Time for Thingamy Thinking?

In the ever-evolving landscape of software development, the quest for increased productivity and efficiency remains a constant driving force. While software has permeated every aspect of business operations, its true potential in enhancing productivity has often been stymied by inherent design limitations. This is particularly evident in recruitment software, where the need for seamless workflow integration is paramount. Drawing inspiration from Sig Rinde's pioneering ideas with Thingamy, it's time to reimagine software as a dynamic workflow enabler rather than a static data repository.

The Story of Sig Rinde and Thingamy

Over a decade ago I was fortunate to meet and become friends with Sig Rinde, the visionary behind Thingamy, who championed the concept of software as a workflow-centric tool. He argued that traditional software models were predominantly database retrieval systems that required users to navigate and manipulate data to accomplish tasks. This approach placed the burden on users to understand complex systems, often leading to inefficiencies and bottlenecks. At the time, my understanding of software was limited, and I would often play the devil's advocate in discussions with Sig. Through these debates, we refined his arguments, ultimately strengthening our ability to communicate the value of what Thingamy could achieve.

Thingamy proposed a paradigm shift: software should model real-world workflows, providing users with the right information at the right time to advance tasks seamlessly. By mirroring actual business processes, software could become an intuitive extension of human activity, minimizing friction and maximizing productivity in the same way as a Henry Ford production line.

The Shortcomings of Software

Most legacy software systems failed to deliver on the promise of increased productivity because they did not embody this workflow-centric philosophy. Instead of guiding users through processes, they often presented a myriad of buttons and menus that overwhelmed users with options but offered little guidance on what to do next. This complexity not only slowed down users but also introduced opportunities for errors and omissions.?

My friend Mark Barlow built an exceptional business with AppLearn by addressing these shortcomings. He introduced an innovative software layer designed to guide users, simplifying their experience and ensuring they could navigate complex systems more effectively.

The Simple Analogy of a Word Processor

Consider the example of word processing software like Microsoft Word. Despite its powerful features, MS Word or indeed Google Docs presents users with an extensive array of menus and buttons—many of which are non-contextual and irrelevant to the immediate task of typing the next word. This abundance of options distracts users from their primary objective which is usually to just type the next word, creating unnecessary friction in what should be a straightforward process.

This analogy extends to most enterprise software, where the interface is cluttered with features that are rarely used while providing minimal guidance on the next steps. Users are left to navigate complex interfaces without contextual assistance, which hinders productivity and leads to a steep learning curve.

The Need for AI Agents and Human-in-the-Loop Thinking

The advent of artificial intelligence (AI) and machine learning has ushered in a new era where software can do more than store and retrieve data—it can understand, learn, and act. AI agents can now perform complex tasks, analyze vast datasets, and even make decisions based on predefined parameters and learned patterns.

AI's Expanded Capabilities

AI has reached a point where it can handle a significant portion of tasks that were once exclusively within the human domain.?

For example in our current world of recruitment with GLYDE? - Recruiting AI, Data & Insights , AI can:

• Screen Resumes: Quickly parse and evaluate resumes against job requirements.
• Match Candidates: Identify potential candidates from databases or external sources.
• Schedule Interviews: Coordinate availability and set up meetings automatically.
• Engage Candidates: Send personalized communications to keep candidates informed and engaged.

With AI capable of handling these tasks efficiently, the role of human recruiters shifts to areas where their expertise has the most impact.

The Time Is Right for Human-in-the-Loop Workflow Software

In light of AI's rapidly expanding capabilities, perhaps what we really need for the majority of computer-related tasks is not so much augmented humans but humans in the loop for a discrete set of tasks that they can do better than software?

In this model, AI handles the bulk of routine and even complex tasks, while humans are involved only where their unique abilities—such as nuanced judgment, creativity, or ethical considerations—are essential. This approach ensures that human expertise is applied precisely where it adds the most value, without burdening them with tasks that AI can handle more efficiently.

If humans focus on the tasks they excel at, the collaboration between AI and human insight becomes more effective. This division of labor allows organizations to maximize productivity while maintaining the quality that only human judgment can provide.

This approach aligns perfectly with Sig Rinde's vision of software as a facilitator of workflows rather than merely a repository of data. In Sig’s world with the widespread use of virtual Kanban type boards, workflows become transparent, enabling teams to visualize tasks and progress clearly. Individuals who are not completing their work would be quickly identified by their lack of activity or by becoming noticeable bottlenecks in the process needing to be unblocked without the need for a Theory of Constraints type analysis.

Software Needs to Guide Users to High-Impact Tasks

Back to the world of recruitment, here’s how this should happen.

1. AI Screens and Presents Candidates -? The software provides a shortlist of candidates who meet the job criteria.
2. Recruiter Reviews and Acts - The recruiter evaluates the candidates, submits suitable ones, and dispositions others.
3. Feedback Loop - If too many candidates are being dispositioned, the software suggests refining the job description or adjusting AI screening parameters which the recruiter does right away.
4. Continuous Improvement - The system learns from these interactions to improve future candidate matching.

This workflow ensures that human attention is focused where it adds the most value, while AI handles the heavy lifting in the background. Notably, it eliminates the need for recruiters to suddenly scramble and waste time contacting ill-fitting candidates to satisfy client demands after a mismatch is identified hours after the initial screening.

By simply refining the job requirements or adjusting the AI screening parameters, the system will automatically deliver more targeted applicants for review in the next cycle, streamlining the process as described in Step 2 above.

Of course, this assumes that the human in the loop has effectively performed their role in a timely manner…but again this is where visibility will help.

The Impact on Autonomy and Job Satisfaction

Although my first self-driving car experience last month came about three years later than I had predicted while working on a transportation project a decade ago, I believe that for better or worse the anticipated shift towards a “factory like” human-in-the-loop model may not take as long to materialize this time around once software starts to facilitate it.

Are Knowledge Workers About to Become the New Production Line Workers?

While this AI-driven, workflow-centric approach promises significant productivity gains, it also raises important questions about autonomy and job satisfaction. As software takes over more tasks, human involvement becomes more focused and potentially narrower—reminiscent of a production line.?

With structured workflows guiding every step, there's less room for personal discretion in how tasks are performed, and the emphasis on standardized processes may limit creative approaches to problem-solving.

I believe the way humans deliver "knowledge work" is about to change significantly, and if Sig were still here, he would undoubtedly have something to say about it and I know I would have enjoyed arguing the counterpoint.