I'm new to Twine and started a complex story in 2.0. However, after 25 passages, I realized that I could not embed an image in this version. Will I need to chalk this up to experience and re-input to 1.4? I tried publishing to file and importing to 1.4 but that didn't work. Unfortunately, I don't know any code either...
The How do I open .html files in Twine 1 and .tws files in Twine 2 thread covers the process of moving a project between the two different versions of Twine.
I am not sure how Twine 1 handles multiple references of the same image but with the manual method you'd probably end up with multiple monster lines like above. And, for sure, those inline images make the "source code" look absolutely messy ...
So it depends ... if you have a few images only the manual encoding looks like a good idea since it allows you to keep Twine 2 and Harlow and you can hope for Twine 2 to implement the same image capabilities that Twine 1 has. If you have a lot of images I'd really fall back to using Twine 1 and rewrite the code where needed.
Using embedded base64 encoded images will increase the size of your story project, which means you will hit the local storage limit faster.
If that is the case then going to Twine 1 will not help at all since Twine 1 embeds the images automatically (if you choose to do so). I've taken a look at a grafic heavy Twine 1 story I did and if you look at the page's source code you'll see quite a lot of those data:image/png;base64 lines that go to the right for miles.
Wasn't aware that there was such a low memory limit so I might reconsider my own approaches
If you are using the installable release of Twine 2 then it should be able to cope with large MB's of embedded images, it is only on-line release that is restricted by the web-browser settings.