Possible but difficult: good work will speak for itself, but there are a lot of network effects which make it easier to publish from within academia. A trivial example might be that it's much easier and cheaper to find and read existing publications while at a university, and being familiar with the literature is extremely important to producing publishable work. Publishing within academia will also make it much easier to find collaborators (extremely important in some fields), find mentors, and learn about the general "style" of the field. Doing without this support structure would make it extremely difficult to publish even if all journals were double blind... which they are not.
You'll occasionally hear a PhD program being compared to an apprenticeship, and there's a lot of truth to that. Completing a PhD seems to be as much about learning how to orient yourself within scientific culture as it is about research or scholarship.
Perhaps one should add that the "apprenticeship" model would represent the ideal. If all goes well, it should then bloom into full research partnership towards the end and after. But again, that's the ideal.
As others have said already, your field matters a lot. If you try to publish in a "softer" field, say history or economics, politics and outer markers count for a lot. If you are in mathematics, there should be zero problem if you are good. Alternatively, if you are in a field where success can be measured concretely, say in engineering, you should also have less difficulties. Good luck, whether you go for PhD or not!
there are a lot of network effects which make it easier to publish from within academia
One of those effects is the intellectual sponsorship of people in legit academia, particularly professors whom you have an IRL relationship with and can recommend you to the editor. If not in school, I think this only works if you directly build on research conducted by contributors of the journal in question.
I would think in CS this would be easier, because professionalism is young in CS. I think it would be an order of magnitude more difficult in the humanities, where accreditation is far more entrenched.
My uni has a couple of public terminals, with full access to all their electronic journals. You can print articles and/or download to thumb-drive.
I don't know if other unis do this, but I would expect that it's part of the contract with journal publishers. Note: they don't publicise these terminals - just ask.
You'll occasionally hear a PhD program being compared to an apprenticeship, and there's a lot of truth to that. Completing a PhD seems to be as much about learning how to orient yourself within scientific culture as it is about research or scholarship.