Slug: interview-tantek-celik-conceptualizing-diso-20 Date: 2010-03-11 Title: “Interview: Tantek Celik, Conceptualizing DiSo 2.0” layout: post

This is the second post in a series on the Future of DiSo (the first post is appropriately titled The Future of DiSo).


I first ran into Tantek Çelik back in 2003 as part of the Microformats community - Tantek has been a key proponent of the HTML-based microdata syntax since the beginning, and his patient guidance along with others in that community has given us XFN, hCard, hAtom, and a number of other widely adopted microdata formats.

Tantek has also been a thoughtful member of the DiSo community, chiming in from time to time to advocate for simplicity over complexity, building on earlier work, and working from consensus where possible. He recently posted on his site about his thoughts on what the next “version” of DiSo might look like, and he kindly agreed to chat with me about it.

Here’s part 1 of my email interview with Tantek. Enjoy!

Conceptualizing DiSo 2.0

Steve Ivy: Before looking more at where you see DiSo going, what is your take on DiSo originally envisioned and as it exists now?

Tantek Çelik: That’s a big couple of questions.

As originally envisioned, defines DiSo as “an initiative to facilitate the creation of open, non-proprietary and interoperable building blocks for the decentralized social web.”

I think this is an excellent high level vision that remains relevant and important today.

The home page also says:

“Our first target is WordPress, bootstrapping on existing work and building out from there.”

That “first target” goal provides a good definition for what I’ve been calling DiSo 1.0 - which was certainly a good focus for when DiSo was originally envisioned back in December 2007, when blogging was still people’s primary means of self-publishing, and only a few folks had started using micro-blogging services like Twitter.

Since then many things have changed, but several in particular:

  • Twitter’s popularity. Its usage has in many ways eclipsed blogging.
  • URL shorteners. They’ve become very popular, and awareness of their problems has grown.
  • The rise of microformats. microformats have continued to mature, advance, and become far more popular than XML publishing efforts with more and more publishers and consumers.
  • Authentication via OAuth.
  • Problems with maintaining your own self-hosted CMS and database(s) have become more apparent (more on that later).
  • WebHooks and PubSubHubbub were created, launched, and have solved many longstanding distributed notification challenges.

I think all of these big changes (and several smaller ones) have made it clear that we need to update the notion of what a DiSo implementation should both look like and do.

SI: It seems like DiSo has been targeted at the hobbyist programmer or technical blogger who knows how to run their own site, etc. But even that user is now often moving to hosted or semi-hosted service where they don’t control every aspect of the site for convenience sake, but might want some control. Who do you see as the target user for DiSo-type technologies now?

TÇ: DiSo, at least with its first implementation focus, was very much targeted at the hobbyist programmer or technical blogger. Setting up and maintaining your own WordPress install is no small task.

Even technical hobbyists who may be able to setup and host their own WordPress may simply be too busy to do so. Nevermind having to pay another backup tax.

I think we need to take DiSo back to its egalitarian roots. - Tantek Çelik

I think Matt Mullenweg himself realized all of this and created to provide hosted WordPress blogs for anyone to easily create and use.

The trend of folks moving to hosted or semi-hosted services is certainly real however, and frankly, as a result it feels like we’ve lost ground from a DiSo perspective.

Even the World Wide Web Consortium (W3C), who is well known to always host their own content, services, mailing lists etc., are using and Twitter to post updates which are often not on their own site: (W3C on, on Twitter).

I’m no exception. I’ve been posting photos on Flickr since 2005, Twittering since 2006, and using PBWiki for my personal wiki pages since 2007.

Eventually I was posting more to other services (mostly Twitter) than my own personal site which itself fell into neglect in late 2008 where it stayed dormant until I began rebuilding it this year with an eye towards DiSo 2.0.

To answer the last part of your question “Who do you see as the target user for DiSo-type technologies now?” I think we need to take DiSo back to its egalitarian roots.

I’d like to see DiSo technologies support any individual or organization who owns their own domain name.

You shouldn’t have to be a sys-admin, a technologist, or even a hobbyist.

Maybe you’re an independent consultant, or a small company, or maybe you just firmly believe in owning your own content. Maybe you worry that hosted services can and will be bought and shutdown, suddenly silencing millions of permalinks (e.g. Yahoo/Geocities, SixApart/Pownce). All those posts lost forever, like tears in rain (with apologies to the respective movies).

All it should take is:

  1. Your own domain name (or names, including a custom shortener)
  2. A Web site hosting service with a fairly average hosting company
  3. FTP/SFTP/scp access to your site

It needs to be that easy.

People quickly understand all three of these:

  1. Owning your own domain name is not much different than owning your own cell number and being able to take it with you (as opposed to sharing a home or work landline with a bunch of other people).
  2. A Web site hosting service is similar to a mobile phone provider - you pay a monthly fee for your number (domain name) to work when people call (browse) it.
  3. FTP access is just like moving files around in folders, which already has broad understanding thanks to the ubiquitous desktop metaphor.

Thus at a high level, with DiSo 2.0 we’re really talking about two big areas of shifts:

  1. Modernization. Many technological changes have occurred over the past several years that need to be incorporated.

  2. For the rest of us. Broader access and usability through a focus on friendlier and simpler user experience, from install, to use, to maintenance.

Fortunately these two areas of shifts are complementary and together help drive a much better design and solution than either would on its own.

SI: When did you start thinking about the “next” version of DiSo, and what was the inspiration behind your list of 10 “features”?

TÇ: I’ve been thinking about DiSo ever since the term was first coined, as it served as a very powerful focusing force to help drive open web standards development. After having my own struggles with WordPress (backups, vulnerabilities, exploits, and updates are all a pain) and witnessing others’ struggles such as Eric Meyer’s problems with WordPress, MySQL, and UTF-8, it became clear to me that for DiSo to reach a broader set of users, it needed an implementation redefinition.

As far as my specific 10 features, I analyzed the above-mentioned technological shifts over the past several years and realized that each of them pointed towards key changes in what a personal site owner needed to do.

Once I started sketching, prototyping, and constructing the building blocks necessary to accommodate those changes, it became clear that what I had started to design and implement very much fit the original DiSo vision, and yet had replaced much of its initial technology implementation stack.

In writing down a concise list of the building blocks it was clear that I had come up with a modest proposal for a DiSo 2.0. Now we just have to finish building it.

More to come, stay tuned.