Hacker News new | past | comments | ask | show | jobs | submit login
How to self-publish a book: A handy list of resources (datascienceheroes.com)
558 points by vharuck on Nov 12, 2018 | hide | past | favorite | 88 comments

Indie publisher here. Nice job - it came out quite well, judging by the images. Curious if the editor and proofreader were working with the same software, or it had to be exported to markdown or some other format.

Facebook ads are not a good investment, IMHO. They are expensive, campaigns take time to set up, and most people on FB are not on the site to browse books. Amazon Advertising is more effective and will be available to Amazon KDP users.

A note about ISBNs: You don't need one if you are just publishing a PDF on Gumroad or an ebook on KDP. If you are creating a print edition through KDP or a service like IngramSpark, you will need an ISBN, though.

For authors based in the states, a warning: The U.S. ISBN registry, Bowker, is a monopoly and prices accordingly so if you do want to go that route the outlay will be significant: $125 for a single ISBN in the registry and $250 for 10 (the last time I checked). Bowker will try to upsell overpriced and unnecessary services, like $25 barcodes and copyright assistance for $80 a pop not including Copyright Office fees. Bowker also left the barn door open on its CC page for six months earlier this year (https://www.publishersweekly.com/pw/by-topic/industry-news/b...).

$125 for ISBN seems like a ripoff to me - the ISBN are given for free where I come from: the Czech Republic; and the system is administered by the National Library

That's part of the reason it's so outrageous - it's free in countries like the Czech Republic and Canada, but some old-school publishing company was granted the monopoly in the U.S., and as a result authors and publishers are getting gouged in the U.S.

The other thing that bothers me is the shameless, mercenary upsells that are designed to trick newbies. $25 for a barcode? Many book designers throw it in for free or you can get a free one on your own at bookow.com/resources.php

When I first got started in book publishing, it was even worse. Bowker was selling website widgets for some ridiculous subscription fee - $120 to start and $60 every year thereafter (https://in30minutes.com/bowkers-isbn-markup-new-authors/). I wonder how many poor souls are still getting charged for that?

ISBNs and healthcare tend to follow similar patterns in the US, contrasted with the most of the rest of the 1st world.

The AMA to both certifies doctors and runs the medical schools. Obviously, they have a vested interest in controlling supply to protect their membership, thus artificial scarcity leads to high prices.

Internet access too.

Thanks for this comment. I recently self-published for the first time and went the Gumroad route. No regrets so far. But I got advice from all over the map, from "you need an ISBN and more" to "don't ever publish on Amazon, they drive the more ethical businesses out of the market." It's a maze for first-timers.

Some of the writing and publishing associations try to point newcomers in the right direction. Science Fiction Writers of America/Writer Beware (https://www.sfwa.org/other-resources/for-authors/), Romance Writers of America (https://www.rwa.org/p/cm/ld/fid=2101), the Alliance of Independent Authors (https://www.allianceindependentauthors.org/) and the Independent Book Publishers Association (http://articles.ibpa-online.org; disclosure: I'm on the IBPA board) have lots of free information on their website concerning best practices and pitfalls.

Hi! I created the post. Will answer your points below:

- There was no editor, or more precisely, I was the editor. What you are addressing is interesting. The two proofreaders, both grammatical and the technical work directly on the ".Rmd" (r markdown). I had to explain -in the grammatical case- "what to change" and what "not to change" about the markdown syntaxes. Even more, if you are working with R package bookdown, you can correct some figure captions, text links, and so on... all of this is r markdown syntax; the proofreader had to change with caution all of this.

- You are right about you don't need an ISBN to publish on gumroad. Just for Amazon (or other publishers).

Is it only me, or does anyone else think that the prevailing business practices of the likes of Amazon are an absolute rip-off when it comes to the low royalties? You go through all the work of writing, editing and producing a book and upload it. Amazon does nothing by the way of marketing & promition, so you end up marketing links to your book on social media yourself. If you want a decent price, you get 35% royalties, maybe another 5% for affiliate marketing. That means that Amazon gets more out of the project than you do, despite the fact that you're the one doing almost all the work (handling downloads & payments seems like a pretty commoditized thing these days).

Compare that to selling music on bandcamp: They take 15%, the rest is yours to keep.

It is absolutely beyond me, how self-publishing authors put up with that.

I GUESS the difference between music and e-Books, to some extent, is how the channel influences perceptions of the quality of the product. People have got very used to low quality writing being available for free on the web. So if your ebook is just another piece of text on some website, many people will not be willing to pay for it, whereas being on amazon creates an expectation that the quality would be in line with properly published work (where, at least, there is a four-eyes principle at work, at shouldering financial risks around a book project etc.)

Someone should do something about that and create an ebook marketplace that enforces certain quality standards and pays decent royalties.

I think it's because Amazon holds most of the cards, so if you don't like it, you can go hang....

My figures (quoted elsewhere in a separate post). Sell my £24.95 book on lulu.com - I get ~£11. Same book on Amazon, ~£4. I did another book which I sold on both platforms, and firstly I had to raise the RRP to £7.95 so I'd make more than £1 on an Amazon sale, and secondly, I didn't sell the ~3x as many as I'd need to to make the same money as selling on lulu.com - so I now only sell the second-latest version of any of my main book on Amazon so that people will hopefully see that there's a newer version on Lulu and buy it from there.

A friend of mine did $120k completely passive with his book (+ audio book) on Amazon. Without Amazon, he would have to get the outreach. Outreach is the expensive and difficult part. When the customers are already there, it's just a matter of converting them.

There are many people who make $25k/month completely passive on Amazon. I'm not a big fan of Amazon, either, but you're not signing a bad deal with them. Most publishers are worse (and don't deliver the outreach).

edit: Another thing to consider: Don't limit yourself to selling books. Add value and place upsells in your books. And for fiction writers: Maybe add merch and other stuff. For non-fiction: Sell online courses or coaching for your expert topic. Usually the book is just the entry.

What was the book's subject? I've always managed to convince myself that the poor performance (relatively to values such as I see you have quoted) is because my market is somewhat niche, but it may not be the case, so any info on such self-published success would be useful for me.

I'm not gonna lie: Daytrading.

It's the definition of a money topic. But I know that there are people who make a fortune with small cook books. I know some other very profitable niches, but I can't talk about them in detail - competition is already extreme. But trust me: There's a ton of money in most topics.

OT: About daytrading (most people think it's basically a scam) - I know people who do daytrading on a daily basis (what a pun) and who live relatively comfortable investing $25-40k. I know a guy who went from 12k to 80k with one investment (was a 2-year investment, though). He's not gambling, he is a trading nerd reading the news and all company reports constantly. But yes, 99% of daytraders lose money because they start to become greedy or have no discipline.

...well I guess the important question is whether or not you think you have a high chance of showing up at a high rank for search terms with enough traffic, whether you have a chance of being featured prominently by amazon's recommender engine etc etc. If you want to make that work for you, you probably have to play a game like being a demand-driven content farm gaming Amazon in a way akin to how SEO tries to game Google.

Maybe there's money in it, but it's probably not the kind of content I would either want to produce or consume.

For most authors, the point of departure is probably, more idealistically, some well written content, maybe with an audience that's a bit niche, and the desire to find a way to get remunerated fairly on the effort that went into producing the content. It sounds to me like that's not really what Amazon is offering.

His book is fairly good and well-written, but I know what you mean.

I'm currently helping an author who wants to write a book and make some money with it (she wrote books in the past and worked with publishers). I explained to her how the self-publishing business works and she also had the same reaction. It's possible to write high-quality content and get fairly compensated, but you should definitely know that most people seem to be content with sub-par books. The average quality is really bad and people seem to like it (e.g. another very profitable niche: erotic books, Shades of Grey is just one of many of them).

Amazon is very generic. If you want to build an audience, you should stick to Facebook, Instagram, Pinterest, Snapchat, Slack/chat groups, write a blog, do vlogs and videos, organize local and offline groups, events and other methods. You can then build a whole community around a niche and sell way more than a book. This is also what I've recommended to her because this is how you build high-quality communities.

> Someone should do something about that and create an ebook marketplace that enforces certain quality standards and pays decent royalties.

Leanpub doesn't enforce any quality standards (it's a self-service), but it does pay around 85% royalties.

I really like their model of being able to buy and sell unfinished drafts, and getting updates as they come along.

One thing that Amazon provides over others that you might find worth the hefty royalties is the "Also Bought" section. Book discovery can be a pain, and I often find myself picking books up that are suggested to me by Amazon. That engine can be pretty powerful.

In practice, Amazon is where people who will actually pay money for your book are.

e.g. Smashwords does a much better rate, but nobody gets much sales.

Founding an ebook store is easy. The hard part is getting the readers who will pay money to come to it - Amazon's completely succeeded here.

I agree with you, but what passes for a book these days is not what it used to be.

Check out how much traditional publishers pay.

Traditional publishers do a LOT more than Amazon does for a self-published author.

That 35 percent is more than you'll get with a traditional publishing contract. There, it's not strange to get 10 percent of the profit.

>Compare that to selling music on bandcamp

The first I've heard of bandcamp - that's probably the difference.

Try publishing via Elsevier and see what you are left with, compared to that 30% is really nothing.

I think there are different dimensions of publishing - publishing for profit or publishing for reputation.

You should not have to choose, Elsevier is taking advantage of you.

Editing your own text can be difficult because when reading it, you'll mentally fill up missing words that aren't on the text.

Use a text to speech reader like Textaloud, balabolka and you'll catch more errors.

There's a setting in Microsoft word to check for grammar, run-ons, verb agreement... and readability rating of your text.

Write first, edit later to reduce writers' block in drafts - Turn off the spell checker or simply write in notepad.

In another life, I edited some things—short memoirs, journals, a couple of works of political philosophy. I picked up a trick that works pretty well, even on my own writing:

Read it backward, last sentence to first.

The vast majority of mistakes you miss aren’t a product of reading your own writing—they’re missed because you’re reading in your flow of thought and you know what you’re trying to say.

This is what a third party offers—they don’t know what you’re trying to say until they read it. But even they can miss things when they’re in your flow. I read others’ work backward to this day, and still catch things I missed on my first forward read.

Edit: I just had to correct a silly autocorrect because I didn’t read my own writing backward. ¯\_(ツ)_/¯

> Use a text to speech reader

Thanks for that, it has never occurred to me. I’m installing a reader now.

> There's a setting in Microsoft word

So… it isn’t in the same realm for quality or language support but diction¹ can be pretty useful for those of us who write in vim. diction(1) will will analyse your text for simple errors, and style(1) will generate some readability stats for you.

Upstream is pretty open to additions if you want more language support, or at least was when I submitted an en_GB support patch years ago.

1. https://www.gnu.org/software/diction/diction.html

Do yourself a favour and get Scrivener. Don't use Microsoft Word or any generic word processor if you're writing a book, sure you can do it but god is it a painful experience.

I have been using Scrivener for a few years now and I couldn't go back to using a generic word processor for writing now.

I try not to recommend closed source software but there really isn't anything else that comes close to it.

I have Scrivener and have used it. The problem is that, at some point, I need to start collaborating and otherwise sharing copies. At that point I need to switch to a common format anyway. So I'm not convinced that Scrivener helps me a lot given that I'll have to export well before I'm finished (in my case). Maybe if I were working on something solo for a long period of time but that's not how I generally write.

Yes I can see the issues in real-time (or near-time) collaboration as that is not something Scrivener really deals with elegantly. I was talking about it more from an individual writers perspective which is how I use it.

I export sections/chapters to PDF, ePub, etc. to send to my editors and technical reviewers.

Actually you got me thinking about how I would need to approach this should a collaborative project come my way. I need to investigate this further as that is now an itch that needs to be scratched :)

For work stuff, I use Google Docs as a matter of course and that works well. It's not perfect insofar as the "it's over to you now and I won't touch it any more" handoff isn't baked into the workflow but once people learn to have discipline around that, it works well.

I do like Sccrivener. I just don't really do the working on screenplay/novel/etc. over the course of months to years that is really it's sweet spot. The one time I used it which was really useful was assembling a book partly out of a series of previously written pieces. It was very good for moving parts around although I suppose outline mode on a standard word processor could have substituted.

Plain text editor with a markup language targetable to PDF is all that's needed.

I don't disagree but Scrivener is basically writing IDE for me.

It allows me to keep all my reference material, notes, documents, pictures, etc, etc. in a single, organised project in a nicely designed project management tool.

I guess I would say Scrivener is my Integrated Writing Environment :P

I've looked at Scrivener and Vellum both and they both look like good pieces of software that take a lot of the tedium out of creating a book and let you just ... write a book.

The only concern I have with both of them is that neither seem targeted towards technical authors so I'm left wondering how they handle things like code formatting/highlighting and such

Grab the trial and see for yourself? I've had no issues writing technical documentation with copious amounts of code and config sections.

I'd say - never publish a book before someone else has read it. (This is how things were done in the pre-electronic era, and there's no good reason to change that.)

Working on a novel now - gave it two thorough passes, then gave it to a beta reader who also happens to be an English teacher.

They found hundreds of little errors (and some big!) that I probably never would have caught or thought about. Definitely worth finding a few volunteers. Lots of people looking to swap novels for feedback.

The only time I didn't was because I wrote a shortish book over a somewhat extended period of time with a collaborator--and with various peer edits. Decided not to hire a copy-editor and it was OK. But I generally 100% agree. You just can't see your own mistakes. Especially if you don't have the luxury of letting it sit for a month and come back to it.

I use Grammarly for editing.

I self-published a book and sold $250k worth over the past year.

I cannot recommend getting an editor/typesetter highly enough. I paid mine something like $500 total and she absolutely saved my life.

I'd go the Gumroad + Stripe route (or maybe Kickstarter if you need some motivation) - Amazon sales have been negligible for me.

>I'd go the Gumroad + Stripe route

Why do you need Stripe? AFAIK Gumroad already handles payments. Something to do with free vs. paid Gumroad plans and features?

They didn't process payments when I set it up. Or rather, they processed them through Stripe

> I self-published a book and sold $250k worth over the past year.

Wow, nice work!

Can I ask what field/topic your book is in?

Any tricks to drive sales that high?

I believe it's the book Secret Sauce: The Ultimate Growth Hacking Guide [0].

Austen Allred is the co-founder of Lambda School [1].

0. https://www.amazon.com/Secret-Sauce-Ultimate-Growth-Hacking/... 1. https://lambdaschool.com

Most of those resources only cover text-heavy books in paperback format. i currently write an astronomy book (many high res images and illustrations) in a large A3(ish) format. the whole process is completely different.

for example:

- you should NOT let a third party print the book for you without any control over the quality of the final product. amazone print on demand does a decent enough job with paperback text books, but should probably not be used for printing anything else.

- unfortunately there is no real (OSS)alternative to CS software like illustrator and indesign when it comes to actually creating the book. yes, you CAN do it with other tools - but if you want to keep your sanity, DON'T. ( I tried and ultimately gave up on the idea of using anything but adobe software....)

- color management is a nightmare. prepare for some real headaches when the first test prints come back and everything that should be "100% black" comes back in different shades and color variations of dark grey.

- if you want any layout-heavy book to be easily translatable into different languages, plan for it from the get go. text like small info-boxes underneath images tend to vary in length a lot from language to language. some languages produce much longer texts than english, others are more compact. this can become a significant problem if all your layouts rely on character-perfect text placement and fixed lengths.

- binding techniques vary from one printing company to the next. it is advisable to choose a company early on, so that you can adjust your layout process accordingly. a simple change to color management, paper format or line characteristics can become a huge problem, when you have to change dozens or hundreds of embedded illustrations.

- you dont need amazons print on demand at all: most printing companies let you put in a small initial order (like 200 books). sell those first, then use the money to order more. margins are great if you actually publish yourself, so you can easily scale up your orders without risk of financial overextension. if you do it with the same company this should not take longer than 2 weeks. that kind of response time is totally acceptable.

Edit: I should add that this only works for books with a high retail price (like 50+ dollar)

I've written a book about a domain I know about - Music Technology - and self-published it about 5 years ago , using lulu.com. It sells for £24.95 [1], and I get about half of that in a lulu sale. This is a 650 page text book which has a LOT of content - it would take hundreds of hours to go through it all for a beginner.

If it sold on Amazon, I would get about £4. I guess I should have gone via Amazon direct? (given that the author mentions 40%?)

Lulu makes about £4-5 on a sale (going on the printing costs that I pay, and me factoring in a bit that I would pay even on a copy that I'm buying myself), whereas Amazon would be taking more like £12-13 going on those numbers.

So, that's why my book isn't available on Amazon - I'd have to sell 3 times as many to make the same money, and that seems unlikely to me. If it was more even, I would sell it on Amazon as well as Lulu as I'm sure I'd get -some- more sales, but not drastically more.

The book is a reasonable seller and a useful source of side income for me, but it's certainly not an 'earner' - I've spend about 3 weeks of very full time work updating it for the latest version, and that will be in print for a year before the next x.5 version comes out (and usually this means changing a large amount of content as lots of little things change). Dividing the year's earnings over the time taken to update it means I get a reasonable rate for that time, but certainly not a good one.

I've thought about writing another book (going into more depth on the same subject, and being a Cubase 'expert'), but I remember that it took about a year of spare time to write the first one, and while it's been a reasonable proposition over that time, it was only because it got picked up by the software company's education rep that it went anywhere and reached any kind of 'critical mass' - for about a year it sat and literally sold only copies to students I was teaching (which was the initial reason for writing the book as there was no text book of worth available). Just to be clear, I didn't enforce purchase by the students, I'd say under 10% of them bought the book.

[1] - http://www.lulu.com/shop/darren-jones/the-complete-guide-to-... (latest version of the book, until Cubase 10 gets released, then it will be superseded)

In my spare time I've been developing a Java-based (don't hate me) WYSIWYM tool that integrates R, Markdown, and a fair bit more. For example, Scrivenvar can also transform XML into Markdown via XSLT, then use an R engine to perform computations that are substituted back into the Markdown document prior to generating an HTML preview from the final text. (For the OOP enthusiasts, it uses the Chain-of-Command design pattern.)

I wrote Scrivenvar because I wanted the ability to use interpolated variables while writing, so as to create documents free from duplicated content (e.g., character names, locations, and timeline calculations in a novel). (My favourite part of the book's R code is integration of a GIS API to compute driving distance based on the lat/long coordinates of two places in the novel, falling back to the Haversine formula if the website is unavailable; the number is then converted to English text using a Chicago Manual of Style function call. Effectively, if I change the lat/long of either location, the value of the book is updated without having to remember where in the text that that particular number was referenced.)

Once a YAML document is loaded, inserting a variable is quick: type a few letters from a value followed by Control+Space to insert the corresponding variable name. This is handy if for deeply nested variable hierarchies.

The software is open-source and very much beta:


Ping me on GitHub if you like the concept and have comments or questions.

Does anyone know how I can self publish on the “bible paper”? It’s the super thin yet strong, high quality, paper.

I want to print a large book but want to have it not be too thick.

Self-publishing print on demand (POD) services like KDP Print and IngramSpark offer two papers: White and a slightly thicker "creme" paper. It's quite nice - I used it for a fiction series - and the cost is not much more than white. But it's not the same grade as the Bible paper you mentioned.

If you work with an offset printing service you can do a short run on whatever kind of paper you like but the cost will be more for smaller orders of 250. At 500 you start to see economies of scale that rival POD costs but you have to take shipment of hundreds of books and store them somewhere.

What your looking for is called onion skin paper and it seems to be pricey, but I agree, great for certain uses

I’m willing to pay. Do you know how to get a book bound using that paper?

Like ilamont said you’ll probably have to go through an offset printing process. There are some online services that offer offset printing for flyers etc. but only with the most common setups. Easiest is to find a few printing companies in your area and ask for offers. You might ask book designers or publishers in your area to recommend printers.

I enjoyed the article and appreciated the information on ISBN numbers in particular. In addition to the suggestions in the article, I would also add Scrivener and Vellum to the list of important writing and publishing tools.

For the most part, I think this article is full of excellent advice. I do disagree with three of the author's tips on how to write well though.


1. There is no need to avoid alliteration. You shouldn't drown your work in it, of course, but skillfully employing alliteration in your text can help your sentences flow and can make certain aspects of your content more memorable. I'd say it is better to learn to use alliteration, and use it to good effect, than to eschew it altogether.

2. Comparisons are not bad, especially in technical writing. When teaching, it is helpful to tie new concepts to existing knowledge and comparisons are one way to create these mental connections.

3. Generalizations aren't bad in technical writing either, especially in code samples. For example, telling a programmer how to print "Hello, World!" to the screen is a helpful generalization. With that knowledge, the developer can now use that same print statement for console output, messages, debugging, logging, and other tasks. Developers can take this deceptively simple generalization and use it for all kinds of specialized needs.

I think the key here is to learn to use alliteration, comparisons, and generalizations in a way that strengthens your writing, rather than detracting from it.

Hi! I wrote the post, glad you found useful!

Also, I'm with you with the recommendations :P It was just a funny image I found on the internet and shared to produce some laughs...

I use a lot of comparisons alongside the book, especially from the "technical to the real world".

Generalizations are also important, otherwise, it is impossible to get to the point. They shape ideas. cheers!

Excellent job! Thank you for your efforts in sharing the information.

I have a (draft) article where I've dumped all of the technical nitty gritty of self-publishing: http://wiki.secretgeek.net/creating-a-book

For papers and presentations at work (ones where i have time to do it right) I write the text, then run it through the xkcd simple writer (https://xkcd.com/simplewriter/), then make a new version with that output, then compare the two side by side and either edit the original down or the simple one up.

It is a fascinating lesson. Some names or terms are understood and should be there. Sometimes a more uncommon word is still clear AND more precise, so it stays. Other times I find complete sentences that I can just remove and my points are more visible.

I try to anticipate counter-arguments and head them off with detail, and/or add qualifiers to statements at the cost of my main point. This technique makes it clear when that happens so I can pull such out entirely or move to a distinct section. Do it a few times and you benefit even when not using the tool, though I need to refresh my brain monthlyish.

I removed "tend to", "most", "some", "often" and three sentences from the above. I also broke my second paragraph into two, just from habits I've learned this way.

A shameless plug for my own.

Publish on Kindle by Joel Dare

Explains how to format and publish a book using Libre Office. The book is short, getting strait to the point quickly. I don't see a lot of sales but readers that have reviewed it seem to like it.


I'm currently in the process of drafting a book about modern data engineering. Data pipelines and data warehouses have changed drastically in the last five years and I feel the few books on the subject are drastically outdated.

I figured I'd just start with content and work my way into the logistics of the actual publishing process. This, and the comments in this thread, are a huge help on the logistics side. Thank you!

Your book sounds interesting! How to keep a track of your project?

> …and a website.

=:) Have you seen Superbooks on web with https://bubblin.io?

Disclosure: I'm one of its developer.

Pretty cool site. FYI the page turn animation is unstable if I grab it near the center of the page (it rapidly rocks between angled-down-from-top and angled-up-from-bottom).

hi, I'm Sonica, CTO and cofounder of Bubblin here.

Yes, the transition is shaky when the curl is near the horizontal axis. We'll fix or remove it on our next iteration of Bookiza.JS [1]

[1] https://bookiza.io

This site looks awesome! Thanks for sharing!

I self published a book with leanpub. I used their online markdown editor. Then I used their preview feature to export in all formats: pdf, epub and mobi.

Then I uploaded these to my gumroad account, created a product and setup a payment widget on my website.

Smooth and simple process.

A few comments. 1) kindle KDP is good, but you need to read up on format issues for print on demand. fine detail around gutter and margin needs to be understood for a high quality product. The specific DPI you render cover artwork in, and sizing has a huge impact. Kindle can be a bit arbitrary on what they accept and what they reject for print.

2) Kindle ePub demands a different set of outcomes. Do your print hardcopy first, then modify it to make the inputs to upload. I used calibre to do all the mods, but people swear by sigil. How you index makes a huge difference. Remember ePub is flow text. All those pagerefs have to be re-calculated into logical offsets and marks, not literal paper counts.

Again, the submission system can be a bit opaque.

If you want PoD hardcopy in Australia.. Avoid kindle. The amazon trade war with Australian taxation has hit hard and they have no local printery.

I recommend Ingramspark, who can do PoD, and manage epublishing into kobo and nook and the like, and who have an agency status in amazon to sell your hardcopy. The PoD rates look competitive with the KDP ones, once you factor US delivery costs in.

(Amazon are pretty cool for worldwide rights)

Ingram demand really tight conformance on the PDF ISO specs for final output. Adobe, craptacular code though it is, will emit the legal form. Sigil and (yay!) libreoffice seem to also do this, but ymmv. Again, the DPI of your images make a huge difference to submission here. Some stuff demands 300, some demands 72 (for eprint covers)

Employ a professional editor/proofreader. They make a huge difference. Seriously, its money well spent. Some of them index too.

Kindle can be a bit arbitrary on what they accept and what they reject for print.

For e-books, too. One of my books was mysteriously blocked without justification, and after several months and attempts at escalation was just as mysteriously re-instated.

Self-published books often have horrible layouts and graphical design, all-around. Thankfully basic LaTeX styles are quite decent and do their job.

Casas's Data Science Live Book looks nice with its basic LaTeX style, I think.

Thanks! All the magic for the layout is thanks to LaTeX and the use of the R package Bookdown: https://bookdown.org/yihui/bookdown/

Bookdown provides a template to produce the pdf, and you can modify later on (for example customizing the table of contents, image positions, etc)

You write the book once, and it can be exported to html, pdf, epub (and kindle)...

These multiformat outputs have come a long way in the last five years, and it's good to see them working. Thanks for the link! The automated approach works best for technical books, just as in your case. There are still some esoteric layout features that are very difficult to automate.

Softcover should be listed here: https://www.softcover.io/

You're familiar with it if you've been through Michael Hartl's Rails tutorial.

I used Softcover to self-publish my own book, releasing it in paperback on Amazon: https://www.amazon.com/gp/product/1980891419/ref=dbs_a_def_r...

Hi! I'm Pablo Casas, the author of the "how-to" post & the Data Science Live Book. I didn't notice the link was here!

Let me know if I can help you, I see a lot of interest here :)

Nice post! I am an Indie Publisher as well and you hit on quite a few things I ran into as well. I went Gumroad right from the get-go and was very happy with my experience. I later added KDP into the mix as well with a Kindle/Printed version of the book, it hasn't done great but it is added credentials in a job interview :)

Does anyone have experience publishing a book similar to a kid's coloring/activity book? I'm not interested in the e-book route and the services I've looked at don't seem to offer the type of binding books of this type have.

A similar example based on my own experience, but for a non-technical book: http://www.gabrielgambetta.com/tgl_open_source.html

Just curious, is it (still) possible to make a living from writing books?

Does anyone have a recommendation for just simply printing a book (epub or pdf) that I'm not interested in selling? Just to print a copy for me?

For short / simple books or "zines", consider a simple rotatable stapler and print at home. Look at many of the zine resources online. This is my own favorite way to publish small one-off or short run books.

I use https://leanpub.com/ and it works for me.

Does anyone know of a PoD service that will do hardcover books for a decent price?

Can you define 'decent'?

ha. OK, how about a PoD service that does hardcover books, at any price!


I've done all my books through Lulu - including doing a couple of projects for clients (schools) who did a 'make a book' project with the kids writing the stories, etc... their service has always been pretty good. The site is a bit clunky, but works. I've done a couple of one-offs for my own use as well.

And looks like they do hardcover (although I've never tried it).

I bought Operating Systems: Three Easy Pieces in hardcover from Lulu.


The book is over 700 pages. It's very nicely done. The front and back covers are just the right thickness. The book is not bound in signatures, it's edge glued ("perfect bound") like a paperback. But, the spine is extremely flexible and not attached to the casing. I was very careful to open it properly by gently creasing down sections of pages starting at the front and the back. If this is not done, there is risk of breaking the spine. I can open the book at any page and it lies flat on the table. The paper is medium weight, light ivory, and non-glossy. The casing is printed nicely with the title, and the dust jacket is first rate. Altogether, I'm quite pleased.

Traditional offset printing company with a plug / PSA: below 500 copies print on demand services are great. Above 500 copies going with a traditional offset printing house will get you a better price. Of course you take on more inventory risk that way.

You also get access to more finishing and cover options, etc.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact