Search interesting materials

Instructions for authors

What constitutes a good article for this blog?


There are four tests:

  1. What is the question? Who cares about this question? Is this an interesting and important question? Is there a clearly posed question and a clear answer, on a question that matters in the world today?
  2. Is it new? What is common knowledge to the (specialised) readership of this blog is not good enough. It's got to say something new.
  3. Is it a logically sound argument? Mere assertions do not suffice. Linearity of argument, where facts and arguments are brought together to win an argument, are critical.
  4. Does it stand a fair chance at persuading a skeptical reader who previously thought otherwise? We do not write for the applause of those who already agree with us; we write in the hope of persuading people who do not agree with us.

Generally, your mind will become more clear through the exercise of talking through the argument, and debating it, with a few others. At its best, you will find a sparring partner who does not agree, and the two of you will write two rival blog articles expressing rival points of view.

Submissions tend to get rejected when: they are essays that meander and do not make a point, when novel ideas are not presented, when they contain unsubstantiated assertions, when the tight chain of reasoning is not in fact present, when they use adjectives and adverbs, when there is lack of clarity of thought, when the author uses turgid complex sentences and impenetrable prose.

Authors of essays express how they feel. That's not what we do here. This blog is closer to the research literature. The article has to be a logical argument. Ideas invented by others should be clearly acknowledged as being invented by them: do not imply they are yours. What is not clearly shown to be invented by others is implicitly your claim to a novel contribution. You may offer novel arguments in favour of an existing proposition, or you may demonstrate the errors of an existing proposition, or you may come up with a new proposition (and rationale thereof). But you have to say something new, and sharply locate your novel contribution as opposed to the things that were known before you started writing.

Look for the "author: " tag on the main page of the blog. These are persons who have authored multiple articles on the blog. Discussing your ideas with some of them may often be quite effective, as would be the step of having them review your draft.

An article that's a review of a research paper


There is a role for articles which review one research paper at a time. These typically have the structure:

  1. What's the question the paper is asking? Is this an interesting and important and unsolved question?
  2. How do they go about it? Is this the right way in which they should have gone about it? Is it a persuasive strategy?
  3. What did they find? In what ways does this change our view of the world?
  4. What was the novel contribution of this paper?
  5. What ideas do you get about related / alternate lines of research which you or others should take up? 

An article that tells a story about the world


Most articles on this blog are original work. Sometimes, it's interesting to write an article which is not a new creation of the author, but tell a story about the world.  It's got to be novel: it should not be common knowledge to this readership. It's got to be interesting to this readership: on subjects of relevance and important enough to merit an article. Here is one example. It's not novel ideas by Swapnika Ramu. But it's a great story, one that is highly interesting to this readership, and one that is not widely known which satisfies the test of novelty.

A few classes of articles that are a recognisable style


We  have a few groups of articles in which there are many instances of a common approach:

  1. Drafting hall of shame. Example.
  2. Watching markets work. Example.
  3. The state of the art. Example.

Language


Read 10 blog posts that are closest to what you are trying to do so as to get the voice right. It must be modern professional English, drawing on the ideas of Ernest Hemingway, Raymond Chandler and Strunk-White. It must be of the writing quality of The Economist and The New York Times. Simple words, simple sentences, short sentences, direct language. The florid writing of creative writing programs is inappropriate here.

It's good to have a quiet, conversational style. Pompous language turns off people (e.g. economists using words like "exercise", "rigorous", "first", "contribution"). Strip out adjectives and adverbs, and that will keep you safe from polemics. Most of what we know is wrong, so let's be humble in our claims. But there's an important distinction between lack of clarity of thought and lack of humility. We should clear in what we claim: there should be sharp clear claims. At the same time, they should be expressed in a quiet, humble way.

Mechanics of submission


It must be a plain, hand-written, HTML file where only the following tags are permitted: p, br, i, b, ol, ul, a, li, table, blockquote, h3. All other tags are forbidden.

Do not write in Microsoft Word or some such GUI tool and "Save as HTML". This does not work. You must write the HTML file by hand.

Generally it's better to italicise what's in the blockquote.

There must be only one kind of subheading: h3. We do not use sections, subsections, subsubsections, etc. Only one deep section titles.

Nested lists don't work properly. Just have bare flowing text and one-deep lists.

Images must be separately supplied as png files, and in the running text, say "Image file xyz.png here". Graphs must satisfy these guidelines. Blogger inserts the caption line; the image must have no title on it.

The first sentence must say "by Author1 and Author2.". If you already have an author page here, then make this a hyperlink to your author page.

At the end, have a para with one sentence about each author. It would be kind to thank a few people there, who helped.

Write me email with a submission where you attach the HTML file containing the article. In the email, propose a title and supply a list of labels that would be appropriate.

HTML syntax checking


If you merely look at your .html file in a browser, it may look fine, but there can be mistakes embedded therein. The problem is that web browsers try to be kind and make things up when you have mistakes. This `Do What I Mean' (DWIM) behaviour may seem okay when you are looking in one browser (it will assuredly break on other browsers). Errors in HTML assuredly break blogger.com when it has to process the submitted HTML. For blogger's publishing to work, the HTML has to be syntactically perfect. Please be super careful about the HTML code that you submit. Every minor detail in it has to be pitch perfect.

There are many good syntax checkers. Here's an example of one in action:

$ tidy -e privacy_UID.html
line 1 column 1 - Warning: missing <!DOCTYPE> declaration
line 1 column 1 - Warning: inserting implicit <body>
line 3 column 609 - Warning: <a> is probably intended as </a>
line 5 column 298 - Warning: <a> is probably intended as </a>
line 1 column 1 - Warning: inserting missing 'title' element
line 1 column 7 - Warning: <a> attribute "href" lacks value
Info: Document content looks like HTML5
Tidy found 6 warnings and 0 errors!

Typesetting mathematics


LaTeX mathematics typesetting works.  . The following raw text:
This is an inline equation: \$e^{i\pi}+1=\sqrt{0}\$. And here is a displayed equation:

<center> \( e^{i\pi}+1=\sqrt{0} \) </center>
generates this result:

This is an inline equation: $e^{i\pi}+1=\sqrt{0}$. And here is a displayed equation:

\( e^{i\pi}+1=\sqrt{0} \)

Typesetting blocks of computer code


Here's a demo of a block of computer code:

<small><small><pre class="prettyprint">
# Make the NPV of an annuity p, where people die off based on the
# survivor function S, when the interest rate is r, l is the number of
# years the pension has to be paid.
value.of.pension <- function(p, S, r, l=0) {
    (1/r)^(1:l) %*% (p * S)
}

# As an example: Price an annuity of Rs.1 per day at age 60:
value.of.pension(rep(365,40), a$cooked.60[61:100]/100, 1.07,l=40)
</pre>
</small></small>



which gives the nice typesetting of the code blocks in this example of an article which interweaves code and text.

Doing references well


1. For many articles, it's sensible to have a bibliography block at the end [example]. My thumb rule is that the materials that would be quite interesting and valuable to the reader of a given article, and matter most in terms of the buildup of knowledge that leads up to this article, should be in the bibliography block. We're not picky about the precise details about how you typeset an entry in the bibliography. But it's important to be consistent - don't change the style from one entry to the next. And, put an href on each title, so the reader can click on it.

2. For the thing which are supporting material but not central to the logic of the article, the traditional hypertext style of underlined links is fine, e.g.:

As was argued in FSLRC volume 1, one member of the board of a financial regulator should be the `administrative law member' (ALM).

Once you have done this, the FSLRC report does not appear in the bibliography block.

3. When you make tangible statements such as the above, or quote from an original source, it's important to give the precise location. Example, the above should preferably be done as:

As was argued in FSLRC volume 1 (page 38), one member of the board of a financial regulator should be the `administrative law member' (ALM).

4. When you work with legal instruments (the Constitution, Parliamentary law, rules, regulations, bye-laws, etc), it's generally a good idea to have a quotation from the law. You must always give the precise identifying information e.g. S.32(a) of the XXX Act (YYYY), and you must always have a hyperlink from the Act to the bare text of the law, where the reader can easily cross-check what S.32 says. For example:

At S.38(1), the draft Indian Financial Code (2015) says The board must evaluate the performance of each member.

This block displays all the required features: It has precise text from the law, it has a hyperlink to the law, it accurately identifies the law, and it tells the reader that the quotation is from S.38(1).

Dealing with referee reports


When a referee raises concerns, you should not disambiguate to me. You should view the referee as a sophisticated reader who did not get what you were trying to say. This calls for rewriting your article in a way that gets through to him. Most referee comments are useful as a tool for sharpening and rewriting in a way that the song that is playing in your head actually plays in the head of the reader of the article.

Frequently asked questions

 
Who can submit manuscripts for possible publication on this blog?
Anyone.
How will manuscripts be processed?
They will be put into anonymous peer review by one or more experts. This can result in one of three outcomes: a rejection, an acceptance, or a set of comments with a revise-and-resubmit.
What's the typical elapsed time from submission to publication (assuming it's accepted)?
Typically, when there is one revise-and-resubmit, it's 10 days from start to finish.
Do I know who the referees are?
No.
Much later, do I get to know who the referees were?
No.
Do I get paid?
No.
Do you accept a payment in return for which you will take an article that we supply?
No.
Do you accept advertisements?
No.
When do I get an author page for myself?
When you have five published articles.
What happens after an article is published
The world gets free access to resyndicate the article, subject to rules about fair use.
Oops, I found a mistake in my published article. What do I do?
Write me email and I will correct it.
Is it okay if I disagree with you?
Yes.
Would you consider a submission which takes up one of the articles on your blog and criticises it?
Yes.
Can I have footnotes in my blog article?
No.
What's the typical word count?
The best readership is obtained by articles which are roughly 800 words with may be 1 or 2 floating objects (tables / graphs). But good articles go all the way out to 3000 words. There is no rigid rule about what is the required word count.
Why do you not flaunt the greatness of the author?
Too many people assume the oldest guy, the richest guy, or the IIT guy, is right. On this blog, we want the quality of the argument, and nothing else, to matter. We want to pay less attention to the brand names and more to the content.

No comments:

Post a Comment

Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticising me is perfectly okay; uncivilised language is not. I delete any comment which is spam, has personal attacks against anyone, or uses foul language. I delete any comment which does not contribute to the intellectual discussion about the blog article in question.

Please note: LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.