Why you should get rid of folders. Right now.

Most current project, product, and knowledge management systems are built on a single hierarchy of folders. A long time ago, someone suggested this option of organizing information, and it is still relevant today. It seems to us that the only and correct way to structure information is to put everything into folders. I am ready to argue with this statement and in this post, I will try to explain why folders as a tool for structuring and building ontologies are an unnecessary element.

To begin with, folders as a structuring tool are useful only when you don’t have many categories and data. You know clearly what should be stored and where, and you don’t have any questions about where to move the document you just created. The size of ontologies in this case is quite narrow. Every member of your team understands the logic of building this ontology and doesn’t make mistakes when moving and searching for documents. In this case, indeed, folders are a great tool that gets the job done.

But what happens when there are more of them? Let’s look at all this with a concrete example.

Example

But if there is more data, the first problems start to appear. It is necessary to understand that the more data — the more rapidly the list of ontologies and folders in the folders will grow. This is logical since the main task of structuring information is to make it easy for anyone to find it later. But the number of folders keeps growing, ontologies become less transparent to everyone except the one who develops them. Documents start to appear in all the wrong places and . chaos ensues. The team starts to expand, and so does the amount of knowledge. Ontologies continue to be developed to simplify the work, but in fact, the opposite happens. And any complication entails not following the rules. People stop thinking about where and what to move. Documents start to get lost and end up in places you couldn’t even think of. New people come in, and the adaptation process of which takes many times longer due to the lack of a system. Current employees lose contact with other departments, and communication through knowledge becomes impossible. The business begins to lose efficiency.

People see this problem and are going to discuss it and take action. And what do you think they solve? Of course, they need to rework the current structure and make it simpler.

The structure has been simplified. The amount of knowledge in the company continues to grow, and of course, the number of new ontologies does too. And everything starts all over again. The endless cycle continues until the company just stops worrying about it, putting the problem at the lowest priority.

New solution

The new solution is to get rid of hierarchies entirely as they now exist. The necessary replacement is much more effective than the tags.

Of course, you can argue that tags are the same thing, but with a different shell. To this I have two strong arguments:

1. Tags are easy to use. Unlike folders — you can put multiple tags on documents at once without duplicating them. If a document isn’t under the right tag, it’s easy to remove and change. Tags are incredibly flexible and can be used with whole documents as well as individual parts of documents.
2. Tags are scalable. In just a few years, we’ll have systems that put tags automatically based on your current tags and document description. It’s not something super complicated, it’s just an adaptation of OpenAI’s GPT-3 technology. The guys at Mem.ai will be releasing a beta version using this technology soon.

Scalability and ease of use are what make tags globally different from folders.

No matter how many tags you have, it does not affect anything, they are not in front of you all the time, unlike folders. If there’s an empty folder, it’s an indicator that something is going wrong. And if the tag is empty, it’s not a big deal.

What’s the solution?

I think that at this point — folders aren’t going anywhere. People are so used to them. And when you implement a new solution, you will encounter a lot of negativity. You have to do it gradually. Get rid of unnecessary hierarchies by replacing them with tags. The evolutionary approach (when you gradually introduce changes) is much better than the revolutionary approach (when you change everything at once).

Conclusion

I believe that current hierarchy development systems are incredibly outdated and in need of modernization. The amount of knowledge is growing daily. And the most important task is to learn how to handle it properly. So that everyone knows what, where it is and what, where to move it to. This is incredibly important. Let’s not ignore new approaches, but try to test them.