November 2006
Mon Tue Wed Thu Fri Sat Sun
« Oct   Dec »

Month November 2006

A podcasting device?

We normally don’t make much room in MDJ for iPod-related items other than those affecting Apple’s financial situation. MDJ and MWJ are about the Macintosh, not about the iPod or about Apple Computer generally (excepting Apple’s corporate health). However, people scoping out new devices still constantly ignore the principles that made the iPod a smash hit.

Today, Dave Winer touches on two while quoting analyst Michael Gartenberg on what an iPod-like device designed specifically for podcasts should look like. It’s hard not to love Winer’s gung-ho attitude; the first thing we’d ask is “what’s the last successful consumer device spec’d out by any analyst?” Analysts tend to argue that future devices should be the best parts of old devices plus some kind of magic new pixie dust to make them “better.” Individuals and small groups do the real innovation. Nonetheless, Dave says:

Since we agree that it should have built-in wifi capability, why should it have USB? USB may be a little faster, but if I can save money and space by only having one communication interface, then I’m going that way. I don’t see myself pushing content from a desktop or laptop to the device, I see it downloading on its own over wifi

Various clueless people have been calling for the iPod to have Bluetooth or Wi-Fi capability since it was released. It wasn’t going to happen, but it may in the next year? Why? Battery life. The first iPods had a realistic battery life of 6-8 hours if you avoided battery-draining options like shuffle play and constant backlighting. Thanks to advances in battery technology, Apple is now claiming audio playback time of up to 20 hours on a full charge (or up to 6.5 hours of video playback), or up to 24 hours for an iPod Nano. That’s approaching the levels where you could get 10 or more hours of playback with the constant battery drain of a radio transmission, either to Bluetooth headphones or to find a Wi-Fi network. It would still require you to think carefully, though – if you leave Wi-Fi on while not intending to connect to anything, you’d be draining the battery in a constant search for a Wi-Fi network that you don’t intend to use.

Winer may not see why to put a USB port on such a device, but it would require a way to configure itself without a network. He lives in Berkeley and has lived in other major cities where comprehensive free-or-cheap Wi-Fi is taken for granted. Most of the world does not – once most people go jogging or start the commute, they have no network available until they return to home or office, and even that network may be modem-based and very slow. A USB port allows for traditional desktop-based loading of content, and as Winer may not have remembered, for charging the battery. When you can combine both control and power into an industry-standard connector, it’s a clear win. Gartenberg is correct in that making a new proprietary connector is a bad idea, and there’s little reason to put in a dumb power port when a smart USB port takes the same space and provides many more capabilities.

The second thing?

I do think I’m going to have to tell it what to subscribe to over the interface, the podcast discovery mechanism really has to be on a system with a rich user interface. Wifi has plenty of bandwidth for that kind of communication.

The key lesson of the iPod is that jamming a lot of controls onto a pocket-sized device is a bad bad bad bad tragically bad idea. The first lesson of the iPod is that you don’t try to use its four buttons and a scroll wheel to load or manage content on the device. You use iTunes for that, since it has a huge screen and the resources of your entire computer to make that task easier.

A device that can find and download podcasts on its own is a fine idea, but how do you subscribe to a new podcast that you know about? Does anyone really want to have to navigate Web sites or enter URLs on a pocket-sized device that should be optimized for playback? Gartenberg says he doesn’t want “syncing” because he can just drag and drop stuff to a standard device, but the market hasn’t really embraced that solution, has it?

A Wi-Fi interface means such a device could automatically update podcasts on its own, perhaps discarding old or already-heard episodes by rule to save space as new episodes come in. But it seems silly to imagine anyone investing big bucks in a device that ignores the lessons of the iPod solely to stamp “podcast” all over everything. Such a device could succeed in its niche not by being more complicated, but by improving upon the strange iPod-iTunes podcast support. Wi-Fi and automatic updating is a cool start, but not if that’s its only advantage and it’s harder to use than an iPod. It would have to be as simple as the iPod-iTunes combination, with more features for podcast listeners (and viewers?). Configuring Wi-Fi on a tiny screen won’t ever meet that burden.


We had some problems trying to get Adobe Acrobat 8 Pro to digitally certify today’s issue (MDJ 2006.11.20) – the program did it just fine, but intead of growing the file size by 5% or so, it bloated it by 44%. We don’t need to add 44% to the size of a single issue for a digital signature, so we signed it in Acrobat 7.

Unfortunately, in the confusion, we kind of forgot to digitally sign the setext version of MDJ 2006.11.20 at all. Sorry about that – little things are still falling through the cracks, but we’re working on it.

Google accidentally sends out Kama Sutra worm

From our friends at Macworld:

Google accidentally sent out e-mail containing a mass mailing worm to about 50,000 members of an e-mail discussion list focused on its Google Video Blog, the company said Tuesday.

“On Tuesday evening, three posts were made to the Google Video Blog-group that should not have been posted,” Google said in a statement, posted late Tuesday night.

“Some of these posts may have contained a virus called W32/Kapser.A@mm — a mass mailing worm. If you think you have downloaded this virus from the group or an e-mail message, we recommend you run your antivirus program to remove it,” said the statement, which was attributed to the Google Video Team.

W32/Kasper.A@mm is better known as the Kama Sutra worm. Discovered in January of this year, it deletes files and registry keys on affected Windows systems. It is blocked by most antivirus software.

Google uses its Video Blog group to let subscribers know when “interesting and fun” videos have been highlighted on the Google Video Blog. E-mail to the group’s mailing list are posted by a handful of Google employees, called Google Video Team

This team was responsible for sending out the malicious e-mail Tuesday night, said Gabriel Stricker, a Google spokesman.

Stricker did not have any more details on how Google ended up distributing the worm code, but he said that internal protocols are now in place to prevent this from happening again.

It’s about time those smug, self-satisfied Google users joined the real world instead of living a “security doesn’t matter” fantasy land.

Google, as you know, is powered through invisible programs that live on every personal computer. And yet despite this massive security hole, Google users have continued to believe that they were immune to viruses, worms, and other security threats that only I, as a columnist, could see coming.

They perform search after search, day after day, believing they’re better than those of us who use Alta Vista, which as you know is the world’s leading search engine and therefore used by all people who are smart enough to read IT columns. Even this blatant example of how using Google can completely destroy your systems is unlikely to awaken them from their cult-induced sense of superiority.

When contacted for comment, Google’s spokesperson, who said things I don’t believe and therefore was obviously lying, said that this heinous security breach had nothing to do with the Google search engine, even though the names are exactly the same. But I saw my good hacker friends take over a system that had Google running on it at the time, so this misdirection doesn’t fool me, and it shouldn’t fool you either.

I can tell you this without revealing details of all the secret demos I want you to believe I’ve watched and perhaps understood: this should be a glass of water in the face of all those insufferable Google users. Using Google makes you just as likely to get a virus (or, as I more correctly call them, “thromboses”) as if you use Windows or ATM machines. I know it won’t wipe the smirks off their faces, but I’ll still go to bed tonight muttering to myself as if it would.

[OK, ZDNet – can we have the column gig now?]

The reality about Steve Jobs and stock options

We’re not going to spend more space right now in MDJ or MWJ refuting items that were obviously false months ago, but it is disappointing to see Dave Winer appear to endorse a story about Apple’s backdated stock-option controversy, quoting a passage that says “Apple will end up being a model case of how NOT to handle such affairs, and Intellectual Dishonesty will have cost the company more than dishonesty itself.”

The actual story, from Mark Anderson of the “Strategic News Service,” claims that Steve Jobs did benefit from backdated stock options because he later traded in all of his options for shares of restricted stock, and they wouldn’t have been worth as much if they hadn’t been backdated.

This isn’t “intellectual dishonesty” from SNS as much as it is “intellectual laziness.” First, as noted in MWJ 2003.04.14, not long after Jobs traded in his options for shares, all of Jobs’ options were underwater at the time. Even the ones that hadn’t vested were still underwater. Jobs exchanged 27,500,000 options that were almost all vested but seriously underwater for less than 20% as many shares – 5,000,000 – that he would not be able to sell or vote or do anything with for three more years.

To argue now, three and a half years later, that Jobs benefited because these options were underwater by US$30 per share instead of US$32 per share doesn’t pass the laugh test. In fact, as of the day Jobs made the switch (2003.03.19), it would have been a net loss for him if Apple’s stock didn’t pass US$44.85 per share by the time the shares vested in March 2006. As of the day of the options-for-future-shares exchange, Apple’s stock closed at US$14.95 per share. If Apple’s stock didn’t at least triple in a three-year period, Jobs would lose money on the exchange. This is the transaction that SNS now says “benefitted” Jobs because his never-exercised, heavily-underwater options could have been two or three bucks more underwater per share.

Second, all this presumes that Jobs is interested in selling Apple stock or exercising Apple’s options. There is not one piece of evidence to support that idea – not a single one. For this, we quote from MWJ 2006.08.05:

Early reports seemed to zoom in on Jobs’ 2000 stock option grant, the one that was for 20,000,000 shares by the time he exchanged it for restricted shares in 2003, thanks to the February 2000 2-for-1 stock split. The famous grant captured media and analyst attention after Apple’s announcement, even though no one could explain why it might be irregular. In fact, it doesn’t look bad at all: Apple granted the options to Jobs on 2000.01.12, and announced it just one week later, on 2000.01.19. Unless the company was scanning a much larger period but chose to backdate the options by only one week, there’s not much “there” there.

Observers looking for some hint of backdating seem to have forgotten about Jobs’ second grant – 7,500,000 options with a striking price of US$18.30 per share (or US$9.15 per share in today’s split-adjusted pricing), awarded on 2001.10.19 because all of the previous 10,000,000 (pre-split) options were underwater. The potential problem? Apple didn’t reveal the grant until an SEC filing in March 2002.

Using prices in today’s shares (after 2-for-1 splits in both 2000 and 2005), an examination of Apple’s stock prices finds that during the window between 2001.10.19 and 2002.03.22, Apple’s stock closed between US$8.78 and US$12.73 per share. For example, on 2002.03.04, Apple’s stock closed at US$12.15 per share, a full US$3 per share higher than Jobs’ option striking price of US$9.15 per share. If Apple’s board actually awarded the options in early March 2002, but backdated them to 2001.10.19, it would have made Jobs’ grant worth US$45 million more than had they been awarded on 2001.10.19.

Before you shout “j’accuse!” at this revelation, you should realize there are plenty of problems with this theory. First, if Apple’s board tried to backdate options to give Jobs more money, then why didn’t the directors pick one of the four other dates in that same window when the closing price was lower than US$9.15 per share: 2001.10.23 (US$9.07), 2001.10.29 (US$8.81), 2001.10.30 (US$8.80), or 2001.10.31 (US$8.78)? Two trading days before the grant date, Apple’s stock closed at US$8.49 per share (2001.10.17); two weeks before that, the stock closed at US$7.49 per split-adjusted share (2001.10.03). If the directors were willing to backdate six months to give Jobs an additional US$45 million, why not an additional three weeks when that would have added another US$24.9 million to the same grant?

Second, you must consider that Jobs has not exercised a single option or sold a single share of his Apple stock since returning to the company, except in March 2006. We explained that at the time on the MacJournals-Talk mailing list, currently on hiatus. Jobs sold almost half of his shares of Apple stock, starting on 2006.03.19 (yes, a Sunday), at US$64.66 per share.

As of that week, Jobs completely owned 10,000,004 shares of Apple stock. Jobs received 1,500,000 shares of Apple stock in late 1996 for selling NeXT to Apple Computer, and according to former CEO Gil Amelio’s book On the Firing Line, Jobs promised not to sell the shares for six months to avoid undermining public confidence in Apple. Amelio says that Jobs professed to understand how important it was to hold that stock longer than six months.

Instead, just as he had sold all but one share of Apple stock after being evicted from the company in 1986, Jobs turned around in 1997 and sold all but one of his new shares as soon as the six-month period had elapsed. The sale, at near record-low prices for Apple stock, hurt investor confidence in Apple and helped usher Amelio out the door nine years ago (MWJ 1998.04.06). That one share turned into four shares with 2-for-1 stock splits in 2000 and 2005.

That was all of the stock Jobs owned until March 2006, when his restricted grant of 10,000,000 shares (not options, and doubled from 5,000,000 thanks to the 2005 split) vested in full. Once they vested, though, they became his personal property. As far as the IRS is concerned, a gain of 10,000,000 shares at US$64.66 per share is income of US$646,600,000. That makes for a hefty tax bill that has to be paid in cash, not in shares of stock.

The SEC recognizes that this happens. Company insiders (executives and directors) are required to notify the SEC and the public when they sell shares of stock so that there’s no secret insider trading. Every year, around the time of Apple’s annual meeting, another round of executive stock options vests, and there’s a slew of SEC Form 4 filings as Apple’s executives turn those restricted options into actual Mercedes-buying cash.

Jobs had never done this, and his own filing showed that his sale wasn’t the usual kind. Form 4 has a “transaction code” in section 3 of Table I, and it’s almost always either “P” (purchase) or “S” (sell). For Jobs’ transaction, the code was “F”: “Payment of exercise price or tax liability by delivering or withholding securities incident to the receipt, exercise or vesting of a security issued in accordance with Rule 16b-3.” That rule lists exemptions to filing notice of insider trading; Apple’s grant of shares to Jobs qualifies because Apple’s board of directors approved it. It means that Jobs didn’t have to file Form 4 when he gained ownership of the shares.

In short, Jobs sold 46% of his new Apple shares to pay the taxes on the 54% that he kept. Even if Jobs had US$295 million in cash lying around, he probably didn’t want to spend it on taxes just to preserve shares that he hasn’t shown any interest in exploiting for nine years. Other than this tax obligation, Jobs has neither sold shares nor exercised options on Apple stock [since taking over as interim CEO in July 1997], not on a single share, even though they were worth hundreds of millions of dollars.

Jobs’ history with Apple shares strongly suggests that he views [options and shares] not as liquid assets, but as recognition of success and as power chits. Jobs doesn’t seek compensation from the companies where he works, and recently made news by refusing compensation for his service on the board of directors of The Walt Disney Company. (Disney’s board had to modify the director compensation policy to accommodate Jobs – the existing plan had no provision for a director declining to be paid. The SEC filing about the change tipped off the media.)

Jobs holds on to the stock of companies where he works, and sells it only in dramatic gestures when he’s not in charge and things aren’t going his way. It wouldn’t surprise us for Jobs to sell all but one share of his Disney stock if he perceives management as fouling up his Pixar legacy. Given all this, it just doesn’t make much sense to believe that Apple’s board would go to such lengths to backdate options for a man who obviously didn’t care to exercise them anyway. As long as Jobs is in charge of Apple Computer, he has no interest in selling stock or options.

Given these facts, what does Mark Anderson of SNS say about it?

Here is my conclusion: I think (and I have no direct evidence for this, other than the behaviors and quotes from those involved) that Steve Jobs was aware of the practice, did personally benefit, and had some role in the granting and dating of those options.

So, Anderson says if certain things had happened that didn’t, and if Jobs had done something that he didn’t do, then the statements that Apple made about what really happened would be false and Steve Jobs is in legal trouble.

We were talking about “intellectual dishonesty?”

Something you can do to help

[Note: This item was originally posted on September 28, but for some reason, it keeps vanishing. Restoring it occasionally bumps it to the top of the home page, but unless you see an “Update:” at the top or bottom, there’s nothing new for those who’ve already read it. Sorry for any inconvenience.]

We’re all extremely grateful for the E-mails and other things that have poured in since Matt first discussed his diagnosis of heart failure, and the new batch that came in today after it was disclosed in MDJ 2006.09.28. Many people have asked if they can do anything, no matter how small, and after reading today’s E-mail, there is one thing everyone can do that would help us tremendously.

Stop using StuffIt 7 or 8. Seriously.

We’ve distributed the PDF versions of MDJ and MWJ as binhexed StuffIt archives for over ten years, because until the days of Mac OS X, that was the best way to save bandwidth while preserving the "PDF " file type and "CARO" creator type necessary to allow double-clicking the issue files. In the past couple of years, readers have requested a switch to Zip archives because they’re easier to decode on other platforms. We would have preferred switching to StuffIt X because, frankly, it makes smaller files, and we’re all about saving bandwidth – but the free StuffIt Expander for Linux can’t decode these files, and a few people do process their E-mail on Linux boxes, so we’ve resisted the temptation.

But for some reason we don’t really understand, a lot of people seem to have stopped paying attention to StuffIt when Apple stopped bundling it. StuffIt 7 was released over three and a half years ago, and a lot has changed on Mac OS X since then. That was pre-Safari, for pete’s sake. StuffIt 9 (released two full years ago, before Tiger) added important new decompression algorithms to keep up with the latest in Zip technology, as well as to support new StuffIt features.

We know that Aladdin/Allume/Smith Micro has not always made upgrading StuffIt easy, especially if you want a new Expander while keeping the functionality of an older paid version. Some versions of Expander installed a “replacement” StuffIt framework that made older paid versions stop working. Even today, Smith Micro requires you to provide your E-mail address to get a link to the StuffIt Expander download page, and notes that by doing so and clicking the links, you are signing up for an opt-out mailing list about new products. We’re glad that you’re no longer forced to download the entire “StuffIt Standard” product and install it for evaluation just to get Expander at all. Even so, this is the kind of behavior that has dropped Expander from a “must-have” to “must-tolerate” product.

Nonetheless, if you use StuffIt Expander, you are well advised to use a current version. If you’re unwilling to try the brand new Expander 11.0, the same download page offers Expander 10.0.2. If you have multiple versions already, we advise that no one use any version of StuffIt Expander older than version 9.0.2. Version 9.0.1 and earlier simply cannot expand all modern StuffIt and Zip archives. If you’re not using at least version 9.0.2, you need to update, or alternately, accept that there are archives in the wild that you cannot decompress – and some of them may come from us.

Paying for StuffIt is no longer a no-brainer (we hope to take a full look at version 11 in an upcoming issue of MDJ and MWJ), but that doesn’t obviate the need to stay up-to-date if you do use the free StuffIt Expander. We try to stay up-to-date on lots of tools to get the smallest files possible, and we simply cannot guarantee that we can create files that old utilities know how to decompress. It’s more of a pain than it should be, but one of the best ways you can help us deliver issues to you is to have a current (i.e., 9.0.2 or later) version of StuffIt Expander.

We hope to make Zip archives that either the command-line or the Finder can decompress, but even that may require current versions of those programs (i.e., Tiger or later). The best way to make sure you can decompress anything that anyone creates is to use StuffIt Expander 9.0.2 or later. Just that simple change would probably drop our support E-mail by 25% per month, believe it or not.

Oh, and if you’re unhappy with current StuffIt offerings or practices, tell Smith Micro. Be specific about what you don’t like and what you’d like to see instead. We know they want to hear from you.