A lack of standards is hindering blogs, feeds, and blogrolls. And I think FeedBurner should fix it.
Rick Klau recently pointed out that my practice of linking to the feeds of those in my blogroll (as opposed to linking to their blogs) may not be the best practice from an SEO (Search Engine Optimization) perspective. I agree. But the problem is much more complicated than that. How to manage blogrolls, feeds, blogs, websites and their interrelationships is a huge problem. I took Rick’s suggestion, in part, and added each blog’s URL along with its feed (which was how I used to do it, until my blog’s homepage got really cluttered, which Rick also didn’t like). As you can see, my blogroll now takes up much more screen real estate, and it leaves me wishing for a usable WordPress plugin for blogroll management. But hogging screen real estate is only part of the problem. Here are some of the other problems that I had as I tried to figure out which non-feed URLs to include in my blogroll.
Sometimes funky port numbers are used.
Accelerated Cure Project
(where the blog is http://www.acceleratedcure.org:8080/)
Sometimes the “best” website for the corresponding feed isn’t the blog.
- BzzAgent
- Kayak
(where the blog is http://www.bzz.com/ but the company is http://www.bzzagent.com/)
(where the blog isn’t http://wp.kayak.com and the company is http://www.kayak.com/)
Sometimes the website and the feed live at different domains, including any blog hosting a feed with FeedBurner.
Sometimes the FeedBurner feed isn’t guessable.
- FeedBurner
- Staring At Strangers
(where the feed is http://feeds.feedburner.com/BurnThisRSS2)
(where the feed is http://feeds.feedburner.com/typepad/YqhW)
Sometimes there is a feed and no corresponding blog.
- SiteSpect
- Rick Klau’s Google Reader
- DomainTools
(where the news page isn’t really a blog)
(where only Rick Klau has access to the website version of Rick Klau’s Google Reader)
(where I looked for, but couldn’t find, the blog)
Sometimes there is one website and multiple feeds.
Sometimes there is no discoverable feed.
Sometimes there is a website, a blog, and a feed, all at different subdomains.
- FeedBurner
(where the website is http://www.feedburner.com/, the blog is http://blogs.feedburner.com/feedburner/, and the feed is http://feeds.feedburner.com/BurnThisRSS2)
Sometimes the feed URL has changed but the auto-discoverable feed has not.
- Staring At Strangers
(where the feed was http://feeds.feedburner.com/typepad/BZnE but changed to http://feeds.feedburner.com/typepad/YqhW)
Sometimes the corresponding website is really hard to find. (Click on the feed and try to find the blog. Go ahead, I dare you.)
Sometimes a guessable top-level domain name redirects to an unguessable URL.
- Boston Red Sox News
- The TTABlog
(where the guessable http://www.redsox.com/ redirects to the unguessable http://boston.redsox.mlb.com/index.jsp?c_id=bos)
(where the guessable http://www.ttabblog.com/ redirects to the unguessable http://thettablog.blogspot.com/)
Sometimes the URL isn’t what you though it would be.
- Cursed To First
(where the blog is http://confessionalpoet.typepad.com/cursed_to_first/)
Sometime the feed is unauthorized.
- Dilbert Comics
(where the official website is http://www.dilbert.com/ and the unofficial feed is http://feeds.feedburner.com/tapestrydilbert)
And these are just the problems with feeds/blogs/websites in my own blogroll. There is no simple tool for managing a blogroll and all of its inherent complexities. So we need a lot of work here. There needs to be a better way to associate a feed with a blog and a website. There needs to be a way to handle the many-to-one and one-to-many relationships (e.g. one feed with multiple websites; one website with multiple feeds). There needs to be a way for publishers to know if they are missing a discoverable feed or if it differs from their actual feed. Feeds created by FeedBurner should have guessable URLs by default. There needs to be a WordPress plugin for managing blogrolls. There needs to be a Firefox plugin for actually discovering so-called “auto-discoverable” feeds which are now only discoverable in theory. And these plugins should be included in the default distributions of these software packages. If blog, feed, and blogroll management is not easy and automatic for both blog publishers and readers, then it will remain chaotic and the full benefits of feeds will not be realized. In fact, if there is one entity that could stand to benefit from creating some tools, standards, and plugins (browsers and blog), it’s FeedBurner. Have at it.
* Dilbert Comic Strip Dropped By US Media Over Creator’s Racist Tirade (2023-02-26)
https://www.bbc.com/news/world-us-canada-64775250
On a more serious note, you’re right re: guessable URLs. Since people can make the URL whatever they want, it was never (as far as I know) a tremendously high priority to tune the algorithm to make the ‘default’ URI be more guessable than it is. On shared blog services, the domains tend not to be useful – as you point out. Point taken… though to what end? It’s not like people are trying to find a feed by appending random strings to feeds.feedburner.com… People find feeds in directories, through lists available at the aggregators, or on the publisher’s sites. I don’t think feeds are like domain names – we don’t host all the feeds (far from it), and mechanisms make it easy to ‘discover’ feeds by simply typing in a site URL and letting the aggregator find the feed(s) associated with the site.
I’m not buying your auto-discovery argument, though. Auto-discovery is a machine-readable standard. To that end, it works well – every major browser supports it. Every aggregator supports it. And that very support makes auto-discovery something that can benefit anyone using a browser…
As far as your argument that you’ve got 2000 bookmarks, that’s a great idea for an app I’d love to see built. (Especially if it could accept different bookmarks inputs: del.icio.us, browser bookmarks, etc.) The app could look at the bookmarks, then crawl the sites to see if there are discoverable feeds available, then build an OPML file to allow for easy importing of the feeds into an aggregator… just like that, you’ve got a dynamically-updating subscription store that’s keying off of what you’re bookmarking. I really like that idea – now to find someone who can build it. 🙂
BTW, I use a ‘subscribe…’ bookmarklet that lets me subscribe to a site (using Google Reader) directly from within Firefox. Details are here. It works great.
I have visions of Dana Carvey reading your comment and pounding the desk. “In my day…”
😉
Greetings Rick,
Regarding the blogroll feature in WordPress. That is not an integrated solution. Think about how you read blogs and built your blogroll. You do it in a reader, not in WordPress. In my case, I read feeds (including feeds from blogs and some feeds without corresponding blogs or websites) in Google Reader. There is no way to automagically make my Google Reader blogroll appear in WordPress. Nor is there a way to say “include a link to the feed and the blog” or “include a link to the feed, the blog, and the website,” or “make this a collapsible list so that it doesn’t hog screen real estate.” Sure, there are blogroll tools, and they all suck (unless I’m missing one that rocks). That’s why I wrote: “There is no simple tool for managing a blogroll and all of its inherent complexities.”
Regarding guessable feed URLs. Do you mean to tell me that http://feeds.feedburner.com/typepad/YqhW was the best URL FeedBurner could come up with for the Staring At Strangers blog? How about http://feeds.feedburner.com/staringatstrangers/, http://feeds.feedburner.com/staringatstrangers001/, or http://feeds.feedburner.com/staringatstrangers999/? Were those all taken? I know that 600,000 feeds sounds like a lot, but it’s not when compared to the Internet’s 88 million registered domain names (see http://www.domaintools.com/internet-statistics/). By that count, there are fewer than 1% of websites with feeds. (This is consistent with my own research.) Now is the time to fix this problem while it’s still manageable. There’s got to be a better solution than a FeedBurner URL comprised of random string of uppercase and lowercase letters.
Regarding auto-discoverable feeds. If a website has a so-called auto-discoverable feed, then a little orange icon appears in the URL bar. So it’s been “discovered” by the browser. So what? What if the user doesn’t notice that little icon? I have 2000+ URLs in my bookmarks.html file. How about a tool to batch process my bookmarks file and tell me which of my URLs have feeds? How about a Firefox plugin that auto-adds feeds to Google Reader? Or that displays a dialog box saying “Do you want to subscribe to this feed?” Auto-discoverable by the browser is meaningless. Feeds need to be discoverable by humans.
Regards,
Erik
Erik – Great stuff. Couple thoughts:
“There needs to be a better way to associated a feed with a blog and a website.”
Actually, I’m using WordPress’s built-in “blogroll” feature, and it allows you to do both.
“There needs to be a way for publishers to know if they are missing a discoverable feed or if it differs from their actual feed.”
I think auto-discovery handles this just fine; go to my site, see the 4 feeds that are discoverable.
“Feeds created by FeedBurner should have guessable URLs by default.”
They are, but keep in mind that with over 600,000 feeds, lots of the “guessable” URLs are already taken, requiring a little manipulation. The guessable URLs are available on a first-come, first-served basis.
“There needs to be a Firefox plugin for actually discovering so-called ‘auto-discoverable’ feeds which are now only discoverable in theory.”
Um, Firefox already supports auto-discovery. Just click the feed icon in the address bar. (Or are you talking about Firefox doing something different? Not sure I follow.)
“If blog, feed, and blogroll management is not easy and automatic for both blog publishers and readers, then it will remain chaotic and the full benefits of feeds will not be realized.”
100% agree.
“In fact, if there is one entity that could stand to benefit from creating some tools, standards, plugins (browsers and blog), it’s FeedBurner. Have at it.”
Interesting. Hadn’t thought about this from that perspective, but you’re right: we position ourselves as a publisher service. This is an area fraught with publisher confusion, and we already provide a bunch of tools to eliminate (or at least manage) that confusion. That said, it’s also touching on things that go outside of what we currently do… but I have some ideas on that front, now that you’ve positioned it this way. I’ll mull this over a bit and get back to you…