A self destructive web

I’m finding myself more and more nowadays interested in content that self destructs, or rather: expires after time and will no longer exist.

This post is more me dropping my thoughts on a page rather than having fully explored the idea (yet). Please note this post is also unedited (which I may use to my advantage and edit over time…we’ll see).

Saving is “forever”?

From an early age I’ve played and worked with computers. I always understood that I was creating programs that would be saved.

That simple idea of saving, be it to a physical tape (I was a Spectrum user) or fast forward to the 90s when I published my first web pages – it was simply naughts and ones and since it was stored digitally†, it could be copied over and over. When it was on a server, all it needed is the server to stay plugged in or even cached on another server and boom, you’ve just unlocked forever.

† Sure, the tape drive wasn’t digital, but you’ll forgive me.

But the simple fact is: we have to make an effort to preserve. Server drives crash, data gets lost, companies disappear (my first site was on lineone.net if anyone remember that – now long gone), domains expire and people die.

So what is forever? Maybe there’s an unspoken expiry? But when the expiry is in decades I feel like I conflate this with forever. I don’t feel like the web community has been going long enough to see that long expiry showing it’s face. Maybe we’ve got another 20-30 years to go.

Should it last forever?

I was just looking at Stef Lewandowski’s tweet about their new hack Linkydin. A nice idea, but I was asking myself: surely these links have context in time.

If you’re sharing a group of links in a team, should that information expire? What good is it a year (or more) later to new team members, and in fact could it share old and busted knowledge that’s been superceeded by better information? I don’t know.

Then I read Paul Neave’s post “Why I create for the web” and he boils it down to the humble hyperlink. Except the way I found his post was through a tweet. A tweet that a) I will unlikely find again (because twitter’s archive search is limited) and b) goes through twitter’s own link shortening service. If that service is shut down, then that hyperlink is sent to the grave (note that this isn’t the point of @neave’s post – but it got me thinking).

I recently looked at Snapchat. Content that purposely destroys itself, I can only assume, because it’s relevant for that moment †. I also remember someone having a blog that would slowly remove all the blog posts one at a time over a period of time (and it would look like the articles were fading out at the bottom of the site).

† Though Snapchat might not destroy the data on the phone it is actually removed from their server, which, to me, means it’ll expire from existence somehow eventually.

I like these later two because they’ve thought specifically about time and how it should be addressed. My own blog, I haven’t thought about time. I haven’t thought about whether it’ll shutdown after I die. I’ve long stopped posting to jQuery for Designers but I’ll keep the domain renewed…but for how long?

Should sites be retired to the archive.org?

I don’t know. A lot of these questions are why I’ve asked Jeremy Keith to speak at Full Frontal whereby he chose the title “Time”. I expect he’ll add a lot of clarity to some of my thoughts, but no doubt raise a good deal of new questions.

Everything ends

Everything has an expiry. Nothing lasts forever, but maybe my expectations need to adjust on the web. If something lasts a few years on the web that’s a good thing. If I can get my data out of a web service, then that’s a good thing too (because I’m now responsible for my content, like photos which I care about).

I’m just very interested in seeing more services taking advantage of expiry as a feature. Heck, we buy and own (small) pets because the expiry is a feature! Why don’t we use this more on the web?

7 Responses to “A self destructive web”

  1. Have you seen http://phmr.al/ by Jonathan Snook?

  2. I have had the same kind of ideas more recently. I used to think that “permalinks” were a thing and that whatever was there, will always be there. Now, I also tend to think that this is probably one approach, but not the only one. Some content deserves to be here forever, some probably not.

    One quote I really like is this one: “We die twice, once in the medical sense, and once when the last person that remembers us dies”.

    I think we can extend that to the web, where something needs to be deleted only once no-one else ever links to it.

  3. I love the concept of stuff being temporary, reminds me a little of this Ludum Dare write-up: http://dominikjohann.tumblr.com/post/59611971605 a game which would only be active once (what actually happens to it is a fun lesson in itself).

    As for the discussion on permalinks, I’m not sure where I stand. I remember being told that “cool URIs don’t change” (http://www.w3.org/Provider/Style/URI.html). The idea that when you publish something, there’s an inherent contract to keep it updated is nice – but it’s pretty scary. “Facts” have a habit of changing, and keeping them up to date across many sites can be a huge task.

    I guess the jQuery for Designers example is a good one, because as the jQuery API changes, the content becomes less and less relevant and at some point might even become counterproductive for users searching for more relevant stuff. So do you try to keep it updated? That’s quite the undertaking. Or do you delete it? Rendering lots of links obsolete (I guess that same contract you sign when you create content is also there when you link to something, trickling down a similar dilemma).

    Putting a hardline on it that this is ‘best practice’, will only isolate people and maybe put them off creating new content or ideas, for fear of pissing everyone else off. So maybe I am sure where I stand on it.

  4. There’s a slight problem here in that the web ‘rewards’ you for adding content. A blog, for example, benefits from a larger web footprint, so allowing old posts to expire would limit its outreach.

    I embrace the idea of expiry, since outdated content can sometimes be counter-productive to retain, but imagine this really calls for a paradigm shift in the way we look at ‘building’ the web. And archives wouldn’t be a bad thing, since old content could be considered history.

  5. @Julien
    The quote reminds me of garbage collection :-) Counting references to content and removing once it hits zero. Maybe the metaphor of garbage collection will indeed be useful as we realize that ‘forgetting’ in huge systems is an essential feature. I suspect we can learn (once again) from looking at how our brains do it.

    Facts actually do not change. They remain always true in their context of time. Once the context changes, new facts are generated from old ones.

    For me the question becomes :”Is expirey something we need to manage actively, or do we let it happen naturally and throw away our illusions/expectations about ‘forever’? “.
    We do not manage death, do we. Also we do not (actively) choose to forget things.

    This suggests to me that ‘ self destruct’ needs to be implemented at the core of our technology stack (IP/TCP for example), because that would be the closest we have to a ‘context’ for digital information.

    As if that wouldn’t be challenging enough, how do we implement ‘forgetting’ in a distributed system? Because if we had a something like an Internet garbage collector, it would have to be a centralized piece, right? Who hosts it then? Google? Would you be comfortable with that?

    This is a nice topic, it gives me goosebumps to think about it although (or maybe because) there are more questions than answers.

    Thank you for bringing it up, Remy.

  6. “I’m just very interested in seeing more services taking advantage of expiry as a feature.”

    There’s a long history of websites taking advantage of this “feature”: Geocities, Friendster, Posterous, Pownce, Vox, Dopplr, etc., etc., etc.

    In fact, just about every web service—over a long enough timeframe—already takes advantage of this “feature”. The problem is that they don’t advertise this feature when people are signing up to their service. They don’t warn people that the thoughts, hopes, dreams, and memories that you’re about to pour into their service will, before too long, disappear.

    Snapchat isn’t interesting because it destroys content after a certain time—every service does that. Snapchat is interesting because they’re honest about it.

  7. @Jeremy – Of course you’re right, it’s not a feature at all, it’s an inevitability of all sites (and indeed all things).

    In that respect, what I mean to say is that I’m interested in services that say up front, for X specific amount of time, your content will exist, but then, it will be gone, forever. I don’t know *why* I’m interested, and I know I have a lot of questions rather than answers.

    In a way it’s similar to a content publishing system telling me I can only use 140 characters to post with. I’m interested in a constraint that could be seen as destructive, but could in turn make me think differently about what I’m doing (obviously my tweets aren’t good examples, but there’s things like the story of the first man on the moon being told over twitter that’s quite elegant over 140 characters).

    Again, unedited – therefore not 100% thought out, but then again, it’s cool – this is my house ;-)

Leave a Reply
Not required

CODE: Please escape code and wrap in <pre><code>, doing so will automatically syntax highlight