The Decline of the Xerox PARC Philosophy at Apple Computers (2011)

This article does not ring true to me at all. The problem with making a big fraction of users into programmers is that we don't know how to do it. The LRG at PARC and then various groups at Apple tried every way they (we) could think of, and other ways have continued to be invented and tried. So far none work. Hypercard was indeed the most accessible development environment, but only a small fraction of Hypercard users ever wrote code (maybe 5%).

Apple has continued to make user programming of Macs as easy as they conveniently can (as far as I can tell having been out of there for a long time). iOS has a different goal -- making devices safe and usable -- which is intrinsically in conflict with maximum programmability. Even on iOS there are plenty of "user programmable" apps.

I think part of the problem here is that many developers take themselves and their developer friends as "typical" but that is totally not the case. This really treats most of the population with (unintentional) disrespect.

I vividly remember the reaction of my fellow students. Given the mockery and jokes from my fellow students, you'd think they were watching a bad sci-fi movie. Most students discounted everything they saw, 'real men and real programmers used C'. I remember being so disheartened that it seemed we'd evolved so little in tools/languages from 1979 - 1995.

At the time, everything was Unix and C programming (DEC Alpha were just being installed on campus, Windows 95 had just been released). There were a lot of reasons Unix/C succeeded, there is a great classic paper about why C beat Lisp, and I agree with the author.

However, what always troubled me, is how my fellow students completely ignored any potential learnings from those videos. In many ways, those early Smalltalk programs were far more impressive than anything they had created, but they just wrote them off.

At GDC 2014, a post-mortem was presented on the classic game Myst. That was written entirely in Hypercard.

Back then, Kay saw simulation as the killer app. He saw discrete-event simulators as a business tool. They had a hospital simulation, where patients went in and went through Waiting, Examination, Surgery, Recovery, Rest, and Discharge. This was visual, and you could click on the little patient icons and get something like "I am a victim of Bowlerthumb". Smalltalk was a descendant of Simula, which was a simulation add-on for Algol. So that was a natural direction.

It turns out that very few people use discrete-event simulators as business tools. Although you can model your simulator or bank branch in a discrete-event simulator and try to fix bottlenecks, nobody does this. That was a dead end as a concept.

Xerox's commercial product, the Xerox Star, was very much locked down. It came with a set of canned applications, and was intended for use by clerical staff. I don't think it even ran Smalltalk, but was programmed in Mesa or Cedar. It competed with a forgotten category of products, shared-logic world processors. These were low-end time sharing systems with dumb terminals tied to a server machine with a disk, used for word processing. Wang was the leader in that field. Those, too, were very locked down.

Back then, hardware cost was the huge problem. Kay said they could build the future because Xerox had the money to spend. (Xerox stock hit its all-time high in 1973). Each Alto cost about $20K at the time. Apple's first attempt at an Alto-like machine was the Lisa, which was a good machine with an MMU and a hard drive. But it cost around $10,000. The Macintosh was a cost-reduced version, and the original 128K floppy-only Mac was a commercial flop. It wasn't until the cost of parts came down and the Mac could be built up to 512K and a hard drive, with the option of a LaserWriter, that it started to sell in volume.

What made the Macintosh line a success was desktop publishing. "Now you can typeset everything in your office" - early Apple advertising. It sold, because it was far cheaper than a print shop.

Before the Macintosh, there was UNIX on the desktop. Yes, UNIX on the desktop, with a GUI, predates the Mac. Three Rivers, Apollo, and Sun all had good workstations with big screens in the early 1980s, before the Mac. The Three Rivers PERQ launched in 1979, five years before the Macintosh, with a big vertical CRT like the Alto. All these had some kind of GUI, generally not a very good one. Those were the first descendants of the Alto. They were used for engineering and animation, not just word processing.

We can bash Apple that they did not turn the masses into programmers at the same time, but it’s far from obvious this is even possible (on the contrary). And, at least for me, a hackable software device is a device that can be broken more easily, thus compromising the first and most important principle of usability. It’s so refreshing to be able to tell people that they don’t have to worry about the device, because there is almost nothing they could break, no matter how hard they try.

Also, this sounds awfully condescending:

I think one of the main consequences of the inventions of personal computing and the world wide Internet is that everyone gets to be a potential participant, and this means that we now have the entire bell curve of humanity trying to be part of the action. This will dilute good design (almost stamp it out) until mass education can help most people get more savvy about what the new medium is all about. (This is not a fast process). What we have now is not dissimilar to the pop music scene vs the developed music culture (the former has almost driven out the latter -- and this is only possible where there is too little knowledge and taste to prevent it). Not a pretty sight.

It’s like someone has got the whole computer interaction thing sorted out and is just waiting for the rest of the idiots to catch up. With all respect to Alan Kay, I’m not buying that.

https://www.youtube.com/watch?v=ubaX1Smg6pY&=

Someone asked Alan Kay an excellent question about the iPad, and his answer is so interesting, and reveals something very surprising about Steve Jobs losing control of Apple near the end of his life, that I'll transcribe here.

To his credit, he handled the questioner's faux pas much more gracefully than how RMS typically responds to questions about Linux and Open Source. ;)

Questioner: So you came up with the DynaPad --

Alan Kay: DynaBook.

Questioner: DynaBook!

Yes, I'm sorry. Which is mostly -- you know, we've got iPads and all these tablet computers now. But does it tick you off that we can't even run Squeak on it now?

Alan Kay: Well, you can...

Q: Yea, but you've got to pay Apple $100 bucks just to get a developer's license.

Alan Kay: Well, there's a variety of things.

See, I'll tell you what does tick me off, though.

Basically two things.

The number one thing is, yeah, you can run Squeak, and you can run the eToys version of Squeak on it, so children can do things.

But Apple absolutely forbids any child from putting a creation of theirs to the internet, and forbids any other child in the world from downloading that creation.

That couldn't be any more anti-personal-computing if you tried.

That's what ticks me off.

Then the lesser thing is that the user interface on the iPad is so bad.

Because they went for the lowest common denominator.

I actually have a nice slide for that, which shows a two-year-old kid using an iPad, and an 85-year-old lady using an iPad. And then the next thing shows both of them in walkers.

Because that's what Apple has catered to: they've catered to the absolute extreme.

But in between, people, you know, when you're two or three, you start using crayons, you start using tools. And yeah, you can buy a capacitive pen for the iPad, but where do you put it?

So there's no place on the iPad for putting that capacitive pen.

So Apple, in spite of the fact of making a pretty good touch sensitive surface, absolutely has no thought of selling to anybody who wants to learn something on it.

And again, who cares?

There's nothing wrong with having something that is brain dead, and only shows ordinary media.

The problem is that people don't know it's brain dead.

And so it's actually replacing computers that can actually do more for children.

And to me, that's anti-ethical.

My favorite story in the Bible is the one of Esau.

Esau came back from hunting, and his brother Joseph was cooking up a pot of soup.

And Esau said "I'm hungry, I'd like a cup of soup."

And Joseph said "Well, I'll give it to you for your birth right."

And Esau was hungry, so he said "OK".

That's humanity.

Because we're constantly giving up what's most important just for mere convenience, and not realizing what the actual cost is.

So you could blame the schools.

I really blame Apple, because they know what they're doing.

And I blame the schools because they haven't taken the trouble to know what they're doing over the last 30 years.

But I blame Apple more for that.

I spent a lot of -- just to get things like Squeak running, and other systems like Scratch running on it, took many phone calls between me and Steve, before he died.

I spent -- you know, he and I used to talk on the phone about once a month, and I spent a long -- and it was clear that he was not in control of the company any more.

So he got one little lightning bolt down to allow people to put interpreters on, but not enough to allow interpretations to be shared over the internet.

So people do crazy things like attaching things into mail.

But that's not the same as finding something via search in a web browser.

So I think it's just completely messed up.

You know, it's the world that we're in.

It's a consumer world where the consumers are thought of as serfs, and only good enough to provide money.

Not good enough to learn anything worthwhile.

http://www.digibarn.com/collections/systems/xerox-8010/retro...

Compare that with a classical Smalltalk UI:

http://research.microsoft.com/en-us/um/people/blampson/38-al...

The programming language of the Xerox Star system was 'Mesa'. Which was much more like what Apple used: Clascal, Lisa Pascal, and then Object Pascal, ...

http://en.wikipedia.org/wiki/Mesa_(programming_language)

How strong the self-defence mechanisms would be if this system succeeds at allowing anyone to be equally able to program anything on the system?

Wouldn't be way "safer" for the statu quo of the IT community to hide implementation power behind the walls raised by the C learning curve?

Interestingly, the divorce between technicians and users/consumers is steady reversing the trend.

I did a double-take after reading that line before I noticed the publish date.

See for example a tweet from Bret Victor a few days ago:

https://twitter.com/worrydream/status/560519372641677312

Or this article by Alec Resnick:

http://alecresnick.com/post/84756789325/from-bicycles-to-tre...

Their argument can be roughly summed up as "We were on the right track with the constructivist approach to HCI in the 70s/80s, but capitalism ruined everything and now we just watch Kim Kardashian on our tablets". There's of course some cynical truth in there- I also spent my grad school years reading Papert and Piaget and Kay and Resnick (Mitchell, not Alec), and I also find their vision of personal computing very enticing. And in many ways I orient my work to fit within the frameworks they built- I have nothing but deep respect for them as academics.

But I don't buy the whole "tablets and phones are just mind numbing dumb entertainment devices". In the past 5 years, I have seen:

- my younger brother start an internet radio from scratch from a bedroom at our parents, that rose to thousands of listeners

- my grandmother use the internet to communicate with her geographically distant friends and family

- teenagers making short movies in their backyard, or learning how to compose music

- kids discovering themselves a passion for photography, able to retouch photographs without needing access to a dedicated dark room, all with a hundred dollar device (when I was a kid, a hundred dollars barely got you a walkman).

- illustrators empowered to work from their home and make a living by working with clients many thousands miles away

- and many, many more.

None of that would have been possible even 15 years ago. Our tools are much, much better - Garageband and Photoshop and iMovie and Fruity Loops and others offer the means to do things in your bedroom that would have cost tens of thousands of dollars and required hundreds of square feet of free space a few decades ago. Sure, you have some purists that might argue that Instagram is to a dark room what a McDonald's burger is to Kobe beef - but I take issue with that as well [0].

Naturally, there are people who only use these devices for watching movies and playing games. That's absolutely fine, and I'm not sure why computer scientists like to take such a haughty attitude towards that. Not everyone spends all their waking hours working on their next masterpiece; and in fact, even the people who produce masterpieces spend idle time doing unproductive things just on par with reading Reddit or watching silly YouTube videos.

Are computers the best they can be? Far from it, and many of us work very hard at it (including the people I quoted earlier). But the attitude that we are now in a tarpit of mind controlling devices and that the golden days of computing are behind us is deeply wrong. We have come such a long way.

[0] https://news.ycombinator.com/item?id=8856780

For example functional languages eschew the use of variables but I think that has set back their adoption because variables merely attach human readability to logic (like comments). Seeing a big blob of functions is great for brevity but disastrous for long term maintenance. A compiler can trivially treat variables (especially immutable ones) like macros and transform them back into the same syntax tree that pure functional code generates. To say that another way: why don’t functional paradigms suggest that we break long blocks of code into shorter blocks with variables? They seem to be discouraging human instinct when there is no cost in encouraging it.

Also if we take a step back and see that the elimination of globals gets us most of the way to functional programming, then the main thing left is the notion of time due to externalities like input/output. Imagine for a moment working in a lisp environment using REPL. When you type some code and hit enter, how is that fundamentally different than a monad? Well, it potentially triggers the entire syntax tree to be reevaluated which is actually more expensive than handling input with a monad. But metaphorically it’s similar - we can just think of monads as places where the logic can’t proceed because it’s missing a piece of the puzzle. If we were using functional programming properly and assumed that we had a near-infinite number of cores and flops, we’d quickly see that monads are the bottleneck.

We should be able to translate between global-less Smalltalk message handing and Lisp monads, and let the compiler optimize away any monads that don’t depend on external state. To me, that suggests that working at a level below human-readable/imperative is generally a waste of time. We should be able to select any code block and use a dropdown menu to select the language we want it presented in. I remember the first time this hit me and I asked myself if it was possible to write something in HyperTalk/Smalltalk/C (or any other imperative language) and covert that to Lisp with these conventions. The answer to me is pretty self-evidently yes. Going the other direction from Lisp to an imperative language is even easier.

A ramification of this is that if you converted a C loop that uses immutable variables and no globals to Lisp, it would be evaluated into the most minimal logic possible because analysis would reveal that (for example) the elements being iterated over have no side effects between one another. In other words, the compiler would independently parallelize the loop and derive something analogous to map/reduce. Why are we doing that by hand? Surely there has to be a better reason than brevity but I struggle to find one.

When it’s all said and done, I think the arbitrary distinctions we’ve made between computer languages are relevant to education and worker productivity, but in the end are fantasy. I would have thought by now that we would have dropped the pretense. If computers worked the way I had imagined they would by this point in history, a tablet would have at least 1000 cores (or drop the notion of cores entirely and go with functional hardware), the compiler/runtime would consider the latency between it and the other devices around it in the mesh and adapt accordingly, all processes from the kernel to userland services to executables and threads would just be functions with the minimum permissions necessary to do their jobs, and most importantly usage would be reversed so that users would write macros through the use of human language rather than try to figure out how to do a task by hand from a locked down sandbox. All this code would be shared out to the world through some kind of hybrid Git/BitTorrent so that the solution for how to perform some declarative programming task would almost always already exist. And all of that would be constantly evolving with genetic algorithms and other software agents.

The tablet may as well not exist, because with the world’s computing power at your fingertips, why is your dumb terminal any better than another except for eye candy? It’s reminiscent of idol worship. Kind of gives me the heebie jeebies actually. And tragically moves us further away from technoliteracy with each passing moment.

At FPL, the large Florida electric utility where I worked, we went all-in, received training, and with pricey consultants, built several client server applications using GemStone as the Smalltalk server.

What most developers loved about Smalltalk was the liveness of the system, in that a just-in-time compiler made it possible to develop in a debugging environment. You changed a line in a method, and immediately you could interact with the revised app.

In 1995, Sun Microsystems released Java whitepapers. When I saw those I thought - Java is Smalltalk with C syntax, but without the image-derived liveness.

FPL switched from Smalltalk to Java, gaining all the benefits of reusable code, and object-oriented libraries.

I miss Smalltalk syntax, in particular named method arguments, and to this day comment my Java code to compensate like this...

    sendMessage(Message.notUnderstoodMessage(
            message, // receivedMessage
            this)); // skill

A particular feature of the Visual Works Smalltalk implementation was a concept known as "Save to Perm" which, after a thorough garbage collection, moved the remaining long-lived objects, e.g. nearly the whole language runtime, to a memory buffer thereafter free of garbage collection.

Even after two decades, no Java implementation has this feature without resorting to off-heap storage.

IBM got on board the object-oriented development wave in the early 1990's and introduced Visual Age Smalltalk. They bought a Smalltalk version control product "Envy" and created the development environment that has over the years evolved into the famous Eclipse (IBM's Eclipse "blocks out" Sun Microsystem's Java).

Or put another way: yes, smalltalk addressed programmability, but it did not address usability.