Android Alpha's First Boot
The day after Android Begins.
“Good heavens.”
Trip Hawkins, president of Electronic Arts, could hardly believe the speed at which ID Technology operated. He had sent a fax containing nothing but positive feedback only the day before. Yet when he arrived at the office that morning, a message was already waiting: ID Technology had uploaded Android Alpha to the company’s FTP server.
“Send for Chief Engineer James Web.”
“Yes, sir.”
Hawkins pressed the intercom and summoned the engineer he trusted most. He wanted to install the software himself, but when it came to evaluating a product of this caliber, he always needed a professional opinion before he could feel satisfied.
“You called for me, sir?”
James appeared in under five minutes.
“Android Alpha has been uploaded to our FTP server. Install it on this machine and give me your impressions.”
“Whoa—already? Do they code from the cradle over in Korea? How are they this fast?”
Yoo Jae-won’s programming speed was nothing short of astonishing to them. Even if Android Alpha was a DOS-compatible system, building an entirely new codebase that avoided overlapping with Microsoft’s code was an extraordinary feat.
“Haha. You still don’t trust that young genius at ID Technology?”
“It’s not that.”
Android Alpha had been the talk of Electronic Arts for days, all because of a single name: Yoo Jae-won. Employees had been receiving generous quarterly bonuses since the end of the previous year, thanks to Keyboard Warrior. The windfall had even allowed James Web to finally acquire the high-end Compaq SystemPro he had only ever admired from afar—a top-tier machine equipped with a 486 CPU.
Nor was that all. Ever since Intel released the 486, experts had begun questioning whether the aging MS-DOS could truly harness the new processor’s power. Running a computer on such an outdated operating system felt like an enormous waste of performance. Interest in a next-generation OS that could finally break free from DOS was intense. Microsoft’s upcoming Windows 3.0 was one candidate, and IBM’s OS/2 was another. Android Alpha had arrived at precisely the right moment. Though still DOS-based, its adoption of a graphical interface for modern hardware had piqued James’s curiosity.
“Then I’ll download it and install it right away.”
“Go ahead. Let’s see it.”
Because Electronic Arts’ internal FTP server was connected via Ethernet, the download finished in seconds. James had already prepared blank floppies at Hawkins’ summons, so he immediately extracted the files and created installation disks.
In a tense silence, he booted from the floppy. The drive was slow, and the process took time. But once the loading finished, the installation screen appeared at once.
“Quite a few detailed options.”
There were three choices: format the hard drive and install immediately, erase the existing DOS and install Android Alpha, or back up the current DOS, create a recovery point, and then install. Since the goal was to evaluate Android Alpha itself rather than test compatibility, James selected the first option—format and install.
The moment he confirmed, the format completed.
“What the—? That fast?”
Normally, formatting meant wiping every sector one by one. Yoo Jae-won had designed the process to erase only the boot sector and file system—a so-called quick format. For corporate drives holding critical data, a full sector-by-sector wipe was preferable, but for personal machines, clearing the file system was sufficient.
System files from the floppy were then decompressed and copied to the hard drive. When the process finished, a prompt appeared instructing him to remove the disk. After he did so and pressed Enter, the system rebooted automatically.
From that point, everything proceeded smoothly. The familiar ID Technology setup screen appeared, automatically detecting and configuring the CPU, VGA, sound card, and Ethernet adapter. Once the settings were complete, the boot finished.
“That’s cute.”
A squat little tin-can robot animated across the screen, spelling out the word “LOADING.” It was an impressive boot animation. A bright fanfare sounded—ba-bam!—and a large logo materialized in the lower right corner.
“That’s our logo!”
It was the Electronic Arts emblem. After displaying for roughly ten seconds, it faded away.
“How do I watch the ad again?”
“I’m not sure. There doesn’t seem to be a function for that.”
“Then reboot it.”
James restarted the machine. This time, the title screen for ID Software’s upcoming fall release, Wolfenstein, appeared. Though limited to sixteen colors, the game’s atmosphere and tone came through clearly.
“Hmm. Better than I expected.”
Hawkins was thoroughly pleased with the way the ad image popped up and disappeared. He was certain that repeated exposure of the logo and products to gamers in this manner would generate strong positive reactions.
“This thing’s got more going for it than I thought.”
James Web offered his assessment from a software engineer’s perspective. Keyboard Warrior ran without issue. Madden 89, Electronic Arts’ recently released American football title, also performed flawlessly. Unlike Keyboard Warrior, it required creating desktop shortcuts after installation, which was mildly inconvenient, but still easy enough. Older games worked fine as well. Most basic DOS functions could be performed directly from the desktop—formatting floppies, copying or deleting files. A media player even allowed playback of WAV files, though the system could not multitask while performing such operations. While Macintosh computers supported simple multitasking, Android Alpha did not reach that level. Attempting to launch a game icon while formatting a disk prompted a warning that the format would be canceled.
“Because it’s built on DOS, multitasking appears to be possible only in very limited cases.”
James sounded disappointed, yet he could find no other significant drawbacks. There was no need for the complicated memory management DOS required, and even complete beginners could easily use the basic functions. It was, without exaggeration, a far superior next-generation operating system. He immediately recognized that if high-resolution desktop applications were developed for Android Alpha, productivity would increase dramatically.
“High resolution, you say?”
“See for yourself.”
The machine was Hawkins’ test PC—a custom-built system from Dell using only premium components. It featured a 486 CPU and a top-tier Cirrus Logic VGA card with 2 MB of video memory, supporting 1024×768 resolution at 256 colors. It also fully supported the Glide X library, ensuring excellent speed.
James entered the settings menu and adjusted the monitor resolution and color depth. The desktop instantly became sharper, with noticeably finer detail. When he opened the basic notepad application, a vast white workspace appeared—large enough to display an entire A4 sheet on screen. Though the program offered no formatting options and was merely a simple text editor, the impression it left on Hawkins was powerful.
“Thank you. That was very informative.”
James’s task was complete. Clutching the Android Alpha installation disks, he left the president’s office beaming like a child who had just received a fascinating new toy.
Hawkins pondered briefly, then reached his decision.
“How much is the advertising rate? Is there a discount for bulk purchases?”
“Of course. Volume discounts are standard.”
For the first time in a while, Yoo Jae-won was speaking directly with Hawkins in the early hours of the morning. Clearly captivated by Android Alpha, the EA president wasted no time getting to the point.
“There’s one thing you need to understand first.”
Yoo Jae-won began explaining Android Alpha’s advertising policy. Simply placing an ad did not mean it would run endlessly until the program closed. Like television commercials, each ad had a fixed exposure period—100 days by default. After that period, the same advertisement would no longer appear. This measure benefited both users and advertisers. Repeated exposure to the same ad would eventually annoy users, and the companies paying for the ads wanted goodwill, not irritation. Therefore, identical ads were prevented from running consecutively, and the duration was strictly limited.
“Understood. Unlimited repetition would be impossible anyway. The effect would diminish over time.”
Once Hawkins accepted the policy, Yoo Jae-won moved to the main topic.
“The rate is ten thousand dollars for one hundred days.”
That amounted to roughly 6.8 million Korean won. It was far cheaper than television advertising and even lower than a full-page newspaper ad. Given that American incomes were higher than Korea’s, the price would feel especially reasonable. Moreover, computers in this era were frequently rebooted—whenever software felt unstable or sluggish, a restart was the usual solution. It was a notorious weakness of DOS, and Android Alpha, being DOS-based, shared the same limitation. Even if the system itself was stable, a problematic application could force a reboot. Each restart naturally increased ad exposure frequency.
“Ho. Cheaper than I expected. What happens after the hundred days?”
“You’ll need to secure a new advertisement. We’ll operate a maximum of twenty ad slots. Once all twenty are filled, anyone wishing to advertise must reserve a spot in the second major patch. Naturally, if competition arises, the price will rise accordingly.”
“Of course. Those who recognize value late must pay a premium. But what I’m asking is how you intend to deliver new ads to users who already have Alpha installed.”
Hawkins had struck at the core issue. In 1989, the greatest obstacle to running adware was the offline nature of most systems. Fewer than ten percent of American PC users were connected to any online service. They simply turned on their computers and worked offline.
“We plan to include updated ad data whenever we distribute major patches.”
“Patches?”
“As you’ve seen, Alpha still has many unfinished aspects. Being a newly created operating system, it has countless shortcomings—network support is rudimentary, multimedia capabilities are weak, and there isn’t even a basic image viewer.”
“We intend to release major patches every hundred days. They’ll incorporate user feedback, fill in missing applications, and improve performance. At the same time, we can refresh the advertisements we’ve secured.”
“Ah, I see. I understand now.”
Hawkins’ voice brightened. As someone in the same industry, he grasped the business model immediately.
“The question is how quickly you can distribute those patches. Even with online services, you can’t reach every user.”
Yoo Jae-won answered with another question.
“How many retailers does Electronic Arts maintain a strong partnership with?”
For Hawkins, the implication was clear.
“We directly manage around a thousand stores. On top of that, we have about forty major independent dealers, each overseeing dozens to a hundred smaller retailers. Altogether, we’re looking at several thousand outlets.”
“That’s substantial. My idea is to turn those retailers into download centers.”
He was thinking of places like Seewoon Electronics Plaza or Yongsan Electronics Market in Korea—computer shops that sold both legitimate software and pirated copies. They would receive patch disks and provide free copies to customers who came looking for them. If the stores were connected to a network, physical disks wouldn’t even be necessary.
“An attractive proposition. No need to distribute hundreds of thousands of disks for free. Of course, since these retailers will be providing a service to our customers, we’ll need to offer them incentives. Still, it will be far cheaper and more efficient than shipping individual patch disks.”
The initial distribution of Android Alpha would follow the same model. For symbolic value, major well-known stores would receive a simple free package—though quantities would be limited. The software would also be uploaded to CompuServe and other major U.S. PC communication services for free download, while retail outlets could provide copies as well.
“The marketing angle is ‘next-generation gaming operating system.’ Any game built with the Glide X library will run perfectly on a smart graphical interface without complicated memory tweaks or manual hardware configuration. And best of all—it’s free for consumers.”
“Excellent. Perfect.”
Hawkins now fully understood both Android Alpha’s positioning and its adware monetization model. Leveraging Electronic Arts’ distribution network for delivery was innovative as well. If executed well, EA could evolve from a pure game publisher into a company that also offered professional PC utilities.
Yoo Jae-won, whose chemistry with Hawkins was excellent, finalized Android Alpha’s launch date during the call: August 5th—ten days ahead of ID Office. The history of Android, which had begun with great promise, was about to reach an even greater crescendo.