[Cool URIs don’t change](https://remysharp.com/2016/12/22/cool-uris-dont-change "Permanent Link: Cool URIs don’t change")

First of all, I made a mistake early on, years ago, I misconstruing "cool URIs don’t change" as "will always be available on the web".

For starters, "don’t change" means that really the original resource indicator works and returned a status code (approximately) less than 400. i.e. It loads, or it redirects (or certainly that’s good enough for me).

What it doesn’t mean is that once you publish something at the end of a URL, that it must always be there. This is where I went wrong. Because it takes a heck of a lot to keep a single domain running, let alone all your projects.

In the year 2016, I have a fairly reliable process for getting a new web site up and running. The costs are relatively low, financially, but the reliance on 3rd parties is high (which is the trade off).

Here’s a breakdown of what’s required to get a new site live (for me):

This of course assumes that the project is just a little stand alone side project. It doesn’t require any special storage needs and doesn’t need to deal with heavy traffic (and need to scale).

Then multiply this by every other little mildly viable idea you have. What if one of them takes off. In my case it was [JS Bin](https://jsbin.com) (running costs per month are around $300-400 per month - and I’m fairly good at being cost efficient).

With this all in mind, is it reasonable (for me) to be disappointed that [alias.sh](http://alias.sh) had shutdown? That their cool URL hadn’t changed, but it had simply vanished.

Or, are websites (and their domains) perhaps actually a living system, something that I didn’t give them credit for?

I’ve not thought of websites like this before, but what if they’re more like living systems?

The web is still relatively young, and indeed so are its makers, so I don’t think there’s a particularly long life expectancy for a web site. And certainly not a predictable one yet.

There’s some that last only a few months of the campaign sites and micro-sites ilk. There are anomalies ranging from [archive.org](http://archive.org) and bloggers running into their late teens of a website.

Just like a living system, they eventually die, and that’s the natural order of things. In spite of being a digital system. Something that from a technical perspective is immortal. Quite simply, our sites are us, and like us, they come to an end.

These endings don’t obviously have to be "in real life" death (though there is some part of me that plans on how to sustain this domain after I’m gone…), but circumstances change too.

Maybe I won’t be able to afford to run JS Bin one day. Maybe it’ll be bought and it owners let it die. Maybe the maintenance and customer support becomes too much of a personal burden. Maybe, I just lose interest. I don’t foresee any of this happening, but it’s possible, right?

In the case of [alias.sh](http://alias.sh), the site had run its course. Now, it’s no more. It served those people who visited, but the individual (or individuals) behind it have retired the site, and only it’s [shadow remains](https://web.archive.org/web/20141216192521/http://alias.sh/) in the Internet Archive. Until, inevitably, the archive vanishes one day too.

Though, it’s entirely possible that the domain gets recycled into a new system (of course then the cool URIs will change…and break), but this echoes our non-digital real lives. New things grow in place of the old.

It’s also worth remembering that even though a domain is purchased, it’s really only rented from the registrar.

I’ll always remember when the CD came to the consumer market hailing the end of tape (similarly DVDs ending VHS). Tapes had a distinct lifetime (though significantly longer than records), but CDs we immortal. WORMs: Write once, read many. The internet to me, once available online, was the same: digital, and thus, immortal.

But now, I think, perhaps this is the programmer and perfectionist in me. In fact, the web, it’s constituent parts: our websites, are us. Just people and our voices. And with that, we are not immortal, neither are our digital counterparts.

Published 22-Dec 2016 under #web. [Edit this post](https://github.com/remy/remysharp.com/blob/main/public/blog/cool-uris-dont-change.md)

Lock Thread

Login

Add Comment[M ↓   Markdown]()

[Upvotes]()[Newest]()[Oldest]()


dtobias

0 points

6 years ago

That’s one reason I’m in favor of the logically structured use of subdomains, where your various projects can be someproject.yoursite.tld and someotherproject.yoursite.tld without needing a separate domain registration (and associated expenses and need for renewals).


Amr Eldib

0 points

7 years ago

I must disagree with your last point regarding Digital being mortal. Digital is immortal; copyable while maintaining fidelity of original copy. However, if we don’t copy it, of course, it’ll die.\ Notice that we’re not looking for a website, we’re looking for the content of a website, and we do a bad job at distinguishing between the two.\ We have incentives to maintain a strong link between sites and their content (making money), and we don’t have good tools to let content live beyond the interest of its creator.\ I think digital is immortal, but we need to make that work.


Tizio

0 points

7 years ago

I recently pulled the plug on a lot of my old websites, including [no-www.org](http://no-www.org). Supporting two dozen domains that are effectively single-serving, special interest, or legacy content really adds up!


penne12

0 points

6 years ago

Really missing no-www D:


rem

0 points

7 years ago

I believe I link to your [no-www.org](http://no-www.org) domain back in 2006 of this blog!


Danny Hope

0 points

7 years ago

Genes are the nearest thing in life to immortality, maybe there’s a software equivalent. That said, not all genes, only some.

[Commento](https://commento.io)