Here’s an article titled I’ve left Twitter. It is unusable for anyone but trolls, robots and dictators. I haven’t left Twitter because I never really went there. I’ve had an account since 2008 (apparently) but I never really used it. There are some old Tweets there because I did hook it up to my own messaging system that I developed until the (proprietary) delivery service failed. But I’ve never logged into Twitter and actively used it as intended. Every once in a while I take a look to see if I’m missing anything, and I never get the feeling that I am. Here are some of the reasons I feel that way.
One of the promises of Twitter is that it can, in theory, provide very immediate information about what’s going on in very specific places. But I’ve never had that work out. For example, my balcony has a nice view of a large urban area and I once saw a massive fire raging below which produced smoke that blackened the sky over much of the city. I wondered if anyone on Twitter nearer to it could provide more information about what it was. Nope. No one I could find even mentioned it. Maybe it was just my Twitter ineptitude, but I was surprised and definitely hoping for better.
Ok, one should not expect highly local timely news. What about other "news" you can get? Problem one is that spurious news is a big topic today. Could Twitter be part of the fake news revolution? Seems so. I liked how this article entitled Facebook, Twitter and Google bosses to be grilled by Parliament over spread of fake news had the compulsory but nonetheless ironic "Follow us: Facebook/Twitter" at the bottom.
The "news" on Twitter is not local, timely, helpful, or even true. What’s else can be found? For me one of the most irritating things about Twitter is the juxtaposed spaghetti graph of bewildering "conversations" with "replies" and stuff. Imagine if you were reading this text and I took a moment to say—Oh and Bob, let’s definitely do that thing we talked about!--You’d be thinking WTF? But then if I just kept doing it—Your arguments are completely stupid, Mr. Squeaky—you’d really want—If you don’t believe me, Tom, check out Russian dashcam videos!--to stop reading. It’s very much like when someone interrupts an in-person conversation to rudely take a cell phone call. I’ve been known to just walk away from that situation.
Let’s assume you can filter that stuff out (though probably only for each conversation). I would be looking forward to a delightful stream of small messages micro-blogged for micro-attention spans. But that’s not what Twitter is, is it? The 140 character message concept is a fiction. Why do I say that? For a while now Twitter has been mailing me six tweets every day. At first I looked at them (in Mutt, my text email client) but soon realized that they would never be interesting. I went ahead and collected a few weeks of these and analyzed their content.
Twitter probably is taking special care to send me only tweets that are meaningless without a marked visit to Twitter (which I scrupulously avoid) but that’s not terribly encouraging. Here’s an idea of what typical tweets with links look like. (I’m including the mystery links for completeness, but I sure don’t recommend clicking on them!)
And where do the links really take you? Perhaps
Who really knows?
Although such links don’t help without further investigation, retweets, can be even more bewildering. This one, for example, doesn’t seem to include a specific link but I assume that it is implied by the very concept of retweet.
RT @SecureThisNow: @_lennart great talk about logging on steroids with full customization at @BsidesSD . #muchlearning #bsidesSD
Finally, here is the one bog standard tweet which had a canonical format.
The ten albums that defined my teenage years were probably all s3ms.
I have no idea what that means but at least it’s the right format.
The fact is that Twitter is just as much about referring people to other things as it is about publishing any content and it is not at all rare that this other material is more than 140 bytes. Sometimes it is weird prose stories, quite long, broken into awkward 140 character packets. Really, Twitter is a baroque URL shortener.
This seems to be a fundamental collapse of the internet. It seems that to a certain extent web search engines have been replaced by celebrity tweets for guiding people to information. Perhaps this is excusable, sort of like Digg or StumbleUpon but with more celebrity influence, but I think it’s really messing with some key principles that made the internet compelling in the first place. For example, it’s definitely shameful how many "140 character tweets" are nothing more than links to non-searchable images containing vast amounts of text. That’s just the worst of all worlds.
Of course if you love celebrity news, and I know that’s by definition really popular, Twitter is pretty great. I just noticed, for example, that Elon Musk does a fairly good job of making his own news so reading his tweets would be like being plugged into what would have been press releases in the old days. But most people aren’t all that newsworthy.
So what of it? Well, I have mixed feelings about it all. If people want to have a way to say short simple things and if those short simple things are essentially links to noteworthy resources, fine. But Twitter, Inc. kind of creeps me out. As I mentioned I had developed my own system, a Bash script, for generating and archiving my short messages and I liked it very much. I used a service called ping.fm to replicate these messages on Facebook and Twitter. Here’s an important fact, however—that company died before I did. Big deal, you might say, but the bigger important point is that if you’re believing there is no chance that Twitter itself can die, you’re not looking at a long enough time frame. Just look at the pitiful state of Yahoo to know that big companies do not equal a safe haven for your data. This is especially true when the true product is you. For example, here is cool hombre JWZ reminding us of this from the comments of a Perl script he wrote for archiving Twitter messages.
# Unfortunately, Twitter seems not to let you retrieve anything more than # about a year old. So for those old messages, you're just screwed. # That's what you get for trusting your data to someone else!
If you visit my main web site today, http://xed.ch, you’ll see that I’ve just added a new section of recent status messages. I also did some other changes like finally moving my whole blog system from "dev" to "live" while leaving a dev version to play with. Hopefully it all still works!
I haven’t hooked the status messages up to Twitter yet, but I may if the API key circus isn’t too onerous. But first, I want it to work where it really matters, on my web site. I rewrote the entire system; basically I edited my Makefile and my main page template, and created a one line sed program. It’s not really that hard! I’m going to see how it goes. I’m also going to think about best practices for posting links now that URL shorteners are clearly evil (including goo.gl and especially the non-optional t.co and facebook hijackers).
Update: Here is another good article questioning Twitter written on the same day as mine.
Update 2: My status system is now configured to publish to Twitter too using Python. The important result is that if you like Twitter, you can follow me at: http://twitter.com/chrisxed