the hardware enabling comes first. what you do with it comes second. there are inventions and then there are what you would call obvious ideas.
most visible advancements come from the latter while it's inventions and things that enabled those inventions which enable them.
for example, if there had not been apple, if there had not been nokia, you would still have that mobile phone - what these companies have mostly done is just applying inventions that enabled their devices to be made, like the transistor and so forth.
information theory therefore comes second, it's about what you do with them - but it is something that was born out of need for it due to other inventions already existing.
as obviously you are working mostly in applications of electric devices and what you can do with them, information theory is more important for you, because you aren't really trying to design a smaller chip and break chip manufacturing minimum size limits, in which case you would find the physics research to have been way more valuable for you - and without those devices that are enabled by science you wouldn't be using them for engineering solutions using them.
shannon seems mostly having been interested in the logic side of things: If you have a machine that does this and this what will be the logical conclusion that you can do with it - people like this are far more likely to pop up rather than the kind of people who come up with the new device itself - for example an internal combustion engine meant quite a lot of changes to the world, once you had that it didn't take quite as much imagination to use it for something as it did to actually come up with a design for a working motor - but once you had the motor it would be obvious to use it to generate electricity, to drive cars, to drive boats and so forth.
incidentally this difference is how you can smell bullshit sellers a million miles away: if someone is selling like a perpetual motion machine that makes water or a car that runs on water, you need to ask yourself: why isn't he applying it to such and such.
the thorium car from a while back for example: who fucking cares if it can run a car forever when, if they had a working model for the power source, they would be using it on a car as the last thing on earth making the whole design and articles about it utterly stupid. Having the power generating unit would change the whole world and the cars would be the last thing to change, so why try to sell the idea as a car engine as the first thing?