hi,

This may not be representative to all blogs, but on mine I have implemented several navigation options: site search, RSS, a date-based archive and tags. When analyzing my statistics I can conclude three things:

1) most users come from Google or another search engine. They jump to an indexed topic directly, read it or ignore it and then leave.

2) the group of rss subscribers mostly check the new stuff and do not look in the archives at all

3) the other portion of loyal readers (or new and curious readers) occassionally browse the archive or do a search. it somewhat makes sense to give them several options for that, but the return in my case is limited. They spent very little time digging through the archives, but that may be in the nature of my content ;)

I'm thinking the average user profile nowadays is somebody using a search engine to navigate. if your site is interesting to them, they may navigate through it once. if they like what they see, they subscribe to it or bookmark it and then monitor it. That's how I got to your site :)

I personally am a big fan of tags and search, especially when a blog or site has lots of content you could otherwise never find.

PS: I'm not sure if this is me only, but my Firefox 1.5.7 crashes each time I try to comment here. In IE it works.

I believe Ferdy has stated it perfectly. It doesn't matter if a site has a search box, most popular links, recently posted links, similar articles links, any sort of archive system: old readers read new articles; new readers may read older artilces.

The only effect I have that works in getting visitors to read archived articles on my site is by placing a reference link in a new article back to an old article. Sometimes they read them; most often not.

When I first came to your site 1 1/2 years ago, I read the article in which I was interested, meandered through older articles - Read some - and left. The last "old" articles I read were those I found on Google whilst researching.

I'll read - from RSS feeders - articles with titles that interest me when accurate or vague; or, link-baiting titles.

I don't believe that anything can be done about truly attracting visitors to site archives. If you use del.icio.us bookmarking as bellweather, new articles are bookmarked for - Maybe - one month (excepting those that are found on sites newly linking back to you) after which they are ignored.

I'll wait for Google to promimently rank my articles so as to get readers reading my archives.

I find that tags or well categorized articles work best out of the methods commonly in place. That's because I tend to surf through many articles within a give topic. This becomes problematic as archives become deeper, however.

A couple of years ago, when I was just starting to delve more deeply into web standards, I would cruise the A List Apart archives on any one topic and just internalize everything. Since then, I feel like a ton of terrific new content has been added that I was able to digest a few articles at a time. If I was just getting into standards-based development now though, I would go to that site, click the code section, then the css section and be greeted with nearly 80 articles. Browsing an archive of that size is simply unmanageable.

As archives grow, search really becomes more crucial. Going back to the ALA example, their search is pretty good, but often even those results are difficult to parse. Titles are crucial to good search. For example, if you're searching for something like Doug Bowman's article "Sliding Doors of CSS" but you don't know the author or the name of the technique, you might search for "css menu". That article won't even come up for that query, though it is most certainly about menus and css. The important thing to take away is that if you have lots of content, you should think about how people might search for it and tailor your titles and content for that.

As you hinted, perhaps Google is the ultimate site archive. Even content with a poor title will often come up in google search, although the searcher might not realize what they're looking at and skip over it.

"As you hinted, perhaps Google is the ultimate site archive."

Rather, as Sean Hinted. Sorry about that, and sorry for the double post.

Good points to raise Dave, I've wondered the same thing, but don't really know what to do about it.

A few times I've found a blog on a topic I'm interested in finding out more about, but invariably topical blogs seem to start off with the basics when they first get going, and head gradually into mroe and more detail, and continue to post about new information. They don't tend to go back over the basics though.

I usually find it very difficult to find the 'start' of some blogs stories, to find that beginner content that helps me understand the stuff later on.

A List Apart has already been mentioned, and I think for it's content, it works reasonably well. Not sure it'd work in all cases, but for them it's ok.

I think the same issues are present in forums as well. Good threads end up disappearing into the ether after a while.

I don't think Google is really a good solution, because I don't think search always meets what people are after. I'd think that several of those people who try seraching through old content give up because they can't find what they are after.

Tagging only works if the tags are good. If the tags used are good, and all content relating to those tags is tagged properly, it can work reasonably well.

Perhaps certain tpoical blogs could have a difficulty rating tag as well. So something might be tagged as CSS and beginner, or CSS and advanced.

And having a system of being able to filter by mulitple tags, so someone could select a specific topic, and a difficulty level.

Regular readers, using RSS are mostly fine as they are I think.

If I follow a feed in Bloglines, I'll keep relatively up to date with it, reading as I go. The only time I'm likely to want to read old content again is if I'm doing something and remember the blog talking about it. Then I would use search, probably on the blog itself, to try and find that post.

I'm not going to casually browse the archives of a blog I've been reading for 6 months already.....

Great discussion. I have two things to add. I don't think tag clouds are completely useless. They are fairly useless for navigation, in that aspects they are identical to categories which we have seen since the beginning of the web. It is the font sizing in the cloud that makes it useful to me. A new user can quickly see what most of the content in the site is about. Slightly useless, but not completely :)

The second thing I would like to add is an idea I read somewhere on a blog. The guy's idea was to have his blog homepage set to a digg.com style overview of his most popular content, and users can vote for it. Only the most popular entries stay on the homepage. Users that would like to keep up with the blog chronologically can still do that by RSS or an additional "view" on the blog. I find this idea to be daring, but highly innovative. I'm too afraid to try it myself, but what do you think?

Perhaps, if you had the mechanisms in place, you could work out rough importance based on the number of page views, or maybe the backlinks to an article in Yahoo! for instance.

Almost like you're own internal Google PR, heh.

More popular, and therefore hopefully better content, you'd expect to get more page views and more links.

Problematic when it comes to quantifying RSS reads etc.... Would be other issues too, but could be another way of working out usefullnes.

I agree, tag clouds say nothing about content popularity, only about sizing. That's why I had this idea of using color codes along with size to indicate both the size and popularity of tags. Never worked it out in practice though.

Yes, the digg.com style blog homepage would require a lot of readers, perhaps too much for an individual's blog. Although I can't find the particular entry where it was discussed, it at least was on this blog:

http://steve-yegge.blogspot.com/

Tags are helpfull after you've ended up on a blog about some topic you're researching.

Say you've searched for something with Javascript and find something usefull on a blog where the are other posts in the same topic or Javascript in general. I'll wast 2 mins to check out atleast the headlines of the posts in the categories that I find interesting on a page.

A voting system is great, hopefully more people will vote then those who comment on a post ;-). Random strangers like myself that come here once a month or so or regular users.

Hi Dave,

After reading some of these posts I have an idea that might work. First off, think of many of the e-commerce sites and attempt to see how they correlate to a blog. They display new products that have been released (frontpage of most blogs), have categories of all of the products available (perhaps tags on posts) and they usually have one other aspect...a clearance section.

The online store is wanting to get rid of some old merchandise so that they have room for the new. In the case of a blog, we might be able to implement a system that shows articles relevant to today. If you are reading the news and realize that one of your posts from four years ago is relevant, you should be able to tag it and have it displayed as a link on the front page. Perhaps it could even be more automated by crawling many news sites and then searching your site (google?) for older articles that might be helpful to readers.

This ensures that the most relevant posts have a chance at making it back out of the archive and it keeps non-relevant posts from ever being seen. This may be good and may be bad. I will leave it up to you to decide.

Interesting.

I'm not a regular reader but have been here before.

Here's how I got to this article: while looking at my del.icio.us page I have some time to stumble around the web, so I click over to del.icio.us for the hotlist and tags to watch. Somewhere there your cheat-sheets category page showed up and I was thinking, "Hmmm, I Love Jack Daniels... been there before and liked it but don't remember exactly the topic." So I ctrl-clicked it for later viewing. Once I got around to it (btw, nice sheets, thanks), I still vaguely remember liking your content, so I clicked the header to see what was new, paying only a passing glance at the sidebar (for now). At the root level I scroll down kind of slowly but not reading anything in depth (there's something intangible that was fouling my reading gears - the whitespace maybe?). Anyway, I see this article and start reading. Good stuff here.

So since I kind of like the article I'm tempted to dig around a bit. I'll head to the sidebar, back to the homepage, and only after that is exhausted will I maybe try to tackle your archive table. I see you have categories and months. That tells me very little and I don't have time to poke around at one site hoping to find something of value. (No offense, just hoping to help.)

What would compel me, a returning but still occasional reader, to read more, or read more often? Well, a list of popular content would give me a fair snapshot of your content, a list of your content picks would help me know what *you* like, a list of recent content would tell me what's occupying your interest lately, etc. It's all important to develop a feel for whether and how frequently I visit in the future.

The big problem is you're competing with a bizillion other yayhoos who want readers; readers with microscopic attention spans. Show me the good stuff, the stuff you like, the recent stuff, ... don't expect me to dig around until you make the case that this is an interesting place to hang out. You got my attention early with the domain name, now you've got to keep my attention.

Also, once someone downloads all your cheat-sheets, they're outta here unless you highlight your writing (which is great btw). Maybe sometime later they come back and get a new sheet, but more likely they forget. I've got a thousand-twenty bushels of links (including yours) at del.icio.us and the only reason I got back here was seeing someone else's link to you. You need to make it to my daily list (which is too big for me to get through daily but you'd be one or two clicks away from my browser start page).

Sorry for the long wandering comment. Just wanted you to know how you could hook me (or anyone for that matter) into coming back.

As a user I'm fond of the good old category tree, related articles and search.

I am about to start at the beginning of your blog, having just discovered it through a link to a cheat sheet on Javascript.

What is most useful for me is a way to keep track of what posts I have read. And I don't mean in my browser history list which disappears when I delete the history or after 10 days. I wouldn't mind having to register (for free of course) so that this information could be stored somewhere.

Since I am interested in all your posts at this point, just being able to start at the beginning and read my way through (over several days) is what I need right now.

If you have posts that are worth going back to (and so far it looks like you do), the category list is one way. But when you start to build up a large amount of posts, important information gets harder to find even with categories. You may need subcategories.

Thank you for your website.