The mobile internet, how the iPhone changed everything in 2007

Field note. About half your players will connect from a phone or tethered to one some of the time. Mobile networks misbehave in specific ways that produce specific kinds of disconnects.

In 2006, "the internet" meant computers. Desktop or laptop. Big screen, keyboard, mouse. Phone-based internet existed but was awful: tiny screens, slow connections, terrible UIs, mostly used for nothing.

By 2010, the internet was mobile-first. By 2020, more than 60 percent of web traffic was from phones. By 2026, mobile is decisively dominant in most countries.

The pivot happened fast. This article tells the story.

Mobile internet before the iPhone

The era 2000 to 2007 had several attempts at mobile internet:

WAP (Wireless Application Protocol). A simplified web for mobile phones. Tiny pages, monochrome text, terrible to use. Carriers heavily controlled what was accessible. Largely a failure with consumers.

i-mode (Japan). A more successful mobile-internet platform built by NTT DoCoMo. Worked because the Japanese market was specific and DoCoMo controlled the experience. Didn't translate to other countries.

BlackBerry. Excellent for email (especially corporate). The "addiction-grade" device of its era. Web browsing was secondary and rough.

Windows Mobile, Symbian, Palm OS. Various smartphone-like devices in the 2003-2007 era. Browsers were available but cramped. Most users didn't browse the web on their phones.

Common thread: mobile internet existed but was unpleasant. People had it, didn't use it much.

The iPhone arrives

January 9, 2007. Steve Jobs introduced the iPhone at Macworld. It was sold starting June 29, 2007.

What was different:

• Full multi-touch screen. Most of the front of the device was screen. Pinch-to-zoom, tap, scroll natural.
• Mobile Safari. A real web browser. Not WAP. Actual websites rendered correctly (or as close as possible at the time).
• High-quality rendering. Pages looked like pages, not stripped-down WAP versions.
• Excellent UX. Smooth animations, responsive, didn't feel like a phone with software glued on.

The pitch was specifically "the internet in your pocket." Jobs emphasized that this was a real web browser, not a watered-down mobile version. The full web. Just smaller.

It was expensive ($499 to $599 with a 2-year AT&T contract). Sales were modest at first. But the device changed expectations.

The carrier control problem

Pre-iPhone, U.S. mobile carriers (Verizon, AT&T, Sprint, T-Mobile) controlled what apps and services could run on phones. Manufacturers built phones to carrier specs. Apps required carrier approval.

Apple's deal with AT&T for the iPhone was unusual: Apple controlled the software experience. AT&T provided cellular service but didn't interfere with how the iPhone worked.

This carve-out broke the carrier-control model. Once Apple proved it could be done, other manufacturers (Google with Android) followed. The carrier-control era ended.

The App Store

The iPhone in 2007 didn't have third-party apps. Apple's idea was that the Web was the platform: web apps would run in Safari.

In 2008, Apple changed direction. The App Store launched July 2008. Developers could write native iPhone apps, distributed through Apple, with Apple taking 30 percent of revenue.

This was the inflection. Apps proliferated. The phrase "there's an app for that" entered the language. By 2010, the App Store had 200,000+ apps and had paid developers $1+ billion.

The model spread. Android Market (later Google Play) launched in 2008. The "app economy" became a real thing.

Android

Google had acquired Android Inc. in 2005, planning a mobile-OS strategy. The iPhone arrived in 2007; Google rapidly pivoted Android to be more iPhone-like.

The first Android phone (HTC Dream / G1) shipped October 2008. Google's strategy: open-source OS, license to any manufacturer, focus on advertising revenue from Google services on the device.

Android grew rapidly. By 2010, Android had passed iPhone in market share globally. By 2015, Android was running on 80+ percent of smartphones worldwide.

The Apple-Google duopoly emerged: Apple's premium iPhone with high margins; Android's distributed manufacturer ecosystem with lower margins and broader reach.

What mobile web traffic actually became

Through the 2010s:

• 2010: ~10 percent of web traffic was mobile.
• 2015: ~40 percent.
• 2020: ~60 percent.
• 2026: ~65 percent globally (higher in countries that skipped desktop adoption, lower in established markets).

For many people, especially in countries where smartphones leapfrogged PCs, mobile is the only internet. The "computer-first" world ended in much of the global south sometime in the mid-2010s.

What this changed about the internet

The mobile pivot changed many things:

Site design. Pages had to render on small screens. Responsive design emerged. By 2015, "mobile-first" was standard advice.

App vs Web. Many services moved to dedicated apps. App stores became gatekeepers. The Web lost some ground to native apps.

Performance pressure. Mobile networks were slower than wired. Sites had to load fast. JavaScript bundles were criticized for being too heavy.

Touch input. Mouse-and-keyboard assumptions broke. Hover states, right-click menus, double-clicks: all reconsidered.

Location-aware services. Phones knew where you were. Mapping, ride-hailing, food delivery, dating: all leveraged location in ways desktop couldn't.

Camera as ubiquitous tool. Phones with good cameras drove Instagram, Snapchat, TikTok, video calls.

Notifications. Push notifications became the interrupt mechanism for the internet. Sometimes useful, often abused.

The internet's character changed. It wasn't a place you visited; it was something you carried.

The new platforms

Mobile created new entire categories:

Ride-hailing. Uber (2009), Lyft (2012). Required phones with GPS, payment, app stores.

Photo sharing. Instagram (2010). Phone cameras + casual photo sharing.

Messaging. WhatsApp (2009), Snapchat (2011), Telegram, Signal. SMS replaced.

Food delivery. DoorDash (2013), Uber Eats (2014), local equivalents worldwide.

Streaming on the go. Spotify, Netflix mobile, YouTube mobile.

Mobile games. Angry Birds (2009), Candy Crush (2012), Pokémon Go (2016). A genre that didn't exist before phones.

Mobile banking. Banks built apps; many users never used desktop banking again.

Each of these would have been impossible or marginal without smartphones.

The platform-as-gatekeeper concern

Apple and Google control the app stores. Together they're effectively a duopoly. This concentrates power:

• 30 percent platform fees. Apple and Google's take from paid apps and in-app purchases. Effectively a tax.
• Approval gatekeeping. Apps can be rejected for reasons that aren't always clear.
• Banning power. Once approved, apps can be removed, sometimes controversially.
• Native vs web app friction. Web apps don't get the same privileges as native ones, pushing developers to native and to paying fees.

Regulators have noticed. The EU's Digital Markets Act (DMA) addresses some of this. Multiple antitrust cases continue. Apple's 30 percent has been reduced for smaller developers (15 percent), but the basic structure remains.

For users, mobile platforms are convenient but constrained. The full openness of the Web exists on the side; the main experience is mediated by Apple or Google.

5G and what didn't happen

5G cellular started rolling out around 2019. The marketing promised:

• "1 gigabit per second to your phone."
• "1 millisecond latency."
• "The end of fixed broadband."

Reality (2026):

• Speeds vary wildly. Some 5G is great. Much is barely faster than 4G.
• Latency improved but not dramatically.
• Fixed broadband is fine and 5G hasn't replaced it.

5G is a real upgrade. The hype was excessive. As often happens with mobile-network generation upgrades.

Mobile gaming

A specific worth-noting subculture: mobile gaming.

For casual users, mobile is now the dominant gaming platform globally by player count. Free-to-play games (with in-app purchases) generate billions in revenue. Genres include puzzle, casino-style, RPG, idle, gacha (Japanese-style randomized loot).

Many traditional gaming companies have shifted resources toward mobile. The economics are very different from console / PC gaming: lower ARPU per user but vast scale.

For hosting companies, mobile gaming has implications:

• Mobile games often connect to backend services (login, leaderboards, social, in-app purchases). These need hosting.
• Real-time mobile multiplayer is harder than console (more latency variability, more variable connections).
• Cross-platform play (mobile + PC + console) is increasingly expected.

The post-app trend

Some signs that the app-dominant era is plateauing:

Web apps. Modern Web (Progressive Web Apps, etc.) can do much of what native apps do. Adoption is slow but real.

Operating-system features. OS-level integrations (Apple's Live Activities, Android Widgets) blur the app/OS line.

Notification fatigue. Users uninstall apps that pester them. App stickiness is harder to maintain.

AI assistants. Conversational interfaces that span what used to be many apps.

Whether the app era is "winding down" or just "settling into maturity" is debated. Likely the latter.

Conclusion

The iPhone in 2007 didn't invent mobile internet. It made mobile internet pleasant enough that everyone used it. Within five years, the world's relationship with the internet changed from "computers connected to it" to "the internet is what your phone does."

This transformation is one of the largest in computing history. Possibly the largest. It made the internet truly ubiquitous, created vast new business categories, concentrated power in two platform companies, and changed how humans interact with information.

For the internet's infrastructure, mobile means: more traffic, mostly through cellular networks, with different latency and reliability characteristics than wired. Mobile drove investments in mobile-optimized CDNs, in mobile-aware routing, in mobile-resilient protocols (QUIC, HTTP/3).

For users: most of the internet you experience is now mobile, even if you don't notice.

Coming up

Next: the cloud era. Starting with Amazon launching S3 in 2006, the rise of "use someone else's servers" transformed how software is built and deployed.