Very much depends on what kind of science we're thinking about.
Sociology and psychology have become much more interesting now that we've ditched the obnoxious tradition of trying to subsume every finding under a theory of everything that ends up explaining nothing – think of Talcott Parsons. The same probably goes for computer science: an interesting algorithm is an interesting algorithm, it doesn't need any sort of wider philosophy to justify itself. And biology. And so on. Often, facts are valuable in and of themselves. It's surprising how little we still know about the world.
But isn't it a problem that you just generate trivial observations without being able to form conclusions? That you missing the perspective that makes the gathered data valuable?
Historically computer science was often bundled with philosophy, because the field of CS wasn't taken seriously enough to warrant a standalone field of study. Philosophy was probably a popular choice due to the overlap in reasoning and principles, which were emphasized by classical philosophers.
Being able to reason about the properties and especially correctness of algorithms sounds useful. Don't worry, I skip that part in my daily work often enough.
It's not a problem to publish novel observations without being able to plug it into a theory right away.
Sometimes, novel observations are important enough on their own without being able to generate conclusions right off the bat. Penicillin, in particular, was published based on the observation of inhibition of bacterial growth in a petri dish which had some fungus in it. It took later scientists to isolate the chemical agent which caused this antibiotic effect and form a theory about it.
In fact, a paper on an observation is probably more important by itself than an observation plus some mediocre theory. The theory paper can come later.
I remember Robert Heinlein had a vision of a future where houses were all built in factories. There was some essay where he ranted about how absurd it was that the average house was custom built like a Rolls Royce. Of course, we do have manufactured and modular housing today, but it's still normal to build structures much the same way as ever, and it's nearly 2020.
I think eliminating handmade software ultimately is a proposition balanced between a tautology and an oxymoron; everything new that is created is "by hand" and everything else is existing software. Anything that can be reused is. Saying we need to reuse more is not right or wrong, it's just empty.
Cool comic. My personal view on the matter is that the advent of globalisation and the internet (and perhaps printed matter before that to a lesser extent) naturally leads to a handful of superstars rather than a basket full of mentors.
You are not just flooded with research, but you are inevitably guided by reported research. Why would you read an obscure paper if you don't already know it's worth it and it's a good paper?
I would as a suggestion say that be happy to do self-discovery of known results as a way to build your core skills, without an obsession with making a contribution and do what you can to keep your grant. If you do discover something new, study it to some level of depth so that your paper is actually accessible to the masses rather than publishing a new result that becomes obscure. (Which is the worst case scenario, isn't it?)
EDIT: In terms of data volume vs. theory, the same holds for you, personally. The more time you spend reading disparate things, the less focused you become on knowing less things but in depth.
Sociology and psychology have become much more interesting now that we've ditched the obnoxious tradition of trying to subsume every finding under a theory of everything that ends up explaining nothing – think of Talcott Parsons. The same probably goes for computer science: an interesting algorithm is an interesting algorithm, it doesn't need any sort of wider philosophy to justify itself. And biology. And so on. Often, facts are valuable in and of themselves. It's surprising how little we still know about the world.