publications - Is publishing on arXiv early a good idea?

I have a lot of unfinished manuscripts in the field of communications engineering and computer science on my computer, and I'm wondering whether it is a good idea to publish them on arXiv before the articles are finished and submitted to a journal or a conference.

One benefit of publishing on arXiv is that it helps to establish that I'm the earliest inventor of something. But I suspect this is a double-edged sword. What happens if I post something to arXiv, somebody else builds on the work and adds more material and publishes it in a journal (perhaps citing my pre-print) before I have a chance of publishing my article in a journal? Does a typical journal accept papers that were novel when posted to arXiv but are not novel anymore? Let's assume that I have an intention of publishing the papers in a journal within a couple of years at most, not within a decade or two.

Another benefit of publishing in arXiv I can see is that many publishers allow one to update previously published material on arXiv with the accepted author manuscript immediately, but do not allow uploading the accepted author manuscript to arXiv if not published prior to submission to a journal within an embargo period. So, by uploading the material to arXiv early I could effectively remove the embargo period.

One potential drawback of pre-publication in arXiv is that theoretically some journals or conferences could consider the material non-novel which could eliminate the chances of publishing in such a journal or a conference. Are there any noteworthy journals that have a policy of not permitting material that has a preprint on arXiv?

I can see the following options:

1. Submit preprints to arXiv as early as possible, i.e. when the work is so complete that it might be interesting to others; when accepted, update with the accepted author manuscript.


2. Submit preprints to arXiv a couple of weeks or a month before submitting the paper to a journal or a conference; when accepted, update with the accepted author manuscript. This would eliminate the risk of somebody building a better work on top of my work and quickly publishing it.


3. Submit preprint to arXiv after publication in a journal. This might in some cases mean I have to wait for an embargo period of a year or so.

Which of these options is the best?