This is a contributed article by Peter Zaitsev, co-founder and CEO of Percona.
Open source is seen as a shining example of how to develop software, particularly for enterprises. The market for selling services and products based on open source code is predicted to be worth $66.84 billion by 2026. However, this success has come at a price and cracks are beginning to appear. How these issues play out will dictate the future of open source in 2021 and beyond.
Prediction #1: There will be more fragmentation around open source and cloud, leading to a battle over value
Our first prediction starts by acknowledging that two camps are emerging in the world of open source as a result of the growth of cloud computing services that provide and use open source solutions. One camp sees the actions of AWS and other cloud providers as a threat to the continuation of open source software development, and the second embraces this competition as a natural progression of open source.
Research and development around open source services relies on a strong and healthy community. However, high-profile open source projects have become popular with dominant cloud providers, as they can repackage these projects as cloud-based services for their customers. This has enabled cloud providers to pocket significant profits. At the same time, it has cut the revenue of the open source companies that actively developed the project, with little given back in the way of code.
These two distinct camps reflect a battle for the ‘soul’ of open source. I predict that this will ultimately become a battle over value. What is best for the customer may or may not be convenient, or it may not give an open source company as much business as it would like. If you are an open source company, in the classic sense, there will be users and organisations that eat at your table for free and who use your software without assuming any development costs.
Over the long-term, I predict that companies will want to see the open source projects they use well supported. This will mean looking at the health of communities and where contributions are coming from. For communities that rely too heavily on single contributors or companies, the signs are not good. Similarly, cloud service providers will need to contribute more back, so that they can continue to benefit.
Prediction #2 - DBaaS will continue to grow and open source will disrupt
My second prediction leads neatly onto where that extra value will come from, as we see the continued growth of Database as a Service (DBaaS). DBaaS enables users to set up, deploy, and maintain databases through a UI or API that abstracts away much of the database implementation complexity. However, DBaaS is generally proprietary - even services based on open source are tied to specific providers or cloud services platforms. If you ever want to move, you may face a big challenge.
This market is ripe for open source companies to provide alternatives if they can meet the demand for ease of use, as well as not locking users in. If open source can provide even 80% of the convenience that developers like, it will acquire a significant portion of the market.
Next year, we’ll see new open source solutions beginning to emerge from a number of vendors, but this move into the DBaaS space is likely to take longer than 12 months. Over this time period, we’ll see signs of dissatisfaction with existing DBaaS and companies will start to look for migration routes.
Prediction #3 - Innovations coming from the open source database world
The move to digital-first and cloud native companies has accelerated in the wake of COVID-19, and we should expect to see more innovations around the management of data.
This will include databases built natively as a distributed system from the ground up, for example CockroachDB or Yugabite. Until recently, we hadn’t seen much innovation in analytically focused databases, and specifically column stores, but that’s rapidly changing. We have ClickHouse emerging along with QuestDB and DuckDB, all of which originated in the open source space.
We should expect to see significant growth for those technologies next year as companies demand (often in real time) faster ways to analyse their massive lakes of data. They aren’t direct competitors to the much hyped Snowflake, as Snowflake offers a far broader solution, but as database engines they will have a strong impact. Already, we are hearing from organisations that are considering ClickHouse, for example, as a way to achieve much better performance than something like Amazon Redshift, which is a proprietary competitor.
Prediction #4 - More standardisation of licences
From next year we should see greater discussion and a push towards a clearer understanding of ‘Not Open Source, Source Available’ licensing. The aim will be to foster greater clarity for users. Already we see the Polyform project, which is similar to Creative Commons and supplies a set of clearly defined licences that any user can understand.
The proliferation of so many Source Available licences has created a complex licensing landscape. This forces developers to spend time researching and understanding what’s involved rather than developing. It creates unnecessary friction and it leads to procurement issues.
A standardisation of licences that are ‘source available’ will give developers and vendors more protection. However, these are not true open source projects and anyone using them should bear this in mind. One of the strengths of open source is that we have a relatively small number of licences, which are reasonably well understood by developers. GPL, BSD, MIT and Apache cover over 90% of released software.
The open source movement was born out of a more pragmatic and business-friendly approach to promoting open development processes. The future of open source now faces certain lines in the sand as it struggles with ethical questions and unexpected business cases. How these battles for ‘the soul of open software’ play out over the next few years is likely to dictate the future of open source for at least the next decade.