The Minecraft Creator Markus Persson Faces Life After Fame | Hacker News Reader

The Minecraft Creator Markus Persson Faces Life After Fame

newyorker.com

148 pointsmitmads101 comments

> Computer magazines of the day would print strings of code on their back pages, which could be transcribed by the reader to create a playable game, and this code-by-numbers task gave Persson his first experience of what would later become his profession. “My sister would read the lines out to me and I would tap them into the computer,” he says. “After a while, I figured out that if you didn’t type out exactly what they told you then something different would happen, where you finally ran the game. That sense of power was intoxicating.”

It's kind of weird to think that no-one gets their start like this anymore.

How could they? You can't do this on your Wii et al. Modern home systems are geared towards a strictly consumerist experience, not a creative one.

This is one of the reasons the Raspberry Pi was created.

What about JS, or writing games in Haxe that compile to run in Flash (which, even if people hate it, is incredibly common).

I can even use AIDE on my phone to write applications for my phone!

I have no idea why anybody claims we have less programming freedom than in the past.

The claim is that it's less accessible for children, not that we have less freedom.

Modern computing environments are way more accessible (and far more readily available).

I'm willing to bet that tons of people get their start this way. Obviously not exactly that way, but in very similar ways.

Access to this kind of information is easier than ever.

All this doom and gloom is self-pitying bullshit.

One of the reasons I recommend Zed Shaw's "Learn Python The Hard Way" to people that want to learn is that it tells you to type out the code in the exercises. It's not quite the same as stumbling into it with the lure of having a game to play, but I think it probably has some of the same intoxicating power.

To me the weird part is that nobody gets their start sitting at a computer with somebody else. Or at least, if you do get a group of kids around a single computer, they're sooner going to watch YouTube (or play Minecraft) than learn to program.

nobody gets their start sitting at a computer with somebody else

My experience tells me that when you are just a bit beyond hello-world, it's way better to sit alone and maybe even disconnect from the Internet. I was starting with nothing more than a book on Delphi and the help system of Delphi, which included Win2k SDK, and feel that only by this way you could develop good debugging skills. I know some people who are learning software development, and the single skill they need badly is troubleshooting complex problems on their own and not relying on colleagues.

edit: now I see that my point is about sitting together with someone more skilled, but you're talking about kids leaning together, which is totally different.

I think one of the most underrated way to learn how to code is YouTube videos -- you hear a lot about books or interactive websites, both of which are great -- but I love following along with random heroes who document the start-to-finish process through video. Cherno's Java game lessons have been amazing thus far -- http://www.youtube.com/playlist?list=PLD1A2E956F00ECC48

That is where, i think, the raspberry pi excels. I bought one for my 11 year old niece and we got busy immediate g with minecraft and python with it. Lots of fun and crazy code has been had. Plus she looks forward to weekends when we sit down and code new stuff for it.

Sure they do. Look at game modding. Replacing sounds with other sounds, changing models are simple forms of that power.

This story is incredibly common amongst my age group (40) in tech. What is the common story amongst those that are 30? 20?

I'm 25, and I started when I was 6 or 7 using QBasic tutorials from around the web (accessed through a blistering 14.4k modem via AOL 3.0).

But I also started earlier than most folks I know.

I found QBasic on Windows 98 CD.

I was bored in class and programmed my graphing calculator.

This was my main introduction as well. (I'm 21.) I remember looking through the TI-84 manual and reading through the commands to see what I could do.

I started when I was 18 in college. 35 now

I'm 19. I started when I was 14 by hosting and modifying a private server (in Java) for a popular mmorpg.

On my playstation 2 I had a demo disc that had a YaBasic compiler. I wrote game code with my controller. I used other people's source code to make pong and pac-man, and had little understanding of what I was doing - but still managed to change the colours. It also introduced me to manually checking the syntax of the code. http://members.iinet.net.au/~jimshaw/yabasic.me.uk/faq/

I am 26 and got started with programming mainly through QBasic and to a degree DOS batch scripts before that.

20 years old, started at 8. Father taught me VB6, got addicted to programming, and built simple games for a few years.

I couldnt say for commonality, but for me writing scripts in notepad for a game emulator to make new items, using a series of tutorials and articles that had been turned into a .chm file because their coverage was so good.

24. Even though we always had a computer at home, I never got into programming and spent most of my time playing games. (The learning curve was too steep, and besides, what could I possibly make that was better than the games I was playing?) Then when I got to college I majored in CS. Now I'm definitely interested in making games. :)

In my opinion, the abundance of dynamic and interpreted languages, powerful tools, programmable hardware, and the internet make getting into programming WAY easier than it was before.

But even you - a smart capable person who likes computers - didn't do so because you already had all the games you needed.

Those home computers? You load a game, or you start programming. Since loading a game often meant fiddling around with a tape cassette and cable, and a few minutes hoping it would load, it's easy to see why people decided to try to code themselves.

I know abstraction is a good thing. I know it's powerful and etc etc.

But there's something nice about being able to squirt data to an address, and know it's coming out the parallel port, and having a hokey resister-ladder DtoA converter hooked up to turn that data into music. Or to have a single instruction to draw a pixel.

As a counterpoint, I'm 20 and I started programming exactly because I was playing around with extending a game (Neverwinter Nights was the game), which involved using the built-in tools to make new levels/worlds but getting elaborate behaviour required scripting in a C-like language.

I remember being utterly confused by what a statement beginning with "while" did, but eventually I was reasonable enough to mostly implement a (very weak) checkers-playing AI.

This then lead to programming Lego robots in C, which was pretty neat. (Actually, thinking about that, the first programming I did was using the graphical LabView thing that Lego provides for Mindstorms robots.)

23. I started when I got bored of the games on the computer my parents had at the time, and I started opening random programs and eventually ended up in the QB3.0 editor. After early attempts at using it as a word processor failed (it kept complaining about syntax errors when I moved the cursor off the line) I asked my dad, and he showed me how to do

  PRINT "Hello from the Computer"

And I worked my way up from there with the help system, and a lot of trial and error.

It helped that our parents limited us to 30min/day of playing games, but were okay with me spending more computer time programming (or playing my own games, though I rarely did that. Making them better was way more fun than actually playing them).

I don't remember how old I was at the time, but it was definitely I was using Windows 3.11 (it was installed, but you couldn't run any decent apps when all your RAM was used up by win). By the time we got a Win95 computer (which in fairness was probably in '98 or '99) I was pretty decent at BASIC.

I continued to play around with it 'til I started learning C++ in maybe 2003? Luckily prolonged exposure to BASIC doesn't seem to have done any lasting damage :)

29. Started playing with JavaScript in 7th grade. Took C++ in 10th grade. Studied visual effects in college. Moved to web after college.

I suppose I can't really speak for everybody but as a 20-something the story a lot of my friends and I had in common was programming our TI calculators. The language was simple (some might say...basic ba dum chh), it was easy to share programs with friends as text or via link and you had a built in packed audience of folks in classes that thought the things you figured out how to do were really cool. We started with simple school problem based programs then went on to more complicated things and then some went on to learn some c or assembly etc.

I think perhaps we might not be starting programming with computers directly all the time, but perhaps there will always be some object that younger folks will learn to program. Heck maybe the next kids will decide to program phones or connected home appliances or something.

I can honestly say TI calculator programming was my first foray into programming as well (I'm 25 now). I remember meticulously typing in a BASIC clone of Pac-man and having it not work the first few times I typed it in (I think it would take me nearly an hour each try). I didn't really program much after I stopped using TI calculators a lot until someone showed me tryruby.org and I found it interesting. Now, two years later, I work full-time as a programmer and have decided this will be my career. Funny how those little side diversions when you were but a kid can come back in a big way.

22. I started at 7 or 8 with Stagecraft Creator, which is a visual programming-ish thing. Then I moved onto to BASIC with Learn to Program BASIC. I have a friend who started with LTPB too.

mid-30's, I have the same story, started programming when I was 8. Had these nice colourful British introductions to programming with cartoonish robots all over the place and full BASIC programs to type out.

23. My foray into coding started by playing with game's level editors (such as Tomb Raider, Elder Scrolls and Warcraft 3). There was a lot of visual abstraction, but you could go deeper down if you wanted to. Then onto Game Maker. It was actually cool to see all my 'code' going from visual blocks in it, to just using the built-in script (over a series of games). Learned a lot from it!

26 here. I didn't get into programming until a couple years ago.

As a kid, I was never really exposed to programming. It wasn't offered in middle school or high school. It just so happened that I was never prompted to try it. I majored in something non-technical, and didn't write my first 'for' loop til I was 24.

Which is a shame, because as a kid I totally would have loved it. If only there was one class, one teacher, one person, somewhere, telling me to try programming -- I would have tried it, most likely, if someone pushed me.

Oh well. As a grown up, I've discovered programming, I have a job doing programming, and I love it.

30 here. For me it was websites and discovering that you could see the source code. I thought it was amazing and instantly wanted to learn how to make one. I found out my ISP offered free web space, stumbled on to Arachnophilia, discovered Usenet and haven't looked back since (this was all around 1996/97).

I started at age 6 or 7 by trying to mimic the command lines printed by my parents' first PC XT's AUTOEXEC.BAT (which lacked "echo off", for which I am eternally grateful), then checking out adventure books from my elementary school library where part of the story required typing in and running BASIC code.

I was also given an Atari 130XE by an uncle that booted to BASIC, explored GORILLAS.BAS when DOS 5.0 came along, etc.

24. Dad worked at HP, so that helped: started at around 6-7ish years old with a 286 that had Qbasic. Moved on to QuickBasic 4.5, then some other Basic variants (Rapid-Q and even CorelDRAW's scripting language), C, PHP and JavaScript, then Python, which is what I still mostly do.

> It's kind of weird to think that no-one gets their start like this anymore.

I've gotten bug reports for Hackety Hack where people say "I can't copy and paste from the lessons. Can you fix that?" I reply (obviously at more length and warmth) "Working as intended."

Reminds me of Zed Shaw's Learn Code The Hard Way approach:

> You must type each of these exercises in, manually. If you copy and paste, you might as well just not even do them. The point of these exercises is to train your hands, your brain, and your mind in how to read, write, and see code. If you copy-paste, you are cheating yourself out of the effectiveness of the lessons.

When I was 14, I started learning C from a book that built around this premise. I think it was called "Type & Learn C." It worked - by the end of the summer I was a somewhat proficient C programmer.

I still think it's kind of amazing how such simple advice can be so effective. I often find that when I'm struggling to grok a new language or framework, that I've forgotten to hand-type at least a few examples.

Here's my current working hypothesis.

The part of the body that learns how to program[1] doesn't speak English. You hear it even in the metaphors programmers use: code smell, this "feels" like a recursive problem, "listen to your code," etc.

That means to learn as quickly as you can you want as many parts of your body involved -- the English-speaking part, your muscles, your eyes, your memories of all the times you've made a stupid mistake only to have it cause you hours of frustration down the road, etc.

[1]: Really, the part of the body that learns anything doesn't speak English. English is just the data exchange format. ;)

See neural network deep learning :-)

Alternatively, maybe you should get your English-speaking part out of the way and let your brain "absorb" code patterns as you read or type them, without overthinking it. https://www.youtube.com/watch?v=HzR8x5MgvDw

I don't know about that, but you do need to at least interact with the code on some level, eg. by cut+pasting, then playing around with the code to make it do different things.

Also, I don't know many professional programmers who type everything in by hand. It'll be something copied from a previous project, or cribbed from the documentation / somewhere online, then hacked into shape.

Of course, in order to do any of that you need to understand what's going on, but there might be easier ways to do that than manually typing in code.

Professional programmer != person learning to code.

Professional programmers do things in certain ways because it's more efficient. It doesn't necessarily follow that you should change that just because someone is just starting out.

Just because they're learning to code, doesn't mean that you should put artificial hurdles in front of them.

I'm teaching my kids to read and write both english and code this way. Something magical happens when you "store and forward" that doesn't happen just with reading and writing drills.

My pet theory is that it has something to do with information compression in the brain, kind of like Huffman coding, except you keep the table for life.

Yesterdays Magazine-typein is todays git clone.

I have a feeling Markus is referring to typing in machine code (in a totally opaque way) using DATA statements combined with a small loader written in Basic.[1] Here's an article discussing this approach.[2] This was very common in the computer magazines during the 80s. You learn a lot about patience and attention to detail, if nothing else. There were attempts to mitigate this..[3]

[1] http://en.wikipedia.org/wiki/Type-in_program

[2] http://www.atariarchives.org/mlb/introduction.php

[3] http://arstechnica.com/business/2012/08/experiments-in-airbo...

But there were plenty of magazine and book listings that had very little in the way of machine code and that were mostly regular program code.

Don't forget that DATA statements just contain comma separated constants, and they were useful for a lot of stuff, not just POKEing a byte into memory. Manipulating these constants gave some people their first taste of 'reverse-engineering' level data. :-)

And the fact that people were typing in machine code, to directly manipulate the hardware, is pretty impressive. It's even more impressive if they went on to learn more about assembler in order to modify the code somehow. Instruction manuals for the machines included information about machine language programming.

From the article: Since the game’s release, in 2009, Minecraft has sold in excess of twenty million copies, earned armfuls of prestigious awards, and secured merchandising deals with LEGO and other toymakers. Last year, Persson earned over a hundred million dollars from the game and its merchandise. Persson—better known to his global army of teen-age followers by his Internet handle, Notch—has a raggedy, un-marketed charm. He is, by his own admission, only a workmanlike coder, not a ruthless businessman. “I’ve never run a company before and I don’t want to feel like a boss,” he said. “I just want to turn up and do my work.”

----

He sounds like a modern day Woz, except he was successful despite not wanting to get involved in business. (by that I mean, Woz was successful, but we don't know how he might have turned out if Steve Jobs didn't push him to cofound Apple)

Reminds me of Veblen's "The Instinct of Worksmanship"

He has something more easily translatable to a product, just add a login system. Woz had blueprints and a few of his own, a physical product with massive costs for each one.

What is the longevity of Minecraft? As a social sandbox game, people can play for a long time. We have a habit of treating games like movies, good for one time through, and perhaps most are. But what if some games are more like Legos, playable for years? Instead of a franchise (King's Quest I, King's Quest II, ...), what if there is some other extendable model, like buying a new Lego kit? I have no idea, but Minecraft is one of a handful of long-lasting games that suggest a different business model entirely. The MMPORGs are like this, too. I'm curious to see what they do with Minecraft Realms.

I just got into Kerbal Space Program. It's another massive sandbox game by an independent developer that's now teamed up with two other devs to build out the game. It's like Minecraft for space rocket enthusiasts.

I love the idea of Minecraft but never got into it, but KSP is irresistible to me. I just can't get enough of it. I wonder what it is about these open ended sandbox games that favours independent devs. Maybe it's the fact they don't rely on a massive ammunt of pre generated content, or even really a designed experience even. Instead they are based on a suite of components you then create content with yoursef. They're really toolkits, just as Leggo is.

> What is the longevity of Minecraft?

There's an unofficial modding API and thousands of mods [1], some of which are extremely sophisticated. For instance, "Red Power" [2] includes an in-game 6502-based computer for which the author wrote a Forth interpreter that can be used to control in-game objects, and somebody else has embedded a Lua interpreter that can do similar things, IIRC.

There are also dozens of user-created game modes (some of which only have informally-enforced rules) and special maps like "Feed the Beast" [3].

The only games I can think of with comparable amounts of content are MMOs and possibly the Elder Scrolls series, though in both of those cases the content is nothing like as varied, nor does it rely on its user's creativity to anything like the same extent.

Minecraft is very much like Lego, only you can also program your own bricks.

[1] http://www.minecraftwiki.net/wiki/Mods

[2] http://www.minecraftforum.net/topic/365357-125-eloraams-mods...

[3] http://feed-the-beast.com/

The cool thing about a lot of the mods is that they don't just add more content to the game, but they add whole new ways that the game can be played. In vanilla minecraft you can build all sorts of structures, and that's very cool and has a lot of room for a creative mind to explore. But then there are mods that add the ability to build automated factories. Or robots that you can program to do almost anything the player can, like gather resources. Or movable structures which allow you to build bases that can fly around the world. Or bees that you can selectively breed or even genetically engineer, which will produce useful resources for you.

This struck me as extraordinary:

“My strongest early memory is of my dad dragging me through very deep snow on a sled,” he said. “I looked up at him and he seemed annoyed at me. Perhaps it was tough work, dragging me, or perhaps I had been crying. And I realized that—hang on—he’s actually a real person, with his own perception of things. It’s not just me looking at things; he is also looking at things.”

I wonder if it's rare for someone so little to experience empathy so powerfully. Certainly I don't recall realizing anything like that until much later.

The whole story about Notch and his father is very moving.

My wife and I have had so much fun playing Minecraft with our two sons and their friends. It really is special.

We have a Minecraft server running on an older Mac mini in the house, and my kids’ have friends — some local and some from other states — that log in and play with them. Sometimes they are all on Skype, talking to each other while they play. It is so cool, and fun to watch them all interacting in their Minecraft world together.

> “I have the ability to get code done, but I’m impatient and it’s scrappy as a result. Maybe that helped me with Minecraft, as it came quickly. But, well, at some point, I’d like to actually become a good programmer.”

It's amazing that he doesn't consider himself to be a good programmer. In what other profession would someone as accomplished as Persson say that?

From programmer's perspective, Minecraft is nothing special (and it's written in Java!). There is much better code out there.

Quality of programming and success are not necessarily related.

Have you actually seen him coding? From what I have heard/seen on video, he is very fast, his code is readable and relatively concise (for Java). And efficiently displaying this giant fully-modifiable voxel world on all types of hardware is no small feat. Also the procedural landscape generation is fairly brilliant. And the whole crafting system that allowed for world modification. I think you should give him credit for doing a great execution of some really cutting edge ideas.

Watched him live coding August 2011 -- it was incredible. I know there are better programmers, especially on a systems level, but it's still just... wow.

http://www.twitch.tv/notch

Not wanting to take anything from notch, but I do concur with the parent. I've seen his live casts, and sure they are great, but let's not kid ourselves, his code is not that great. He is fast cause he knows the libraries he uses very well (which is admirable) but at the time, put me in front of DX9 and my speed was the same. Now put me in front of iOS code and I can probably match it. I think he just got to that status where people idolise him (heck, I remember when his first livecast was out, 100s of blogs doing their own 'what I learn from watching notch code' like he was some super coder, which weren't much more than typical stuff you learn in the first 2-3 years working at a company(at least in games).

And from other sources (can't verify) minecraft code is reported to be quite bad, which is the main reason many of the bugs weren't fixed very fast.

Having said that, he had a vision, executed it well and made a lot of money. My hat is off to him for that.

That's probably true. Just to expand on this idea: Notch's priorities when he wrote Minecraft were quite different from most of our priorities as an employee, I would wager.

I dunno about you, but writing truly readable, intuitive, easy to maintain code is quite difficult. And you'll likely never get it right the first time. Refactoring for readability takes consultation from other human beings, and lots of time.

And when you're writing software for a business, it's really important to put in that time. You know what you're writing is going to last five or ten or more years and will be seen by dozens of people. It _needs_ to be good.

But for a guy working independently on a novel project? Fuck all that, I say. Just iterate as fast as you can towards the vision that you have while keeping some minimum standard of readability/maintainability. It doesn't need to be very good code, it just needs to be good enough that you can go back to it later without being completely confused.

A musician might.

There aren't that many professions were a decent piece of work that makes one person happy can scale up to make a million people happy with minimal marginal effort.

Excellent example. Most famous rock musicians aren't extraordinarily talented at playing their instruments - Richard Wright of Pink Floyd couldn't even read sheet music. Jimmy Page is quite a sloppy guitarist.

Pure technical ability is a dime a dozen. Check YouTube for a zillion teenagers who can play shred guitar as well as anyone. But inspired creativity is rare and valuable, and execution requires only a certain level of technical ability, not mastery.

Jimmy "I'm not the greatest guitar player" Hendrix. https://www.youtube.com/watch?v=6gwLQAuHJv8

I thought he was charged with murder or something.

Yeah same here. It's obvious that this kind of headline could be interpreted as sensationalist, but perhaps it was just incidental. We'll never know.

Yup. Felt the same way.

no - it's abundantly not trying to be sensationalist.

Life typically means just that - it's not some shorthand for a term of imprisonment.

Not sure if sarcastic?

"Life typically means just that - it's not some shorthand for a term of imprisonment."

Can you show me an actual statistics on what percentage of "X faces life" in English texts relates to severe criminal charges, and what percentage means something "non-sensationalist"? Because as a non-native speaker, I've always seen it in the former context.

eh? It's all about context, isn't it?

‘Life after x’ never refers to a life sentence—neither I nor anyone will have a statistical reference to back them up on this assertion—it's just not needed.

shrugs

- ed

well, i suppose 'faces life after quadruple homicide' would.

Where the heck would anyone find a corpus of text to study that phrase?

Oh, https://www.google.com/search?q=faces+life+after

(Incidentally, Google's de-duplication logic fails horribly on that query. 6 of the top 10 results are the same blogspam-copied story, and it's not even fresh news.)

Haha, what do you mean 'no'? That I didn't think that?

sorry, i meant to 'reply' to VexXtreme - i figured you were simply being light hearted.

Same here. (Faces life in prison.)

Another finely written newspaper article about indie game development / mindsets involved, titled "Where Do Dwarf-Eating Carp Come From? / The Brilliance of Dwarf Fortress", centering around the two brothers, mainly focusing on Tarn Adams:

http://www.nytimes.com/2011/07/24/magazine/the-brilliance-of...

Compared to the other story of a $100M GOOG Exec, this is a story I really like, as a programmer.

Apples and oranges.

How so? The GOOG exec story is about how to get rich by climbing the "Corporate Ladder", while Persson's way is to become successful by going indie and producing games people want. The interesting fact is that their wealth are comparable number-wise.

I guess I wasn't really focusing on the wealth as it seems to me to be only mentioned in an effort to get people's attention. I think both stories are fascinating.

Agreed on the link-bait factor. Though the perfect timing of these two stories got me think why so many great programmers go indie: if you are great at building stuff, try to be next Persson, or if you have great vision, PG will be your model. Otherwise, you're screwed.

"Last year, Persson earned over a hundred million dollars from the game and its merchandise."

If you made that kind of money, why would you even bother to do anything but just retire and enjoy life?

I guess he enjoys programming and making games.

Boredom?

I'm not saying it doesn't ever happen, but often the people who have this kind of success don't get there by aiming for retirement.

In his case retire and enjoy life might be the exact same as what he's doing now. Programming minecraft :D

>If you made that kind of money, why would you even bother to do anything but just retire and enjoy life?

>“When you have the kind of success Minecraft has brought, you can just choose yourself the way you want to do things,” said Persson. “I don’t want to feel like I’m in charge or anything... I try to have a studio where people go to make games for the fun of it, not just because some investor has said we have to make money.”

He pretty much is retired. Making the games he wants to with other people is how he enjoys life. Profit is not at the top of the list.

Sounds awesome to me!

In some sense, enjoying life is hard when you have lots of money. It gives you lots of options, and you cannot do all you would ever want to do.

Say, you are a programmer, and you are enjoying your work. Would you enjoy yourself more riding a Ferrari today? Sitting on some Caribbean beach? Following your favourite NBA team by going to all their games? Study medicine? Start a charity? How do you know you would stay happy doing that? Would you feel guilty spending your life lying on a beach, where you might have written some useful software instead?

Most people would shrug of those ideas or not even think about them because "that's life". If you have a hundred million in the bank, though, all of these are realistic options.

If money is no objection, some feel that they really are themselves to blame if they aren't happy for a minute or don't accomplish anything they feel valuable in the rest of their lives.

He sounds pretty sad.

Most people are pretty sad when their family members commit suicide.

http://kotaku.com/minecraft-creator-we-did-not-pay-anyone-to...

Wow, some real schadenfreude going on there.

I really don't understand some of the bile that follows Persson around.

Are people really that fucking bitter that a guy who by his own admission is a complete klutz (programming and business-wise) can become so successful?

Either they're bitter or extremely gullible. Or both.

I think that from the programming side he is just too modest, not klutz at all. Maybe he is also good at business in some way he don't even realize.

Hmm, I didn't really see any bitterness in the article (even in the tweets they list). Are you talking about the comments?

Yeah sorry, should have made that clear.

Someone tagged the picture of the DJ with the comment "This just in : Mojang might have paid a famous DJ to help people feel more inclined to party. This is disgusting."

Apt.

And exactly why do you think that's apt?

To me, it seems like a superficial comment that misses any of what actually matters about the topic for the sake of a smug joke.

Reply on news.ycombinator.com