Marjolein Hoekstra lives in The Hague, Netherlands and writes the blog CleverClogs.org. In the magical world of RSS power use, Marjolein is the High Priestess. She is my mentor and my friend. She is many peoples’ friend in the industry.
Marjolein is both a pioneer and a cynic. Her latest project, an advanced implementation of the Grazr OPML platform for podcasting professionals, called Podcasting Professionals, is just the most recent example of her work. She also built One Pipe, a handy feed filtering bookmarklet that was featured on Life Hacker and her influence is baked into many other peoples’ RSS services online.
She is also a cynic, though. She doesn’t have a lot of hope for the future of RSS magic by power users. I disagree with her on this forecast, but will reserve my arguments for a later post based on a recent consulting case study of mine.
For now I will leave you with Marjolein’s own thoughts on reading thousands of feeds, the RSS vendor landscape and the future of RSS.
Marshall: I’d like to interview you about the free RSS related tools landscape; like about your list of services that have dissapeared (that’s fascinating) – about some of your favorite tools online and why you like them.
Marjolein: What about how I use a feed reader, not to read feeds but to bookmark them.
Marshall: How does that work?
Marjolein: Well, most people complain that they don’t have time to read all those feeds. I can’t possibly imagine that I could read all the feeds I collect. I collect them because I want to be able to catch up with what’s new should I need to.
So I have categories of feeds about my core interests, and when I come across an interestig web page with a feed – I ‘bookmark’ it by dragging it onto my feed reader. Over time, a pretty solid representation emerges from who the major players are in that field.
Marshall: Because they are the ones you’ve added just as you’ve discovered them?
Marjolein: Well, I trust my own experience. If I see a blog I can tell pretty quickly if it’s meaningful and authoritative – as much as you can.
When I’m on a research spree about ‘widgets’, for example and I want to know what authoritative bloggers have said about a certain new service, then I log on to the BlogRovR account that monitors those feeds for me.
And when I visit a page that’s been blogged about by that person, I’ll know about it.
Marshall: That sounds like it could be a big step less complicated with some advanced Blogrovr functionality, but I follow what you’re saying.
Marjolein: This is particularly useful for people who are active in a certain field; for example marketing, or PR, or knowledge management, or SEO.
Marshall: Yes, I can see that.
Marjolein: Answering the question “has any of my peers covered this site yet?”
Marshall: I use a custom search engine similarly but I can see how this method would have certain advantages.
Marjolein: I get carried away.
Marshall: No no, this conversation inspired me to try putting a blogsearch feed from ask.com into Aiderss.
Marjolein: In the past I’ve experimented with creating hundreds of search feeds, thousands actually. I should be more precise: hundreds of search feeds, tracking thousands of keywords. Each feed would typically search about 6 keywords. There’s no mechanism to really facilitate this easily; keywords might be deprecated, no way that you can edit a text list somewhere to reflect this. You’ll have to rerun the query and grab the feed again. It now turns out that Google doesn’t like multiple simultaneous queries – it times out.
I wanted to create a river of news from hundreds of keywords related to RSS. So one thing that would be nice is if you could set up a list of preferred keywords somewhere, hundreds if necessary. There’s no mechanism for that anywhere yet.
Marshall: Yes, that is the kind of thing that keeps me up at night as well, honestly. Fwiw, it worked – ask.com blogsearch plus aiderss is fairly interesting.
Marjolein: Awesome. Have you got this live somewhere where I can see it?
Marshall: [It worked at the time but it’s broken now, AideRSS is down, unfortunately – a real, real shame.]
Marshall: So one thing that would be nice is if you could set up a list of preferred keywords somewhere? What do you mean by that?
Marjolein: I maintain lists of keywords, four lists actually. Right now they’re on my hard drive, but typically they’d be on the web somewhere – preferably behind my credentials, unless I choose not to. The idea is that you can collect feeds from individuals but I’ve found that it’s very refreshing to find stuff within my interest range from people I’ve never heard of.
How to find them? Keyword search makes sense but searching for “RSS” won’t cut it – way too broad, noisy.
So I’ve defined lists of keywords and if any of these occurs within the title of the post, together with the words RSS, aggregator, or feed (including variations) then it’s likely that that post is of interest to me. Among the keywords that I’d monitor are names of vendors of course, technologies and techiques, names of people.
Marshall: How do you run those queries? As long strings of OR?
Marjolein: I run those queries like this “inposttitlerss|feed|feeds|aggregator|aggregators inposttitleaccept|addict|agent|alert” That’s one query.
Marshall: In Google Blogsearch?
Vendor Failure in RSS Edge Cases
Marshall: That’s great, but I want to ask you about your list of companies that have taken a dirt nap, as they say.
Marjolein: Ok- “seemingly out of business” the list is called in my document.
RSS companies seemingly out of business
Marshall: What’s your criteria for being on that list?
Marjolein: I should probably have put notes behind each, what I did, and why I came to the conclusion. The likely procedure that led to this was that I opened the homepage and read a ‘we’re moving on’ note or domain taken by someone else etc.
Marshall: And what does this list mean to you? Are there any theories this fosters in your mind about the state of the industry?
Marjolein: Well, if you look at my list of RSS tool vendors 250 of them on that list have a feed. Let’s assume this is half, or one third of what’s really going on -maybe 750 RSS tool vendors at the moment? Then ‘only’ 25 ceasing is really not that much. But the other metric is how active are these others? One good way to measure, is to look at their news feed. It turns out only a handful really update their feed regularly.
Marshall: News like company news or their blogs?
Marjolein: Both, either – release news!
Marshall: And what does infrequent mean to you? Like less than once every three months?
Marjolein: Hh, I’m talking about companies that haven’t updated their feeds in years; it’s annoying. After the first year they lose interest.
Of those 750 companies in the RSS sector only 250 produce a feed; of those I think only 30 regularly update their feed. I should probably research that to make sure. I almost dismiss a new RSS vendor immediately if they don’t have a feed. I mean how can you…
Marshall: lol, yeah
Marjolein: Eat your own pie
Marshall: Eat your own pie! That’s what I always say.
Marshall: So who’se making money?
Marjolein: Within the RSS scene? I presume the enterprise guys. [This was a depressing tangent to go down, so I deleted it here. Fact is, Marjolein does not believe many RSS vendors are or will make any money at all.]
The Future of RSS
Marshall: Ok – so what’s the future of RSS Marjolein? Will the freakiest, screen scrapingest services go out of business and the big, simple ones get bought up? Will that slow innovation way down?
Marjolein: I think we won’t see many new RSS services arise like we’ve seen in the past few years. It’s just not sustainable for them.
RSS will just be a commodity, a feature, of existing start-ups. Look at what Yahoo came up with: Pipes. It’s great for geeks but not normal people, so to speak.
Services like ZapTXT have a chance if they can keep it simple.
This is the contradiction that Nick Bradbury once wrote about. As a power user, I’d want more and more features; the developer follows up, thinking these are all good things – creating a monster that new users find daunting.
What we do, feed splicing, feed manipulation, scraping -there’s really not that many people in the world who do that. Maybe 250 altogether? And of those a lot are developers themselves and they are capable of writing their own routines.
So what’s the market anyway? There may be people who merge a feed here and there. But not on the scale I’m doing – the effort [by developers to build the software] just doesn’t justify the goal.
It probably doesn’t surprise you that I maintain a list of RSS freaks. This weekend I sorted out their email addresses, if I didnt’ have them already. One of the things I like doing is to send out interesting stuff to people. Thus I need to know who to send things to, who are the players.
Marjolein: Back to that quote that you entered before.
Marjolein: “Although there’s a lot of attention data that could be stored in OPML, my recommendation is that we keep it simple – otherwise, we risk seeing each aggregator support a different subset of attention data. So rather than come up with a huge list of attributes, I’ll start by recommending a single piece of attention data rank.”
Marjolein: That’s pre-APML. [Attention Procesor Markup Language]
Marshall: Is APML generally accepted now?
Marjolein: I like to think that it is and I act like it is and I talk about it like it is.
Marshall: So what do we do Marjolein? Those of us who love to mess around with feeds using 3rd party services in creative ways?
Marjolein: They’re at a loss, at least I am.
Marshall: What do we do if that looks like it was only a short period of time and not a move into the future?
Marjolein: We shift focus, Marshall: we find other things and just accept things the way they are. Or we sit down and write detailed specs on “what is the perfect feed manipulation tool” and get funding.
Marshall: And then we put up the money for a bounty and make it open source?
Marjolein: lol, yes.
This conversation with Marjolein reminds me of one of my favorite childhood movies, The Last Unicorn. No matter what happens to the edge cases in RSS magic – at least some of the Web 2.0 world’s magic is here to stay. Stay tuned to my blog here for forthcoming case studies.