This page will explain, in the simplest possible terms, why many use a period or other character before a @Twittername. While I once wrote a widely circulated explanation, it was rather dense, and I find that at least a third of my stream still doesn't understand the issue. Thus, I felt a simpler explainer might help.

What is the period (or other character) before the "@twittername" for?

In the simplest possible terms, it "breaks" Twitter's native (built-in) reply threading (a fancy techy term for connecting tweets together).

Without it, your Tweets beginning with @someName will NOT show up to your ALL of your followers, UNLESS they happen to follow both you and the person you are replying to. 

Will only a period work?

Nope. Almost any character will work just as well.

All that matters is that the ?@ character combo be the very first characters in your tweet text.  That said, the period, because I and a few others hammered Twitter streams with it for over 2 years, has become the de facto standard. I would not deviate from it because a) it will just confuse people, and b) there's no reason to. The period is so small, that while anyone can still see it, they visually just tune it out. It has no impact on Twitter readability.

Doesn't that waste a character in my text?

Yes, but LOLs waste 4 (including the space after), and we know you ain't giving them up, right Spanky?  Now STFU about one lousy character and keep reading.

Why would Twitter want to hide my replies from my followers?

The logic is that by only showing replies to those who follow you and the person you're  engaging, innocent bystanders who are not interested in your conversation will not have to see it.

Twitter did not always do this. All replies were seen by everyone no matter what.  While this forced change of Twitter behavior caused a huge furor at the time, as with many Twitter changes, the userbase had no choice but to eventually learn to accept and live with it.  In fact, even I have come to welcome the change. But then, I never did mind the logic. It was the very confusing way they chose to implement it (and spring it on us without much diplomacy) that irked me (and others).

Again, back in May of 2009, I wrote about Twitter's "Replies Issue."  Read it if you like the grim details. Otherwise, find something better to do—such as keep on reading.

But doesn't using this trick make it hard to follow a conversation?

Sometimes, but rarely. The downside is that it breaks Twitter's native connections, so users of Twitter.com (and some clients), cannot "thread together" the stream of messages between the two participants. This is often called "reading the conversation."

Why don't I (and possibly you) care?  Because the best "conversations" usually involve many more than two people anyway, and the threading never worked for that at all.  So I simply use "search" of all the names I care about. It's an extra step, but since most of my frequent engagements are all with people I follow anyway, we all see each other's updates. So it's only only when the discussion is with someone I don't follow, or when  I want to see what the whole herd is saying that I bother to use the search method. But I do it so often, it's not a bother. I usually have a browser search page open all the time, and I just change the names and press go.

Should you always use this technique?

Absolutely not. The rule of thumb should be "Is this something of interest to enough of my followers that I want them all to see it?"  If the answer is no, then simply reply normally without the period.

So why does @Shoq appear to use it so often?

Three reasons:

  1. Because the majority of my tweets are about political or social issues that I feel are of interest to all or most of my stream or they probably wouldn't be following me in the first place.
  2. I may be responding to some wingnut with a dozen followers and I want to amuse or inform my stream about the idiot.
  3. I use it so often for 1 and 2 above that it's become a habit I don't always break when I should. Me so bad. I hate me for it. You can too.

Doesn't Tweetdeck (and other clients*) give you the option to see all replies?

Yes, it does. But Tweetdeck is only one of dozens of popular clients*, and support for this special feature is very rare at this time. And Twitter.com (the client which most people still use), has no such support.

Why doesn't Twitter just build an option into their client?

Because neither software development nor software developers are very rational entities. They do things in their own way, in their own time frame, and for their own reasons. They really don't give a rodent's rectum what you think about it.

But not all developers are total asshats. Hell, if I ruled the world—or at least Twitter— I would implement a "Reply and Reply to All" feature. Just like email. Yeah, it's simple. That's why they missed it.

Where can I learn more?

You can't. All of human knowledge on this topic stops right here. You could Google for it and prove me wrong, but nothing good could come of that. This issue is confusing enough, and you have already been well-armed with all that really matters. Learn to be content with the easy answers. There are so many hard ones that we all need to worry about.

*Oh, and WTF is a "Client?"

Something I get asked about often, so I think I should finally explain it here, even if it has very little to do with Twitter replies.

I can do that, because, at least within the narrow confines of this document, I rule the world :)

A client is a term programmers and us techy types like to use to describe a program that communicates with another programmed service on a remote computer somewhere. Tweetdeck is a client for Twitter, Google Reader is a client for news feeds, etc..  Often, many clients will exist for the same service. (Twitter has hundreds, but only a dozen or so good and well supported ones).

How do I use a Bidet?

Now this is something you could have Googled for yourself. But as you have so often come to expect from me, I've saved you all that time and effort. You're welcome.

How to use a Bidet Properly (Video)

 

 

Note: This post was written in early 2011. But nothing has changed. Millions of Twitter users still don't understand that while it has some value, Twitter's built-in Retweet function has all but ruined a key benefit of Twitter for discovering and recommending information to our followers.

You need to understand why in order to understand why you miss a lot of stuff on Twitter, and why others miss your RT recommendations... routinely,

It will also explain why my followers so often see me using the old style (copy and paste) RT method instead.


 

Well, it's been well over a year since Twitter implemented its built-in "new" retweet feature, which for many of us, immediately wrecked what had been one of Twitter's best features: the old retweeting style which evolved organically from the user-base, and with no input from Twitter whatever.

Naturally, something that millions of people loved just had to be replaced by something confusing and destructive to the Twitter experience, just for the sake of their business model and investors.  So, replace it they did. And in so doing, they created a huge level of completely preventable confusion for both new and old users alike. Confusion that still wreaks havoc on your timeline, even though you're probably not even aware of it. I am just fed up with explaining this over and over again, and still kicking myself for not writing this post over a year ago.  My purpose here is NOT to rehash all the details of what Twitter did, or why. You can click here for that. My purpose is simply to tell you clearly, and without all the confusing details, just…

Why you should NOT (always) use Twitter's built-in retweet feature

There are three very big reasons:

Reason One: Your followers won't miss important stuff you want to recommend to them.

To help you understand why, consider this scenario involving you and two other followers:

@Newbie is some follower you just followed recently. He has about 250 followers himself, and you decided to follow him because he tweeted some funny cat pictures now and then.

@Oldbie is a follower you've liked and trusted since you first got on to Twitter.

One morning, before you woke up, @Newbie retweeted a tweet concerning a really big story in your field of interest. He used Twitter's built in RT feature. It was a story you would really want to know about.

Later that day, @Oldbie retweeted about that the same story, also using the built-in retweet. As you respect @Oldbie, you would have clicked to read his tweet immediately had you seen it. Unfortunately, you never saw @Oldbie's tweet because Twitter's built in RT logic doesn't show you items that:

  • have already been retweeted by any of your other followers. 
  • you follow the original tweeter, and thus, Twitter assumes you don't need to see a RT of it—FROM ANYONE.

If you do see a RT of someone you don't follow, you only see in one time. Twitter simply increases the tally of how many users retweeted that first tweet that you received earlier from that relatively unknown @newbie person. Of course, that tweet came in hours ago, and is now so far down your timeline, you were never likely to see it anyway.

But wait, there's more bad news: Suppose the Tweet that both @Newbie and @Oldbie are new-retweeting was posted by someone else that you follow. In that case, you won't see either of their retweets because Twitter logic assumes you saw the original tweet, even if it went by long ago while you were sleeping, nodding off, having sex. This is a massive fail when big stories are involved. Unless you saw the original when it appeared in your timeline, you will NEVER know about it no matter how many thousands of people RT it. And this is absolutely horrible. It squashes one of Twitter's greatest values: allowing people you trust to filter what you see.

So, now do you see the problem? Twitter decided that you only wanted to see a single instance of a tweet, no matter who sent it, or when, and without any regard for how long you have known a user, nor whether you even recognize, respect or trust them. Thus, because you like some of @newbie's cat pictures, or didn't notice the first time an original tweet showed up in your timeline, you completely missed all the secondary reminders from other friends concerning that story. (And this explains why we often don't see something all of our friends are talking about, too.)

Now you might say, "but I hate seeing dozens of retweets of the same thing." And you're right, sometimes that can be annoying. But keep these things in mind.

  1. Repetitive tweets tell you a story mattered to a lot of your followers. You might ignore the first few retweets you see, but when the 3rd, 4th or 5th come in, you're going to notice, and may well be glad that you did.
  2. Repetitive tweets tell you something is really important, or you wouldn't be seeing so many of them. Even on my mobile phone, all those retweets just help reinforce for me what people find important. And that counts. A lot.
  3. The repetition is not nearly as annoying as missing a story that could be important to you, nor missing out on all the comments that can sometimes be as important as the story itself (or at least funnier). And this leads nicely into the next reason you don't want to use built-in Retweets…

Reason 2:  Users can't add their own comments to the built-in retweets

This one you probably know something about already. But you should not underestimate its importance. Comments embedded in old-style Retweets are one of the very best ways to learn more about a story. More importantly, they help you to discover new users whom you would never know about had they not added a funny or informative comment to a manual, old-style retweet. 

For example, consider this retweet from @newbie, who is now using the old RT method so you WILL see it regardless of who else retweeted that same item already:

@newbie: This story increased my profits! RT @nerdish Study says hire only very good spellers. http://butt.ly/3das

Not only did you discover a new interest that you shared with @newbie, but at the same time, you discovered a new person named @nerdish. Without that all important old-retweet comment, it would probably have sailed right on by you if you're timeline was even modestly busy.

Reason Three:  New RT minimizes the impact of Twitter campaigns

Often, we want to let people and organizations know how we—and the crowd—feels about something. With Old style RT, they'll see thousands of tweets aimed at them. But with the new built-in RT, they may not even see a single one.  Let that sink in. If you want @piersmorgan to know how pissed off you are about something, he'll never see it in his mentions unless you use the old RT.  Be lazy and use the New RT, and it's like saying "ditto" on some lone tweet he may never even see. You scream, but no one hears.


Now, you may be saying to yourself…

"What good is all of this if those followers don't play by the old retweet rules as Shoq has laid them out here?" A good question:  If it all depends upon what THEY do, and has nothing to do with the choices that YOU make, then why bother? 

Three good reasons:

  1. It does affect you. The same rules applied to what you can see will have bearing on what others see from you when you use the new RT. Sure, if you're the very first person among your followers to retweet someone, your RT may get seen. But if not? Then it's just one more number added to the tally on the first Retweet showing up in the timeline of people following both of you.
  2. Because only if you understand the impact it has on you, can you ever explain it to them so that they don't do it. And of course, so you won't either. 
  3. As I demonstrated when I created the #MT and the shorter #FF (for #followfriday) tags, change happens on Twitter over time, as the crowd discovers a better mousetrap.  If YOU stop relying on new RT, others will too.

So now you know

Twitter has a lot of reasons to want you to use the new retweet; reasons that have everything to do with them making money by reporting tweet metrics to advertisers, but absolutely nothing to do with the value that you and your followers will get from Twitter.

For me the choice is simple

Old Retweets are the very essence of Twitter. Sometimes I will use a new RT when I'm in a big hurry and just want to be sure at least some of my followers will see a tweet. But this is usually when I really don't care enough about it to give it special treatment, or when the tweet is so densely or specifically worded that I really can't easily condense it down to something that I can fit a comment into. Alas, if the tweet was from earlier in the day, only those who don't follow the original tweeter are even going to see my recommendation.  Again, for me, this aspect of the issue is a massive fail. It impedes our ability to recommend things to our friends, or those who rely on us as a filter of news and information.

As for you, well…

You're gonna do what the hell you do no matter what I say anyway. And that's how it should be. But at least now you know the stakes. Twitter is a very powerful tool for communication and knowledge sharing. Sometimes it seems like the only people who don't quite understand that well enough are the fine folks at Twitter.

Related

Back story on the "New" Retweets (by Shoq)

About the MT signal (by Shoq)l

UPDATE1:  If this article is accurate, then Twitter is now claiming that their new search—which prompted the post below—was just an "experiment."  (Conducted live with 100 million users? Hmmm)

If true, then all of this drama was about nothing at all. As I said initially, for all we know, it's a bug. When Twitter doesn't share with us these little tests, nobody can know what's really going on. It's annoying that I had to write all of this about something that lasted only a few days.  But so many were hysterical, there wasn't much choice.  
Notice that they say they will look for other ways to remove duplicates.  I think they should simply provide various kinds of filtering options in advanced search, and leave the rest alone.

UPDATE2 (5/23/10): I have seen no changes to search. The new method is still in force, making me greatly doubt the story above claiming RTs would be returned to home/mentions search.

ORIGINAL POST

Yesterday, Ray Beckerman wrote a post titled: Has @Twitter Declared War on Traditional Retweets? #TR

In my opinion, the change he was concerned about is actually a very good change.  But as usual, twitter has not explained it before implementing it.  So Ray's post has brought needed attention to it, but his rather irritated tone with them springing it on us this way has lead many of his readers to take a very hostile stance toward it before they actually understood it.  Many panicked, and have run all over Twitter announcing that the sky may be falling.

Yes, it is a change, but it's hardly a huge change and the sky is not falling. It just requires a bit of knowledge about the new options that you now have when searching for tweets and #tags. So let me explain it before it gets even more out of hand. (I've been getting one question about this every 20 minutes).

The problem that isn't a problem

Twitter has decided—in their typically unannounced way—to make a  change to the way search works, but only IN YOUR HOME (and @mentions) page search boxes. The general search box (at search.twitter.com) is not affected by the change.

As Twitter UI staffer, @charles tweets here, all they have actually done is change the search parameter to be "exclude:Retweets." 

This is actually a GOOD THING. Previously, searches always included every single retweet, and the result was often that the ORIGINAL tweet would be buried beneath many pages of search results. This is known in the tech world as search "noise."  The original tweet is the "signal;" the thing most people are interested in finding FIRST.

So what does this practically mean for your searches?

That you have options in how you search for stuff that you never had before. This change is giving you something, not taking something away.

All it means is that IF you want to use the little home page/@mentions box, you have 3 choices in how you can search:

You can search for:

1)  #tag or Keywords:  example: "#shoqstag"

This will get you only original tweets without duplicates (which is any tweet containing the "RT" code anywhere in it).

Why? Because this eliminates what can sometimes be dozens—or hundreds—of duplicate tweets. That include tweets where people may have appended comments to their RT (using old style RT), but those are derived from the original tweet, and generally much less important for most types of searches.

If you REALLY want to see every single RT-bearing duplicate, and all those possible comments (most rarely do), you simply use #2 or #3 below.

2) tag or keyword PLUS "RT" — Example: "#shoqtag RT"

Do this at the home/mentions box, and you will get everything with your terms plus a RT in it—but NOT the original tweet. This is a small annoyance at first, but actually makes sense when you get used to it.

Just remember that adding RT gives you the retweets, and omitting it doesn't. And if this really still annoys you, for some reason, you can always…

3) tag or keywords entered on the standard  twitter search page — where everything works as it always did.

What about a #tag search/click?

This is the only confusing part of the new methods, and it's not a big deal either.  A tag search is when you click a linked #hashtag.  As it works now, which search method is used depends on where you were when you clicked a #hashtag.

Tags in your a) Home Timeline, b) @mentions, or c) viewed tweets* — will use the HOME/mentions style method without the RTs included.

* Viewed tweets refers to when you isolate a tweet by clicking on its "view tweet" link or its time stamp. Example,

Tags in a regular search.twitter.com. search results page — will include everything.

What About 3rd Party Apps?

Tweetdeck treats #tags as it always had, and probably most clients will do the same. As for Tweetie2 (Twitter's own mobile client), who knows? They are barring old style RTs, reportedly, and that will probably crush that product fast.

What about the "TR" suggestion

This was Ray's idea to work around what he perceived as a bad problem. The idea is that if you really feel a need to indicate that you're RTing someone, and ensure that your tweet still shows up as an original tweet when someone clicks a tag, you'd be sort of saying it's a Retweet without actually using the RT code (which would be excluded with the new method).

I just see very little point to this fake RT/TR idea, and in fact, it will just add yet more code bloat that nobody understands, and really look like you're trying to game a system designed to serve most people with a generalized solution. If everyone started doing this, you'd have even more confusion, as some duplicates would show, and others would not. This would NOT be a good outcome, in my opinion, and I would urge you not to do it.

That's all folks

I am glad Ray brought this up. I've provided all his references herein if you'd like to read more on his view. For a few days, I'd be happy to discuss it with Ray and anyone else interested at tag: #searchchange.

Please use the green button to retweet this to all the people who were confused enough by Twitter weirdnesses before this change.

As always, please follow @shoq for updates on this issue.

Related Info

@RayBeckerman's original post.— which created all this concern, and referred to it in fairly dramatic terms. I understand that. These changes can be annoying when Twitter doesn't announce them. But I do wish he would edit that post and soften the language so people see this as something to understand and not fear.

Twitter's @Charles responded to Ray's concern here, and here. In both tweets, he confirms what I've explained above.

NextWeb Reblogs Ray's Post — This was the more Googled post, getting too much play, but it was simply a reblogging of Ray's post.