How we crowdfunded $750k for a giant book about keyboard history | Hacker News Reader

How we crowdfunded $750k for a giant book about keyboard history

glennf.medium.com

219 pointsingve86 comments

One of the more interesting bits (as one of the backers, staring at my copy on the shelf now), is that the author asked for pictures of the book once you got it, and just the ones he's been comfortable posting in the updates are amazing. People's whole collection of antique computer history with the book open to the page about it, or propped up on a typewriter, etc. I sent one of my Model M, Unicomp, Thinkpad portable keyboard, Ergodox and recent dactyl manuform with the book.

How is the quality of the actual writing inside the book? I suspect high - looking at the supplementary text, but everyone seems to judge this book by its cover

Pretty high, though I have only been skimming. It's kind of a coffee table book for me, since the part I really care about (modern keyboard cambrian explosion) is very short and way at the very end. I did read the entire small booklet "The Day Return Became Enter" and that was wonderful.

One item that interested me was the tax issue he raised - getting income in one year, and having to pay tax on that, but then expenses in the following year.

In my jurisdiction this is known as future-income, and essentially in year 1 a simple accounting provision is made (hence reducing profit in year 1) which has the effect of moving the revenue into the year where the order is fulfilled.

This is especially true for subscription revenue where "annual amounts" are almost always falling into 2 financial years.

This is enormously important from a refunds point of view - it's hard to give out refunds if the tax has gobbled 30% of the amount collected.

From the article he's suggesting this practice is not universal? Which somewhat surprises me...

Many jurisdictions do not allow this sort of deferals for cheating.

Problem is with kickstarter, they should structure this thing as a loan, that gets written off by delivering or failing.

It's possible to do this on an accrual basis, but when I have spoken at times to accountants about it, they have discouraged it: the IRS can be dubious, and it makes the most sense to run your business that way if you're entirely organized around it (i.e., you have a profound mismatch in your basic function between income and expense within calendar or fiscal years).

On the contrary, most places enforce this kind of 'deferral' (the accruals concept) to stop cheating. Imagine i took all of the cost first and the revenue later. I could offset the loss from this against my other income (which is taxed at source) and get a refund out of the tax authorities! It would be like an interest free loan!

It's also enforced here to keep reports of profits sane. If I report revenue as profit (without adjusting for future expense) then I'm artificially reporting "higher profit than what actually happened."

This artificially boosts share prices, and other things, which can have unfortunate effects downstream.

Take Worldcom. They defered expenses (by capitalising the phone bill to be depreciated in future years) thus boosting reported profits. With, um, unfortunate results.

So yeah, in my jurisdiction its enforced. But I can imagine that in other places its not, or revenue is taxed, or whatever.

Tax result <> bookkeeping result <> cash flow

In many countries tax office does not care about your bookkeeping result. Tax is calculated using separate rules (and then becomes part of bookkeeping result).

Also if we go to philosophical discussions: you can have great bookkeeping result yet no money, or lots of money now, but unprofitable projects (e.g. construction companies).

Taxing revenue is very useful with multinationals who use cunning tricks to move profits back to their home countries. So yes it remains a valid tax

you want to make kickstarter an international bank?

They sort of are already, no?

In the US, for an individual/LLC, my understanding is that the default is that Kickstarter funds received would be income pure and simple reported on a 1099-K past some threshold. https://www.kickstarter.com/help/taxes

There may be ways to defer some of that but you'd probably need to work with an accountant.

It is called the 'Accruals concept', expediture must be matched to it's revenues. This is a fundamental principal of accounting.

It would be extremely novel for tax to be on a 'cash basis', but I'm sure there are places it is true. I think it might be an option on micro entity accounts for instance.

A classic example of this in action is with stock/inventory. Things still in stock at the end of a period don't hit the profit in that period, only things that are sold.

It's not at all novel. Small businesses in the US typically have a choice (C corps do not AFAIK) to use cash-basis accounting rather than accrual.

Which makes sense. If they're not making large capital investments that they hope will result in revenue over time, they're largely cash flow businesses.

Tax result <> Bookkeeping result. In fact bookkeeping result as per US GAAP van be different than the result under IFRS or regulations of other countries (for example due to different approach to depreciation).

Many countries calculate tax excluding accruals to stop fraud. When you have the cash now and need to pay tax next year, you could as well "run away with the money". Or even not run with the money, just spend it - and then end up unable to pay tax for two years.

Technically this fraud could be committed at any moment, but when the cash would lie on your account for a year, you could think abiut it more often.

Also conceptually: if accruals are allowed, you can basically shape your tax result to never pay any tax? Make bigger accruals every year.

>> if accruals are allowed, you can basically shape your tax result to never pay any tax? Make bigger accruals every year.

That's not really how they work. You don't get to just make up a number (at least we dont). You can only defer income that's for work not yet done. For work you did do you get taxed.

But I appreciate all the replies, and it goes to show how tax policies change from place to place. Which is why you don't get advice from the Internet- you get it from local accountants and lawyers.

It might be a naive question, but why don't they just keep printing the book, if it was a hit?

There is a section on the website that says: "The book is almost sold out. A small remainder of books will be available to order in late 2023 or early 2024"

I was looking for some beautiful books about technology, so I'd love to buy the book, but I can't, and now I have to hope that I won't miss that email.

Or is it just a sales tactic like booking.com's "there is only one room left" banners?

A book like this is likely not print on demand, and the cost to do another run is high enough compared to rate of sales that it’s not directly worth the risk.

You have to make 10k books at a time and warehouse them, for example. Selling 100 a year wouldn’t cut it.

This feels like a perfect use case for a RFP marketplace that I think the world desperately needs - and then perhaps there can be a few individuals or organizations who would do such small orders, and maybe that's all they do - and they get enough work where it's probably not as cheap as mass printing 10,000 but at least it could get done and at a reasonable price; maybe with a "by donation" option for individuals who feel compelled or able to support the manufacturer even further.

It would be difficult to manage the quality. That's OK if the price reflects the quality, but if I'm to spend €100 (say) on a large photo book like this I don't want thin paper, imprecise binding, or other corners cut.

It's likely that you have a weird situation where making ten books costs about the same as making a thousand, because ten would have to be done via quite expensive printers and a bunch of hand setting, but a thousand is done via complex machines that once setup, are relatively cheap to run.

As an aside, print-on-demand from Amazon and others has been pretty amazing for black and white text, if you quality check it, but it quickly goes sideways for hardback color books.

People have even reported bad printing on simple text, but it's not clear if that's bad files (as the user, you have quite the leeway in what you send them).

Indeed, you'd have to have very clear specifications and perhaps sample prints that can be ordered to get samples of various qualities available; and then public review/rating system would be necessary for people to build up cred.

You can easily print far, far smaller batches. That’s not the issue.

The article literally talks about how the printer had never seen a project like it before.

If the small run costs the same as doing a 10,000 book run, it may have been easy to print but is likely not economical to sell.

Project manager and editor of the book here! Marcin has posted a lot of updates and newsletters about the printing process. It was fairly arduous to get the quality for the print run we did. Marcin hasn't disclosed the total, but you can tell through the Kickstarter campaign it was at least 4,500. Each of the book set’s hardcover volumes comprise 38 large sheets of paper printed on both sides, then 10 for the softcover volume 3, then sheets for the endpapers and covers and slipcase. Each of those sheets takes from an hour or two to much longer with the degree of attention we paid.

We spent about 100 hours on press with the printers across about eight days, sometimes 12 to 14 hours a day in July with 80°Fs and 90°Fs outside (the printing plant is conditioned somewhat for heat and humidity, and sometimes we had to stop printing for the day as it was just too hot).

It was glorious and exhausting. But we were almost at our physical limits for overseeing the print run and our printers were close to their ability to simply house the raw, printed, and then bound materials. This is an intensely physical object to its scale and the number of pieces.

Of generic paperbacks, sure; this is a collector's item multi-volume hardcover photobook though.

Ok that’s a fair point. You can still do small batch, but I agree, the results wouldn’t be anymore more than what you’d get from a laser printer and a glue binding.

Are you saying it’s cheap to print and bind small batches of high-quality books with color photos?

The setup cost and batch size are likely both too large to justify it without a large number of prepaid orders.

It might actually be easiest to run such a second printing as a second kickstarter campaign.

No need to actually do KS, just organize it as a group buy: If enough people commit and put down a deposit, the run is funded and it happens. If not just unwind it.

Isn’t that effectively almost exactly what Kickstarter does?

It’s a fundamentally different problem because you’re not, from either side really, worried about the classic Kickstarter “funds raised but we can’t keep our promises” issue. You probably already have a working relationship with a print house, print ready source files, and on the financial side you probably know exactly what all your cost basis is, e.g. setup costs of $x, $y printing costs per copy and fulfillment costs of $z. Kickstarter’s fees aren’t nothing.

As others have said, these things take a lot of upfront cost, so they have to be done in batches. Actually there's quite a lot of products that are done this way, I would argue pretty much anything that takes "pre-orders" for a long time instead of "order now and get yours in a day or two", for example the Playdate console.

Thank you for these amazing comments and insight! One element I might not have emphasized as much in the linked piece is just how much of "team" you accidentally assemble. I project managed this for Marcin, but he hired other folks (proofreader and indexer) and commissioned some photography and 3D illustration. But beyond that, likely 100 people across our printer, the outside bindery (which did the hardcover binding), and the slipcase maker (an intensely handmade product as much as flawless as it might seem) were involved in bringing it to life. Then the many 1,000s or 10,000s of people who handled the books across the shipping process in the USPS and worldwide. Lots of people involved whenever you start working with pesky atoms.

In regard to determining pledges/pricing tiers. Rather than go by a few universal rules of thumb, it might be better to look into what people had been prepared to pay in the past for crowd-funding projects similar to yours.

A few years ago I helped a friend raise 300k on Kickstarter for a board game project. One thing I did was scrape (gently) pledge and backer data from ~100 or so projects that seemed to cater to the same kind of gamer he wanted to attract. Then I plotted to see which pricing range tended to bring in the most revenue.

It's still a judgement call in the end, because each project is unique, but at least you're not shooting in the dark.

This is an extremely good example of why any market research is infinitely better than no market research. It can inform product, technology, budget, and business.

Speaking of research, I'd love to know if anyone's ever done any research into why tabs nowadays almost universally expand to 4 or 8 spaces. Why haven't 5/10 spaces won out? Why not 3/6/9? Why is it powers of 2? It would seem like counting to 8 isn't much more difficult than counting to 10 or to 6. Each of them use less than 4 bits, which is a typical minimal cell of information even in the most primitive computers, if you look at right after the primordial stone age of holding one bit per tube or per magnet.

Early terminals - and the typewriters they were converted out of - did not have tabulation fixed to these amounts. Early code editors did not seem to have such tabstops either. So what gives? Did at some point everyone sit down and decide that tabs were 4 spaces, like in some sort of UN meeting?

I've been trying to get an answer for this for years. Maybe decades. A good answer still eludes me.

Fun topic pivot... =) the emacs `lua-mode` has three spaces as its indentation default and it still blows my mind on fresh emacs installs.

https://github.com/immerrr/lua-mode/blob/d074e4134b1beae9ed4...

Maybe look into the number of characters that can be shown in a line. With 5/10 tabs and 80 chars per line you need to break code into two lines more often than with 4/8.

Although modern mobitors and resolutions are huge, so you could have 200 character lines. But arent they harder to read than shorter lines?

nice try, but code wasn't even indented back when 4 space tabs became uniformly accepted as the only standard.

100% agree. We looked at comparable books in terms of how they were pitched to an audience, size, quality, features (slipcase, color, binding, etc.), and went rather low on the retail price in the end, hoping to make it up with quantity, which happened!

Definitely looking at comparable projects helps you figure out potential final dollars you might raise, average rewards, and distribution of rewards.

I just received my copy. Truly one of the best books in my collection. I’m looking forward to years of enjoyment.

The memory game is more enjoyable and less challenging than it may initially appear. https://shifthappens.site/#test-keyboard-layout-knowledge

I would like a giant book about virtual worlds in games. Games that have a free-roaming element. So things like Zelda: OOT as well as full blown open world games like GTA V. There's so much work that went into them, and also so much collective human memory of time spent immersed in those worlds. Wish someone would crowdfund that.

If you're familiar with the sandbox mmo EVE Online you might like Empires of EVE. They're very well put together and very thorough historical accounts of some massive and long term player driven content/conflicts throughout the history of the game. It was so much fun to read about why my side's leadership made certain decisions and about the other sides' perspectives of some of the space wars I was part of when I still played.

https://www.empiresofeve.com/

Oh man, I'd really like a copy of that and I see they have a sale on ($40 for both hardcover volumes), but then... $50 for international shipping is a bit steep.

It's not at the same scale but I recommend the zine called Heterotopias.

https://heterotopias.itch.io/

There's a book called Designing Virtual Worlds from 2003 that might be interesting to you; since it's 2003 it predates World of Warcraft and some of the information is dated, but other elements are timeless. It was re-released under a Creative Commons license in 2021: https://nwn.blogs.com/nwn/2021/08/richard-bartle-designing-v...

And another recommendation if you're interested in worldbuilding is the Encyclopedia Eorzea books from the Final Fantasy XIV game, three of them; they contain the full backstories, timelines, summarised in-game stories, locations, characters, races, etc etc etc. However, it's written "in-universe", not so much from a developer's point of view.

That said, you didn't specify which part of said games you would like to read about, I can imagine more information about the design and build would be interesting too, a look behind the scenes. The video game resources I mentioned get you there partway though.

I know this is definitely not what you’re talking about because it’s a website and not a book but it’s a link always worth sharing https://noclip.website/

My son has a 500 page book about Zelda BOTW. It's more of a guide book / walk through. It's incredible how much stuff there is in those games.

Hope it has a KTV and its descendant the space cadet keyboards. TTTT the space cadet was too mushy for me, but the 70s keyboards of the KTV (and most of the keyboards of that era) had perfect key travel and solid metal base plates. Selectric was great too; shame about eh PC keyboards which were OK but inferior.

I don’t see KTV in the index (or Knight TV), but the space cadet keyboard has a lovely photo spread and two pages describing its effect on vi and emacs, its many modifiers and math keys, and some other facts. There’s two other small mentions; a pair of photos comparing its depth to an thin Apple Magic Keyboard and a homebrew keyboard inspired by it.

I took a flyer on this kickstarter and I’m really impressed by the results. It’s a beautiful labor of love. Incredible design and detail.

Emacs was developed on the KTV in TECO on ITS, and the space cadet keyboard was only on the lisp machine. How did the latter affect vi, which ran on different hardware and has a different model? Or Emacs, really? I wrote an awful lot of code using it but as I said it was unfortunately spongy.

Does anyone make a keyboard where the caps lock button stays down so you know what state it's in?

Not really anymore nowadays. The Apple Extended Keyboard[0] did have that feature. Similarly, Cherry used to produce a lock variant of their MX switches[1] that could be used for a such purpose. However this switch really only ever saw use in industrial applications.

[0]: https://en.wikipedia.org/wiki/Apple_Extended_Keyboard

[1]: https://deskthority.net/wiki/Cherry_MX_Lock

The short answer, no. The state of Caps Lock is entirely in host software, to the keyboard it is just another key. To get what you’re looking would require some very expensive mechanical contraption to reflect the state of the LED. And then what if you have reassigned Caps Lock, as is common? As cool as it might look it’s going to be a very limited niche.

USB HID does have codes for locking versions of Caps Lock, Num Lock, and Scroll Lock¹. OS X might support these, given that older Mac keyboards had locking Caps Lock. Linux doesn't².

¹ https://www.usb.org/document-library/hid-usage-tables-15

² https://elixir.bootlin.com/linux/v6.7.3/source/drivers/hid/h...

Having written USB HID code for barcode scanners where this came up (thrilling I know). None of the major OSes support it including OS X. It’s sort of a homage to a historical artifact (that table dates to the early 90s) but useless because even if you use those old keyboards you will have to emulate with the usual scan code.

Barcode scanners are cool. I'm working on setting up a barcode system for my personal library and then probably home inventory. Do you have any recommendations?

I found 4 used scanners for cheap (Datalogic Gryphon GD4400).

I mean sure, in theory something else can activate caps lock. In practice, it's only ever activated through a button press on the keyboard used. Caps lock being activated by something else or being reassigned to another button is the edge case here.

A laptop with external keyboard is not exactly an edge case. In Windows the caps lock state is per session, not per keyboard and I believe this is the case with most Linux desktops, the Mac is an exception here. Even there, the other case you are missing in your truth table is button depressed when not connected (ie the initial state). None of the common OS have locking support out of the box.

> Caps lock being activated by something else or being reassigned to another button is the edge case here.

BS. There is more variety to the assignment of the caps lock key then I think you realize even if in isolation some of them are edge cases (input method switching is a big one). https://support.apple.com/en-euro/guide/mac-help/mchl84525d7... Windows has similar including chords in the standard configuration for Japanese on non Japanese keyboards.

I think pointing out edge cases while ignoring that the request is for something that doesn’t exist with no evidence of high demand is an edge case is quite silly.

I was going to say just get a Cherry MX lock and pop it in, but those have been discontinued for what seem to me to be obvious reasons.

Why does the keyboard send caps lock? I assumed it would either send capital character or lowercase character.

That’s not how it works because it is more flexible and sensible to do the translation of a physical key location to a logical function within the OS. While fancier keyboards do allow alternate key maps and macros, your basic standard keyboard is just a stateless matrix of numbers (scan codes). Any hardwired button state handling or latching diminishes straightforward programmability on the OS side.

Understand that you can configure any modern OS for any arbitrary keyboard layout regardless of what type of keyboard is connected. The keyboard itself has no concept of what is a capital letter and what is a lowercase letter, it’s just buttons. This makes sense, because at the end of the day touch typists don’t really care what is printed on the keycaps. If I want to use Chinese (eg pinyin), Japanese (which is a cluster of layouts), or Dvorak on a US keyboard why should it matter? This can be handled easily in the OS, the keyboard is dumb.

And the concept of uppercase and lowercase as it would apply to a QWERTY layout doesn’t make even make sense for many languages.

When you press the physical caps lock key you’re just sending a number just like the Q key or something. The OS decides what to do with it. I have virtually no use for caps lock personally so it’s a control key. Using it as a dual purpose key for input method switching is not uncommon. More generally another reason for not baking in functionality in the keyboard is provision of accessibility features such as Sticky Keys and Filter Keys.

As an alternative, my Surface Pro 3 has an LED on the Caps Lock key to indicate on/off. It is a terrible keyboard otherwise, but perhaps there's other keyboards out there with this feature.

Being able to feel the state is best. It means when you hit caps lock you know if it is in the wrong state and can hit it again.

It saves you from typing a couple characters and then removing them, fixing the caps lock state and re-typing them.

Ahh, my blind spot - tactile feedback. My hands aren't very sensitive to touch.

Hope they release an epub once the campaing winds down/get their pocket money.

I mean I've got one on arcade pixel based font design. Don't underestimate the market for a niche book/thing where the fans involved are nuts for it

A coffee table book about Keyboards that is also a functioning Keyboard /s

(we did talk at one point about incorporating a working keyboard into the book and decided it was madness on top of the complexity of a book this big)

Hehe. That's lifetime funding if you go buy some Bitcoin.

Considering that all that information could be accessible in an easier way online (and with better resolution), a book doesn't make much sense. They are heavy, you can't copy/paste, you can't zoom on the images, you can't share it easily... however, they are beautiful. I hope people keep making books for years - not only this kind of books (that are beautiful), but also paperbacks, etc. Books are a beauty on themselves, even if they are not the best format for some purposes.

I guess, however, we are going into a future that resonates with a quote from HBO's The Newsroom, that says that books are the new form of art. They will soon be like paintings.

You bring up good points against the physical medium (especially when it comes to note-taking), but I'd respectfully disagree.

Books I've read multiple times, the great ones, nothing beats a well-made physical edition.

The way light pushes off the pages, that smell...the book changes over time as you read it more. It almost takes on a life of its own with yours. And, if you need to be completely offline or Amazon decides to one day Google-drop your entire online library, that book will still be open to you.

Great work deserves to be supported. Buying a physical copy is my small way to show my support.

Not only that, a much read book is like an old friend when you pick it up again. The wrinkled edges, the spine giving away with ease when turning, the small tea spill stain you made on accident.

You can smell another book while reading an ebook, and those are readable offline

You can smell another person while caressing your significant other, but it probably won't be a satisfying experience in the round.

Are there scratch n sniff book smell stickers for ebook readers?

A book makes sense because lots of people want a book. Sounds like they are planning an ebook version as well, so that solves some of your problems with the format.

>you can't zoom on the images

I enjoy picture books, which went through a golden age of illustration that ended when 32 bit computer graphics became available. Often I look at scans of picture books, and the experience is disappointing compared to printed paper. It would help if I had a giant A3 size tablet, so that I could see the whole illustration at once at the intended size and hold it like a book, but even if the scan has good resolution there's still the matter of gamma or contrast or color balance. I don't want to adjust these to some theoretical perfection, I want to adjust the actual lighting under which I look at an actual picture. So maybe a gigantic e-ink tablet could be ... as good.

Searchability, it's true, is an advantage. Physical books offer an advantage in discoverability, though, because of the way flicking through the pages works: not only is it very fast, but you can see multiple pages at once while keeping the places with your fingers, utilizing basic monkey hardware to great advantage in ways which swiping hasn't yet replicated.

What would be really cool is if all of this information was readily formatted and readable on something like Wikipedia, with all of the sourced images, and then people could "print" their own personal book copies through a great printer. I can already order custom books for my toddler with his name woven into the books illustrations from places like Uncommon Goods.

It sucks there is no inbetween from <send webpage to desktop printers> and <buy a niche book that needed $750k in crowd funding to even exist>.

I'd love to be able to ship myself beautifully printed and bound books on all sorts of free information.

The original kickstarter was for $150k, not $750k. They went way past what their minimum target was.

The issue with doing print on demand is that it basically limits you to doing industrial inkjet printing.

“Real” offset printing will produce a sharper, higher quality result, and is more amenable to using nicer paper stock too.

You can absolutely zoom in on the images: You hold the book closer to your face.

Also, fundamentally printed images just look different than displayed ones.

a PDF on my tablet is going to disappear from view within days. A coffee table book that takes up space on a table or looks good on a bookshelf provides an evergreen visual reminder, but not in an annoying way like a phone notification or home screen shortcut.

The problem is epubs and the like didn't evolve shit since invention.

It's more that small e-ink devices are completely unsuitable for reading content that demands a fixed layout designed for large pages, often with multi-column text and lots of photos/artwork/diagrams.

Those are the best kinds of books. Epubs haven't evolved since 2010 because it's done it's job. It's provided a scaffolding for selling .epub files for $9.99.

I prefer physical books for this reason. If I'm paying $10, I'll buy a used copy over a DRM file.

Reply on news.ycombinator.com