One compelling reason to not date your articles is that age sometimes destroys perceived value even if it doesn't destroy actual value.
For example, I have a bunch of stuff writing about customer service written in 2006/2007. That field has not evolved so much that "Don't curse at your customers" is bad advice in 2009. However, if one of those were posted here, the first comment would likely be "Dude, mark in the title this is from 2006", because the audience here is socialized to assume that the new and shiny is what they want to be exposed to and the old and crusty is largely not valuable with the exception of a few classics.
I have sites where I have no visible dates for exactly that reason: I intend the content to be evergreen, and if it is evergreen you don't need to know what year it was planted. (Edit to add context: elementary school teaching activities. I am not a textbook publisher so I have no incentive for ripping up my multiplication activity every year and coming out with New Shiny Multiplication 2010 Now With Numbers You Totally Couldn't Have Multiplied Last Year!)
Shouldn't the judge of evergreen-ness be from the reader, and not the author? If it really is timeless, people will come back and link to it again and again. I see the Norvig posts on learning to program in 10 years come up again and again here.
It sounds like if you wrote books instead of blog posts, you'd rip the covers off so readers can't judge it by its cover.
Why rip covers off when I can instead obsessively A/B test them and remove elements that hurt the business value of the IP I spent so much work creating, like dates?
Here's the fundamental difference between me-as-employee and me-as-businessman: I will get paid for my labor of January 19th on February 30th, and will never see another penny of it even if what I do today has lasting value. What I do after I get home, though, I get paid for every year if it has lasting value. I will not cheat myself out of that lasting value by setting an auto-expires header on it.
I've had the same frustration as the OP where the date wasn't included when I was looking for time-sensitive information. It was a disservice to me-as-reader, and I ended up not reading the article and looking elsewhere, since I couldn't be sure of its timeliness.
If that article was timeless, it's because I hear about it again and again from other readers linking to it, not because there was no date on it.
Of course, I don't know what you write, nor have I seen the results of your a/b testing. So shrug.
I disagree. You are effectively saying elementary school teaching activities do not change and are the same now as compared to 20, 30 years ago. I admit there might be some valid cases of "evergreen" articles, but I would argue that for these cases, showing the date will not take away anything from the article.
Readers are smart, trust them to put the date in the proper context rather than hide it from them.
"... What happens when you don’t date your articles? ..."
I read a lot of articles and going through through them is a bit like undertaking an archeology dig for useful bits of information. The best sites include subtle hints that convey context. Useful if you are evaluating, "do I continue reading?"
How?
Authoritative sites are good because they convey more meaning than just an article. They tell you "who" is writing, "when" it was written and most important their viewpoint and intent, the "where". You find this out using subtle hints.
Date is an important clue for understanding context. Writers who don't including perceived affordances such as date, mark themselves as the writer equivalent of a first time ice skater constantly tripping over. A writer newbie, who doesn't grasp the concept of readers who use context to search and evaluate. Some sites where I have found a motherload of stuff:
I read these sites again and again, because the value of the posts are increased by the "perceived affordance" of details such as the date a post was written.
Now I want to display a random date to each viewer of my blog, and record that date when they make a comment. Then I will see if the commentary changes based on the date they think the article was published. (The dates on the comments will also be changed randomly.)
Depends on the content. If you're dealing with broad timeless subjects dating isn't a big issue. You could read an article 50 years old dealing with male/female relationships and get some value out of it. If you're dealing with technical issues, or any subject that moves fast, it's more than a little irresponsible not to date your work. I've personally hosed multiple Linux systems trying to follow old out of date HOWTOs. I know better now of course but I still have to filter through hundreds of outdated results for almost any technical search I do these days.
I'll add here that more people should date comments in code! Don't know how many of you are collaborating on projects, but It's a personal pet peeve of mine to see something like:
// Updated for the new login system
...which of course has other problems too, but it's really annoying to try to figure things out when comments are out of date.
I also do this. But, theoretically, it's not entirely legal. In other words, if you have a piece of content which has not been changed, but for which you've "automatically" updated the copyright date, you are basically lying about when the content was created. Not sure how this affects any future standing you may have. It's also not really a concern for a reasonably dynamic site, including things like JTV.
I wonder: if the team of web site caretakers gets together several times a year, looks at the web pages, sometimes decides to change it, sometimes not -- is it legal enough to put the current copyright?
Or, coming from the other side: one could put forward an argument that the contents of a page is regenerated every time anyway (definitely regenerated at the client, and often regenerated on the server as well). I wonder if this argument holds water, I am not a lawyer in any way.
Untested hypothesis: maybe adding some nearly-trivial randomized content, such as a quote from a public domain quote database, can allow you to claim a new copyright each time the page is served?
I completely agree and I can't help but think that it relates to "The Last Responsible Moment" approach to decision making where delay = more current data, so if an article has no date, its value is seriously suspect.
For example, I have a bunch of stuff writing about customer service written in 2006/2007. That field has not evolved so much that "Don't curse at your customers" is bad advice in 2009. However, if one of those were posted here, the first comment would likely be "Dude, mark in the title this is from 2006", because the audience here is socialized to assume that the new and shiny is what they want to be exposed to and the old and crusty is largely not valuable with the exception of a few classics.
I have sites where I have no visible dates for exactly that reason: I intend the content to be evergreen, and if it is evergreen you don't need to know what year it was planted. (Edit to add context: elementary school teaching activities. I am not a textbook publisher so I have no incentive for ripping up my multiplication activity every year and coming out with New Shiny Multiplication 2010 Now With Numbers You Totally Couldn't Have Multiplied Last Year!)