Cliff's Notes: Style Guidance for the IBM JournalCliff's Notes: Style Guidance for the IBM JournalNote that papers should not have the slightest trace of informality, idioms, jargon, or vague writing. Write cleanly, simply, and precisely. An example of vague, idiomatic, or informal writing follows. Clear writing is important because we have a broad international readership from many different disciplines.Example: "This has high costs. The subroutines transparently achieved good throughput, but this was expensive for the cores that do not scale appropriately. Hopefully, the idea will take off around the world, due to the new LDAP module."
Problems with this sentence include:
a) What does "high costs" refer to? Financial cost? Computational cost? Some other kind of cost?
b) What does "transparently" mean? Does it mean something is done in a simple manner, a straightforward manner, or something else? Does it mean an easy-to-understand manner?
c) What does "good" mean? Does it mean "high", "adequate," or something else?
d) What does "expensive" mean? Does it mean "time-consuming"? Costly from a financial standpoint?
e) "Take off around the world" is an idiom that wont be understood by the readership all around the world.
f) What is a "core"? Does it mean "processor core"? The first time you use "core," please use a more full reference to the word.
g) What does "scale" precisely mean in this context? Usually something "scales" with respect to something else, e.g. with respect to number of processor cores.
h) What does "the idea" refer to?
i) LDAP. Spell out all acronyms at first use: LDAP (lightweight directory access protocol).
Thanks again.
No "Leveraging": Authors from large companies love going wild with "leveraging" everything throughout their papers. Over and over again, they leverage this and that. Or they do solutioning.
Write: "We will make use of the power management mode," not "We will leverage the power management mode." There are lots of other kinds of "jargon" you should try to avoid.No "Good": Do not use the word "good" in your paper. It is far too vague and informal. Replace "good" with a more precise word such as: accurate, useful, relevant, appropriate, effective, etc.
No "Rich": "The processor made use of rich data." Do not use the word "rich" in your papers. Be more precise. Does it mean diverse, valuable, comprehensive, varied, something else?
No "Transparent": "The processor behaved transparently." Do not use "transparent" or "transparently" in your papers. Be more precise. Does it mean: simple, straighforward, easy to understand, something else?
No "Last Decade": Try to avoid this phrase, because the word "decade" often refers to a particular set of 10 years, like the 90s. Does it mean 2000 to 2010? Sometimes people use the word to imply the 10 years prior to the date the paper was written, e.g., 2002-2012. Since the phrase has no clear meaning, avoid it.
No "Driving": Authors from large companies love going wild with "driving" everything throughout their papers. Over and over again, they drive this and that. Does the word mean "encourage"? Facilitate? Something else?
No "Right": Authors often suggest that something is "right." Be precise. Does it mean accurate, correct, relevant, useful, appropriate...?
Use "Cognitive" as little as possible: Use this word very sparingly. This word is so tremendously overused that its meaning is becoming obscure. Do not use it merely to indicate you are performing analytics. "Cognitive" usually involves URL (understanding, reasoning, or learning.) It can involve natural language processing, pattern recognition, simulation of human thought processes in a computerized model, etc.
This is not a Powerpoint presentation. Please avoid using bulleted lists (and also numbered lists) in your papers, as much as possible. Use full sentences and paragraphs. Look at a journal, such as Science or Scientific American. Notice nearly zero use of bulleted lists and numbered lists.
Many IBM divisions have an official procedure for clearance. For example, the IBM Research division has a database for this. People from IBM Research, look here.Your division may or may not have a procedure in place. The first step would be for you to ask your manager. If your division does not have an official procedure, just send me an email from your manager saying something like:
"I approve the paper for external publication so that the IBM Journal of Research and Development can obtain external peer reviews of the paper and then externally publish the paper after completing the peer review and editing process."
You'll be asked to send me your original editable digital figure files.Please leave images of figures embedded at the end of the paper for reference with their figure legends, one figure per page. However, we also need separate figure files for each figure that actually contain the information in the figure.
For flow charts, graphs, diagrams, etc.: please send the separate figure files in a format that contains the data points, lines, and text (e.g. PPT, Excel, EPS) -- and not in an "image capture" format because image captures don't contain actual data information and can't be manipulated by our art vendor. Also, simply pasting the "image" of a diagram into Power Point will not be useful as the image does not contain the original data and thus cannot be manipulated.
For screen shots: Files that have been resaved in *.jpg format at any time often do not produce usable (sharp/clear) results and therefore are often not acceptable for screen captures. Try PNG, TIF, etc. However, don't use screen shots for ordinary diagrams, charts, etc. Send me the original data file in PPT, EPS, Excel, etc.
Please make sure that the figure files match what is embedded in the paper, as I've had authors send me figure files that sometimes look different than what's in the paper. Thanks.
Maximum of 4 figures: placed at the end of the Word document. Each figure should really be just a single figure, and not an attempt to cram part a, b, c, d, etc. into one multi-part figure to avoid the 4-figure limit.
The moment I receive your paper (if it is cleared by IBM for papers with IBM authors), I'll start getting the paper peer reviewed. To help supplement the possible reviewers I may find on my own, could you suggest five reviewers (include at least one from outside of IBM) whom you feel would probably be willing and able to review your paper in a timely manner? Please send me names, affiliations, and e-mail addresses.Reviewer identities will be kept secret from you, but I find it useful to give authors the opportunity to make suggestions because authors are quite familiar with people knowledgeable in their fields and who are likely to give useful reviews. (You don't have to contact these people.... I would invite the people, as I see fit, to review the papers, and then they would return their reviews to me.)
If any of your figures has been published previously, it may be a copyright violation if we do not get permission for reuse. Please write to the original publisher for permission to reuse, and let me know when you have the permission. Please credit the figure source appropriately.By submitting to us, authors implicitly agree to a number of policies that are central to our mission. Your manuscript must be a report of significant new work, a new review or overview, or a combination of the above. Although a shorter version may have been published (or submitted for publication) elsewhere -- for example, in a rapid publication such as Applied Physics Letters or in a conference proceedings -- the manuscript must differ significantly from the shorter version. If the paper reuses significant blocks of texts that have been published elsewhere, this could be a copyright violation. Please remember to reword and change the text to avoid copyright violation, and also clearly credit the original source. Your paper should not have been published in a conference proceedings prior to publication in our journal. If you feel that your manuscript is not a report of significant new work or a new review or overview, please let me know here:
The figure captions should clarify acronyms and abbreviations used in the figures, if they have not been defined elsewhere. Even if the acronyms are excruciatingly obvious to you, please add a parenthetical indication in the figure caption, e.g.: "Figure 1. Block diagram of the amplifier circuit. (WL: word line; SG: select gate; Accel.: accelerator.)"As you look at your figures, ask yourself these questions: Are all aspects of the figures explained sufficiently well? Do you feel that a general technical audience will understand every term in the figures (and tables)? Look carefully at the figures and tables. Is there an abbreviation, term, or acronym that is not 100% clear in meaning?
In the figures and tables, if the reader will have even the slightest question about: a term, piece of text, symbol, acronym, axis legend, line, arrow, use of color, or other figure element -- then please let the reader know the meaning.
Please refer to each figure and table in the main text. For example, "Figure 1 shows...." Place all figures and tables at the back of the paper.
Example first line of an Introduction:
"Services-related exchanges [1] form a growing component of the IT (information technology) global economy, often in the form of professional services engagements [2-4]."
or
"Since the late nineteenth century, researchers have studied EBTs (efferent biological tissues) using a variety of microscopic stains [1, 2]."Your manuscript must be a report of significant new work, a new review or overview, or a combination of the above--and be targeted toward the general technical reader. Please cite key references to related work in journals, books, and conference proceedings in your Introduction.
Note that the references at back should be used for references only and not for commentary and explanations.
Please cite all references in order, first [1], then [2], etc.
Please end your paper with a "Conclusion" section.Please add the author bios at the end of the paper when you submit your paper. This is important because we often find that the author bios are littered with English errors and undefined acronyms -- and some of our reviewers actually point out these problems during their reviews.
Recall that this is a technical journal, not a marketing journal for your company. The goal is to provide sufficient background to the subject in the Introduction, by referencing the work of others (outside of your company) -- by citing related books and journal papers.Increase your chances of having the paper accepted:
Example of a weak paper: "My company is a great company, planning interesting work, but we have no results yet and we can't place our work in perspective with the work of others outside of my company."
Example of a stronger paper: "We have results. We're aware that people outside of my company work in this area. Here is how our work can be compared to the work of others. Here are references to the outside work."
New review papers are also acceptable if the Introduction and Abstract make it clear that the paper is intended to review a field.
Please follow the instructions in the Microsoft Word Template given to authors.Thanks!