Building Rumors, Fooling Slashdot

"As seen on Slashdot."

8/18/2K: You can fool some of the people all of the time, and you can fool all of the people some of the time.

We've been publishing groundless rumors on The Rumor Mill since last December (New PowerBook in January, predicting a pink cobranded PowerBook Pepto Bismol). We've learned that no matter how hard you work at it, some people just won't read between the lines.

We've tried to make it more blatant. At the top of the page, it says, "Unsubstantiated Mac rumors and more." That's an understatement. In the footer, we claim The Rumor Mill is "an alternative to real rumor sites" and provide a link to our About page.

Within each article we try to drop hints that this is all in fun, something we also state quite clearly on the home page. Hints in yesterday's article include:

• "John Doe 14" as our "unknown (to Apple) source," a reference to Apple's infamous John Doe 1 with up to 24 helpers suit
• calling the whole thing "idle speculation" in the last line

Thus, it really surprised us to see Slashdot pick up the story and give it credibility. You'd think my pseudonym (Anne Onymus = anonymous) would be a dead give away. Sigh.

Building Rumors

As Dan Knight has commented before, it's remarkably easy to create plausible rumors with absolutely no basis in reality. That's what we do here - and I guess we do it a bit too well.

There are rules for building rumors, but the two key ones are to address the hopes or fears of your audience and be in touch with current developments.

Let's look at how this article was constructed:

Intel has announced that the Pentium 4 will hit 2 GHz next year, forcing Apple to completely rethink its game plan. It's decision: Abandon the G4 except on graphics workstations.

The Intel stuff is true, lending credibility to what follows.

Rumors that Apple will adopt Pentium, Transmeta, Itanium, Alpha, and any number of other processors always seem to be making the rounds. We build on the hope/fear that Apple will abandon the G4. That's our hook.

Working closely with IBM, Apple will adopt the G5 processor, which should be shipping in quantity by Macworld San Francisco in January 2001. The move validates IBM's long-standing claim that AltiVec was the wrong way to go in improving chip performance.

It's self-evident that G5 is the logical name for the next generation CPU, just as G4 followed G3. We're just using a trend to project a plausible future.

Dan Knight has been one of the leading voices crying out, "The emperor has no clothes" as far as the G4 is concerned. His G4 Insignificantly Superior to G3 is one of the best articles explaining why most users most of the time won't see any benefit from the G4 over the G3.

At the same time, it's true that IBM has resisted AltiVec. Souping up just part of the CPU has less overall effect on performance than speeding up the whole CPU.

And, of course, we all need new models at Macworld San Francisco. That's where the hope angle comes in.

Instead of a souped-up video section coupled with an otherwise pedestrian 500 MHz or slower CPU, the G5 architecture is what the G4 should have been all along: an optimized successor to the G3 with full multiprocessor support (like the 604e and G4) and a large, fast on-chip cache (like IBM's PowerPC 750CX).

Here's where we push IBM's agenda of faster, better CPUs, not just enhanced graphics processing. And we tie it to the PowerPC 750CX (a.k.a. G3e), which Dan Knight is convinced will be an excellent CPU for iMacs, iBooks, and low-end PowerBooks (see Should Apple Use the New G3?)

We're basically building on trends here. If IBM were to design a G5, this is probably how they would do it. Since they don't own AltiVec, they would have to license it from Motorola and pay them a royalty. Since the whole idea of AltiVec is anathema to IBM, they would ditch the technology in any CPU they designed.

And the whole Wintel world laughs at our 500 MHz processors, not realizing how powerful they really are. We call it pedestrian, possibly raising your ire, and promise something better.

Using IBM's commanding lead in copper chip technology, the G5 will include a 1 MB cache running at full CPU speed, which will range from 700 MHz to 1 GHz upon the chip's release. It will also have less power draw than the G4, making it a candidate for top end PowerBooks.

The copper chip stuff is true. IBM also demonstrated 1.1 GHz PowerPC processors over a year ago, but Apple's decision to go with Motorola's AltiVec cost them that kind of speed (see Dan Knight's Honest for more on the dysfunctional AIM consortium).

It's pretty widely known that IBM could easily be producing 700 MHz G3 processors today, and possibly 600 MHz or faster G4 processors. The speculation on real rumor sites is that Motorola is preventing IBM from selling faster G4s until Motorola can produce them. True or not, it's part of our rumor mix.

Best of all, the G5 "has legs" and will easily reach 2 GHz by the end of 2001, according to John Doe 14, our unknown (to Apple!) source on this story.

That was Dan's addition, the "has legs" phrase. It has a nice ring to it. My favorite part was referencing "John Doe 14" from Apple's unknown leak lawsuit. Even if we had an inside source (remember, these are completely fabricated stories), there's no way in the world anyone could know he or she is John Doe 14.

This means Apple will have three processors in the hardware. The IBM PowerPC 750CX (G3e) will be used in lower tier models such as the iBook and iMac, ranging in speed from 533 to 666 MHz and using UMA2 architecture, which includes a 133 MHz system bus.

Now we throw in all sorts of authoritative sounding numbers and phrases, like UMA2 (apparent code name for the 133 MHz motherboard). It's plausible technobabble that addresses your hopes.

Some Slashdot readers did remark that nobody would ever release a 666 MHz CPU in a country with any level of Christian background - mark of the beast and all. That was just a bonus; neither of us thought of 666 as anything more than a reasonable processor speed on a 133 MHz bus.

The Power Mac G4 will come in dual and quad processor models using 500 MHz G4 processors on the current Sawtooth motherboard.

This is 100% plausible. Apple could probably do this today, but except for people using Photoshop or one of a handful of other AltiVec-enabled applications, what would be the point? Sure, graphics geeks would pay the long dollar, but until the OS really supports multiple processors, even today's Power Mac G4 MP is overkill.

At the top, the Power Mac G5 will ship in single, dual, and quad processor configurations with 733, 866, and 1,000 MHz processors on a new 133 MHz UMA2 motherboard. Apple will also have a PowerBook G5 running at 733 and 850 MHz, as well as a G5 Cube in 866 and 1,000 MHz versions.

Well, if there were a G5 and it did run at these speeds, these are probably the speeds Apple would use. This is just following trends.

Details are sketchy, but there may even be an iMac Special Edition with a G5 clocked at 666 MHz. Further details remain unknown.

One more bit of wish fulfillment, along with a bit more product line confusion (something Apple is getting very good at with four iMacs, two iBooks, two PowerBooks, three Power Mac G4s, and two Cubes).

Motorola was unavailable for comment. Sources at Apple and IBM refused to comment on this idle speculation.

Almost every good rumor ends with a line like this. We especially like the phrase "idle speculation" at the very end. That should tip off every reader who gets this far that we're just having fun.

Of course, we can't leave out the matter of style, which I picked up from John Dvorak's news column in PC Magazine. It's very simple: use a lot of bold text. For some reason, that seems to make rumors seem more valid. ;-)

Conclusion

As you can see, creating a good rumor is an art. What surprised us was seeing Slashdot link to the article as a plausible rumor and then seeing Slashdot readers debate the whole thing. We'll leave you with a few of our favorite quotes, although you really should enjoy the whole thread (there is some strong language).

• Whilst we know that Apple rumour sites are usually pretty lame, this does have a string of credibility to it.
• I wonder which Apple employee will now be fired and sued for damages for releasing these trade secrets!
• Apple is running out of options--I've even heard some talk about them leaving PPC/AIM, but not for the cpu you all are thinking of.
• It may also be worth mentioning, that IBM's cpus typically run cooler than Motorola's, and IBM has been pushing the clock cycles rather than Altivec option for a while now.
• ...overall performance of the machines should benefit by actually being able to up the clock speed again.
• As for whether AltiVec was a bad decision, I'm not an expert on the subject but the impression I've gotten is more that Moto is incompetent than that AltiVec fundamentally limits clock speeds.
• This rumor has some good logic behind it, truth on the other hand is arguable.
• The fascinating thing about the 666 MHz G5 iMac SEvil, is that it actually runs infinitely faster than computers running at much higher clock speeds, but also infinitely hotter (Optimal operating temperature is 3 million Degrees F). The downside is that "Eternal Damnation" clause in the EULA, and when running Linux the Daemons can occasionally get a little out of control, spawning evil child processes, and attempting to take over the world.
• Apple's in a sticky situation... they want Altivec because it speeds things up, but they need higher clock rates to compete with AMD and Intel. Motorola hasn't been able to get high clock rates and the G4 has been stuck at 500 MHz for a year.
• No marketing department in the Christian(and maybe elsewhere) world would *EVER* market a processor as 666 MHz.
• Will the new 666 Apple Special Edition be available in black with flames? will the cost show up on the invoice as: (1) soul + 4.95 shipping?
• Good for the die-hard Mac folk. They will be getting some serious speed at last. Too bad the cost will be astronomical.
• My friend works at Apple and told me that the reason they dropped the Altivec from the G5 is because they wanted to use the space on the die for a new on-chip LED light show that you will be able to see through the computer case.

At least one poster got it right: "Dan Knight (the operator of Low End Mac) makes no claims of truth about his "rumors". I fact he freely admits that his rumors are made up."

Anyhow, Slashdot, thanks for all the extra traffic. We hope everyone got a good laugh out of the whole thing.

We hope you've enjoyed the ride.

- Anne Onymus