In response to a previous post where I asked “Where is the ‘open web’ now?” I received some of the most interesting feedback I’ve ever gotten in all The Millikan Daily’s years online. In attempting to pinpoint where this site lies in terms of web classification, I learned it’s definitely not on the closed web, but it’s still not totally on the open web. Maybe nothing can be in either place in absolute terms, but we can definitely be working to make more things tend toward the open web.
The first comment hails from Matt Terenzio, who I thought about when I wrote the initial post because we used to talk about this stuff at a weekly web discussion hosted by Dave Winer at NYU. Matt wrote:
WordPress, even the hosted stuff on WordPress.com is more open because you can export your content and move it. https://en.support.wordpress.com/export/
But you raise a serious point. Even if you have control of your data, it doesn’t mean it stays alive on the web after you die. No business can guarantee that. A library, educational institution or the government seem more capable to pull something like that off, but as of now, we don’t have a great solution.
He raises a few important attributes of online content in the scope of the open web:
- Exportability – Can you take your content and move it somewhere else? With WordPress, the answer is yes, technically. WordPress offers this, as Matt points out, though maybe you have to pay something if you’re using the .com version (as I am) rather than the .org version. When you export, WordPress wraps all your content up and spits out a nice, zipped-up file that you can send elsewhere.
I’m going to add to this a few other related aspects to consider:
- Archive transfer – Once you have your data exported from the CMS, you can transfer it somewhere else a few different ways. The old-school way would be physically via a hard-drive transfer, but more likely you’d do it online. This poses an interesting conundrum though, as various entities are at work to clamp down on the ability of average users to transfer large files. Why? Maybe because the assumption is that a large file transfer will be used for malice, like ‘illegally’ downloading a movie file. I’ve never downloaded anything illegally, to my knowledge, so it’s unfortunate that the tools I have to work with are limited in this respect. Once a friend gave me a chunk of the server space he owned and maintained so I could transfer whatever to and from it without paying extra. If I wanted to do it now, I would probably have to pay for some cloud hosting service who then might technically be able to access the data I was transferring. Right now we’re talking about being open so this doesn’t matter at the moment, but later on it will.
- Importability – With facebook, I haven’t ever heard of someone importing their timeline to a different online framework where archives and data is displayed in any meaningful way. Which isn’t to say it couldn’t be done, I’ve just never seen it. If you have, please comment below.
Moving on, Matt also brings up an important second topic:
2. A shelf-life of content dependent on mortality – Even if you are diligent in following all best practices to maintain your own slice of the open web, there’s no guarantee they will continue to be maintained after you die.
Of course there are services that make it more likely your content will be maintained and accessible online at the original hyperlinks post-mortem. I met a woman in Iceland who founded a start-up to do something similar, though I’m not sure if it is still operational. I just emailed her to see what’s up.
Then of course, there is the Internet Archive which hosts the Wayback machine. While there are ways to get them to prioritize the archiving of specific websites, I believe their methodology is mostly random, and they will save “snapshots” of certain sites at various moments of time.
Which brings me to our next comment via Scott Rosenberg:
Hey, Arikia — I think of “the open web” as more of a spectrum; some sites and services are more fully “open web”-ish and others less so. If you own your own domain and pay for hosting for your site then that gives you the most control/ownership and puts you in the best position to preserve your work. (I’m still hosting pages I first posted in 1994!) The IndieWebCamp people and their work are definitely worth checking out — building open source tools and protocols for self-owned and maintained sites to thrive and connect with the various silos. David Weinberger’s book “Small Pieces Loosely Joined” put together a lot of the strands of what made the “open web” work and become valuable during the first wide flourishing of blogs beyond the tech scene in the early 2000s. Openness/permeability to links is so central here — one reason Facebook feels so closed is that you can put a single URL into a status update but you can’t simply add links to your content the way you can anywhere else on the Web. To me that’s what really makes it a closed system…
One thing is for sure, I wouldn’t have gotten this feedback had I posted this question on facebook! Two main points here:
- The open web as a spectrum – Cool. I like spectrums. Many of the human systems we are taught to think of as binary for the sake of mental simplicity are indeed spectrums—why wouldn’t the web be the same? Of course this introduces another layer of complexity to the answer to the question: Is any given website open or closed? It would be easy to lump everything into a bucket of open or closed, but it seems the answer will usually be “neither—it’s some point on a spectrum between open and closed.” Does the spectrum have endpoints? If so, what are they?
- Control/ownership – Scott owns his domains and pays for hosting, so he inherently has more control to make his content more open OR more closed, as he wishes. Let’s focus on the open side here, because keeping closed content closed is another can of worms I’ll want the crypto folks to chime in on.
Remember PicPlz? If you don’t, it was a photo-sharing application similar to the Instagram we all know now, but it was mainly marketed to Android users, which is why I used it, because I used Android smartphones for years before I was given my first iPhone in 2013. I uploaded a bunch of pictures to PicPlz, which were shared to other users through links created by the app that were anchored on its domain. When PicPlz folded, all those links evaporated. Who knows what happened to the images themselves, probably deleted. I don’t remember if they provided an option for the user to export their data, or if I chose to if they did. But it goes to show that if you depend on an entity outside of yourself to maintain the integrity of the links to content you create over time, well, you probably shouldn’t care too much about that content, because you’ll have no control over if it stays where you put it.
Next, Eas provides a recipe for their online content maintenance:
My general approach is:
1. Register my own domain name, separately from the publishing/hosting platform (so I can move things even if the publishing/hosting service goes out of business suddenly, or we end up at loggerheads).
2. Use a publishing platform that makes it easy to export data, including comments. I’m using WordPress.org, hosted on a virtual linux server from a established web hosting company with a seemingly sound business model (I pay them every month).
3. Publish under my own domain.
4. Make daily backups.
5. Switch platforms/hosts as needed.
I don’t use wordpress.com, but it would fit into my approach, since they let you bring your own domain, and provide a way to export everything. The exported data can then be imported into self-hosted WordPress, or another system that supports the format format.
In the longer run, I’m thinking of exporting dormant sites into a static format and hosting them on something AWS S3, with the knowledge I could move them to any other static file hosting in the future.
This all requires some ongoing effort on my part. At the very least, I have to keep paying the bills, and I have to move stuff when companies and product offerings rise and fall.
This sounds great! I am lost. I consider myself relatively tech-savy, so if I’m lost, I can assume most other people who aren’t specialists in online hosting and probably just want to write things and share them with other people who write things will also be lost. I’ll make it a point to understand what Eas is saying through online research, but my point is, there is a point where access to the open web breaks down for “ordinary users,” and this resistance is what feeds the establishment of closed systems.
Luckily, as a general principle of life, I always dig into the resistance as much as possible.
Finally, Bob Mottram writes:
I’m a firm believer in the open web, and I think in the not too distant future it could enjoy a new expansionary phase. I run a project called Freedombone, which was inspired by an earlier project called Freedombox. These and similar things are intended to help people take back ownership and control of their data and online presence in a more convenient manner (sometimes also known as “userops” because it enables users to do what previously only systems administrators could). As the hardware and software combinations are further developed it will be easier to run your own blog, wiki or social network node and so you’ll be able to decide what happens to your old photos, whether you want to license your content in particular ways or what happens to your data if you’re no longer around.
So a possible solution to the multitude of privacy and data ownership dilemmas is to “be the web”. My project has the concept of the “web of backups” in which friends can help to ensure they never lose data via mutual automated encrypted backups. The more you get into hosting your own services the easier it is to see how little value the big companies actually provide and how expensive their services are in terms of privacy.
Userops? This sounds like something that should exist, albeit something that those who favor from closed online systems won’t like very much. Count me in.
I want to be the web. Bob, I’ll be in touch.
These comments have been inspirational. If the open web is a spectrum, I’d like to push as many of the people I care about as possible over to the ‘open’ side when it comes to their online behavior and where they’re depositing their mental nuggets over time. I’m working on a few projects right now to do just that, so you can expect more posts like this from me, here and elsewhere on the open web :)