Rixstep
 About | ACP | Get Stuff | Industry Watch | Learning Curve | Newsletter | Search | Test Drive
Home » Learning Curve

I Want a Gold Mac

By Brendon C Bleebwart. All rights reserved.


Get It

Try It

Look at this screenshot. Just look at it. Read the text. Absorb it.



It says the file 'could not be saved'. It then goes on to ask 'do you want to save your changes to it anyway?'

Which is it? Can you save the file or can't you?

There are no ambiguities in Unix. There is nothing but embarrassing ambiguities in Apple.

The text goes on to explain 'the file is locked'. What is 'locked'? You can search through Unix documentation for mention of 'locked' and you'll find nothing. There's no such thing. Files are open for reading, writing, running, or they're not. There is no 'locked'.

Yet this OS, a product not of the Unix community but of the hermetically sealed anti-Unix Apple community, states that a file is 'locked'.

Does anyone understand what that means?

Ask 100 Mac users if they understand and at least 99 of them will fumble for an answer. They don't know. They're too stupid.

So what does 'locked' mean? I'll get back to that. If you've followed this site at all for the past umpteen years, then you'll know. But I'll get back to it - I'll explain later. For now, let's look at this fabulous Gold Mac.

For it's the gold one I want. $999. That's enough power. One of my former bosses used to tell us to never buy high-end machines. That's what they did at Microsoft, he told us. The programmers at Microsoft used high-powered hardware and never had any idea how weak their code was for ordinary users. You need to see how ordinary users - how all users - fare with your shitty code, he told us. So the low end for $999 is good enough. In fact it's optimal.

But it has to be gold. That's just a fetish of mine. A gold Mac. For only $999.

$999 is actually a lot when powerful laptops can today be had for a third of that price. Apple computers (and devices in general) are very highly priced, and this for several reasons.

One is that research and development (R&D) for Apple products is always high. This is always true because Apple keep oversaturating their markets. Apple need to keep oversaturating their markets because their markets don't expand. Apple's markets don't expand because Apple don't let them expand. This is an artificial constriction. Bill Gates warned Steve Jobs about this on two occasions, once when Jobs was still at Apple, and once again when Jobs had moved to NeXT. Steve Jobs didn't listen.

So we pay a high price for the privilege of obtaining a gold Mac.

What, aside from the colour, is so good about a gold Mac? With this one it's the engine inside. To borrow from Intel, it's 'Apple M1 Inside'. This new gold Mac no longer has an Intel processor. That in itself should raise eyebrows.

Intel processors are jalopies. They're like single-cylinder lawnmowers. Compared, that is, to Rolls-Royce engines such as found in Motorola and PowerPC CPUs. Both of those latter CPUs have dozens upon dozens of on-board registers. This is important because moving things in and out of memory takes time and cuts down on processor efficiency. The Intel architecture is extremely impoverished in this regard. Intel has AX, BX, CX, DX, SI, DI, and so forth. Intel has an accumulator register (AX, so named because it's now 64-bit as opposed to the original 8-bit A register). CX is a counter register. CPUs need to be able to count stuff. SI is the 'source index' register. It's used to iterate through matrices. It has the counterpart DI register - the 'destination index'. Intel also has a register for the instruction pointer - this is the register which holds the address to the next instruction to be performed. Intel also has a register for the stack pointer - this is a register used for the 'stack', a part of memory used for passing data from one part of the code to another. Intel has a few registers at its disposal, but Intel has nothing like the army of data and address registers of the Motorola and PowerPC CPUs.

IBM's PowerPC CPU was also the world's first CPU capable of executing all instructions in a single clock cycle. CPUs operate after a 'drum beat' of a sort. They have a sort of 'clock' which tells them to perform the next instruction. But some instructions take more than one clock cycle to complete. Traditionally Intel memory transfers - from CPU register to RAM or back - would take twelve (12) clock cycles. IBM's PowerPC could complete them in one (1) clock cycle.

The original PowerPC had a relatively slow clock speed. This was noticed and commented on at the time. Physical constraints made this necessary at the time. But, as anyone could attest who'd taken the time to study the PowerPC schematic, it was only a matter of time before the PowerPC left the competition in the dust. The actual design was nigh-on perfect. It couldn't be made better. Intel had been beat. Intel was yesterday's sensation, the home of the world's first complete on-board 4-bit processor, the 4004. But that was so far back that, in the rear view mirror of history, it was at most a tiny speck, and getting smaller all the time.

But marketing was and is a big thing. Perception is everything. Windows computers were running at faster clock speeds. Just as lawnmowers run faster than Rolls-Royce. One must keep in mind how astoundingly stupid Joe Mac User is. The marketing departments know and never forget. Apple lost the clock speed war. There was another war as well, a decisive situation.

For IBM didn't want to go further with their Power architecture. They had other ideas instead, but those ideas created rather 'hot' processors, much too hot to be used in a laptop computer. IBM could sell these new processors for games consoles and, for them, that was enough. To continue their research and to try to create comparable laptop CPUs would have cost them another $250 million in development costs, and no one save Apple would ever use such a processor, and Apple, with their neurotic market constrictions, were never going to use that many 'cool' CPUs. So IBM told Apple 'no' and Apple had to look elsewhere. And Apple finally got in under the covers with Intel, the company they'd insulted for so many years, correctly pointing out that their own Macs, running PowerPC CPUs, were vastly superior.

Nothing was ever wrong with PowerPC. The 'wrong' was Apple's alone.

The move to Intel meant additional security threats. Apple's use of the PowerPC meant Mac users enjoyed a level of protection not available to Windows users or even Linux users. Malware payloads are often written in 'shell code', direct machine code CPU instructions. But the PowerPC doesn't speak (or understand) 'Intel', thankfully, so Mac users were impervious. Up until Apple's move to Intel.

Apple's new move away from Intel should be a good thing, but the Rosetta 2 translator, which takes Intel code on the fly and turns it into Apple M1 code, might be a way for hackerdom to get around that.

Other than that, other than the unacceptably high price and the barriers to entry into Apple's ecosystem, it could be a good investment to get a gold Mac. But those barriers to entry are indeed daunting. If you're planning on marketing software for Apple's platforms, there are a number of things you have to think about before making that thousand-dollar leap. As for those not involved in software development: who are you? How many are you? Aren't you - your thin demographic - part of the reason that prices are so high?

But for those of you planning on making a killing in the Apple fields, there are myriad things to consider.

1.

How big is your market?: Most likely you'll target mobile devices. Software for the Mac has never and will never find a market. That market essentially doesn't exist. There are a great many great developers out there, but few of them see the Mac as a platform of the future. There was a time when Apple looked to be the 'Great White Hope', but that was only a marketing ploy. The 'open' nature of Apple code was a chimera. Open Darwin project leader Rob Braun closed down in disgust. He finally understood he'd been used by Apple's all-powerful marketing to create the illusion that Apple had become part of the open source movement. Nothing could have been further from the truth. Without the great inventive minds from the wider community, the Mac will never be a platform of promise.

2.

So you're going to sell software for Apple's mobile devices? Take a good long look first. Their market is small and aggressively constrictive compared to Google's Android. Google today has become, for the mobile, the Microsoft of old for the PC. There are few hardware constraints, few constraints in general, and the market is huge - five times the size of Apple's. Think about that.

3.

Constrictions. If you're a sucker for punishment, you might want to try creating software for Apple's iPhone/iPad. Basically you won't have a chance even if you do get something online, as Apple will only work with and promote the 'major players', and you'll get lost in the rubble. Given all the constrictions, with half the Unix APIs being outlawed because they make Apple look bad, you'll be hard put to find anything worth bringing to market. You simply don't have the creative elbow room.

4.

Sustainability. Years ago, my colleagues at Rixstep contacted a major Mac house in the US. This was back in the day when they and others still saw Apple as the unexpected road to safety and freedom for Windows refugees. They met in person with people from this reseller and happened to be there when a member of Apple management in Cupertino popped in for a visit. This upper-echelon Apple manager, described by them as decked out to look a complete clone of Steve Jobs himself with buzz cut, designer beard, black polo, denim jeans, and white trainers, was actually told by the reseller to come back later, as the question on the table was more important. The question on the table? How to equip an office with Apple system-ware, software, and hardware - and nothing else. How to configure an Apple network using Apple products and nothing else. And so forth. The reseller ultimately told them that he'd get back to them later. He never got back to them. For you cannot, to this day, equip a company of any size, large or small, exclusively with Apple products. There's no such thing. Not that you'd want to of course: Apple won't give you an exit strategy. You keep using Apple products or you abandon it all. And that's the chief reason the corporate world will never embrace Apple.

For your Apple investment, there's no exit strategy.

Software that runs on Unix platforms may run on an Apple platform, but there's no guarantee that it will, and, more and more, there's less likelihood it'll run well, if at all. Herein lies a serious security threat as well.

Software that runs on a Mac will mostly not run on another Unix. Those other Unix platforms generally have weak GUIs and aren't even in the same ballpark as the Mac.

Acknowledgements

The key to Apple's insecurity can be found in the file /Library/Documentation/Acknowledgements.rtf found on every Mac.



Super-hacker Charlie Miller, formerly of the NSA, who's won prizes by the carload for cracking Apple computers and devices, who also showed how Michael Hastings could have been assassinated, has made it clear why Apple platforms are so vulnerable. It's all in Acknowledgements.rtf.

The world of open source is a good world. It's built by people who collaborate on projects they can use in common. Ultimately a group will take charge of a particular project, and changes and submissions to that project will go through a vetting process. The Linux kernel is perhaps the most famous example. Other well-known examples include the FreeBSD OS on which NeXTSTEP was based.

The essence of open source is that one group can assume full responsibility for a project and everyone else can leave it alone. If a bug or vulnerability is discovered, then it's reported to the project's group and they take care of it. No one else need worry about it, and, most importantly, no one else need waste their time.

Not so with Apple.

Apple continue to go their own way, to their own detriment and to the detriment of their fanatical customers who pay for this indiscretion. The price they pay - the damage - can be seen in Acknowledgements.rtf, as Charlie Miller realised.

All you need to hack a Mac, explained Charlie Miller, is read through Acknowledgements.rtf. Acknowledgements.rtf is a list of all open source components used by Apple. But what you'll discover, said Charlie, is that Apple's versions always lag behind the rest of the industry. Oftentimes far behind. Why? Because Apple can never take any of those components that everyone else uses off the shelf and use them the same way themselves. Why not? Because at Apple they have to first make them more 'Mac-like', whatever that means. And before the dedicated skunk group can find the time for that silly task, new bugs and vulnerabilities are inevitably discovered.

All Charlie had to do was compare Apple's version numbers with the project's version numbers. If he saw a discrepancy, all he had to do was check the project's change logs. There he could find the vulnerabilities described in detail - vulnerabilities that Apple still hadn't even looked at, much less remedied.

That frightening gap exists to this day. Leave it to the hackers to tell you where the dangers are today, but suffice it to say that they're there, they'll always be there, so that Apple's version of Unix, to the extent it can still be called a Unix, is comparatively wide open to exploit. The one thing that saves Apple is their neurotic attitude towards marketing. As Bruce Schneier more or less expressed it: 'Nobody gives a shit about Apple'. No substantial market share, nothing to see, folks, move on and leave them to their own Mac-like devices.

Do you still want to develop for the Mac? You can perhaps be forgiven if you've been around a while and remember the glory of NeXT. The NeXT hardware platform suffered from Steve Jobs' neurotic obsessions but, once the dust had cleared, the true jewel of that project, as Wikipedia once described it, was the software and the NeXT approach to software development. Build it in layers, as Doug McIlroy or Brian Kernighan would have said, let the other guy do the hard work. Stand on the shoulders of giants. Collaborate.

NeXT built their stuff on FreeBSD. It's hard to assess today so long after the fact, but it's likely that chief engineer Avie Tevanian saw no reason to muck about with that code. If it works, don't fix it, as the famous IBM motto goes. What Avie and his team did was build a magnificent layer on top of FreeBSD, a layer that no one had yet conceived of and no one to this day has come close to emulating, far less surpassing.

The modularity of this layer was beyond comparison, right down to the semicolon. The kind of structure that, for example, was totally lacking in Microsoft's Windows API, save for the parts that IBM wrote of course. The layer was actually divided into two layers, or frameworks as they called them. Shared libraries. What Microsoft called 'DLLs' - dynamic linked libraries.

At the very bottom Avie had the Foundation framework. This was a collection of abstract functions, written in the Smalltalk language Objective-C which turned out to be perfect for the task. Everything here was abstract and broken down into logical steps. How to store a Unicode string. How to construct an array. And so forth.

Then, on top of that: how to change a string, how to change an array - how to add to or remove elements from it. And so forth. And still none of that is visible anywhere. For that you needed the AppKit framework. This is where the abstract Foundation became reality - visible reality.

In terms of structural beauty, it's got to be the greatest achievement in computer science since Ken Thompson started work on his own OS in Murray Hill.

But what's left of that today? The NeXTies weren't finished when Redwood City moved to Cupertino. And the world's changed since then as well. But the nihilist way that Apple engineers have plucked at and wrecked that body of work has no parallel.

Apple will survive. Big companies always do. They almost always find a way. They throw plumber's pipe at things when they need to. The disintegration sets in when there's no longer a way more plumber's pipe can be added.

So is it a good move, Apple abandoning Intel? Absolutely. Intel is not good, never has been good. Not since Motorola/PowerPC came in.

Should one run out right now to purchase a gold Mac? Perhaps not. See what happens to the early adopters first.

And, finally, is there a future for the gold Mac, for Apple in general?

At least it's gold. And that much is cool. Otherwise you can ask Reese and Jennifer.

 - BCB Oyster Bay November 2020

PS. Re locked files. There's no such thing but Apple made it into a thing. What Apple did was bastardise the NeXT code to make it possible, except it doesn't always work. Rixstep pointed this out for years, and, some ten years later, Apple acknowledged this in a slight change in their diagnostics, but they never fixed the underlying issue, and most likely never will. Their users are too stupid to notice, and too afraid to complain.



Note that the diagnostic is still wrong - it still leads users in the wrong direction.

Fuck 'em.

About Rixstep

Stockholm/London-based Rixstep are a constellation of programmers and support staff from Radsoft Laboratories who tired of Windows vulnerabilities, Linux driver issues, and cursing x86 hardware all day long. Rixstep have many years of experience behind their efforts, with teaching and consulting credentials from the likes of British Aerospace, General Electric, Lockheed Martin, Lloyds TSB, SAAB Defence Systems, British Broadcasting Corporation, Barclays Bank, IBM, Microsoft, and Sony/Ericsson.

Rixstep and Radsoft products are or have been in use by Sweden's Royal Mail, Sony/Ericsson, the US Department of Defense, the offices of the US Supreme Court, the Government of Western Australia, the German Federal Police, Verizon Wireless, Los Alamos National Laboratory, Microsoft Corporation, the New York Times, Apple Inc, Oxford University, and hundreds of research institutes around the globe. See here.

All Content and Software Copyright © Rixstep. All Rights Reserved.

CONTACT INFO:
John Cattelin
Media Contact
contact@rixstep.com
PURCHASE INFO:
ACP/Xfile licences
User/Family/Business
http://rixstep.com/buy
About | ACP | Get Stuff | Industry Watch | Learning Curve | Newsletter | Search | Test Drive
Copyright © Rixstep. All rights reserved.