Monday 29 August 2016

It's time to say goodbye

On a recent Slate Culture Gabfest, the hosts bemoaned cutesy 404 pages - the responses you get when your device is able to make contact with a server, but that server is not able to deliver the page you've requested. Perhaps the most famous 404 page was Twitter's "fail whale" (discontinued in 2013), and for a period there web people got very excited about making playful, pun-filled and beautiful 404 pages. It was seen as part and parcel of caring deeply for the user's experience ("We can't give you what you want but we'll make it beautiful!"). Pages and pages of results can be generated for a search for a phrase like "best 404 pages".

So it was a little disconcerting to hear people on this podcast who I really like seeing all this effort as going to waste. Perhaps no-one really wants to be delighted when they're told they can't have what they're asking for: maybe they just want things to be fixed.

To my mind, the artisanal 404 page is one of the great artefacts of Web 2.0, and I really hope academic study, plushy print publications, and considerable archival effort are being devoted to its documentation and retention for future generations to enjoy and learn from.*

Right up there with the 404 page is the shutting-down announcement. Nothing reminds you that the web is an ephemeral space with a metabolism akin to a hummingbird's like trying to remember all the web services you've used at some point and then had to transition on from when they close (here's me three years ago being overly upset about Google Reader shutting down: reader, life has gone on).**

I've been mulling this since clicking through a couple of links to the news of - a location-based micro-blogging-ish site I'd never heard of before - setting to close at the end of this month.

I read about this in a post by Russell Davies, where he talks about the pleasure of writing on corners of the internet that are public, but not highly frequented. Starting on Gowalla Davies had written little fictional histories about places he was spending time; when Gowalla closed down he moved on to, which is now, too, shutting down.

But they're shutting down in the most elegant way. As site founder Craig Mod writes in a Medium post, they are taking all due care of their users that they can, but also trying to record the existence of for posterity. As he writes:

Web projects often lack hard edges. They begin with clarity but end without. We want to close with clarity. To properly bookend the website. Sometimes web projects exhaust themselves. Outlive themselves. Are allowed to stagnate, be forgotten. Resources dry up and then one day — poof — they’re gone. This has happened countless times, Geocities being one of the foremost examples. We don’t want this to happen with

So, is enabling users to download their archives, and to promising to keep an online archive alive for at least a decade. And they're also producing a small number of physical storage artefacts, inscribed with the content of the site, that will be deposited with institutions like the Library of Congress:
We’ve partnered with Norsam Technologies and Los Alamos Laboratories to utilize a special ion-etching process, capable of printing tens of thousands of pages onto a 2" × 2" plate. 
The process does not produce “data.” It is not like a CD. It is not a composition of 0's and 1's representing the information. It is the information itself. The nickel plate is a medium, not media. And everything printed on the plate will be readable with an optical microscope. 
The nickel plates have an estimated life span of 10,000 years. They’re fire resistant. They deal well with salt water. And because they’re printed with our pictures and words — assuming contemporary language is decipherable in the future — anyone who finds this and has access to fairly elementary technology (an optical microscope) will be able to read our thoughts and experiences as mapped to city and place.
It's such an interesting question. As with video games (see MOMA on acquiring their first batch) so many questions exist. What are you collecting when you try to memorialise a website or service? The code? The visual appearance? The user interaction? The history of the making and marketing of the site? Or the described experiences of those people who made and used it?

Businesses shut down all the time, and have for as long as they've been around. Traditionally they've been captured for the record - often incidentally, lurking in the background of the main focus) in photographs, newspapers, gazettes, phone books, correspondence, government records. Sometimes - when they're old enough, big enough, flash enough or loved enough - their closing is a cause for reminiscence and celebration. But often they just slip out of sight.

But websites and services seem to take their closures with the same seriousness that they take their launch. (Three thoughts on this: (1) people who work in web design are pretty much the most self-analytical and self-descriptive that I know - what other industry devotes so much time to examining and publishing in real time upon itself? and (2) People who are shutting down things on the internet are doing so in a medium beautifully designed for people to tell them exactly what they think of their decision; and (3) like 404 pages, the attraction of the graceful failure is as potent in this part of web operations as in any other.) Here for example is John Gruber announcing the shutting down of Vesper, a notes app, last week. Although the app is shutting down due to low usage (and, from my deduction, accompanying low interest) and a notes app is hardly a crucial life tool, Gruber's post is a detailed and thoughtful retrospective on what they could have done better, and why they made the decisions they did. No simple CLOSED sign here.

Here, for example, is a long and admiring article on the grace with which Glitch, a massive multi-player online game, was shut down. The game was built out of this almost obsessive love for the end user that powered early Web 2.0 properties like first-iteration Flickr (not coincidentally, the lead designer was the co-founder of Flickr and founder of Slack, both of which place/d a tremendous emphasis on surprise and delight of users). From the Glitch post announcing their shut-down, the list of FAQ titles:

  1. Why why why why?
  2. What will happen to the team building Glitch?
  3. What will happen to Tiny Speck?
  4. How long will the game remain open?
  5. Can people still sign up to play?
  6. But I spent money! What about that?
  7. I'm really angry about this!
  8. Why can't you sell the game so someone else can take it over?
  9. Why don't you give the game away or make it open source or let player volunteers run it?
  10. Why can't you just _______________? 

Or for another example, look at how Matt Webb announced the closure of design company Berg through a poem on the company's blog (week 483, RIP). And here's his post on the Little Printer blog that shares that announcement, and explains that Little Printer, Berg's experiment in the Internet of Things and a more human interface with the web, will also be shutting down. And here's his post - redolent of a late night and too many cups of coffee*** - explaining that he's "trying something:
Hi everyone. Hello from where I’m sitting at home in London. It’s Sunday night, I’m Matt, my personal homepage is over here. Till recently I was CEO at Berg, now I’m the last employee and trying to wrap things up nicely. To do that I’ve got about a day a week cos I also now have other commitments. 
I’d like to try something…
I really hope someone out there - an obsessive individual, an academic, or an institution - is collecting these artefacts. By their very nature they're usually written in places that will soon disappear from the internet, and then all we'll have are the GigaOm and FastCompany rehashes (until they too degrade away.) and gravestones in the Internet Archive. But they're such a visceral and telling part of this period of design, communication, and business.  They deserve to be held on to. Maybe there's more nickel-plate books to be made.

*Seriously, there don't seem to be any print publications on the history of the 404 page, and I think that's a real oversight. And of course, all my bookmarks on this topic are long-lost in the endless shuffle of bookmarking sites I've been through over the past decade or so.

**Also, of interest to me but probably not for you, looking at what sites I opened first when I got up in the morning, I've gone from Email > Twitter > Reader > Work email to Feedly > [work] Facebook > Twitter > Email > Instagram .... work email when I get to work.

*** I have no idea how late it was or whether Webb even drinks coffee - I suspect not - but you get what I mean.

1 comment:

andrew morton said...

Small nit pick correction: the Twitter fail whale wasn't a 404 page, it was more likely a 500 or 503 error page indicating the server wasn't able to process the request.

To add another link to the pile there was a great radio story on the shutdown of the sims online: