(This was originally posted to my Slashdot Journal, which allows comments and is available here.)

So I recently ran across a new site, courtesy of the fine folks at MetaFilter: Rememble. In a nutshell, it’s a sort of ‘digital scrapbooking’ site. It describes itself as “a ‘washing line’ for your digital bits and pieces. Thread together texts, photos, videos, sounds, scribbles, scans, notes, tweets… so they’re not drifting in a digital wasteland.”

As a compulsive digitizer, I’ll go first and say that it sounds great.

There are a lot of services that provide the ability to save little text snippets for later (Google’s Google Notebook, when coupled with the appropriate Firefox Addon, comes to mind), and Flickr is the gold standard for digital photo organization and sharing, and there are similar single-media sites for other purposes. However, there’s a distinct lack of a single site that allows you to collect, view, organize, scrapbook, and share various types of digital media in a cohesive format. And that’s a darn shame: as more people get online and involved in modern interactive services, as they get more of their lives online, it’s only natural that they’ll want to be able to save parts of it for later, just like they do in the physical world. (And, of course, being virtual lets you do things in an online notebook that you can’t easily do in a dead-tree one, like suddenly decide to view all your clippings by date instead of by subject.)

Unfortunately, Rememble’s execution — at least at the moment — falls flat. For a site that treads on being almost postmodern, its approach seems driven by a desire to create a vast silo of exploitable content.

First major gaffe: you can’t see anything without registering for an account. That’s right, nothing. So let’s say you set up an account, dump a lot of stuff into it, and then want to share it with some friends? Nope, sorry, they all have to sign up for accounts. This is such a major, deal-breaking limitation, it’s hard not to immediately think of one of those ubiquitous “FAILURE” image macros. I can only hope that this is some sort of limitation due to the service being new — I mean, they can’t really be that stupid, can they?

Similarly, you can’t deep-link to content that you upload. That’s right; you can’t embed things you upload to Rememble on your blog. While this isn’t as obvious a death-wish as the lack of sharing ability, it’s potentially more damaging. Flickr succeeded in its early days mainly because it became popular with bloggers looking for an alternative to services like ImageShack that didn’t suck quite so badly.
Flickr offered one-click tools for resizing an image and embedding it into a blog post. It was slick, people loved it, and they got a community of users rather quickly.

Beyond that, there doesn’t seem to easily be a way of getting content out of Rememble once you’ve gotten it in. This bothers me, personally, although it may not be the sort of thing that a casual, non-backup-obsessed user might think of. (Though, in my opinion, they should.) A service like Rememble could, over time, end up being a significant repository of information and digital relics; having your Rememble store disappear would be like having your family scrapbooks torched.

After taking a casual look at Rememble, and comparing it to a successful service like Flickr, a number of concrete steps come to mind for, if not actually ensuring the success of a community-oriented “Web 2.0” media-sharing site, at least making it slightly less prone to sucking:

1) Sign-ins should only be required for content creators, never viewers. Even a free, one-minute signup procedure is one minute too long to expect random people I might want to share content with to go through. It’s unnecessary and borders on arrogant.

2) Prohibiting blogging and direct linking may seem like a good idea, but it’s not. Really. The people who are going to want to blog and direct-link are also the ones who are going to make or break your service. Don’t alienate them 30 seconds after they upload their first bit of media. Yes, it may burn you to spend money on bandwidth so your users can use you like ImageShack, the Internet’s cheap village whore, but chin up: everybody has to start somewhere.

3) Expose your APIs, and encourage third-party development. (To be fair, I’m not sure what Rememble is doing with their APIs; maybe they expose them and just aren’t obvious about it.) Use standard interchange formats whenever possible. Since exposed APIs are considered one of the keys to useful, modern web services, they really need to get this right. Luckily, Flickr has a good model. Follow it. Also: Let users export content, not just import it. Acting like the NSA, hoovering up stuff and never letting anything slip back out, makes people justifiably nervous.

4) Provide a way for backups. Also: nobody likes commitment. Don’t expect users to trust you, your datacenter, your RAID array, or your backup strategy. For all we know, you’re running this thing on a spare server that your boss could repossess at any time. Provide users an easy way to grab a snapshot of everything they’ve created (a big tarball of media files and XML metadata) for their own peace of mind. Also, people like knowing that they have a way out if things go sour.

If Rememble took those four steps, they would probably have a service that I’d use right now — at least for trivial stuff. From there, the sky’s the limit.

Of ‘second tier’ features, an ability to encrypt content using an open-source, client-side applet (so that it gets encrypted by me, not by the server on the far end) would be nice, particularly when you’re talking about automatically archiving text messages and other communications that may be sensitive now but nice to have later — perhaps this could be offered as a premium service? If you do it right, with full auditability, you might even get corporate interest.

What really would make a service like Rememble outstanding are the interfaces. Imagine plugging a service like this into your SMS/text-messaging service from your phone, your email reader, and your IM client (archiving both conversations and status messages): you’d have a single online archive of all your communications. Privacy nightmare? Quite possibly. But it would also be handy; no more trying to remember how somebody sent you a bit of information. Plug it into your address book, so that you could cross-reference other people’s online identities, and you’d be able to see all communications with a particular person over time, regardless of medium. Or run a quick search and you could see all the people you discussed a particular topic with.

I find the possibilities for a Rememble-like service pretty exciting; for someone who really likes compiling and managing information, it’s just oozing with potential. And more than anything else, that’s why Rememble is painful: it takes something that should be mind-blowing and renders it in a form that’s lame and unimaginative; without an obvious grasp of what web services are all about.