The reason we can’t just wish away or “fix” complexity is that every solution—whether it’s a technology or methodology—redistributes complexity in some way. Solutions reorganize problems. When microservices emerged (a software architecture approach where an application or system is composed of many smaller parts), they seemingly solved many of the maintenance and development challenges posed by monolithic architectures (where the application is one single interlocking system). However, in doing so microservices placed new demands on engineering teams; they require greater maturity in terms of practices and processes. This is one of the reasons why we cautioned people against what we call “microservice envy” in a 2018 edition of the Technology Radar, with CTO Rebecca Parsons writing that microservices would never be recommended for adoption on Technology Radar because “not all organizations are microservices-ready.” We noticed there was a tendency to look to adopt microservices simply because it was fashionable.
This doesn’t mean the solution is poor or defective. It’s more that we need to recognize the solution is a tradeoff. At Thoughtworks, we’re fond of saying “it depends” when people ask questions about the value of a certain technology or approach. It’s about how it fits with your organization’s needs and, of course, your ability to manage its particular demands. This is an example of essential complexity in tech—it’s something that can’t be removed and which will persist however much you want to get to a level of simplicity you find comfortable.
In terms of microservices, we’ve noticed increasing caution about rushing to embrace this particular architectural approach. Some of our colleagues even suggested the term “monolith revivalists” to describe those turning away from microservices back to monolithic software architecture. While it’s unlikely that the software world is going to make a full return to monoliths, frameworks like Spring Modulith—a framework that helps developers structure code in such a way that it becomes easier to break apart a monolith into smaller microservices when needed—suggest that practitioners are becoming more keenly aware of managing the tradeoffs of different approaches to building and maintaining software.
Because technical solutions have a habit of reorganizing complexity, we need to carefully attend to how this complexity is managed. Failing to do so can have serious implications for the productivity and effectiveness of engineering teams. At Thoughtworks we have a number of concepts and approaches that we use to manage complexity. Sensible defaults, for instance, are starting points for a project or piece of work. They’re not things that we need to simply embrace as a rule, but instead practices and tools that we collectively recognize are effective for most projects. They give individuals and teams a baseline to make judgements about what might be done differently.
One of the benefits of sensible defaults is that they can guard you against the allure of novelty and hype. As interesting or exciting as a new technology might be, sensible defaults can anchor you in what matters to you. This isn’t to say that new technologies like generative AI shouldn’t be treated with enthusiasm and excitement—some of our teams have been experimenting with these tools and seen impressive results—but instead that adopting new tools needs to be done in a way that properly integrates with the way you work and what you want to achieve. Indeed, there are a wealth of approaches to GenAI, from high profile tools like ChatGPT to self-hosted LLMs. Using GenAI effectively is as much a question of knowing the right way to implement for you and your team as it is about technical expertise.
Interestingly, the tools that can help us manage complexity aren’t necessarily new. One thing that came up in the latest edition of Technology Radar was something called risk-based failure modeling, a process used to understand the impact, likelihood and ability of detecting the various ways that a system can fail. This has origins in failure modes and effects analysis (FMEA), a practice that dates back to the period following World War II, used in complex engineering projects in fields such as aerospace. This signals that there are some challenges that endure; while new solutions will always emerge to combat them, we should also be comfortable looking to the past for tools and techniques.
McKinsey’s argument that the productivity of development teams can be successfully measured caused a stir across the software engineering landscape. While having the right metrics in place is certainly important, prioritizing productivity in our thinking can cause more problems than it solves when it comes to complex systems and an ever-changing landscape of solutions. Technology Radar called this out with an edition with the theme, “How productive is measuring productivity?”This highlighted the importance of focusing on developer experience with the help of tools like DX DevEx 360.
Focusing on productivity in the way McKinsey suggests can cause us to mistakenly see coding as the “real” work of software engineering, overlooking things like architectural decisions, tests, security analysis, and performance monitoring. This is risky—organizations that adopt such a view will struggle to see tangible benefits from their digital projects. This is why the key challenge in software today is embracing complexity; not treating it as something to be minimized at all costs but a challenge that requires thoughtfulness in processes, practices, and governance. The key question is whether the industry realizes this.
This content was produced by Thoughtworks. It was not written by MIT Technology Review’s editorial staff.
These robots know when to ask for help
A new training model, dubbed “KnowNo,” aims to address this problem by teaching robots to ask for our help when orders are unclear. At the same time, it ensures they seek clarification only when necessary, minimizing needless back-and-forth. The result is a smart assistant that tries to make sure it understands what you want without bothering you too much.
Andy Zeng, a research scientist at Google DeepMind who helped develop the new technique, says that while robots can be powerful in many specific scenarios, they are often bad at generalized tasks that require common sense.
For example, when asked to bring you a Coke, the robot needs to first understand that it needs to go into the kitchen, look for the refrigerator, and open the fridge door. Conventionally, these smaller substeps had to be manually programmed, because otherwise the robot would not know that people usually keep their drinks in the kitchen.
That’s something large language models (LLMs) could help to fix, because they have a lot of common-sense knowledge baked in, says Zeng.
Now when the robot is asked to bring a Coke, an LLM, which has a generalized understanding of the world, can generate a step-by-step guide for the robot to follow.
The problem with LLMs, though, is that there’s no way to guarantee that their instructions are possible for the robot to execute. Maybe the person doesn’t have a refrigerator in the kitchen, or the fridge door handle is broken. In these situations, robots need to ask humans for help.
KnowNo makes that possible by combining large language models with statistical tools that quantify confidence levels.
When given an ambiguous instruction like “Put the bowl in the microwave,” KnowNo first generates multiple possible next actions using the language model. Then it creates a confidence score predicting the likelihood that each potential choice is the best one.
The Download: inside the first CRISPR treatment, and smarter robots
The news: A new robot training model, dubbed “KnowNo,” aims to teach robots to ask for our help when orders are unclear. At the same time, it ensures they seek clarification only when necessary, minimizing needless back-and-forth. The result is a smart assistant that tries to make sure it understands what you want without bothering you too much.
Why it matters: While robots can be powerful in many specific scenarios, they are often bad at generalized tasks that require common sense. That’s something large language models could help to fix, because they have a lot of common-sense knowledge baked in. Read the full story.
Medical microrobots that travel inside the body are (still) on their way
The human body is a labyrinth of vessels and tubing, full of barriers that are difficult to break through. That poses a serious hurdle for doctors. Illness is often caused by problems that are hard to visualize and difficult to access. But imagine if we could deploy armies of tiny robots into the body to do the job for us. They could break up hard-to-reach clots, deliver drugs to even the most inaccessible tumors, and even help guide embryos toward implantation.
We’ve been hearing about the use of tiny robots in medicine for years, maybe even decades. And they’re still not here. But experts are adamant that medical microbots are finally coming, and that they could be a game changer for a number of serious diseases. Read the full story.
5 things we didn’t put on our 2024 list of 10 Breakthrough Technologies
We haven’t always been right (RIP, Baxter), but we’ve often been early to spot important areas of progress (we put natural-language processing on our very first list in 2001; today this technology underpins large language models and generative AI tools like ChatGPT).
Every year, our reporters and editors nominate technologies that they think deserve a spot, and we spend weeks debating which ones should make the cut. Here are some of the technologies we didn’t pick this time—and why we’ve left them off, for now.
New drugs for Alzheimer’s disease
Alzmeiher’s patients have long lacked treatment options. Several new drugs have now been proved to slow cognitive decline, albeit modestly, by clearing out harmful plaques in the brain. In July, the FDA approved Leqembi by Eisai and Biogen, and Eli Lilly’s donanemab could soon be next. But the drugs come with serious side effects, including brain swelling and bleeding, which can be fatal in some cases. Plus, they’re hard to administer—patients receive doses via an IV and must receive regular MRIs to check for brain swelling. These drawbacks gave us pause.
Sustainable aviation fuel
Alternative jet fuels made from cooking oil, leftover animal fats, or agricultural waste could reduce emissions from flying. They have been in development for years, and scientists are making steady progress, with several recent demonstration flights. But production and use will need to ramp up significantly for these fuels to make a meaningful climate impact. While they do look promising, there wasn’t a key moment or “breakthrough” that merited a spot for sustainable aviation fuels on this year’s list.
One way to counteract global warming could be to release particles into the stratosphere that reflect the sun’s energy and cool the planet. That idea is highly controversial within the scientific community, but a few researchers and companies have begun exploring whether it’s possible by launching a series of small-scale high-flying tests. One such launch prompted Mexico to ban solar geoengineering experiments earlier this year. It’s not really clear where geoengineering will go from here or whether these early efforts will stall out. Amid that uncertainty, we decided to hold off for now.