How to Know Whether a Topic is Publishable in Computer Science

Question

Forgive my ignorance, but I am very new to the publishing process in Computer Science. I have some research that I have done involving analyzing data formats used in web-based image processing applications being presented in various IEEE papers. Some are conference papers and some actual IEEE transactions. They all seem to proposing the use of XML or in very few cases JSON as the data formats of choice.

I have read a few IEEE papers done on data traversal speeds involving comparisons made between XML and JSON or one of those two with that of CSVs, but not all three. And I have not found anything that give clear direction for one's reasoning for selecting a particular data format and why. Have I simply missed something? Tis likely, but I wanted to confirm with others first.

I have developed a small proof of concept here, and found some interesting results that I wrote about it here and wanted to see if this is worth trying to publish. Is this a good place to get insights, and share ideas that could be publishable? If so, does anyone have any insight into whether the material I have here is of interest to the scientific computing community? Or is this just a clever IT solution?

I am in the process of refining the paper to include tests done in compiled as well as an interpreted languages. Not to mention the reasoning behind selecting other data formats like CSV or TSV, which appear to much faster to traverse than JSON or XML, but less flexible. JSON appears to be the best choice on average, but more testing is required.

If this is not appropriate, let me know and I will remove this posting. Please also let me know if it is indeed appropriate and if you need more information or less and more of a synopsis.

Thank you in advance.

Answer 1

I'm not from this field, but I believe the following advice holds in any field.

First of all, I will suggest reading related literature; always a good idea. What do people in the field say about this topic? Is there an active community working on it? If not, you may find it hard to publish this work.

In my discipline it is quite acceptable to publish workshop papers. These do not "formally" count as a publication (they don't normally publish proceedings); however, your paper will get reviewed (not very deeply) and if accepted for presentation you will get even more feedback at the workshop itself. You can of course submit it to several workshops: this is perfectly acceptable practice, and I've seen it done with papers that contain only preliminary results.

What's more, this will tell you where your paper stands. If it gets rejected from a workshop on the topic, this is probably a good indication that it's not ready for publication yet.

Important note: it is absolutely fine to submit a paper to more than one workshop; it generally -not- acceptable to make concurrent submissions. That is, submitting to one conference before the outcome of a previous submission is decided. This is highly frowned upon in most communities I'm familiar with (tantamount to self-plagiarism), and will almost certainly make you look very bad. What's more, this will help you for two reasons:

1. It'll give you a chance to review the feedback you got from the reviewers, and possibly improve your paper.
2. Communities tend to be small; if you double-submit a paper, there's a non-zero probability that it'll be reviewed by the same people repeatedly (I have had the chance to review the same paper more than once, and I'm fairly junior in my field). Not only will you piss off anyone adversarial to your work by making them read it again, it will also indicate to them that you are not serious and are just "spamming".

So, submit your paper to one workshop, wait for the outcome, then submit to another.

Of course, the tradeoff here is that it'll delay the publication of the work. That said, you may want to acquaint yourself with the field and what's going on by going to a workshop or two, even if your paper is not accepted.

Another alternative is to simply send your working paper to a number of experts in the field and ask their opinion. My experience is that even prominent researchers will seriously answer such queries if the work is comprehensive enough.

Good luck!