In 2020, the Covid-19 pandemic has demonstrated just how fragile we are. The world has never experienced a global pandemic like this one in the last 100 years. Ever since the Spanish flu, we haven’t been faced with a disease that was able to paralyze the entire planet. Many facets of human lives were completely transformed by the pandemic and we had to find new solutions on the fly in the face of strict measures imposed by governments all over the planet.
In some areas, that proved to be a blessing in disguise, as we discovered that we are able to do things in a much easier way. For instance, the percentage of jobs that were performed remotely skyrocketed. One area that suffered the greatest changes in software development. Faced with the crises, many solutions had to be questioned and rethought. That left us with a rather interesting software development perspective for 2021.
Google has always dabbled in deep learning with their TensorFlow library, but the game was completely changed when Facebook released PyTorch, completely re-envisioning both deep learning and neural networks. PyTorch was created out of necessity, as Facebook needed a tool to handle images, text, and video in real-time. PyTorch does that, and plenty more. Google retaliated with TensorFlow 2.0 in 2019 and, according to some studies, it is the most popular deep learning library.
Some sources say that 2021 may be the year we see Facebook’s answer and heir to PyTorch. Regardless of that, deep learning will be a mainstream field in the upcoming years.
Artificial Intelligence (AI)
Artificial intelligence, as a term, has been around for decades. Many people, when they hear it, first think about Skynet and Arnold Schwarzenegger coming back from the future to save mankind. Many experts agree that projects like ethical AI can go a long way in preventing anything similar. In fact, some forms of AI are already in use.
In the gambling industry, for instance, AI has been powering various pieces of casino software for years and doing it quite successfully. The latest trend in AI development is explainable AI. That is an AI that has to explain every decision it has made. That should provide researchers with plenty of data to understand it and further improve its decision-making process. EU has already passed legislation that regulates this area, and many other governments will soon follow suit.
Cloud computing offers so much flexibility and convenience over traditional solutions that it is not surprising that it will be a major trend in 2021. Depending on the needs of users, it will branch out into two diametrically opposed directions, centralized and decentralized networks. Since this is not a “one size fits all” solution, each user will choose the approach and degree of it according to their needs.
According to Forrester, global cloud infrastructure will increase by 35% in 2021 alone. In terms of money, the industry will be worth some $120 billion, with companies like Amazon, Alibaba, and Microsoft taking a huge chunk of that money. The Great Cloud Migration is in full swing and cloud engineers are in huge demand right now. For users that demand low latency and high bandwidth,
As a technology, blockchain offers many potential benefits to its users. However, the implementation is drastically behind the potential. We could see that trend reversed in 2021. Ironically speaking, blockchain’s greatest advantage is also the thing that is holding it in place – anonymity. Big players are reluctant to invest in something they can’t fully control. The very idea of anonymous users is appalling to most CEOs since they equate anonymity with lack of control.
Fortunately, there are plenty of independent companies that are ready to carry the torch and hopefully, that will be enough to push blockchain to mainstream technology and mass adoption.
Big Data Democratization
Modern data platforms are incredibly complex, including components like a data lake, data sources, ETL pipeline, data warehouse, and many more. This has allowed a relatively small number of players to dominate the market, effectively cutting off all potential competitors. Snowflake changed all of that. Their revolutionary approach with one platform and centralized data source has been a massive success with customers.
It is not a perfect solution for everybody, but the majority of users will be able to find some use for it. The democratization of data is something we sorely need and Snowflake has made a big step towards it.
One thing that keeps it from losing its place is its proverbial backward compatibility and regular updates. Every six months there is a new version, allowing it to stay with the times. One program that may interfere in the fight is C++. C++20 was just released and developers promise that it will change the way we program completely.
By the end of 2021, there will be 3.8 billion smartphone users in the world. That is roughly half of the planet’s population. All those phones need apps to function properly and the app market is skyrocketing. However, not all apps are created equal. Among various types (native apps, cross-platform apps, hybrid apps, and cloud-based apps) native apps will remain dominant. Their dedication to a single platform makes them flexible and easy to use, something end-users value highly. In fact, short of massive government intervention, this trend will remain dominant not only in 2021 but in the entire decade.