One way to rate science fiction is by how hard or soft it is. Harder sci-fi holds more to the rules of science as we know them, and softer sci-fi does the opposite. This isn't necessarily a rating on how good the piece is, (though people may tend to have a preference towards one side) but is instead a way to compare or categorize it.
For me, I use the following categorization of objects and technology when thinking about how hard the science is in a particular piece, and listed from hardest to softest.
Current Tech:This is anything that we have currently built and know how it functions. Sometimes hard to use too much of, depending on how current the piece is. For SEAC, this is what you start out with. An example for this is the chemical rockets we currently use for getting to space.
Future tech is anything that we've got plans for and have a reasonable idea of how to build, but haven't actually built. As with most technology, the actual thing is likely to have some quirks that weren't thought of with the initial plans. The nature of what we have now though means that most tech that I see in this brand of sci-fi comes from this section, rather than current tech. NASA isn't exactly eager to outfit their shuttles with weapons, after all. This is what you'll tend to see in SEAC. For an example look at one of the myriad number of nuclear rocket designs made during the cold war.
I name anything where we don't have a good idea of how to build, but do have an idea of the limits and possibilities of, as theoretical tech. Here we stretch things a bit, and it's something that I want to stay away from if possible. Because the limitations aren't really known, there's a lot of leeway in their capabilities, which means there's plenty of temptation to stretch the limits that I think will keep the game good. I'll certainly add some, but I want to keep these in check as much as possible. The fusion drive is a good example, as it's a possibility for a very strong drive if I'm willing to handwave away all the limitations that currently make it a Very Bad Idea.
Anything that doesn't have a basis in known science falls here. These are the fun things that can often make each piece of fiction unique, but come with all sorts of dangers. For me, in designing a new system like this I need to make sure that there's not too many unintended consequences and that it fits without taking the fun out of something else. A common example that you'll see is anything that enables faster than light (FTL) travel.
For magic in SEAC I'm going to stick as much as I can to only including FTL travel and energy shields. FTL because, as established, the universe is just too big for slowboating it around if I want to have a good colonial empire. Shielding is there to keep the setting to a more maritime model. Without it, I'd need some sort of super-armor for the ships if I want them to withstand the damage that the weaponry of the time could likely throw at them. It's also a handy tool for dealing with radiation nicely without needing to worry too much about the excessive weight required.
On a side note, I saw this video today and thought I'd share: