software

Apple’s Next Macintosh OS

by Jean-Louis Gassée

Operating systems don’t age well. Some have better genes than others or they have more competent caretakers, but sooner or later they are stricken by a cancer of bug fixes upon bug fixes, upgrades upon upgrades. I know, I lived inside two OS sausage factories, Apple and Be, and was closely associated with a third, PalmSource. I can recall the smell.
The main cause of OS cancer is backwards compatibility, the need to stay compatible with existing application software. OS designers are caught between yesterday and tomorrow. Customers want the benefit of the future, new features, hardware and software, but without having to jettison their investment in the past, in their applications.

OS architects dream of a pure rebirth, a pristine architecture born of their hard won knowledge without having to accommodate the sins of their fathers. But, in the morning—and in the market—the dream vanishes and backwards compatibility wins.

Enter the iPhone.

The iPhone OS, iOS, is a Macintosh OS X derivative…but without having to support Macintosh applications. Pared down to run on a smaller hardware platform, cleaned up to be more secure and tuned for a Touch UI, iOS is the dream without the ugly past. Tens of millions of iPhones, hundreds of thousands of applications, and billions of downloads later, this is a new morning without the hangover.

And now we have the iPad, another iOS device. (I’ll omit the newer Apple TV for the time being.) 8.5 million iPads were shipped by September, a mere six months after its introduction. The installed base will reach 14 to 15 million units by the end of this year.
To paraphrase the always modest Apple PR boilerplate phrase (“Apple ignited the personal computer revolution in the 1970s …”) the iPad re-ignited the marginal tablet category.

After more than 30 years of stalled attempts, the tablet genre has finally gelled. We see a flurry of tablet announcements from Asus, HP, Samsung, Dell, Archos, and many others, using Windows 7, WebOS, and Android. Surprisingly, we have yet to hear a pundit declare 2011 ‘The Year of The Tablet’. It’ll come.

On the other hand… Apple held a Back to the Mac event at its Cupertino HQ last week. As the name implies, Apple wants to make it clear that it’s still committed to personal computers. (You can see the full keynote here…but that’s 90 minutes. A tongue-in-cheek, adjective-laden 104 second montage gets to the essence here.) The iPhone may generate half of Apple’s revenue, but the event reminded us that Macintosh desktops and laptops are a $20B/yr business—a business that’s growing faster than the rest of the PC industry. Apple made a point of showing how the iPad, after taking its genes from the Mac, was feeding DNA back to its progenitor by way of the Touch UI that will appear in the release dubbed “Lion”, OS X 10.7.

During the Back to the Mac presentation, two prayers of mine were answered: A Macintosh App Store and a smaller laptop. The App Store has received the expected “walled garden” critique, but having seen how difficult it is for small Mac software developers to get retail shelf space or to make money selling their wares on line, I like the idea. A few days ago, I downloaded a neat little utility to silence the startup sound on my new 11” MacBook Air. How much did the developer make? Zero, it’s freeware; the programmer didn’t want to spend the time and money to set up a commercial site. How much would I have paid for it from a Mac App Store? Less than $5, more than 99 cents.

As for the 11” MacBook Air, Walt Mossberg, WSJ’s tech guru, penned an insightful review that’s neatly summed up in its title: “MacBook Air Has the Feel Of an iPad In a Laptop”.

So: A clean, fresh iOS; we’re not abandoning the Mac…What are we to make of these competing messages? My theory:

  • Today’s PC operating systems have advanced cancer
  • Personal computers as we know them are here to stay
  • Apple will move to something like an iOS Macintosh

Easier said than done. Steve Jobs remembers well the trouble Apple had getting apps for the first Macintosh, the painful failures of Lotus Jazz, the lame Mac software from Software Publishing Corp., creator of the best-selling PFS: series for the Apple ][. Ironically, some of the best software came from Microsoft—the word frenemy hadn’t been coined yet but retroactively fits. So, just like the iPhone App Store made the iPhone, the Macintosh needs a marketplace, an agora in preparation for the transition.

But a transition to what?

An evolution of the iPad? Certainly not something I saw at Il Fornaio, one of the local Valley watering holes. There, a very serious woman had her iPad standing on the official Apple keyboard dock, writing and, from time to time, raising her hand and touching something on the screen. As Jobs pointed out in the keynote above, it’s an ergonomic no-no.
Now, turn to the laptop. As one of my colleagues says: “It’s dark inside the box.” It’s what the machine does that matters, not what’s inside. Indeed. Imagine a port of OS X on an ARM, or A4, or AX processor, or even a Loongson CPU for that matter. If the right applications have been ported or adapted or, even better, created de novo for the platform —and made available through the App Store—would we object?

But, you’ll argue, “Aren’t these processors much less powerful than Intel’s?” Ask an iPad user: The machine feels swift and fluid, much more than a conventional PC.

Yes, there are no heavy-duty apps such as Photoshop or AutoCAD for the iPad. (AutoDesk publishes an AutoCAD companion app for the iPad and the iPhone.), but who knows? Adobe might be tempted to do for Photoshop what Apple has done for its OS: Scrap the past and build a modern Photoshop that’s written from the ground up.
Intel processors suffer the same type of cancer that afflicts operating systems. Their instruction sets and, therefore, their hardware, power consumption, and cost are beset by the tortuous need to stay compatible with existing code while offering an endless procession of new features. Intel has tried a fresh approach at least three times: the iPAX 32 in the early 80s, the Itanium (promptly renamed Itanic, a political compromise hammered out to keep HP’s PA architecture out of contention), and a brief fling with ARM called the XScale. Each time, the company (or the market) decided backwards compatibility was the way to go. Intel’s position is transparent: They believe that the might of their technology and manufacturing will bulldoze the cost and power consumption obstacles of the x86 architecture.

(We’ll note in passing that there is no Wintel in smartphones. For its Really Personal Computers, for its Windows Phone 7 devices, Microsoft is all ARM.)

Compare the bulldozer approach to what Apple did when it designed the A4, the “dark inside” of the iPad. Apple’s next Mac processor could be a multicore (or multi-chip) ARM derivative. And the company has proven time and again that it knows how to port software, and its support of the Open Source LLVM and Clang projects give it additional hardware independence. We all know the Apple Way: Integration. From bare metal to the flesh, from the processor to the Apple Store. Hardware, OS, applications, distribution… Apple knows how to control its own destiny.

Tomorrow’s MacBook Air might have even more of the “Feel of an iPad in a Laptop” that Walt Mossberg detected. The tablet and the laptop could run on the same “dark insides”, with the same software, and the same Touch UI interface. And, for a desktop machine, an iMac successor, we already have the Magic Trackpad for touch input.

(IMCO, the current Trackpad doesn’t feel magical enough: on the two devices I own, the touch input isn’t as reliable, pleasant and “second nature” as it is with existing mice or a laptop trackpads. I gave up after two weeks. I’m not the only one with that view, I’ve asked. And the local Apple Store doesn’t push appear eager to push the device either.)

All this doesn’t mean the x86-based Macs would disappear overnight: high-end Mac Pros, for example, might continue for a while as they do today for applications such as Logic Studio or Final Cut.

If this sounds farfetched, one question and an observation.

The question: Would you bet the longer term future of your $20B Mac business on an endless series of painfully debugged x86-based OS X incremental releases? Or would you rather find a way to move that franchise to a fresh hardware/software platform fully under your control?

The observation: Last week, the other Steve, Ballmer, was on stage at the Gartner Symposium. There, he was asked about Microsoft’s “biggest gamble”. Without missing a beat, as this forceful public speaker never does, he answered: “The next revision of Windows.” Not Windows Phone 7, not the Kinect game device, all near and dear to his heart, but Windows 8. (See here and here.)

He, too, is thinking about the future of the PC business.

JLG@mondaynote.com

PS: As I edited this note, I found this TechCrunch post dealing with the same iPad-Mac convergence.

HP’s Board Gets No Respect

.

And rightly so.

You recall: Last August, HP’s Board of Directors dismissed its wunder-CEO, Mark Hurd. Well-loved by Wall Street, although not so much by employees, Hurd turned HP around after the lackluster Fiorina years. He made acquisitions, cut costs, and put the company at the very top of the IT industry. But HP’s fearless leader was accused of having entangled himself, carnally and emotionally, with a female “marketing contractor”, and of having engaged in a few financial peccadilloes in the process of covering up the relationship.

I’ll hasten to add that Hurd reached an amiable—and solid—settlement with the former soft-porn actress. By “solid settlement” I mean we’ve heard exactly nothing from the aggrieved woman, or from Gloria Allred, her highly expressive Hollywood attorney. (As a self-described “Fearless Advocate for Justice and Equality”, Ms. Allred appears to dig gold on behalf of the rejected/dejected paramours of media and sports celebrities.)

While Hurd tried to do the right thing after his alleged mistakes, HP’s Board and management repeatedly and needlessly pilloried him, barely stopping short of accusing their former CEO of fraud. (See more sorry details in this Monday Note.)

All this led Larry Ellison to publicly lambaste the HP Board for kicking Hurd to the curb—and to promptly hire him as co-president of Oracle.

Ignoring the “when you’re in a hole, stop digging” maxim, HP doubles down and sues Hurd. Their complaint? As Oracle co-president, Hurd will inevitably misuse HP’s confidential information and cause his ex-employer grievous harm.

Larry chuckles and lashes out again. He calls HP’s suit vindictive, which is true, and adds that it will make it impossible to continue as business partners, only somewhat true as each had already recently moved into the other’s business. Oracle bought Sun and HP got into software and services by acquiring EDS.

A few days later, on the eve of Oracle’s OpenWorld, the suit is settled. HP’s pain is salved by a few million dollars, and the threat of the misuse of confidential information is suddenly, mysteriously no longer an issue. One wonders about the damage HP’s Board did to the company’s reputation by treating this alleged sinner in such a bullying and ultimately lame way.

While Hurd stays out of the limelight plotting Oracle’s next moves, HP directors keep stoking the coals for their critics. In their quest for a new CEO, the Board rejects internal candidates for the third time and pick an outsider: Léo Apotheker, ex-CEO of SAP Germany. This leads to another salvo of Ellison jibes. (When Larry calls himself “speechless”, you know he’s having a good time.)

But wait, there’s more.

What does the Board do besides recruiting Apotheker? They hire Ray Lane as Chairman. As the link to his Kleiner Perkins bio proves, Lane is, without a doubt, an “industry figure”, the type Kleiner Perkins, one of the largest VC firms in the world, likes to co-opt. But the slick KPCB bio (there is, significantly, nothing on him on Wikipedia) omits an important episode: Ray’s acrimonious departure from Oracle. The more charitable souls among us hope that everything is forgiven and forgotten. But knowing the protagonists, Larry and Ray, a more realistic view is that HP’s Board brought Ray in with a specific intent: They want to strengthen the team for a fight against Oracle.

There are three problems with such a move.

First, we now have two muscular venture capitalists on HP’s BoD: Lane and Marc Andreesen, from Andreesen Horowitz (as an aside, admire the firm’s spartan site). While some argue that it’s great that HP has such connections in the VC world (as if any executive or Board member couldn’t get us VCs to return their calls), there’s a governance problem. There will be many situations in which Mark’s or Ray’s existing investments and connections will raise conflict of interest questions; they won’t be deemed independent directors. More

The OS Doesn’t Matter…

by Jean-Louis Gassée

Once upon a time, operating systems used to matter a lot; they defined what a computer could and couldn’t do. The “old” OS orchestrated the use of resources: memory, processors, I/O (input/output) to external devices (screen, keyboard, disks, network, printers…). It’s a complicated set of tasks that requires delicate juggling of conflicting constraints, and every OS handled them differently—or not at all. In those days, the OS was married to the hardware and only a handful of “wizards” with Electrical Engineering degrees—and a deep understanding of circuitry—understood (and invented, and protected) the arcana of OS construction.

Over time, the secrets of these illuminati leaked out. About 20 years ago, the OS lost its mystery. We had learned enough about writing an OS kernel that it became a college-level topic and a Christmas break hack.

Today, there’s only one operating system: Unix. (Okay, there are two, but we’ll get to that.) This is why I contend that the OS doesn’t matter—or that we need to take another look at the word’s content, at what we mean when we say ‘Operating System’.

When RIM decides to go with QNX for its upcoming tablet, the PlayBook, tongues wag. After calling its Blackberry OS the “best of breed” (a tired markitecture phrase), RIM is surrendering to reality: The “proven OS” foundation proved to be unfixable. Because of the layers of software silt that had accumulated over the years, the edifice couldn’t be modernized. Better to quit and make a fresh start. QNX is based on a Unix OS for embedded applications that dates back to 1982(!) when it was first released for the Intel 8088 microprocessor.

The same thing happened at Apple…twice. The Apple ][ OS (or lack thereof, purists will say) couldn’t be brought up to modern standards, so the Macintosh had to be built on a fresh foundation. The original Mac OS foundered on its own beachhead and was replaced by OS X. Based on the Mach kernel, OS X is another Unix derivative, co-authored at CMU by Avie Tevanian. Mr Tevanian improved the system during his tenure as head of software at NeXT and was instrumental in convincing Apple that their purchase of NeXT would breathe new life into the company.

Open the Terminal application on a Mac and what do you see? A noble and worthy Unix “shell”, a program that geeks use to interact with the OS. Terminal uses the bash shell (for Bourne Again Shell. Created by Brian Fox, bash is based on the sh shell, which was invented by Stephen Bourne. Unix mavens love their word-play acronyms).

And now we have the Apple iOS, an OS X derivative that uses bits from the same kernel.

Regard Palm. The sine qua non of handset makers saw that their PalmOS couldn’t be fixed, so they pressed the restart button and created WebOS, a Linux derivative.

Android? It’s based on a Linux kernel. Nokia’s MeeGo? Ditto.

The list goes on. We have the spiritual children of Unix living inside the Cloud, powering the millions of Linux servers running at Google, Facebook, Amazon…

The only exception is Windows. Initially built on top of DOS, Microsoft painstakingly added version after version, always striving for backward compatibility while, at the same time, adding new features. It didn’t always work well (who wants to remember Windows Me and Vista?) but it worked well enough because Microsoft never gave up. They fixed mistakes that they claimed didn’t exist, and now we have the well-respected Windows 7. (Inevitably, critics will say that Microsoft wouldn’t have gotten away with such a tortuous path if it weren’t for its vigorously enforced monopoly.)

Windows will live on — in a PC industry now at a plateau. But otherwise, in the high-growth Cloud and smartphone segments, it’s a Unix/Linux world. We need to look elsewhere to find the differences that matter.

The technical challenges have migrated to two areas: UI (User Interface, or the more poetic—and more accurate—UX, for User Experience) and programming tools.

Now that all “system functions” are similar, the game for hardware and software makers is to convince the user that his/her experience will be smooth and intuitive. Your device will walk on water (with the programmer right under the surface), catch you as you fall, make sure you don’t get your feet wet.

For the developer, what we now call the OS must supply ever-growing expressive power—think a fife versus a twelve-keyboard organ. To wield that expressive power, the programmer needs software tools. The industry uses acronyms such as API (Application Programming Interface), IDE (Integrated Development Environment) or phrases such as Application Frameworks. They define the rules and conventions—which ideas are allowed and how to express them—and the software tools that programmers need to develop an application.

This is today’s OS. User experience. Development tools.

One last element that is and isn’t the OS: This new creature called an App Store (or Marketplace, depending upon the…OS). In my non-technical view, the App Store must be considered part of the OS totality, part of its gestalt. Applications have always been in a feedback loop with the OS. A program can only do as much as the OS allows it, so it played tricks to create multi-tasking, to allow smooth audio/video playback. These “tricks” were incorporated into the OS (and the hardware—think GPU), which then bred another generation of apps that wanted more, and so on.

The App Store genre, invented or not in Cupertino, is now part of that loop, a killer OS component, one that deserves a Monday Note of its own.

JLG@mondaynote.com

Science Fiction: An Apple-Curated App Store

In an alternate universe, Apple has announced the App Store Guide and Blog. Choice morsels from the PR material follow.

“We came to realize that a quarter million apps meant worse than nothing to Apple users”, said Apple’s CEO. “I get confused too! Reviews are often fake, lame, or downright incompetent. PR firms have been caught astroturfing reviews, publishers have resorted to flooding the App Store with shameful clones of successful applications. I won’t let one of Apple’s most important, most imitated innovations sink into anomie.”

[Remember, this is sci-fi.]

“So…Today we’re proud to introduce the Real App Store Guide, written and maintained by Apple experts. We’ll review new and existing iOS apps. We’ll tell you which ones we grok (and that grok us) and give you the straight dope on the offerings you shouldn’t touch, even if they’re free. In our Guide, you’ll find a series of paths: For the Traveler, the Gamer, the Music Lover, the Graphic Artist, the Oppressed Enterprise Windows User, Teachers, Parents, Doctors… The Guide will also feature a blog, a running commentary on the iOS App landscape with intelligent answers to cogent questions. And in keeping with our usual standards for decorum and IQ, the blog will be moderated…”

And so it is, the App Store is fully curated, at long last.

As always, this doesn’t please everyone…at least on the surface. In reality, the usual naysayers are thrilled: More pageviews! Ryan Tate jumps on the opportunity and frenetically fires at Steve Jobs’ inbox, trying to start another late night email séance. But this time the Emailer In Chief doesn’t bite.

Customers, on the other hand, like the Real App Store Guide. Users can finally find their way through the twisted and confusing maze of programs. They learn to adjust for a particular writer’s opinions, much as we’ve all learned to compensate for the biases of, say, movie reviewers. The blog gives civilians a forum where they can argue (politely) with the named authors of the reviews—there’s no anonymous corpospeak here.

App authors…some of them aren’t so keen on the idea. The ones that get tepid reviews are understandably furious and threaten lawsuits (in vain…their attorneys are told to re-read the App Store T&Cs). With a modicum of care with words, that’s what the Guide’s editors are for: Safe negative opinions. More

Curious Summer

by Jean-Louis Gassée

Nothing much happens in August, we thought. Wrong. Our three-week break has been filled with a number of “interesting” events.

Curious Yellow

Let’s start with Mark Hurd’s exit from HP after five years of great financial performance as CEO. If you missed the fireworks, you can get a refresher in this Business Insider post by Henry Blodget, or this excellent NYT piece by ace columnist Joe Nocera.

In twitter terms, it looks like this: A “marketing contractor” claims Hurd sexually harassed her; an inquiry fails to substantiate sexual harassment but finds “an inappropriate close relationship”; the investigation also reveals that expense reports were fudged in order to conceal a tête-à-tête with the female. Mistakes were made, Hurd is fired. End of story.

Not quite.

When a CEO gets the boot, a modicum of decorum is usually observed . Not this time. From HP’s General Counsel we hear that “Mark demonstrated a profound lack of judgment that seriously undermined his credibility and damaged his effectiveness in leading HP”. And that’s on the record.

In her memo to the troops, Cathy Lesjak, HP’s CFO and now interim CEO, accuses Hurd of “misusing corporate assets,” referring to the illegitimate expense reports and alleged payments to the erstwhile soft-porn actress for work not performed.

But forget the salacious details; there’s always Google for that. What puzzles most of us is the exit package story. HP maligns Hurd, accuses him of what lay people call fraud… and then grants him an exit package worth tens of millions of dollars, $35M according to unverified estimates. Attorneys, less puzzled than supercilious, sue HP’s Board on behalf of despoiled shareholders.

In the next few weeks we’re certain to get a clearer picture of the inside animosity directed at the cost-cutting, Wall Street-pleasing CEO. His alleged misconduct may turn out to have been nothing more than a convenient pretext, a word that resonates in HP’s history.

Curiouser and Curiouser

This one’s harder to explain: Intel’s acquisition of McAfee. If you own a Windows PC with Intel Inside, there’s a good chance your computer came with bundled anti-virus/anti-spam/anti-spyware software from companies such as Symantec or McAfee. Microsoft entered the fray a few years ago and provides what they call Security Essentials—for free (Microsoft also offers a free safety scan here). PC Tools, AVG, Kaspersky Labs and many others provide the now customary combination of free and paid-for software security products.

In short, this is an active, thriving scene: Symantec’s revenues are at the top of the $5B range and McAfee’s are close to $2B, despite the competition with “free” products from Microsoft and others.

So what possessed Intel’s CEO Paul Otellini to risk his reputation—and more than $7B of his shareholders’ cash—by wading into such a complex, competitive sector? Seasoned Valley observers such as the WSJ’s Don Clark are politely puzzled (see here and here). Otellini intones a new mantra: Security Is Job One. This marks “Intel’s move from a PC company to a computing company”. Sonorous words, certainly, but without a story of higher revenue and profit for the combined companies, there’s not much to back them up.

More

Under the hood: Google Apps and Apple

With its Cloud Apps, Google tells a nice, simple story: All you need is a browser. Life is simple, we take care of everything, no more fighting with fat, expensive desktop bloatware.
You can access your data and our apps Anywhere, Anytime…if you have an Internet connection. If you don’t, as we’ll see in a moment, things become more complicated. More like yesterday.

Let’s start with a simple Web app. How does it work?

Somewhere, a computer runs a Web server. In turn, the Web server runs an application whose job is to pull the strings of the browser marionette hiding inside my computer at the other end of a Net connection. The app tells my browser to display ‘Monday Note’ at these coordinates inside such-and-such a window, using this font, that size, and this color. Or the Web app sends a file and tells the browser where and how to play it, and so on.
But what happens if I lose the Net connection? The server no longer pulls the string, the marionette collapses, my Web application is dead.

To achieve its strategic goal of displacing Microsoft Office, Google knew it had to provide an off-line version of Google Apps. Off-line capability is implemented by dropping a replica of the Cloud—a Web server, the application code running on that server, and a local cache of my data—into my computer. My work will be uploaded to the Cloud when the Net connection is restored. With today’s software technology, with abundant storage and computing power on desktops and laptops, Google’s goal isn’t unreachable.

But…the Cloud can be replicated inside my laptop?

It’s not as fantastic as it sounds. While the Cloud evokes images of Google server farms and Big Iron, even the flimsiest of netbooks now provide ample RAM space (at least 1Gbyte, often 2), plenty of disk space (160 Gb or more), and an Intel processor running at 1 GHz or faster. Recreating the server, storage, and applications is well within their power.

Furthermore, your PC/laptop/netbook already contains a Web server. Every Mac carries a copy of the Apache Web server (“the most popular HTTP server software in use” says the Wikipedia article), as so do most Linux “distros” on netbooks and DVDs. Windows provides a Web server called IIS, Internet Information Services, the “second most popular web server in terms of overall websites…” (Wikipedia). If you want Apache on Windows, it’s free and easy, go here. The Windows Installer package (née MSI) weighs in at 6Mbytes, that’s all. More

The Adobe – Apple Flame War

The short version:

Who, in his right mind, expects Steve Jobs to let Adobe (and other) cross-platform application development tools control his (I mean the iPhone OS) future? Cross-platform tools dangle the old “write once, run everywhere” promise. But, by being cross-platform, they don’t use, they erase “uncommon” features. To Apple, this is anathema as it wants apps developers to use, to promote its differentiation. It’s that simple. Losing differentiation is death by low margins. It’s that simple. It’s business. Apple is right to keep control of its platform’s future.

The longer version:

The upcoming 4.0 release of the iPhone OS will come with licensing language that prohibits the use of Adobe’s Flash-to-iPhone compiler. The compiler is a clever way around the absence of a Flash interpreter on Apple’s smartphone OS. It takes Flash code in and outputs iPhone OS code, allowing Flash content and apps to run on the iPhone (and iPad). Problem solved.

Not so fast, says Apple, we’ll only allow applications that are written “natively” with our tools. No cross-platform tools, no Flash-to-iPhone compiler, no Flash.

Less than 24 hours later, an Adobe employee, Lee Brimelow, posts a virulent critique of Apple’s latest prohibition, titled “Apple Slaps Developers In The Face”. He concludes with a vigorous ‘Go screw yourself Apple’ and then adds a postscript: ‘Comments disabled as I’m not interested in hearing from the Cupertino Comment SPAM bots.’ Ah, yes. The one-way mirror…
[What the irate gentleman fails to say is this: The only developers slapped in the face are those who don’t use Apple development tools because they want to write a cross-platform app that may or may not use the particular features of the iPhone OS.]

He’s not alone in condemning Apple. In his blog, sunnily called “Why does everything suck?”, Hank Williams asks if “Steve Jobs Has Just Gone Mad” and wonders about “Insane Restraint of Trade”.

Adobe appears to be worried. In its latest SEC (Securities and Exchange Commission, the stock market regulator) filing, the company admits that its ‘business could be harmed’. If Apple succeeds in turning developers away from Adobe’s tools, a new version of which, CS5, is about to be announced, well, the money pump will stutter.

There are calmer minds, however. In his highly-recommended blog, Daring Fireball, John Gruber explains why Apple changed the iPhone OS licensing agreement. It’s strategic, really: Apple doesn’t want anyone else to have control over which OS features the applications have or don’t have access to. I’ll explain in a moment why it’s rational for Apple to fend off cross-compilers, and why it’s not too rational for Adobe employees and others to criticize Apple for keeping control of its future.

But, first, a bit of history.

More

Software and Brakes — Part II

by Jean-Louis Gassée

This week, no iPad disquisition, no large companies engaged in contorted Kama Sutra embraces, no Google-Apple-Microsoft love triangle. We’ll revisit these topics in due course but, for the time being, let’s go back to a geeky topic unadulterated by geopolitics or markitecture: software and brakes. Last month, we looked at the software invasion of automotive braking systems. More specifically, we looked at the interplay between braking and kinetic energy recovery in the Toyota Prius.

Today, we’re going back to “soft brakes” for another set of applications: differentials and stability control.

Differentials. Fifty years ago, they wore out, they made noises, they had to get careful periodic checks and special lubrication. Now, with progress in metal allows, high-precision machining and modern lubricants, they’re rarely seen or heard of. Yet, they perform and important role and their basic design suffers from one no less critical flaw.

(Here, we’ll assume a rear-wheel drive car. The concept applies to all drive configurations.)

The important role goes like this: when the car turns, the two wheels on the same axle draw circles of different radii, smaller radius for the inner wheel, larger for the outer. As a result, the outer wheel must turn faster that the inner one. This is no problem for the front axle whose wheels are “free”, not driven. But, for the rear axle, we’re in trouble: the drive shaft attached to the gear box connects to the axle through a 90 degrees angle gear. This causes each wheel to be driven at the same speed. This is fine in a straight line but causes wheel slippage when we turn as the wheels must rotate at different speeds.

This was the arrangement when the very early automobiles mimicked horse carts. On carts, wheels on the same solid axle did slip in a turn, but said wheels didn’t have to provide any traction, the horse did. In a car, the axle provides traction and wheel slippage works against stability and comfort, to say nothing of tyre wear.

So, the differential was invented. It’s a little counterintuitive at first but it works beautifully.

See this touchingly kitsch Chevrolet video. Or this learned Wikipedia article. Relatively simple, once you get the hang of the planet gear’s role. And universal.

But trouble starts right away.

More

Honey, I shrunk the Tax Code!

It’s really about another kind of code, but read on a bit…

This is an old dream: making the tax code shorter, simpler. From time to time, a politician of the populist persuasion comes out and promises to get things right. The ultimate expression for this drive towards simplicity is the Flat Tax movement: a single fixed (as opposed to today’s progressive, accelerating) tax rate for everyone. Google ‘‘flat tax” and you get more than 40 million hits. A popular fantasy. And not one just held by kooks, conspirationists and other End of Times prophets. Stanford University hosts the very serious right-wing Hoover Institution, a think tank that publishes scholarly papers such as this one. Or we have Steve Forbes, as in Forbes Magazine, proudly posing, postcard in hand, on the cover of his book: Flat Tax Revolution: Using a Postcard to Abolish the IRS.

Fun stuff, or sad, you pick your perspective. (Speaking of fun, this note’s title is a reference to a cult movie: Honey, I shrunk the kids!)
The basis for the agitation is a feeling of hopelessness. Year after year, we add articles to the tax code, to adapt to new circumstances, to new projects, to plug loopholes. As a result, the code grows and buggier.

This latter phrase, “the code grows fatter and buggier”, also applies to our high-tech spaghetti code: operating systems. Every new rev served with fresh bugs!

Software is… soft, malleable. Next to hardware, adding/changing features is comparatively easy, tempting. With features added upon features, bug fix patches over earlier generations of patches, the whole edifice soon starts to look like the accumulated layers of a Babylonian archeological dig.

Over time, the mess resulting from this “organic” growth gets worse and worse because OS developers are caught between the past and the future.
Looking forward, happily fueled by relentless technical progress, our industry constantly comes up with new hardware features and more modern software techniques. These improvements result in new ways application programs “talk” to the system. We use the acronym API (Application Programming Interfaces) for the way application software accesses OS resources.
Looking back, companies and individuals have huge investments of money and people in application software. OS improvements can’t come at the expense of backward — an ominous adjective — compatibility.

When making improvements, the system must add new APIs and keep the old ones around. Add but never subtract: he result is bloatware. More

Soft Brakes on the Prius

Once upon a time, I took my Wehrmacht staff car to the Palo Alto service shop. As I mentioned a barely perceptible change in the feel of velvety autobox when it shifted gears, Ernesto, the all-knowing, all-seeing tech nodded: ‘Yes, we need to load a new revision of the software in your automatic transmission…’

When I mentioned this to my Monday Note boss, he jumped: ‘Why don’t you write a piece on cars becoming soft, that is an exposé of the total invasion of software in today’s cars?’
We’re both geeks, you see. And, yes, I could sing the praise of Toyota’s PSD (Power Split Device), the clever Continuously Variable Transmission (CVT) used by the Prius. (That type of transmission was invented in the 60s by TRW, Thompson Ramo Wooldridge, a Texas company. The patent has expired.)

But there is more urgent than software-driven transmissions in the news. Toyota is hit (or has hit itself) by one problem after another. After recalling something like 8 million cars because of a combination of pesky floor mats and sticky gas pedals, we now hear Toyota’s crown jewel, the lated model Prius has buggy brakes.
How come?
Software, of course. And the need to score high numbers in the mileage tests.

Here is how the problem builds.

First, brakes.

Once upon a time, brakes were simple: you pushed on the pedal, the pressure was hydraulically transmitted to the brake itself. Assuming a disc brake, the pressure caused calipers to squeeze the disc between two sets of pads, thus slowing the car. For convenience and safety, your foot’s pressure is multiplied by a brake booster.
So far, no microprocessor.

Then we invented anti-locking brakes, ABS. As you know, when wheels lock, the car skids, braking is impaired, stopping distance increases. ABS uses sensors to monitor wheel motion. If the wheel abruptly stops moving, it’s locking. A software program then directs the brakes to lower the pressure on the disc pads, the wheel starts moving again, until it locks again and the unlocking cycle restarts. You can sometimes sense a vibration when the ABS quickly repeats the sequence, trying to keep the car just at the edge of maximum braking without locking the wheels. Nice. Modern implementations have faster actuators, the device that temporarily reduces hydraulic pressure, and smarter software, to better deal with road surfaces with gravel or snow where some amount of wheel lock is the better strategy. There is even newer code that detects a panic stop and forces the kind of maximum pressure normal drivers are reluctant or unable to apply. More