If you run a self-hosted WordPress/Movable Type blog, you’re already there. Instead of S3, you’re hosting it on your hosting provider’s space (which could well be S3). In fact, this is how rahulgaitonde dot org works.

Now Joshua only alludes to this, but these pieces aren’t coupled loosely enough to move to plug out one component and fit another in. For example, I can’t take the Feedburner RSS component out and replace with another – my RSS feed URL is tied to Feedburner. I can’t move my template transparently between Blogger and Movable Type.

Back in May 2008, I had similar thoughts about separating the email interface from email storage:

There is a market for start-ups that provide only an interface for existing email. For people who are willing to pay for (cheap) storage of their email and for bandwidth. Users will be able to migrate from and to such services without needing to copy huge amounts of email to their new email provider.

I wonder whether in the future we’ll eventually build such a decoupled email system, or find an alternative to email altogether.

The aggregate results of every search ever entered, every result list ever tendered, and every path taken as a result. … a massive database of desires, needs, wants, and likes that can be discovered, supoenaed, archived, tracked, and exploited to all sorts of ends… this artifact can tell us extraordinary things about who we are and what we want as a culture.

That phrase made it into his book “The Search” and quickly became a popular way to demonstrate how enormously important and powerful Google might eventually become.

This last Saturday, the New York Times ran a piece on a possible new monetization idea that Twitter was considering: to “offer shopping advice and easy purchasing”. People already solicit their Twitter followers’ opinions, and it is also already possible to identify real-time trends related to a particular product, company or event. Put those together, and you get an extremely powerful (and, the founder hope, lucrative) tool.

Viewing this piece of news in the context of “database of intentions” you can see how the web has evolved since Battelle propounded that idea:

One, Twittter is now a database of actions, of people announcing by-the-second what they have tried, used, bought, rejected, liked and disliked. I see an attractive opportunity for an analytics firm to help companies make sense of what people are saying about them, what events caused this conversation, and the results of a company’s actions/response on the conversation and subsequent sales/signups.

Two, it is still a database of intentions, but at dizzying, real-time speeds. From the New York Times article:

“Commerce-based search businesses monetize extremely well, and if someone says, ‘What treadmill should I buy?’ you as the treadmill company want to be there,”

While it’s certain that companies can use these intentions to snap up customers before competitors, it’s unclear as yet how companies will be able to scale and respond if and when Twitter achieves Google’s adoptions levels. There is definitely an opportunity for another business here.

Which is all very fine. Until the day you die.

What happens to your digital possessions after you’re no longer around? It’s a question without a good answer, mostly because it hasn’t been asked often enough. Understandably. The Internet’s only been around some 15 years, and we’ve only begun putting personal info on the Web (Here’s a broad list) for about 5 or 6 years. In other words, very few of us from the Internet Age are dead yet.

But we’ll need answers in the next few years. Archiving and preserving a departed loved one’s online possessions is going to be a huge opportunity not so long from now. I say opportunity because things aren’t as straightforward in the online realm as in the physical, and there’s plenty of scope for smart thinking and innovative solutions.

Imagine you’re a startup that specializes in archiving digital creations. You’ve been commissioned by the departed’s relatives to preserve digital memories. Consider three issues you’d face:

Tracking:

What did your client (well, client’s loved one anyway, let’s just call him/her the client) create on the Internet? You can cover the obvious - email/chat/blog/microblog/photos/videos/social network. Then you get to the hard stuff: all the comments he/she’s posted on websites, forums he/she’s been active on, scraps/wall posts on friends’ social network pages, old email accounts he/she might have had in the past. and so on. 

Right now there is no reliable way of tracking this. How will you go about this?

Ownership:

Who owns data that your client had put up? The answers for some of these are straightforward - does Google own your videos on YouTube? Does Yahoo own your photos from Flickr? Read the fine 
print. But what about the scraps/wall posts your client wrote on his/her friend’s Orkut/Facebook profile? Comments on his/her friend’s blog? One view is that since they’re on the friend’s Orkut profile, they belong to the friend. The counterview is that scraps belong to whoever wrote them. 

Matters are further complicated if the client had stated before death that he/she wanted this sort of data deleted post-death. Will the owner of the blog that your client had commented on allow it?

Another question is about transfer of ownership. If Alice has had an email conversation with Bob that she would not want anyone other than Bob to view, should she have the right to veto the transfer of his email account to his next of kin? Perhaps she revealed her birthday and birth year to Bob. Could she veto the archival of his calendar?

Context:

This is closely related to ownership. Often, data by itself is useless without the context it was originally created it. A comment your client left on a blog post has very little value without its original blog post. A scrap/wall post or a “reply” tweet even less so. A pretty picture your client clicked and uploaded on Flickr is greatly diminished in value, significance and memory without the comments it sparked. A social network profile without the accompanying network is hardly social. But archiving the context along with your client’s content will raise the above ownership issues.

These are problems we haven’t faced with physical possessions because these problems never applied to them. How we sort them out is a both a tricky business and a business opportunity.

…a licensing program that gave hardware manufacturers the ability to build Kindle clones, along with an incentive to sell them at near-zero margins. Amazon would give those manufacturers access to the core Kindle hardware specs (there’s no real magic there anyway) and the right to call it a Kindle device so long as they also put the core Kindle software on the device. That software links the device to Amazon’s store, meaning downloads revenue flows through Amazon.

Amazon would then share a percentage of net margin generated from downloads with the hardware manufacturers.

Techcrunch has put into words what I’ve felt since the day the Kindle was announced. After all, Amazon isn’t in the hardware business at all; it’s in the product and content retail business. I can imagine that in the initial days of the Kindle launch, Amazon needed its own device to build a strong association between Amazon’s brand and the mobile ebook model. Now that that purpose is served, manufacturing and selling the Kindle hardware is an overhead that Amazon could avoid.

Just like Associates?

This isn’t very different from the masterstroke that Amazon played years ago with its Associates affiliate program. Before Affiliate Marketing became the wild jungle that it is today, Amazon launched a series of innovative tools – aStore, Omakase Links, Product Previews – to let publishers (people who owned websites/blogs/suchlike) add links to Amazon’s content onto their web pages. These publishers then earned a cut of the sale generated by clicks from the links on their web pages.

Kindle is Associates all over again, except instead of web-based tools, we’re talking hardware specs.

For instance, Amazon’s aStore let developers build their own focused online “stores” (which displayed Amazon’s books). (A religion-focused website would be able to draw viewers and sell that category of books better than Amazon.com itself.) In the same vein, a student version of Kindle with access to e-textbooks and additional bookmarking features would be better marketed and sold by a third party which is focused on only that market.

With such an Affiliate/Franchise/Licensing model, manufacturers would fall over themselves for a chance to access Amazon’s massive ebook and newspapers database – and a cut of the subsequent revenues.

The Mobile Opportunity

Once third party manufacturers have licensed the Kindle specs, they are no longer restricted to building anything that looks like the Kindle today. I can readily think of well-designed iPhone/iPod Touch ebook applications like the New York Times app. This fits in with American universities doling out iPods Touch and iPhones to their incoming freshmen.  A market for Nokia’s S60 devices would be many times larger.

What do you think? Would you purchase a Kindle application for your mobile device?

Aside: Of course, manufacturers would then be free to choose the carrier of their choice for wireless content delivery. That sure isn’t going to make Sprint-Nextel happy.

The Background

It’s no secret that the inbox paradigm we use today wasn’t designed to handle today’s volume of communication. Several start-ups have ideas to fix this, notably Xobni and Xoopit. Others want to go deeper and rethink the email client. Their biggest barrier to their entry is that data stored with an existing email providers (esp. webmail) is impossible to get at, short of copying that entire data. Zenbe, for instance, is an email client startup that is importing email from other POP3-enabled email accounts (including Gmail) to its servers.

The Opportunity

I’ll admit this is far-fetched, some way into the future, and has a relatively small target audience.

If users were to move their existing email into a database in an online storage service like Amazon’s S3, then several email clients could access this data, perhaps even simultaneously. If a user chose to move from one email client provider to another, the data itself – email, contacts, calendar, tasks – would stay intact; the new client would only have to create a new index for that data, and the old index could be deleted.

For users with email addresses from their own domain (such as rahul at rahulgaitonde dot org), even the email address would remain the same. The user would only pay Amazon for the storage and the bandwidth; the email service is still free, typically ad-supported. You could even use IMAP and access this from a desktop or mobile client.

The Upside

Innovation: Start-ups will be free to “fix” the traditional idea of an email client without having to worry about storing or importing tons of email.
Data ownership: Since the storage of is distinct from its interface, the user has far more control over their data than with today’s webmail services.

The Downside

Market size: The notion of having to pay for email storage seems anachronistic today. Most users also don’t care about what client they use. I contend that there will still be sizable numbers of users willing to pay, enough to make such an email service work.
Loyalty: While it is easier for such start-ups to acquire new customers, it is also easier to lose them – the cost of moving to a new interface is almost zero.
Failure point: Should the online storage service experience downtime, email will be unavailable.

Whenever I use del.icio.us I simply save Web pages from the plug-in on my browser, and rarely actually go to the site. I’d estimate that my ratio of saving things to going to the site is 10 to 1, maybe even 20 to 1.

Quite right; I save to del.icio.us for a number of reasons: reading articles when I have time later, archiving howtos, and so on. But the web interfaces of both GB and del.icio.us aren’t very well-designed, and don’t lend themselves too well to retrieval. So “social bookmarking” is, in practice, a one-way street for several of us.

Enter Firefox 3′s AwesomeBar. By adding the ability to search bookmarks by tags, it is already a readily-accessible, ubiquitous interface to your bookmarks.

Those words to the right of each URL are tags for local bookmarks

Prima facie, it’s a look-no-further solution to the major current online bookmarking woes.

Though, of course, it takes the “social” out of social bookmarking :)

Right now, I still need to navigate to the website (out of my RSS reader) and comment in the little form at the bottom of the page. That is *so* old-world Web!

Of course, several blogging platforms (WordPress for one) offer RSS feeds for comments, but having separate feeds is non-optimal for two reasons. One, comments are only useful in the context of the parent post. Viewing it as an independent feed doesn’t fit in with that. Two, it still doesn’t solve the problem of adding comments.

An RSS reader is still a one-way street, funneling content from several websites into one location. Ideally, all of a user’s interactions with a website (read/write/view) could be done via RSS feeds. Because, after all, that was the intent of publishing a feed in the first place – to save the user the tedium of having to actually *visit* the website.

Imagine. Posts. Comments. Advertising. Live chat. All through the RSS reader.