gregdoesit

A few interesting pieces of information on the publishing industry, as shared in this trial and article:

On profits:

> PRH’s Dohle said that the top 4 per cent of his books produce 60 per cent of his profit. The best books, he says, are the ones “that you don’t pay a lot for and become runaway bestsellers.” He excuses the fact that so few of his big bets pay by comparing himself to a venture capitalist. The book industry, he says, is “the Silicon Valley of media.”

On marketing:

> The big houses are skeptical about book marketing. One publisher said, “I don’t think marketing money can create a success,” it can only amplify or extend a success. That may be true but the big houses, we also heard, are not really trying. They spend little on marketing: roughly 2 per cent of expected revenue per book. (…) Some agents ask publishers to guarantee how much they will spend on marketing a book before they’ll sign. Most won’t.

On advances:

> about 70 per cent of all the money paid on advances by the Big Five goes to $250,000-plus books. Judge Pan asked if $250,000-plus books comprise 70 per cent of Big Five book sales. The answer was no, not even close.

(Disclaimer: I both self-published books and also write a paid newsletter on Substack)

The author of this article makes the point that Substack pays ~90% in revenue, versus traditional publishers which do ~10%.

While this is true, I do think the author seems to deliberately avoid mentioning the self-publishing platforms like Gumroad, Podia etc which allow authors to keep 95-96.5% of revenue when publishing books, and also have a direct connection with people who bug the books through emails. I say this as someone who published through one of these platforms. (Also, there are other newsletter providers which charge even less of a commission, like Revue charging 5% or Ghost or ConvertKit charging a monthly fee based on the number of subscribers, which could come out to below 5%)

A real (financial) pull of publishing a book via a monthly, paid newsletter is the higher revenue per customer. A customer won’t pay more than, say, $40 for a book. But they’ll more likely pay ~$120 by paying $10/month for a year. See also how Steven Sinofsky is publishing his book series Hardcore Software on Substack with the option to pay up to $300/year to get a final, printable PDF edition of the series once finished. I paid $100 to read this series for a year and one of my friends went for the $300 option.

Holding on to publishing rights is a huge difference when going self-publishing or newsletter-first. When I worked with a traditional publisher, only then did I realize I have no rights to publish my words without publisher consent: this was when I asked if I can share drafts of my writing in public, and the publisher said this is down to approvals on their end. This realisation was a reason I ended up self-publishing instead. As the author of the article notes, by keeping publishing rights, it’s always an option to sign a contract with a publisher for a print book.

What is not written in the article is how self-publishing makes a lot more (financial) sense once:

1. You have enough of a reach to not benefit much from the - sometimes surprisingly little - marketing that a traditional publisher does

2. The association with a publisher brand is no longer a priority.

3. You don’t need a publisher helping you finish a book pushing you with deadlines (they sure help!)

4. You are ok paying for - or doing yourself - editorial things (development editing, copyediting, proofreading)

When I took my kid to Disneyland for the first time, I got the question on the way “dad, can we go to Netflixland as well?”

This question from a kid made me realize how much of an entertainment empire and revenue generator Disney is versus Netflix.

With Netflix, buying the monthly streaming service is the most you’ll spend. With Disney, it’s likely the least: the toys (with Disney royalties), the movie tickets for the family and finally - for some - a Disneyland trip where we dropped more over a few days than my lifetime Netflix spending will be.

The amount of investment that went into making Disney what it is today is incredible, and Netflix has a lot of catching up to do.

Will there ever be a ‘Netflixland’? Does Netflix even want to compete with Disney, head-on-head?

Seven years ago, I started a blog called The Pragmatic Engineer. For three months, I wrote a post every two-three weeks. But no one really read them. So I took a short break, which break turned into a month, then another, another… and I did not have motivation to write. Barely anyone read it, after all.

A few months after starting the blog, I saw a small traffic spike - maybe 30 visitors - from this site called Hacker News after a submission [1]. I never heard about HN before but it looked interesting so I started to visit it, and eventually registered to upvote stories I like.

Then, another few months later - when I had not published anything for 3-4 months - as I was checking out HN, I could not believe what I saw: a months-old post of mine was on the front page titled “Move Fast Without Breaking Things”. Traffic was so high, my site could barely keep up on shared hosting with MediaTemple, my blog running Wordpress.

There were 50+ comments discussing my post, versus the 2 that were on the blog itself (it had a commenting system). The comments were more insightful than I’ve ever seen for any of my blog posts. This was the HN post [2].

Now, this post did not “make me money”. But it WAS a turning point where I got validation that strangers on the internet find what I wrote interesting. And it gave me a motivation to write when no one read it, knowing that just because people are not interested in what I wrote today, people might find it interesting months or years later.

I kept writing my blog for years. Eventually, this blog served as the basis of The Pragmatic Engineer Newsletter which I launched a year ago, and which is my new fulltime job and one of the most popular technology newsletters with over 120,000 people reading it. More than enough people chose to pay for it to make this a a viable fulltime career on the long run.

Thank you, Hacker News, for those first two submissions, all the comments, and the motivation it gave to keep writing. It was a turning point. I still remember how I could barely believe in March 2016 that so many people I never met can be interested in reading thoughts I put in writing.

[1] https://news.ycombinator.com/item?id=10692734

[2] https://news.ycombinator.com/item?id=11374221

> “ Despite cease-and-desist actions against Forsage for operating as a fraud in September 2020 by the Securities and Exchange Commission of the Philippines and in March 2021 by the Montana Commissioner of Securities and Insurance, the defendants allegedly continued to promote the scheme while denying the claims in several YouTube videos and by other means.”

Even now, the site Forsage.io is live, and claims that 2,559 people joined the past 24 hours. Who knows if it’s true.

But how does this work: when a site like this is charged with fraud, can it operate until there is a verdict, defrauding more unsuspecting victims? Needless to say, visitors to the site will have know knowledge of this lawsuit.

Yes, how can we take it off? I’ve sent an email to dang.

I had no idea. Always assumed it was limited to these countries listed here: https://www.levels.fyi/locations/

The point stands that I’m building an Europe-focused site, being a former hiring manager based here. Eg in Hungary, the compensation data is in HUF for most companies with local contracts. On TechPays, this information is in HUF. On levels, it seems to be in USD.

It’s a different site. I’m building it because I’d like to have something with more details on this market and because it’s a very fun side project and an excuse to keep coding.

Working on more generic anonymized names! The goal is not to guess which ones these companies are.

Thanks for the suggestion! I do want to do a better job visualising things in the overview. I’ll think about how to do it, without being too confusing.

Thanks! This is a tricky one as you already pointed it out.

I’m voting for simplicity: capturing the number in the point of time, and also capturing whether the equity component has appreciated or depreciated.

My bet is that having the date of the submission plus data on whether this was a new offer plus the indicator of how the equity changed will give a good enough sense for the sake of using this data.

An interesting question here, on Hacker News.

Why did Paul Graham build Hacker News rather than encourage the use of Digg? It already had submissions of similar types that Hacker News started with. I assume it was because things were missing off Digg that Hacker News built, and the two sites ended up serving different audiences in the end.

Why build something new rather than encourage using existing services?

Because many of us have this entrepreneurial sprit to build, and do it the way we think is a good approach, filling in for gaps on the market. Time and people then decides if it was worth it.

The most obvious answer: TechPays caters for countries not present on eg levels.fyi - see how there are no data points for eg Estonia, Hungary, Belgium etc.

Then, this site is built with an anonymous-first mindset and offers additional context on data points - like whether or not equity is liquid - which I’ve not seen before, and has localization like being able to search for The Randstad in the Netherlands.

While there is data on levels.fyi, I expect this site to get significantly more data for European countries as it’s the main focus of it. Eg it already the has the most data for the Netherlands across any data source online - more than 2,000 entries -, and I have not promoted the site, this came from running in stealth. Same with eg Hungary which just launched but it’s hard to fine any data on this country - and impossible to submit data on most similar sites for it.

https://techpays.com/europe/netherlands

Thanks! Another reason this is still in alpha - I was not ready for someone to submit this to HN just yet.

Wow, how the world changed from using Google's CDNs to having to host it yourself.

I just pushed a fix to move the script to the local domain. It will be live in a few minutes.

I'm building the site (but I didn't submit this Show HN).

The private part of the shares are clearly marked as private in details, and there's an explainer eg at [1].

I do think it's important to include private equity in packages. My Uber stock package turned out to be worth a lot of money once the company went public. I now have friends at Databricks with massive packages - which can go up or down, and are dependent on an IPO. But this is very important information in my view.

[1] https://techpays.com/details/bbkzjx1e6mhj6ynm8yhhu

Hi! I'm building this website, but I did not submit this show HN here. I don't think the site is ready for such a submission, it's currently in alpha - very clearly marked. It's not even beta-ready yet!

The data set so far is small - limited to a few countries - and we're working behind the scenes. There are also lots of known issues, and a few data issues.

I'm hoping this site will help people in tech get a better sense on tech compensation outside regions where there's better compensation transparency like the US. I started this site after I saw the impact of my article on the trimodal compensation model had [1], but got many questions on what companies pay "Tier 3" or "Tier 2" packages in various countries.

So, to reiterate: this is heavily WIP, we've got lots of work to do, and we're planning to add more countries, one at a time. Eventually covering all of Europe, and then beyond! One day I hope it will be ready for a proper "Show HN" submission. That day is not today :)

Also, I reached out to dang to remove it from Show HN as it’s not submitted by someone who did the project (and not ready for such a submission)

[1] https://blog.pragmaticengineer.com/software-engineering-sala...

Thanks. The book ended up as 61,000 words - which is not that much compared to all-text books - but lots of pages of resume examples. A "normal" 200 page book could have up to double the number of words (100-120K words). Just want to add that detail, as a "normal" 200-page book in two months would be much more impressive.

I originally aimed for a ~40 page book - a PDF initially -, assumed it will have a few chapters at most. As I started writing and pulling in more advice from recruiters and hiring managers, the chapters added up.

In what is pretty usual for projects you do the first time, I under-estimated the scope, as well as the time it will take. My original budget for time was a month. Not too different from some software projects in a completely new domain :)

Just want to push back on the note that my newsletter "basically summarizes and quotes from a different paid newsletter as a primary source".

The last issue had the most number of The Information quotes [1] plus a bunch of original reporting and opinions.

The one two weeks ago had one such quote [2] and a bunch of original reporting.

I hadn't quoted The Information three months before, the last time I mentioned anything from that source was late April [3].

All linked issues are un-paywalled.

Point taken though on quoting from other publications and the perception a few such quotes can create in a paid publication, not to mention when these external quotes are how an issue starts. And especially when burying original or "exclusive" reporting from original sources.

I'm learning the ropes on this one as I go and will iterate on the format and the notion of "reporting on reporting".

[1] https://newsletter.pragmaticengineer.com/p/e0cb6dd3-0527-421...

[2] https://newsletter.pragmaticengineer.com/p/fff15093-0d83-41d...

[3] https://newsletter.pragmaticengineer.com/p/1d8672a5-a2ab-48e...

Thanks - I keep evolving the publication; The Scoop is barely a few months old and was (still is) a bonus addition to the long-form weekly articles.

To quote or not to quote from other publications is an good question. Re-reading the last two Scoop issues, they both, indeed started with The Information quotes as you rightfully noted. I didn’t mark the information that was unreported outside my newsletter clearly either. The ones that were all based on directly talking with sources at companies - like the internal controversy on Slack access at Stripe, compensation rises and the Board's role at Adyen, the Meta PSC process as it’s playing out right now, the details on Google’s hiring freeze as confirmed with managers at the company, the Graphcore hiring freeze, reporting on the upcoming Apple hiring slowdown - and then sharing when Bloomberg shared information on the same two weeks later.

Here is an un-paywalled The Scoop issue from two weeks ago [1] where the first article was, indeed referring to The Information as a source, the second news referencing Bloomberg - confirming earlier reporting - and the next six scoops mostly from my own sources.

I don’t like the idea of calling out “exclusive” details, but I can also see how quoting from other sources can de-value this publication. Also, the reality as a on-person publication is there will be relevant news I won’t get to, but might still be relevant to share as a summary.

[1] (removed paywall) https://newsletter.pragmaticengineer.com/p/fff15093-0d83-41d...

Thanks for elaborating. You're right: the title is misleading. I changed the title to reflect what I meant to write about: "Want to learn about entrepreneurship, as a software engineer? Sell something online".

And yes, your point is a valid criticism on "selling advice" without having been successful the space you write about. It's also a reason why I'm trying to steer clear on offering advice on running or scaling a business - since running my own business, ironically. I made one exception so far on this, sharing my thoughts on the term 'creator'. [1].

I also noticed how writing about business topics pulls in people who want to succeed in a similar way, and can go down the "publish self help content" in ways that you described. It's not appealing to me.

Appreciate the caveats on the publication! I'll also think of how to make it more clear to indicate my sources are direct and the information is "exclusive" without over using the "exclusive" part.

[1] https://news.ycombinator.com/item?id=32233437

The Pragmatic Engineer here :) And here is the list of side projects and their status I posted about you might have referred to [1]. I never posted about failed startup ideas.

As I wrote in another comment, this post is from late 2020, when I planned to start a startup, and when I had no idea I would be successful with my writing - it was maybe a month or two after publishing my first book. The context of this article was the skills I thought I'd use when going down the startup route.

On some assumptions of my recent writing:

> "I suspect a lot of the quotes where he says “I interviewed an engineer at ____ might be just be sourced from random Tweets"

When I write that I talked with someone at a company, it means I connected with them directly on various channels (e.g. over DMs, Signal, video call etc). I talk with lots of people, and incorporate some of these into the "The Scoop" articles and sometimes tweet about them (like this [2]). If I base something on a tweet, I'd share that. If I use something else as a source of information I share that as well.

If I come across things that are not true, I aim to correct it immediately (like this [3])

In my new career, integrity is the most valuable asset I have, and I aim to make it very clear when I share information that comes with directly talking with people, versus taking things from other sources.

> As with all content marketers selling advice, please take this all with a grain of salt.

Agree on this, and please see the context on when I wrote this article.

[1] https://twitter.com/GergelyOrosz/status/1481195527467319297?...

[2] https://twitter.com/GergelyOrosz/status/1549867691346956288?...

[3] https://twitter.com/GergelyOrosz/status/1534884178428366848?...

I'm the author of this article. Please note that 2020 should be added to the title, as I wrote it late 2020, after I left Uber and my plan back then was to start a startup some months later.

I had recently left Uber, and launched this e-book, and wrote this post with notes on what I thought might be useful skills as I go ahead with that startup.

Note that I did not go ahead with founding a startup, so most of this post should be taken as "skills I thought would have been useful if I launched a venture-funded startup."

I do think you can learn things better if you have some skin in the game. Specifically, learning about entrepreneurship can be easier if you start a small venture, even if on the side.

Also, as PragmaticPulp pointed out, the title is misleading given my advice is based on a book launch, not a startup. I changed the title on the original blog post to "Want to learn about entrepreneurship, as a software engineer? Sell something online."

Author of the article here and just came across this comment. Here's how I did it. By working at Uber in Amsterdam for 4 years where I made more than twice as much as a senior engineer as I did before when I was at Skyscanner, in London where I was a principal engineer.

Saved the difference and I had about 3-4 years of savings. Decided to "risk" 6 months of it and dip into it.

https://twitter.com/GergelyOrosz/status/1552615944446525442

Congratulations! There's something special about the feeling of you "earning" your first revenue that comes from an online customer you've never met, and not from an employment contract, consulting, or similar "traditional" relationships.

I'm not sure why, but when this first happened with me - it was selling a book I wrote online - it started to change my way of thinking, opening a realm of possibilities I never considered before.

Because if you can sell something to a customer for €10,72, that's €107,20 for 10 customers and €10,720 for 1,000 customers. And, unlike with employment, where there is a limit on salaries, there is no - theoretical - upper limit on your revenue.

Congratulations and hope you keep building your entrepreneurial muscle. Best of luck with the side project and beyond!

> “The sad truth is that the path of an architect or principal / staff engineer is closed for a pure android developer”

When I worked at Uber as an engineering manager, this was partially the case: at Uber’s scale, there was a path to grow to Staff-level just with pure mobile, but it was a crowded place and difficult to get to as a mobile-only engineer expert in one platform only.

I encouraged native engineers wanting to get to that next level to go broader - venture into areas like iOS or web or backend, to become a more well-rounded engineer and ultimately someone who can lead and influence wider groups.

If you don’t work at a company with large mobile platforms, switching stacks to eg backend typically has more path for growth.

I also wrote a book called Growing as a Mobile Engineer where my advice is basically the above: to grow beyond the senior level, either you’ll need opportunities to work at places with large enough mobile scale - typically ones with mobile platform teams -, or leave native mobile to further your career.

Wow, I never knew of dial.uber.com!! m.uber.com was built by an engineering team in India, but for global usage.

I formerly worked at Uber. I’m surprised that Bolt blocks the web app from working. Uber also built a web app which is incredibly lightweight - here’s a summary on how it was built [1]. It’s a life saver especially if e.g. at an airport and you pay for data over roaming. Uber doesn’t exactly advertise m.uber.com, but there’s no blocking of it.

My suspect is that Bold wants you to use their app as they can eg send push notifications, and gather more metadata from users - some of which can be used for eg more signal on determining if a request might be fraudulent.

[1] https://eng.uber.com/m-uber/

Here is a screenshot [1] of the page served, which indicates there's a DNS misconfiguration issue.

[1] https://twitter.com/GergelyOrosz/status/1533825620148469760?...

Additional context on the waste: gas fees are not only charged to those who have successfully minted an NFT - so those paying ~$5,500 worth of ApeCoin for the NFT plus ~$2,000 - $6,000 worth of ETH for the gas fee.

If a mint is failed because of the wallet ran out of ETH, those people lose all their ETH and get nothing in return. An outlier, but this how this person lost $6,900 of ETH to get nothing in return [1]

[1] https://twitter.com/existentialenso/status/15205751405636649...

While the post-mortem is thorough, it misses key details on what companies experienced who were unlucky enough to be caught out by this outage. For example, it fails to mention how impacted customers lost access to certain Atlassian services for up to ~2 weeks: JIRA, Confluence, OpsGenie. But not others like Trello or BitBucket.

Of these, losing access to OpsGenie for this long was a massive problem, dwarfing most other systems. OpsGenie is like PagerDuty in the Atlassian world.

I spoke to several engineers at impacted companies who could not believe their incident management system was “deleted” and had no ETA on when it would be back, or Atlassian could not prioritise restoring this critical system ASAP. JIRA and Confluence being down was trouble enough, but those systems being down for some time was things most teams worked around. However, suddenly flying blind, with no pager alerting for their own systems? That is not acceptable for any decent company.

Most I talked with moved rapidly to an alternative service, building up oncall rosters from memory and emails - as Confluence which stored these details was also down. Imagine being a billion dollar company suddenly without pager system: and no ETA on when that system would be back, your vendor not responding to your queries.

I talked to engineers at such a company and it was a long night to move rapidly over to PagerDuty. It would be another 7 days they could get through to a human at Atlassian. By that time, they were a lost customer for this product. Ironically, this company moved to OpsGenie a few years before from PagerDuty because OpsGenie was cheaper and they were on so many Atlassian services already.

The post-mortem has no actions on prioritising services like OpsGenie in reliability or restoration, which is a miss. I can’t tell if Atlassian staff are unaware of the critical nature of this system or if they treat all their products - including paging systems - as equals in terms of SLAs on principle.

Worth keeping in mind when choosing paging vendors - some might recognise these systems are more critical ones than others.

I wrote about this outage from the viewpoint of the customers as it entered its 10th day and it was discussed in HN, with comments from people impacted by the outage. [1]

[1] https://news.ycombinator.com/item?id=31015813

If you take away “universities where everyone wants to hire from but very few can”, I am personally dubious of what else there is left.

Also you might want to look to understand what went wrong at Triplebyte which wanted to provide a one interview-only experience for experienced candidates but it never worked. Companies still did their interviews. The company also had to pivot their business model from this and explained in detail why they did it [1]

Triplebyte was targeting a very valuable and hard-to-hire group of experienced engineers. You are targeting the easiest to hire group. Consider what will make you succeed as a business where Triplebyte could not, even though they solved for a much bigger pain point for hiring managers.

[1] https://triplebyte.com/blog/rethinking-triplebyte

It’s true that if you can say “you can hire from {top N colleges like MIT}” then this would be attractive. Trouble is, this won’t scale beyond the top 10: the ones which get far more recruiter reachouts than students.

The other thing that could be a sell is retention. Retaining new grads beyond a year or two is notoriously difficult. If you’d have data or proof that this approach significantly increases retention and its because of the group setting, that could make the premium laid worthwhile.

You still need to convince companies to trust your assessment method: not a problem while you only have new grads from the very top schools, but would be a problem if you open up to all colleges.