weighs only 2.45 kilograms. Steven Barrett and his team have been working on this design since 2009. Barrett has said that the science fiction phenomenon, Star Trek, which he often watched during his childhood, greatly inspired him when developing the aircraft. Is it possible that the spaceships we have seen in Star Wars and Star Trek can soon become more than just geeks’ fantasies? While Barrett’s design has still got a long way to go, especially when compared to the USS Enterprise, Millennium Falcon or TIE Fighter, it is still a fantastic accomplishment.


In theory, Ionocrafts could be a game-changer in transportation of cargo and people. Still, one has to remember a revolution is still many years away. The aircraft was tested on a distance of about 60 metres, in a closed gym, under close supervision. Still, Barrett and his team have repeated the test 10 times and found that enough ions have been produced to sustain flight the entire time.

Source: Nature YouTube channel

What struck me the most is the fact that the concept of ionic wind propulsion has been around for quite some time –  it was mentioned as far back as 1709, when an English scientist Francis Hauksbee published Physico-Mechanical Experiments on Various Subjects, and in the 1920s by Thomas Townsend Brown, whose research and work have influenced further discoveries. But it was not until the 1960s, when Russian pilot Alexander Prokofiev de Seversky introduced his idea of Ionocraft , that it became to feel more realistic. Still, it was not until earlier this week, in November 2018, that an actual prototype was built and tested.





Tagged , , , , , , , ,