Getting Started with Journal article Writing

Its year 2018, I just stepped into third year of my Ph.D. and pressure of publishing a paper started. Although, i had works at my disposal, how to start writiing, what to write first, and planning the structure was way beyond my understanding. Being in academic lab comes with support for lab members but a lot of time, these skills one have to develop eventually but trial and error-no way out. The struggle that i went through in mmy writting journal articles (specially technical) inspiured that why not to share it with others. This article is a collection of my learnings and experiences in technical writing. The first paper i wrote, i still don't read but i think it was most important paper in my writting journey. After struggle of 1 month I had 1st draft of paper and reached out to my mentor to review, I can barely find any line in the manuscript which is not marked as red. It was not because the content or grammar is bad but more because of it was missing the information that is required to be delivered in effective and structured manner. After another couple of month of struggle, it passed the review of my mentor and is submitted for publication. Over the time, i have several journal papers now and few things that i observed that i think are important to be shared with others. I hope it helps you in your technical writing journey.

Journal article writting sometime i feel is more dificult then writting a fiction and specially when you have a lot of information to deliver whereas in fiction you have to create these informations. In matter of fact, journal article writting is the process of documenting facts and information in manner that is easy to understand and follow. It is not about writting a story but more about delivering the information in structured manner. Here is my take on what to keep in mind when starting to write

1. Establish that you have a peper (may be in mind) and not just the result a. Write the conclusion sentence of the abstract before starting anything. If you can't write it, its just an study which is not finished and need more work b. Name what contribution is there with this article- new method, new dataset, new analysis, new insight, new application, new tool, new framework, new theory etc. 2. Freeze before you draft a. Set the results-freeze date. writing while experiments are going on will most probably result in rewriting the whole things. late results go in next paper or in revision b. Build the figures and tables first. A paper is connected through figures and tables with captions and description which is last piece and acts as connective tissue 3. Pick a journal, then read it a. Check scope statement, word and figures limits, structure of submission, APC, and time to first decision. b. Read 5-10 papers published in that journal and check the structure, style, and tone of writing. It will help you to understand what is expected and how to write in that style. 4. Set up the mechanics on day 1 a. Set up reference manager with clean library. This is important to keep tract of references you are going to use while writing the paper and should match correctly to avoid reviewers critic on that b. One master version, tracked version, a naming convention and eveything follows after that

Once you have decided these before writing, your onward journey becomes easy and mopre organized. While these help you in getting ready for writing, writing itself comes with challenges which can be mitigated or organised by keeping following in mind:

1. Sequence of writing a. Ugly 1st draft first and fast: once you have 1st draft, you have more clear vision on what all you want to write. If you target to write and edit together, it will drag and nothing will get acomplished. b. Order: figures -> methods -> results -> discussion -> introduction -> abstract -> title. This order is important because it helps you to build the story and connect the dots. c. Write the limitations section early: it usually helps to weigh your claims correctly d. Decide the word budget for each section 2. Sentence and paragraph structure a. One paragrapgh, one idea, topic sentence first, supporting sentences next, and concluding sentence last. This is important to keep the flow of the paper and make it easy to read. b. Put the action in verb: "performed an analysis of" becomes "analyzed". Normalization are the main cause of dead academic prose. c. Tense convention: past tense for what you did and found, present tense for what is generally true and for what a figure shows. d. One term per concept, repeated without variation. This is important to avoid confusion and make it easy to read. Avoid synonyms for the same concept. e. cap abbreviation at about five. Every extra one cost reader's memory

3. Revision done seperately a. Margin rule: write the purpose of each paragraph in margin in one line. If you can't summarize, re-check the paragraph and rewrite it. This may mean the paragraph is either not clear or not required. b. Go through the discussion and check if all the claims maps to results. Fix if not able to map. c. Read aloud: this is important to avooid the mistakes which you are not able to recognize due to repeated ready of the manuscript and possibly, you are reading the wrong wording as right. this also helps to identify the rhythm error. d. Number check: All the numbers in the manuscript should be fomatted in the same way and follwed through out the paper e. No discussion in results and no new results in the discussion. 4. Final check a. A point to point reference check: Eac h refernce should corresponds to claims to which it is attached to. b. Format check: The format of the final draft should be similar to what is there in requirement of the target journal.

At this stage, i do agree that a lot of LLM-based tools are available which can write the paper for you but use those as assistive tool instead of asking it to write specially technical articles where you are betting and writing your work. due to any reason if the writing includes false information and get passed, you may be passing wrong information to the whole community. this may lead to non-reproducibility of the work and loss of trust. Hope this article helps in the writing journey and you can share your experience and learnings with others.