Hacker News Reader

Comment by kqr | Hacker News Reader

It 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.

View on HN

I think the greatest preponderance of technical books are written with a useful lifetime of at most 5 years, and frankly probably not that long.

I 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.

Sure, 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.

Which 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 Modern electronics is definitely more sophisticated and much of classical physics has been superseded by modern physics.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.

I'm used to that from math, but I wonder if math books have many urls as citations. I guess I haven't read any modern ones but I wouldn't expect it, so do modern electronics books have urls as citations?

From what I have seen, which is admittedly limited, textbooks tend to push additional services from the publisher but very little is lost if you don't use those services. Then again, the books I have seen are older (even though I may be reading a recent edition).

Anything directed towards more casual learners, I would have a hard time commenting on since it tends to be more the history of science rather than being technical in nature (where references tend towards primary sources anyway), or embedded development (which seems to suffer from a similar mentality as software and web development since it has a software bias).

The person you're replying to didn't mention anything about the useful lifetime of technical books specifically. URLs appear in e.g. footnotes and bibliographies of plenty of non-technical works.

you're right, for some reason I took it as being in reference to technical books.

I have a printed book on AspectJ. Pretty sure it's past its "useful lifespan".

As technologies age, so do anything that talks about them.

This 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)

Yeah, but a printed book well outlives an author's good intentions, too. People underestimate the lifespan of books.

What else do you want them to reference?

WebCite [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/

"Perma.cc helps scholars and courts create permalinks of the web sites they cite" <https://news.ycombinator.com/item?id=10948033>

I 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.

The 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.

The 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.

That 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.

The 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.

I'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.

> I can't count how many times I've tried to track back a source from a wiki and end up with a 404.

Yes, that's really frustrating. I often have the situation where I click on a reference for a claimed fact and get redirected to the front page of a news website.

Typical citation formats have a date referenced so you're free to try that yourself.

Personally 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.

That's only possible if you own the material, otherwise you may be distributing copyrighted content.

Honestly? 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!

I 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.

Reply on news.ycombinator.com