This morning SlideShare launched a complete rewrite from Flash to HTML5, and a new mobile web app. You would think that as the world's largest social content network, with 60M monthly unique visitors (you too can be unique) this would be a big deal. Here's Liz Gannes from All Things Digital:
SlideShare Does Biggest Redesign Ever… But It’s Invisible!
Presentation-sharing service SlideShare today joins the HTML5 craze, and will transition its largely Flash-dependent site out of Flash into HTML5, making it significantly faster and able to be accessed by iOS devices. Multiple executives at SlideShare described this as their biggest project undertaken to date ― but said users can expect the site’s appearance to change little, if at all.
That is the entirety of Liz's post. A classic. Love it.
But I think there are two important points here for entrepreneurs. The role of rewrites and, more importantly, the web is back.
Pivot 2.0
I am so very tired of the word pivot. Yes, entrepreneurs need to change course when things aren't working. But a better definition of entrepreneurship is experimentation at the margin. You take your hypothesis and prototype and make small moves at the boundary of what you know and do.
The hardest move in many cases is a rewrite. Scrapping old code for new. Chosing to redo old features instead of developing new ones. Investing in better architecture over fresh paint.
What makes it especially hard is customers are more likely to pay for new features. Markets for startups are asymmetric, with your buyer having less information about your product's current and future state and the viability of your temporary organization. The only time buyers care about your architecture is when it fails so badly they will cancel on you. So you can't just listen to customers on this one, you have to listen to yourself.
One of the worst decisions I ever made was telling Sun they couldn't fund a rewrite in Java and delay our market entry by another six months, so we could keep the first six months of Perl. Longer story, that one.
In this case, the decision was easier -- listen to Steve Jobs. Steve decided to kill off Flash. Which, if you are written in Flash, kindof sucks.
Still it wasn't an easy decision. Five years ago the only viable way to build SlideShare was Flash. Today it still works, but doesn't on iOS. This year we developed Zipcast, which was made in-part possible by HTML5, which wasn't possible five years ago. This year the team made the very hard call to forgoe new feature development and do a 100% rewrite to HTML5. Read CTO & Co-founder Jon Boutelle's post on how the team did it.
As we got into the project we learned we could make SlideShare 30% faster, reduce page load by 40%, retain document fidelity in the conversion and redendering process, and consolidate on a single platform. The web.
The Web is Back
HTML5 is largely a marketing term, but it represents the evolution of HTML, Javascript and CSS standards which are now adopted by every modern browser. Thanks, in-part, to the marketing term Ajax coined by the same guy who had just coined Blog, innovation flourished in that term Web 2.0.
Now I don't think we are at the beginning of such a platform shift on the web. The real platform shift is people, and how their behavior with the web is changing at mainstream scale. But there are new things you can do, like how Pusher uses HTML5 Websockets for real time messaging. And more importantly, HTML5 is a more than viable platform for mobile apps like SlideShare.
Yesterday Marshall Kirkpatrick blogged: Firefox Creator Says the Web is Dead Meat; Android Creator Disagrees. Joe Hewitt argues that web technologies need an owner to compete with the curated app store model. This goes against core principles of the web (nobody owns it, everybody can use it and anybody can improve it), but those are more under attack by governments than standards bodies that move slowly.
The Web is the Biggest App Store
I really don't see entrepreneurs heading towards iOS or Android for innovation sake, more for the distribution opportunity. And when they do, they face the hard decision of developing on multiple platforms that made software such a pain in the ass before the web. Let alone potential revenue share.
And I don't think the web needs an app store, let alone an owner. Putting your app in a store only gives you distribution through the owner's model of curation. On the web, you market your app by giving back to the web. Sharing information and links that benefits the web's structure.
Think of your startup as a band and your app as music. The real distribution opportunity isn't happening in app stores. It's when your music is shared as a social object by people. And increasingly in seamless ways like Facebook's latest Spotify integration (yes, proprietary P2P, its a metaphor). You want your app to to be sharable across the social web in a way where people can engage with it in less than 2 seconds, not in over 2 minutes after downloading.
Five years out I'd bet on the web being the most successful mobile platform thanks to improvements to come, and its more organic distribution opportunities.
