I've struggled for a long time with the deletionism/completionism divide. I was initially strongly completionist for the reasons many others have said: this could be the repository of all human knowledge, no matter how trivial, and why would we give up that dream? To save on hard disk space?
But my viewpoint these days is that Wikipedia's size has the same kind of problems that code size does. More code isn't bad in itself. After all, if that code doesn't interact with your code, you can just ignore it. Something like a zillion-line OO project with tightly defined interfaces gives you that property.
However, there is an inevitable maintenance burden that comes with the size of your project. If you want to change code conventions, or update some library that is used throughout the codebase, or even just try to keep the project to a certain standard, those actions are all difficult in proportion to the number of lines of code.
In Wikipedia's case, they don't even have access to the same kinds of push-around-huge-mountains-of-code tools that developers can use to manage this problem. They have tools, sure, but a fundamental part of Wikipedia's model is that it is unstructured (or at best semi-structured) data, just one big text field, and so any automated transforms are necessarily limited.
So Wikipedia makes up for its fuzzy data model by just throwing people at the problem. For it to be the sum of all human knowledge means it needs proportionally many editors to maintain that knowledge. If there ever could be enough editors to do that, and if there could be a structure that would allow them to organise themselves, I'm not sure Wikipedia is it.
The problem is that Wikipedia can't regulate the number of volunteer editors in the project. If the encyclopedia gets too big for the editors to manage, large chunks of it will just atrophy and there'll be nothing they can do about it. Worse still, editors will leave because they are unable to handle the burden which only makes the problem worse.
So I'm not sure that completionism is actually feasible, at least not with the structure as it is now. I still dream of that repository of all knowledge, no matter how trivial, but I just don't know how we get there.
But my viewpoint these days is that Wikipedia's size has the same kind of problems that code size does. More code isn't bad in itself. After all, if that code doesn't interact with your code, you can just ignore it. Something like a zillion-line OO project with tightly defined interfaces gives you that property.
However, there is an inevitable maintenance burden that comes with the size of your project. If you want to change code conventions, or update some library that is used throughout the codebase, or even just try to keep the project to a certain standard, those actions are all difficult in proportion to the number of lines of code.
In Wikipedia's case, they don't even have access to the same kinds of push-around-huge-mountains-of-code tools that developers can use to manage this problem. They have tools, sure, but a fundamental part of Wikipedia's model is that it is unstructured (or at best semi-structured) data, just one big text field, and so any automated transforms are necessarily limited.
So Wikipedia makes up for its fuzzy data model by just throwing people at the problem. For it to be the sum of all human knowledge means it needs proportionally many editors to maintain that knowledge. If there ever could be enough editors to do that, and if there could be a structure that would allow them to organise themselves, I'm not sure Wikipedia is it.
The problem is that Wikipedia can't regulate the number of volunteer editors in the project. If the encyclopedia gets too big for the editors to manage, large chunks of it will just atrophy and there'll be nothing they can do about it. Worse still, editors will leave because they are unable to handle the burden which only makes the problem worse.
So I'm not sure that completionism is actually feasible, at least not with the structure as it is now. I still dream of that repository of all knowledge, no matter how trivial, but I just don't know how we get there.