10 ms·
Most websites are implicitly designed with a short lifetime
- kqr 5y agoIt always strikes me as hilarious when printed books reference web URLs. Granted, URLs are meant to be permanent (the cool ones don't change, as they say) and any webmaster worth their salt will strive really hard for this. But c'mon. You think a website will outlive the useful lifespan of a printed book? Hah! You should see the age of some of the books I've found useful in the past few years.
- dgellow 5y agoWhat else do you want them to reference?
- notRobot 5y agoWebCite [1] was built for this in the past, but using archive.org is a better idea nowadays. Save a copy of the webpage to the archive if it doesn't already exist and link to it. Has the added advantage that anyone who goes to the link sees exactly what you saw when you linked to it, with no modifications. [1]: https://www.webcitation.org/ https://www.webcitation.org/
- cxr 5y ago"Perma.cc helps scholars and courts create permalinks of the web sites they cite" <https://news.ycombinator.com/item?id=10948033 https://news.ycombinator.com/item?id=10948033>
- bryanrasmussen 5y agoI think a service like webcitation is better because 1. Archive.org follows robots.txt rules - also used to remove content on request from owners (not sure if still do but I suppose they do) making it less sure than webcitation. 2. webcitation could also provide other services like - export citations for book or collections of books into one book of citations. 3. Similar citations tracking stuff that you can get on google scholare, how many other publications were citing this text or part of this website. on edit: obviously this is if webcitation or a similar service was still a going concern, pragmatically yes archive.org is better because functioning and taking new content to archive.
- notRobot 5y agoThe only reason I recommneded archive.org is that webcite is no longer taking submissions, and their HTTPS cert has expired. Other than that, I agree with you.
- KeepFlying 5y agoThe Internet Archive at least since it'll keep a URL alive and consistent a lot longer. At least plain URLs can be manually passed to the internet archive for a lookup, but I'd love if publishers/authors made a habit of maintaining their source links (via shortlinks, published errata, etc). That said, even with how easy it is to update, Wikipedia suffers from this immensely. I can't count how many times I've tried to track back a source from a wiki and end up with a 404.
- dusted 5y agoThat rests on the assumption that IA will outlast everything else, which is unresasonable, if nothing else because, while it may be supported by the community, when Kahle becomes unavailable to the project for one reason or the other, there's no saying what will really happen to it.
- chongli 5y agoThe IA has well over 100 employees, possibly over 200 at this point. With a very strong mission statement, I have to imagine they have a very good culture dedicated to Brewster’s goals. I’m sure he’ll choose a very good successor. I’m sure he’s already taken considerable steps to codify his mission into a charter. He’s a very smart guy and he’s worked too hard to allow his vision to falter when he retires. Furthermore, archivist is a full blown profession with its own professional society. I’m sure IA will always be able to hire people who are serious about preserving and cataloguing information.
- Mezzie 5y agoI'm curious who it's going to be and what field his successor will come from. As a coder/developer with an MLIS + a decade of library experience, in my experience there aren't that many people who grok both the tech and the politics/people aspect around preservation, which you would need to head IA. I wouldn't be shocked if, if IA is around in a few decades, they start expanding into educational partnerships to ensure the vision continues. Or institute fellowships. Or they may just inspire more people; IA's work was one of the things that convinced me to pursue an MLIS (as a girl, I was discouraged at the time from CS). Or we'll start seeing more degree specializations that mesh the two, like HCI concentrations.
- VelNZ 5y agoPersonally I would save a copy in a standard format and make it available on a domain I control. I've done that in the past for scientific publications.
- dgellow 5y agoThat's only possible if you own the material, otherwise you may be distributing copyrighted content.
- kqr 5y agoHonestly? Go through the trouble and ask the author for a license to print the relevant parts along with the referencing material. The most egregious cases is when the author references their own web content!
- notapenny 5y agoI chuckle at something similar where books on programming languages are titled "The Definitive Guide to <Insert Programming Language>". Our industry changes, and fast. Everything is chaos.
- bryanrasmussen 5y agoI think the greatest preponderance of technical books are written with a useful lifetime of at most 5 years, and frankly probably not that long.
- bzzzt 5y agoI divide technical books in two categories: - Those explaining a programming language, framework or application with a specific version. These books clearly have a lifespan related to the thing they are explaining and will often end up in second hand or wholesale shops. - Books explaining principles. It doesn't matter if they are 2 or 20 years old since the things they are explaining don't really change.
- bryanrasmussen 5y agoSure, but the greatest preponderance of programming books explain the first, also some books that seem to be explaining the second are really more tightly tied to a particular programming language or platform than they like to let on - like Clean Code with Java.
- II2II 5y agoWhich is almost entirely unique to programming books. Take a book on electronics, and a book discussing principles and a 20 year old book will be nearly as relevant as one published today. Take a book on physics and a 100 year old book will be be nearly as relevant as one published today. That doesn't mean the fields are unchanging. Modern electronics is definitely more sophisticated and much of classical physics has been superseded by modern physics. One definitely needs more modern treatments in order to work at the frontiers of these fields. Yet the principles are less ephemeral. On the other hand, software development is much more fad driven. The most basic principles may still apply, yet the application of those principles change (sometimes subtly and sometimes dramatically) due to a tendency to throw the baby out with the bathwater.
- bryanrasmussen 5y ago
- aspyct 5y agoI have a printed book on AspectJ. Pretty sure it's past its "useful lifespan". As technologies age, so do anything that talks about them.
- anchpop 5y agoThis is a use case of IPFS I think. Any IPFS "website" will remain available at its original IPFS content ID as long as at least one person is seeding it, so if a book referenced one the author could seed it to guarantee that the URL will continue to work as long as they're willing to keep it up (and any interested 3rd party could do the same)
- kqr 5y agoYeah, but a printed book well outlives an author's good intentions, too. People underestimate the lifespan of books.
- stickfigure 5y agoWhat seems a lifetime ago I wrote the backend for the website of Sprite's Sublymonal ad campaign. It mashed up a bunch of wikipedia data and fed it to a slick flash frontend. The whole website was expected to last two weeks. Funny story: I was expected to use Coca-Cola's Websphere infrastructure, maintained by IBM at some IBM datacenter. It was going to take them months to set up, and I had a week before the campaign launched. I ended up running the whole thing on a couple (rackspace, IIRC) VPSes and never told the client.
- Brajeshwar 5y agoI still maintain my personal website which was once pretty popular. It remained neglected for a very long time, and now I just keep it for the archives and for my personal rambling. Some of the popular links are the likes of "/archives/000129.php", "/archives/2004/10/the_elements_of/", etc. I have tried my best to take care of the top 20-odd links and redirect them to the new/better URL, which is likely to survive my naming method of "/YYYY/title-of-the-link/". Many of these has survived DCMAs, still enjoys inbounds from the Russians, and even the Chinese. One of the many reasons for me to get rid of the Analytics is that I feel bad and always wanted to solve the errors. Now, I just don't have a clue besides the basic analytics that comes with Cloudflare, which I see once in few months.
- russianchinese 5y ago
- MisterSandman 5y agoYou can't say all of that and not link your personal website :P
- Brajeshwar 5y agoPsssstttt, DDG or Google or Search my name "Brajeshwar". ;-) Fun fact, some people in the political higher ups in India wants to use that domains and thinks they have rights to it. I have 20+ year of data/info to prove them wrong.
- MisterSandman 5y agoI am so jealous that you were able to get a .com name. I'm Indian as well, but I'm younger and my name is common so by the time I was looking into a .com, it was being sold for thousands of dollars ://
- Brajeshwar 5y agoAnother family fun-fact -- I also own the .com of our family name "oinam.com". Oinam[1] is a community and the name of a place in our hometown. Well, there are occasional events where people will ask me to buy it, email for questions about stuffs of that place, etc. 1. https://en.wikipedia.org/wiki/Oinam https://en.wikipedia.org/wiki/Oinam
- ChrisArchitect 5y agoYes, and? We're talking about information technology here right? It's been constantly in flux for 30 years when it comes to the web and keep going as far as PCs and data storage systems before that. Not saying I wouldn't like to see more long-lasting designs and content archiving is a huge deal/challenge/lost histories real problem, but hard to expect 10-20 year design choices and infrastructure to hold up or be planned for.
- Bancakes 5y agoMost websites have a few kilobytes of zipped useful text, and a few useful jpegs. By all means, most social media notifications could fit in a single ICMP packet each. Hacker news could be checked over ARPANET. Wrt. classical probability, useful data divided by total data/metadata, the ratio has steadily been decreasing. Our infrastructure is changing for the worse and 'probably' about to fail.
- sekao 5y agoThe reason websites aren't designed to last is not due to ever-changing design trends; it's because the web is inherently ephemeral. For a website to continue to be available requires continuous effort and expense to run a web server, update SSL certs, etc. This automatically shortens the time horizon when you're developing for it. I think the way to get more "timeless" software is to build a platform where the code and the data can be brought offline. PWAs do the first but not the second. The data needs to be in an enduring format like SQLite that can be queried offline. With that kind of system, an app or game (and the accompanying data) could remain usable 20 years from now, long after its author stopped distributing it.
- jstummbillig 5y ago> The reason websites aren't designed to last is not due to ever-changing design trends; it's because the web is inherently ephemeral. For a website to continue to be available requires continuous effort and expense to run a web server. This automatically shortens the time horizon when you're developing for it. Interesting thought, but I don't think that's quite it. Everything is ephemeral to some degree. Everything will require a combination of resources to keep something available continuously. For example keeping books around is not exactly free. Yeah, we usually "just have it in a bookshelf at home", but if you think about what that actually entails you'll probably start wondering if that book actually appreciates all you've done for it. When running a home server becomes as simple as building an IKEA shelf, and when making a website is a native as writing on pen and paper, we will have a lot less bookshelves and more websites just being around and with that the expectation of them being around and zero understanding of how or why. No consideration being paid to routing, backup, space, bandwidth, because there is nothing to consider. Something that is magically and reliably available for money, like electricity. If you move you just bring the box and it works again as soon as you plug it in, like any desk lamp would. Technology is still way too hard and inaccessible for that to be a thing.
- enobrev 5y agoYou could potentially keep a website "on a bookshelf at home" as well, provided the database is small enough. It doesn't even need to be SQLite. No matter the data format, you could get a fully running website on a raspberry pi, maybe with a subset of the data. I'm picturing a bookshelf of raspberry pis in custom "boxes" like a collection of 8-tracks. Pull one out, plug it it, turn it on, and connect to it. Considering the majority of my 20 year web-building career is long gone, this idea kind of intrigues me.
- approxim8ion 5y agoThe thesis is fine, whatever, but do they actually prescribe a solution here? What can someone do to "design for a long lifetime"? What does a website that is "designed for a long lifetime" look like? This all very hypothetical. My personal site is dead simple, about 20kb uncompressed (50-odd pages), and uses semantic HTML, minimal CSS (~2-3kb) and no JS. But that's for simplicity and to keep my costs down. There's no saying whether even this basic structure will be supported 20 years down the line because it works on browsers now.
- WA 5y agoWebsites are shopping windows. They change all the time and that is okay. But somehow, people have the expectation that the entire web is more like a permanent archive of written content. That is simply not true.
- superkuh 5y agoThis analogy works for commercial websites. But that's hardly comprehensive. There are two mutually exclusive views of the web. As a set of protocols set to allow individual humans to share information about things they love and the web as a set of protocols to make a living. Profit motivated web presences want views, they want attention, they need nine 9s uptime, need to be able to do monetary transactions absolutely securely, and they want to be an application not a document. They live and die on the eternal wave of walled garden's recommendation engines because that's the network effect and that's where money flows. They're ephemeral and only last for a couple years. It doesn't matter if this means extremely high barriers to entry because money solves everything. Individuals' websites are freeform presentations about the things that person is interested in. They are the backyard gardens of the mind and the most important thing is lowering the friction from thought to posting. There's no need to get tons of traffic instantly (or ever), they're mostly time insensitive and often last decades.
- Semaphor 5y agoHeh, the website I work on went online in 1996, there was a design update in 2006 (when I joined, still as a student), and we are planning to make it mobile-friendly this year. How’s that for longevity? ;)
- onion2k 5y agoI don't think people building websites explicitly think about the expected lifetime and plan for whatever specific lifetime they expect. On the contrary, people do that a lot. Every abstraction, every 'separation of concerns', every micro service or micro-frontend, is a response to the belief that at any moment a developer might need to replace some part of the whole. Pushing content into markdown files that can be loaded with any build process that understands markdown is an attempt to be build something where the website, eg the content, can outlast the current code. Every template-driven design populated by values from an API that's loading a dictionary from a database is decoupling the pages from the code, which greatly extends the life of the website. The website is not the code. That's just the part of it that you see. The website is the ideas, creativity, content, effort and understanding that went into defining what the code does. Some parts of the code might change relatively quickly, but that's the nature of code. We find better ways of doing things. Browsers change. We choose to optimize for different things. That's a good thing, even if it means throwing some old code away. It's a bit like the Ship of Thebes. Everything might have been replaced a few times, but the website is still the same website.
- franze 5y agoI design websites with the mindset of changing requirements over time. So the website must be created decoupled and scaleable. Start with the URLs. URL-rules URL-Rule 1: unique (1 URL == 1 resource, 1 resource == 1 URL) URL-Rule 2: permanent (they do not change, no dependencies to anything) URL-Rule 3: manageable (equals measurable, 1 logic per site section, no complicated exceptions, no exceptions) URL-Rule 4: easily scalable logic URL-Rule 5: short URL-Rule 6: with a variation (partial) of the targeted phrase URL-Rule 1 is more important than 1 to 6 combined, URL-Rule 2 is more important than 2 to 6 combines, … URL-Rule 5 and 6 are a trade-of. 6 is the least important. The most important part in this context is rule nr 2. permanente, no dependencies with anything. no dependencies with the navigation, not with categorisation, not with the content the URL hosts, if not 100% within the organisations control. So basically you end up with something like https://www.example.com/%short-one-letter-namespace%/%unique-slug% https://www.example.com/%short-one-letter-namespace%/%unique... - or - if you want to keep future market and language expansion in mind https://www.example.com/%market-language-namespace%/%short-namespace%/%unique-slug% https://www.example.com/%market-language-namespace%/%short-n... The URLs must be also decoupled from the CMS or any technology choice underneath. The next things is "don't do redirects". Redirects are only done when there is systematic change of the website (relaunch), but not a daily business tool. Interlinking only to canonical URLs. No parameters. Each (Web)Application (that mostly interacts with user input) gets its own subdomain. i.e.: checkout.example.com, map.example.com Throw away things that you do not need any longer. Cleaning up is an ongoing process of a website, similar to creating things. Responsibility, 1 person is responsible for the whole website. 1 person is responsible for each section of the website. If a section of the website has nobody who is responsible any more, somebody must be found or the section will be deleted. Just a few of the rules I put in place. It's a mixture of decoupling and responsibilities. Change is inevitable - and manageable.
- titoasty 5y agoAs a creative developer, my job is basically to make short life websites. Her is the way most of the websites I do are considered: when you make a communication campaign, the website is one of the medias used, like tv ad, print ad, ... So the websites only last for the duration of the campaign. Consider them as ads like any other print/tv/radio/.. ads Let me tell you that all the considerations on accessibility have never been even discussed or even a few times laughed at.
- hnfong 5y agoA website is just a medium to publish and present information. It generally has as long a lifetime as the underlying project. The short-livedness of websites is a symptom of short-lived projects and ventures in general. To me, this phenomenon is a parable for perseverance rather than a technological cautionary tale.
- cainxinth 5y agoMy Wordpress hobby blog is 12 years old and is still using the same theme as day one. A few years back my favicon suddenly stopped working. I called Wordpress and they said my theme was deprecated and to get the favicon back, I’d have to choose a new theme… but once I did, I could never revert back to the old one. My blog has thousands of short form posts and I very much like how they are presented. I just left it the way it was. I sincerely hope they never make me try to switch, but realistically I have to assume they eventually will.
- GreenWatermelon 5y agoQuestion: if it's a 12 years old blog, then I assume it's simple. Why don't you rebuild it with the theme you choose? Granted, I don't know how WordPress works, or any if its technical circumstances.
- cainxinth 5y agoI have 6000 little posts, each with a 300x300 image... and it's set up the way I like it. I just know that if I tried to import it into a new theme, something would be off that would require fixing manually, post by post, which I definitely wouldn't do.
- schrijver 5y agoTheoretically, the way to go about it would be to save several generated static HTML pages from your blog and work backwards from there, adding the necessary Wordpress PHP. Granted it would be intense work, but then you would no longer have the impending cancellation of your theme hanging over your head.
- schrijver 5y agoSo your Wordpress host doesn’t allow custom themes? Because theoretically, you should be able to make a ‘modern’ theme that creates the same HTML + CSS as your beloved vintage theme. I sympathise with wanting to keep the presentation. It was made in and for this context. It’s like reading an old magazine, you get so many extra clues about the context from the design. If a magazine does a redesign, it only affects the issues going forward. In my ideal world, blog redesigns work the same way. I’ve done my share of maintaining and archiving older websites, where I tried to keep the design in place while updating the underlying infrastructure (or converting it to static HTML). The one thing I would sometimes change is to ‘zoom in’, multiplying all pixels in the CSS by a small amount, because the web moved to larger text sizes and screens, so older websites tend to look tiny.
- irrational 5y agoI work on a website that has been around in some form or other for nearly 22 years. In that time it has undergone about 7 major revisions. The last revision was about 5 years ago. But we are still finding bugs and the site requires two of us to work on maintaining it. The site is interesting in that it is far from being our core business site, yet is useful enough that the company will probably keep it around until after I’m retired. There is always work that needs to be done and we have a large backlog list of things that we need to do to keep it running. Our biggest issue right now is the site is running on a cluster of 2008 Windows servers and we want to move it onto AWS EC2 Linux servers. We’ve been trying to get to this work for years, but there is always something else that needs doing that has higher priority.
- cmusfan 5y agoDisposable websites are good for job security. There are other factors to consider, once a site turns into a desktop app replacement with dynamic content it is expected that it will change and grow, features dropped and added, styles change, the tech stack is replaced. Building a single use product to last doesn't make a lot of sense, the only part that does is the idea of maintaining "permalinks" where the content is stable and "append only", but the style may change. The world has not placed enough requirements to preserve a historical friendly web, which is why archive sites, print to pdf, and other options exist. It is expensive to plan, and maintain sites with longevity. You don't know what the future holds and longevity might be at odds with relevance.
- aethertron 5y agoPerhaps neglected web content should die. Improvement via a kind of Darwinian mechanism would depend on that. Web content that people that people care about could last forever if we adopted the Federated Wiki patterns, i.e. don't just link, but copy content from other sites and put it on your own (everything there is, in principle, licensed CC share-alike).
- rietta 5y agoShould books be destroyed after the life of the author as well?
- aethertron 5y agoNo. In some cases they should be destroyed during the life of the author (e.g. a book filled with CSAM photos).
- emodendroket 5y agoLet's be real here. Tons of books are pulped all the time. Pruning their collections is a major activity that occupies libraries.
- tomhunters 5y agoBut there are some special exceptions, except Facebook and other tech giants.
- zokier 5y agoIt is unfortunate that mirroring is not first class concept on web. There are some concessions like relative urls, but largely mirroring a modern web site is a non-trivial operation. If mirroring was easier and more prevalent, then the longevity of individual websites would be less critical.
- fleddr 5y agoYou can generalize this conclusion into almost anything being designed with a short lifetime. Our society is fast and getting ever faster. It's unstable and exponentially changing. In these condition, nobody can afford to do anything foundational or timeless. We're in constant survival mode, just trying to make it to the next financial quarter. Why worry about 20 years from now when we're not even sure our entity survives the coming 3? Why invest in something "proper" when that proper thing may be entirely obsolete or lose relevance? Why develop a product that lasts if you can win more when you make one that doesn't?