The Follower Slot Syndrome

The social messaging service Twitter, which has been called the Swiss Army Knife of online communications, has seen a few changes under the hood since the inception of the service. Among these are the hardcoded follow rule which was introduced on Twitter in late 2008. Those most hit by this rule are users who are unaware of the limits and are very generous with their follows. When I myself bumped into that limit, it gave rise to thoughts about how I use this service, what kind of value it has and how I need to follow and unfollow others.

Important stuff? No, not really. But then again it touches on some pretty important things, like our ability to speak and be heard via the online architectures we use. And that warrants some lengthy attention IMHO. I hope some influencers and “high profile” Twitter users will take note, reconsider their stand and build up their capacity to deal with larger information intakes.

The follow rule

Chances are you won’t have bumped into this limit if you’re new to Twitter, but if you follow many people, and specifically if you follow more users than follow you back (typically celebrities or other high-profile influencers), you’ll likely bump into it when you hit 2000 follows.

Before Twitter introduced this rule, following was free game. Everyone could have as many or as few followers and follows as they liked. Everything was open and one could be generous with one’s attention without fearing that one would “run out” of slots. This changed dramatically with this rule.

The basic rule is this : you can follow only +10% in excess of your number of followers after you hit 2000 follows.

Basically, if you’re followed by 2000 users, you can follow 2200 yourself. If you’re followed by 10.000 you can follow 11.000 yourself. This rule, while good-intended, has some bizarre effects when you take a closer look at it. Among other things, it raises significantly the value of the commodity on Twitter known as a follow (i.e. attention), and even more that of a mutual follow (mutual attention), i.e. someone who follows you where you follow that someone back too.

Background : “bait-following”

This rule was introduced to combat “bait-following”. This is also sometimes referred to as Twitter “spam”, but I don’t acknowledge there is such a thing as “spam” on a service like Twitter. Many users have experienced this particular obnoxious phenomenon. Some users, either by themselves or using tools which utilize the Twitter API will track you down depending on your profile description or keywords in your tweet track record and follow you. What makes them different from users who genuinely want to connect with you, is that they do this en masse, following thousands of users, in the hope that some percentage will follow back. Hence the baiting. Those who do follow back can then be exposed to the advertising or other spam-like messages such as affiliate links to products you’re really not interested in or links to services which “help” you get “more followers”.

In Twitter’s early days, it wasn’t uncommon to browse around and follow other users somewhat randomly and sometimes stumble over interesting profiles and make new genuine connections. But automated tools made it considerably easier to “exploit” the fact that most Twitter users were generally willing to follow back others who were interested in connecting with them (and maybe still are, to a large degree).

These tools and the users who employ them (I’ve experimented with some myself at one time) use Twitter as a broadcast platform. It is the same logic applied to the online medium as is daily applied to television. It doesn’t matter if you waste 99% of your audience’s time, if you can sell something to the remaining 1%. That may be enough to make it worth it. Trouble is, the 99% still think it is a waste of their time, and therefore using methods like these to “increase following” is doomed to dry out sooner or later, as most will quickly see through the scams and unfollow such scammy attempts at gaining some attention.

After the hardcoded follow rules, scammers must now unfollow all those users who don’t follow back (but this is comparatively easy with automated tools), but then they are free to repeat the stunt. In other words, this particular type of use of Twitter persists. It’s still very common, and there’s very little the hardcoded rules can do to prevent it, because basically Twitter is a very open platform which grants access to it’s data to a wide host of third-party tools (which, among other things, make it great).

Twitter misconceptions

I wouldn’t care about follow or follower numbers so much as I do here, but because I feel Twitter nurtures some misconceptions about their own tool, which will make it less valuable, to me and other users – and ultimately to Twitter too.

First, as I stated, it is hard for me to accept that the crude misuse of Twitter described above is spam. For anyone to deliver a message to someone on Twitter that someone has to follow that anyone first. So messages on Twitter are always solicited. That you have been tricked into soliciting the messages doesn’t make such messages unsolicited.

In my humble opinion, Twitter should have kept their service pure. They should have butted out. They shouldn’t have started becoming involved with determining what kind of interactions took place using their service. Twitter would have survived fine, in spite of the crude attempts to undermine it’s usefulness. They should have worked to ensure it stayed a strong platform, which could make it as reliable as a phone line, but way more powerful. Twitter is a strong versatile platform and people used it very creatively on their own, blocking users they didn’t like and following those they did like. It was brilliant.

But they did. Twitter as a company couldn’t just look quiet at the many paths it was conceived scammers went to undermine their service. Fear started to kick in, and demands came from some users that Twitter needed to regulate and filter conversations and connections. They started abolishing user accounts whose following behaviour patterns made them suspicious. And they introduced hard coded rules, with the aim to stifle that particular kind of baiting spam as described above.

Twitter has a perception of it’s own service as a stream of information, which has to be managed. Noone can manage an intake of more than 2000 followers. At least not without losing out on many messages. So the argument for such hardcoded rules goes. However, this perception is wrongheaded as an attempt to figure out how Twitter data is used. The truth is Twitter has no idea whatsoever about what creative ways users may take in the data in their streams. One user taking in a lot of information may analyze it with a piece of software Twitter knows nothing about. Another may write a tool which filters the incoming stream according to criteria Twitter wouldn’t ever understand. Fundamentally Twitter is optimized for filtering at the receiving end, as the information intake will almost always be much much larger than the outgoing information stream. What we need in other words, is not better ways to restrict access, i.e. hardcoded limits on the posting end of the information loop – but better filters at the receiving end.

Filtering the information intake

I’ve often had Facebook friends complain about the massive stream of messages from me coming their way, when I send my tweets via in that direction. True, some nerdy stuff in there which they could care less about, but I want to include them, not exclude them from my information circuits, that’s why I send it their way. If I know the precise recipient of a message, I will usually send a direct message or email to that person. But more often than not, there’s no direct recipient but the aim to strike a chord or strike up conversation and input, like when I write a blog post. Social messaging is sometimes referred to as microblogging, and that is perhaps a very accurate description of the way I use Twitter. I send it their way, because I hope some of it may create new connections, from where the conversation may rise. I may discover new things about my friends doing this, because I never quite know who possesses the information I seek or share my interests and concerns.

Increasingly, as recipients of large information flows, their job then is to learn how to filter what they take in (if they do not choose to block me or unfriend me because I am “too loud”). We all need to do this. We all need to learn how to filter out incoming streams, i.e. prioritize what is more important than something else. What we need to read before something else. What emails to reply to first. Etc. Increasingly, we also need to learn to code and use aggregation tools on our own as well as free licensing, if we want to be independent of the filters offered us by proprietary service providers.

A large information intake or large information stream may be overwhelming, but it has nothing to do with spam. Spam is unsolicited messages sent to a lot of people in the hope that a small percentage responds and buys something. Information streams can be managed, filtered, analyzed, put from one form into another form.

The hard-coded follow rule imposes a limit in the wrong end. To get the best possible dataset, you don’t limit the intake, you take steps to make it easier to process the intake, to make it easier to get the desired data out. Twitter has no real idea if their users have need for a small or large intake of information for their data needs. But this is not the only place where Twitter don’t _get_ Twitter. I’ve often come back to how Twitter displays a huge failure to understand the value of their own data, when they don’t allow access to the full archives of tweets. You can go back only to what corresponds to three months worth of tweets. This means that all this data cannot be retrieved, filtered, analyzed and brought to use by clever people who want to know something about social behaviour patterns, particular brands, viral effects and all other things thinkable and mentionable. Twitter has a pre-conceived notion of what Twitter is, and if users don’t use Twitter that way, they are wrong and must be corrected with hard-coded rules to use Twitter as Twitter was intended. But the truth is the versatility of Twitter has made it much larger than itself – it has outgrown it’s initial purposes by milelengths. If Twitter doesn’t get that (and the true value they can offer as a business), they risk running their service into the ground, because they don’t make it profitable.

Following back

Now, I recently provoked some debate and diagreement among some of my followers when I provocatively asked why they didn’t follow me back. Actually, the message was not really aimed at those who do follow me, but at those who don’t. By those I mean the wide host of celebrities and influencers which are known to have a large following on Twitter, but only follow a small host of people themselves. I follow a wide host of them, but hitting the 2000-follow limit forced me to re-consider a lot of them. In fact, I unfollowed at least 800 users who didn’t follow me back, in order to allow me to follow others, who do follow me.

When someone follows me and I feel they are real people who are interested in what I have to say, I usually want to follow them back. Not only as a token of courtesy and respect, but because I feel strange when talking to someone and I have no idea what they are like. I want that influx of ideas from others and I honestly don’t care so much if I manage to read _everything_ but it’s there and I can take that data, do a search, create a filtered feed and other things if I want to, when I want to. You can too, if you want to, and if you want to learn how to do it.

What stopped me from following others back? The 2000-limit and the many many users that I followed, who didn’t care to follow back. It says I can only have 200 “non-follower” users I follow, if I want to follow everyone back who follows me (and I usually do). So what provoked me is that while high profile Twitter users such as Barack Obama, Scobleizer and Guy Kawasaki follows me back, why can’t others? If they can, why can’t you?

To me not following someone back is a message saying “I don’t care what you have to say” or “You’re less important than me”. Less worthy of attention. I’m worthy enough to be in your stream, but you can’t be in mine. That is the wrong message to send out, no matter what you want to communicate using Twitter, it’s a bad way to start a conversation with anyone. Don’t get me wrong, there’s nothing wrong with being very selective about who you follow, but if you overdo it, you also risk coming off as arrogant and disrespectful, because you do not take part in the exchanges on an equal footing.

I don’t care if Obama, Guy Kawasaki or Scobles actually reads what I have to say. I care about the gesture. I care about them saying with that gesture that if you give me attention, I will give you mine back. Even if it is not true. They will not occupy one of the most valuable rare 200 slots I can allocate to only information intake. These may be reserved for others, typically high-profile users whose opinion and information is so important to me, that I don’t care if they listen to what I have to say. As a company or as most people using Twitter, you don’t want to bet on yourself being in that category. You should follow back. Why reach out (have a Twitter account) and then don’t want to listen to what people have to say? Indeed, what those few who’s already decided they want to give you their attention, have to say (if anything).

I don’t consider myself an atypical Twitter user. There are many bloggers, companies, organizations and other users who use Twitter because they have a message they want out. We want to reach other people, make connections with others who are interested in what we have to say and offer. But I just unfollowed a lot of startups and internet professionals, who didn’t take the time, were too disinterested or too lazy to follow me back. They lost what tiny piece of my attention they had. They didn’t need to. With a small gesture, they’d still be in. Would it matter? I don’t know. Nobody knows. But they’d have given a small but important gesture, which doesn’t cost them much but may – just may – give them something of value back some day.

If attention matters to you, i.e. it matters that you reach someone out there with whom your message resonates, you can’t afford to throw away the tiny bits of attention you’re afforded when you’re afforded them.

flattr this!

Tags : , , ,  

An Open Letter to LinkedIn

I’m getting increasingly pissed off by this message, which appears every time I want to add someone as a connection using LinkedIn :

Please note: You should only invite people you know. Several recipients of your invitations indicated that they don’t know you. If enough recipients indicate they don’t know you, then you will be required to enter an email address to invite classmates in the future. More info…

Tonight, I just had it. So I wrote this email to the LinkedIn Customer Services Dept., which is shared below. Wanted to share it with you here, in case they do not comply. As I’ve said before, I’m not a big fan of these types of closed social networks, such as LinkedIn and Facebook and others. In fact, we don’t really like them, but we use them anyway, as long as they are beneficial to us and can connect us with people we wouldn’t otherwise be able to reach. But obviously there are limits. I don’t want to be insulted, and the above message comes pretty close to feeling like an insult to my intelligence.

So I wrote this email :

from	Morten Blaabjerg 
date	Sun, Oct 19, 2008 at 9:50 PM
subject	Please remove annoying message

Dear Sir,

Can you please remove the message saying “several recipients of your invitations indicated that they don’t know you etc”… from my “Add connections” page? It is kind of annoying to see it there every time I check into that page, and there’s no imminent way I can remove it myself, it seems.

It simply spoils my good mood. If someone doesn’t remember me, is their business, and they can elect not to connect with me. I don’t care if they do not connect with me. If they don’t remember me, I can live fine without that particular connection.

But I take insult from repeatedly getting my good mood spoilt by being spoken down to like a baby every time I want to connect with someone using your service.

Alternatively, if you don’t want to remove that message, I would like to cease to use your services and have my complete account erased from your servers. And I’d like to have our LinkedIn group removed too. This stuff just gets on my nerves.

Yours Sincerely,
Morten Blaabjerg, Kaplak

Kaplak has chartered unknown waters and reached strange shores :

If LinkedIn does not comply and remove their system message from my screen, I’ll simply demand my accounts be erased and leave LinkedIn. I will also cease to recommend others to use it. I’ll focus on other networking services such as Plaxo Pulse or others, where I don’t have to be spoken down to every time I want to connect with someone. LinkedIn is useful and a fine tool, but it’s not life support.

Also, I forgot and should have given Customer Services a heads up, that I have never worked in that big company called Rubicon, which they constantly recommend I connect with employees from. I once co-edited a students’ periodical of that same name, though. It seems strange to me that LinkedIn cannot see, that these are very different entities.

flattr this!

Tags : , , , ,  

Aggregation Tools For WordPress: The Pros And Cons of FeedWordPress and WP-o-Matic

We’re in the process of setting up our Planet-like website Kaplak Stream. I’ve done some extensive reading and testing of the two most prominent aggregation plugins for WordPress and WordPress MU : Guillermo Rauch’s WP-o-Matic plugin and FeedWordPress by Charles Johnson (aka RadGeek) of Feminist Blogs. This article will examine the pros and cons of both these plugins, in their present state.

Both aggregation tools are open source and distributed under a GPL license, which means that anyone may adjust the workings of these plugins and re-publish their version. They are each however developed and pioneered by one developer only, and rely heavily on the committment of their developers.


WP-o-Matic is developed by 16-years old Argentinian wunderkid Guillermo Rauch, who has done a remarkable job. Schedules are very easy to organize. They are called campaigns, and each campaign can fetch as many feeds as you like. Campaigns are executed by cron, which runs on the server and executes the fetching script at specified intervals. If you can’t get cron from your web host, the WP-o-Matic script can be executed by Webcron. Webcron has been a free online service until recently. Now, the service must be paid for, however (at a very low price, one may add).


  • Wonderfully flexible customization options of each campaign, directly accessible from a brilliantly designed WP admin interface: specified expressions or URL’s can be transformed, and additional custom text or code added to each post in the campaign (such as ads). Great stuff.
  • Uses cronjobs for executing the script, which should provide the greatest reliability, if you can get it.


  • Doesn’t use timestamp of fed posts, if they are older than the time window set for the campaign. I.e. if a post is months old and you’ve set your campaign to fetch every hour, posts will be timestamped with the time of feeding it, rather than the original timestamp. This sometimes means older posts are published in the wrong or opposite order of the feed, which messes up the chronology of a blog. This, combined with the bugs which makes it difficult to re-run fetches without completely removing the campaign, makes correcting the timestamps a very tedious affair. If timestamps are important to you, this is a no-no.
  • Uses Unix/Linux cronjobs for fetching feeds, which is good if you can get it – and know how to set it up, but not all can or do.
  • Seems unreliable when used without Unix cron. Campaigns are not processed at all, or processed at the wrong time intervals.
  • Bugridden – small bugs such as campaigns not resetting properly, when reset. Complete campaigns and posts have to be deleted if one wants to re-fetch a feed to test a new configuration.
  • Uncertainty if the plugin is supported and developed further by it’s developer. Last release is from October 2007. Guillermo (who has now turned 17) recently announced his continued support for WP-o-Matic and the release of a new version in the near future, along with a new website specifically for this plugin.


I initially had problems with feeds from Google Reader (and Twitter, for that matter) – titles showed, but content disappeared. At first I thought this was a general problem with Atom feeds, but it turned out it’s because WordPress (even the latest versions) comes bundled with an outdated Magpie RSS parser. At first glance, the problem wasn’t fixed by exchanging the rss.php and rss-functions.php with the updated ones bundled with FeedWordPress, but reinstalling these files and re-entering the feeds did in fact solve the compatibility problems with Atom feeds. At first, coming from WP-o-Matic’s advanced campaigns setup, I wasn’t impressed with the interface provided by FeedWordPress initially, and the hazzle I had with Atom feeds gave me the impression that this plugin was no match for WP-o-Matic. But as I worked with it, FeedWordPress turned out to be an extremely competent agent for the job.


  • Extensively well documented
  • Seems to be the more stable and reliable candidate of the two. Works great with WordPress’ built-in cron alone.
  • Built-in API for WP themes and plugins to use
  • Maintained, supported and seems to be actively developed by the developer (last build 8 May 2008)
  • Works great with timestamps – fetches all timestamps from feeds 100% correctly.


  • Can’t add custom text or code to the posts of each particular feed, except if one utilizes the API. If one utilizes the API from a WP theme, custom changes will apply to all syndicated posts, when they are displayed on the site. This is a solution in cosmetics only, in that the custom layout and text is applied only in the visuals – and not reflected in the actual contents of a post. One has to access the API from within a plugin, which hooks itself up with an action or filter in WordPress, to actually ‘inscribe’ posts with custom text or code, which stays with the post, no matter how it is skinned or re-published by other sites. This requires a bit of PHP coding/hacking skills.
  • Can’t import tags. Tags can be imported by FeedWordPress as new categories, however, which somewhat alleviates the problem, but forces you to go with the category system over tagging or both.


Both these plugins reviewed here possess tremendous power, at the point of your fingertips. None of them are perfect, however, and both still need work, but I’m impressed with both. What they can do, and the power and speed of which these plugins work, is impressive. I’d love to have FeedWordPress feature the powerful customization scheme of WP-o-Matic, and I’d really like to have WP-o-Matic use the WordPress cron so reliably and steadily as FeedWordPress does. And I’d really really like to have WP-o-Matic just get timestamps right, with the ease of FeedWordPress.

However much I adore the flexible and powerful customization interface (the ‘campaign’ setup) of WP-o-Matic, we have to go with the more stable candidate of the two, which is FeedWordPress, IMHO. Especially since we can’t get cron right now, and are reluctant to pay for it right now, if we can get something which works great at this level, without paying for it.

We’re going with FeedWordPress, for these reasons mainly :

  • It works well, even without setting up cronjobs (using WordPress’ built-in cron).
  • It deals well with timestamps. There’s no messing around with the chronology of posts.
  • It is the best documented plugin of the two, and it has an API which makes it easy for us to tweek it for our uses.
  • And we have greater trust in it’s developer Rad Geek/Charles Johnson to continue support and development for this plugin.

When using free software plugins, I find picking the ones you want to use comes down to what killer feature you really want and which developer you trust the most to deliver it and continue development and support.

flattr this!

Tags : , , , , , , , , , , ,  

Get WordPress MU To Stop Worrying And Love Embedded Stuff

Kaplak Stream is based on a WordPress MU install (currently v2.6.1), where a network of niche sites are fed one or more feeds on a particular subject in the ‘stream’ or from particular online services, using feed aggregation tools.

Building the setup for Kaplak Stream so far has revealed a path ridden with challenges (as one might expect). WordPress MU, which is a tremendously powerful package, is not as widely used as it’s popular little sister, and therefore is less well documented and supported, which goes too for the compatibility and effects of various plugins.

One initial thing which gave rise to some trouble, was to get WordPress MU to stop worrying and love embedded stuff such as YouTube videos and widgets. WordPress MU was designed for great environments hosting thousands of blogs, with thousands of different users, and has a higher security threshold than regular WP. And there’s no way to turn this filtering of tags off in the Admin interface.

Now, there’s a plugin called Unfiltered MU which will remove this filtering of posts and thus allow the embedding stuff. Unfortunately this plugin works only with posts actually published using the Admin interface editor. It doesn’t work with imported posts (from your old single-WordPress setup), and apparently it doesn’t work with aggregated posts either. So if you setup MU and want it to import an old blog or set it up to aggregate items from a feed, you still got trouble.

I found out one has to manually edit kses.php to enable the tags used by embedded stuff, at one’s own peril. For our purpose, however, we’re not concerned with security in the sense that we are the only users of our system, for the time being.

At your own peril (I underscore the fact that you may put your setup at risk enabling these HTML tags, but hey, life is dangerous) : Put in these tags and something along the lines of the below code into your “allowed” arrays in kses.php : object, embed, param, script.

'object' => array (
			'id' => array (),
			'classid' => array (),
			'data' => array (),
			'type' => array (),
			'width' => array (),
			'height' => array (),
			'allowfullscreen' => array ()),
'param' => array (
			'name' => array (),
			'value' => array ()),
'embed' => array (
			'id' => array (),
			'style' => array (),
			'src' => array (),
			'type' => array (),
			'height' => array (),
			'width' => array (),
			'quality' => array (),
			'name' => array (),
			'flashvars' => array (),
			'allowscriptaccess' => array (),
			'allowfullscreen' => array ()),
'script' => array (
			'type' => array ()),

Pick the ones which you need for your videos or other embedded media to work. Allowing the ones listed will allow video embeds from most providers, incl. YouTube, Google Video, Viddler, and others as well as widgets from a lot of sources. It works on posts aggregated by FeedWordpress for instance, which was my problem with the “Unfiltered MU” plugin.

flattr this!

Tags : , , , , , , , ,