I enjoyed this take on the question so much at the time that I spent quite a while playing with Twitter's advanced search function to find it...

The reason I'm a historian in a (nominally) professional sense is precisely because the boundaries of what 'history' actually is is so fluid, which means that you can enjoy an astonishing amount of intellectual freedom as a historian to pursue new ideas and interests. Most historians would limit it slightly further than the linked tweet thread in that we're concerned specifically with human history, but even that is potentially an incredibly broad category - not just the history of human politics and societies which probably makes up the bulk of histories you encounter at school, but the way we think, play, love, hate heal (or fail to heal) ourselves, try to understand the universe using science, philosophy or religion, and plenty more besides. The study of history is basically infinite in this sense - anything which has ever happened (real or imagined) is nominally fair game, no matter how distant or recent.

Obviously, there aren't infinite historians studying infinite histories (the monkeys with typewriters presumably take up too much of the university budget), so the question remains: how do historians choose what histories to explore? While every historian has their own answer to this question, we can point to several factors which influence the shape of the histories that we tell:

1. Significance. If you work as a historian, by training and by the cold logic of academic and/or commercial market forces, you are concerned about significance. When defending a PhD thesis, one of the most common first questions is 'So what?' - in other words, why should anyone else care about the research that you've done and presented? Significance can be defined in a lot of ways - it might be histories that affect today's society, they might be histories about which people care a great deal, they might be histories which have been misunderstood or misrepresented. But there needs to be some kind of analytical higher purpose - historians differentiate themselves from antiquarians in that they aren't looking to catalogue the past for it's own sake, but in order to better understand it and explain it. Being able to clearly and convincingly explain why finding out more about a seemingly obscure subject is vital - but also limits research topics to stuff that you can convince people is worthwhile. While these boundaries do change over time, they certainly shape the kind of history that gets researched and written.
2. Knowledge and training. To write the history of a particular subject, you need a certain set of skills to do it effectively. History teaches some of those skills, but is very focused on certain kinds of analysis over others. While historians tend to be very good at using and understanding human sources (writing, testimony, art, literature, that kind of thing), few of us are trained in, say, medicine or geology That doesn't mean we can't write about the history of medicine - doctors and hospitals leave behind plenty of the kinds of human sources we know how to use - it's less likely that we're going to be able to confidently write about the history of the human spleen as an organ. Equally, while we may be able to write about the way that people understand and relate to landscapes, we might be less well trained to discuss the merits of one particular rock over another. This also applies to the kinds of analysis we do in our work - relatively few of us are trained in advanced statistics, so history writing often tends to avoid analysis and arguments that rest too far on statistics (to be clear, this isn't entirely because we're all lazy or innumerate - a lot of historical questions either aren't suited for answering using statistics, and reliable statistical analysis using historical data becomes a very iffy proposition the further back you go).
3. Tradition. Today's historians have inherited a millennia-long tradition of how to do history, defined by the work of ancient scholars such as Herodotus, and refined by writers and scholars ever since. This has led to engrained expectations about what proper history looks like - the kinds of topics (generally politics and states), the kinds of subjects (often powerful men) and the kinds of approaches (using official records and archives). The thing is, this wasn't a neutral process - history has always been written with purpose, archives and documents tend to best preserve the perspective of those with power, what is accorded significance isn't decided democratically. Take the classic example of the nation state - history writing as an academic discipline was codified in the same period that nationalism started to become a dominant ideological framework, and nineteenth century historians tended to be nationalists, writing histories of a national past to justify the existence and legitimacy of the nation in the present. As a result, nations were for a long time the unthinking building block of history writing - we wrote histories in national-sized chunks, or the relations between them on an international stage, eventually arriving at the point where we kind of automatically reads nation-sized spaces back into history before they actually existed, and making it very difficult to imagine other ways for the world to be legitimately organised in the present (and states have any number of ways - from heritage institutions, to official archives, to commemorations and ceremonies, and university systems themselves - to reinforce this). While many historians do try to think about and challenge these kinds of constraints, they still implicitly or explicitly shape what we write about.