Subscribe to RSS
Posts Tagged ‘twitter’
Another Twitter cynic
Serial CEO Peter D. Csathy says:
Twitter’s revenues will pale compared to the ever-growing expenses needed to scale the service. And, my prediction is that the Twitter nation will lose steam in its passion and the bloom will be off the rose.
So, Twitter — here’s a tweet in less than 140 characters — “SELL NOW! Follow YouTube’s lead!”
It’s interesting when I see people talking about how they wish they had thought of Twitter first. As I see it, that misses a critical point. First, you have to come up with the $55 million it has taken to create and operate the service (so far). Do you have access to that kind of money? If not, then the idea might not be enough.
It’s a myth that if just any schmo came up with a Twitter-esque idea that VC money will come rolling in. There are plenty of other ideas that could have lost $55 million in the same amount of time on the cutting room floor (that never receiving funding and therefore went nowhere). @Ev was not just any schmo – he was already rich and had already made a bunch of VC’s a lot of money before he started Twitter. He may not know much about building scalable infrastructure – guess what it doesn’t matter – he knows how to raise money. And that’s what matters.
Likewise, it’s a myth that the users (insane growth) come first “without spending money” and that the money comes after. This happens one in a zillion times but even in most of those cases, at least SOME money came in early. Many startups that have created the mythology that they had “amazing growth without any money” are BS. They raised a lot more money than people think before they had significant growth (Skype, Facebook, and Twitter all come to mind).
Twitter is cool (but also has issues) but it’s taken $55 million (and 45 people) to get here and they still don’t have any revenues or a published plan to get revenue, so the costs and requirement for more and more other-peoples’ money goes on. The founders themselves recently said “We’re one percent into Twitter.” Does that mean they will need $5.5 billion? 
They also said “if we described Twitter in three sentences, the first two would be about not putting too much fidelity on it, and the last sentence would be ‘we don’t know.’“
So before you start having Twitter-envy, you need to ask yourself, “do I have access to $50 million (or $5.5 billion)?” If not, then don’t worry about it – you were never going to build a high operating cost company like Twitter or Youtube anyway, whether you had the idea first or not.
BEWARE: twittercut is a password stealing SCAM!
A quick search on twitter will show this site is bogus: http://search.twitter.com/search?q=twittercut
DO NOT sign up there (or even visit the site). If you did sign up, change your Twitter password asap.
Here’s what the home page to the site says:
Welcome to TwitterCut.com.
TwitterCut.com is the best place for you to grow your twitter network and gain a ton of followers. We recommend giving it a shot, it’s free and will help you get the followers you need. This system is brand new, so the quicker you get involved the better it will be, fill out the form below and get started right away…
It looks like they stole some Twitter graphics there. I note that the Home, About etc. links are all no-ops.
The WHOIS record, gives the following:
Registrant:
Jordan EMbry
1646 thompson drive
owensboro, Kentucky 42302
United States
Domain Name: TWITTERCUT.COM
Created on: 21-May-09
Expires on: 21-May-10
Last Updated on: 21-May-09
Administrative Contact:
EMbry, Jordan jembry13@gmail.com
I have not found a white pages listing for an Embry at the above address. The domain is registered with GODADDY.COM, INC.
May or may not be the same person as:
- Jordan Embry – Bowling Green, KY | Facebook
- Jordan Embry – LinkedIn
- Jordan Embry (JordanEmbry) on Twitter (account suspended)
- Another Jordan Embry (ijordan89) on Twitter (also in KY)
Twitter #fixreplies and “intelligent” networks
A few weeks ago, I said “Twitter is a “Stupid Network” (the good kind)”. It turns out I was wrong.
Twitter created the #fixreplies problem (see here and here) for themselves when they decided to treat these tweets specially from the start, breaking the end-to-end principle.
These are tweets by people I follow – they are public tweets, not direct messages. They are in the public timeline. If I follow that person, they should be in my timeline like any other tweet from that person (whose tweets I follow), regardless of the content of that message.
I follow this person. That means I follow what they say, the updates they post. If Twitter never treated a post that starts with @ special in the first place, we wouldn’t be having this discussion. In that case, if I wanted to filter such tweets, I could do that in my client, at the “edge”.
Now Twitter has pulled those tweets out and so I can’t “filter” them – I have to go looking for them.
This is the kind of mess you get yourself in when you go around breaking the end-to-end principle all will-nilly.
#fixreplies hack – shows what Tweets you’re missing
UPDATE May 18 2009: fixed some bugs, removed Twitter API rate limit issues, and simplified usage to one simple HTML file
I put together a little hack that shows the tweets you’re no longer getting.
It’s pretty hacky and requires some manual labor to use, but some people may find it entertaining. I decided to do it all with Javascript so it requires nothing more than your own browser. Also, that way I can provide all the code and let people see exactly how it works and perform whatever beautification they may want, etc.
Basically, it produces output like the following, showing the @reply tweets you didn’t see in your Twitter stream becaue you don’t follow the person the message is directed to. In my case, it showed me 63 interesting tweets tonight. The output is something like this:
Each entry is shown with a link to the original Tweet on Twitter so you can click to get the full details of the tweet.
How to use it
It is all in one HTML file: whatumissed.html.
You can copy and paste the above into a local file on your computer and open in your browser, or simply execute the version at mrblog.org directly (and “View Source” to see how it works):
http://mrblog.org/whatumissed.html
- It should be pretty easy to turn off the diagnostic data if you like by using
visibility: hiddenfor the “statusmsg” div in the html/script. - There is no little CSS or other styling here – if you hack the code with cool CSS styling, please post back to me so others can benefit too.
- If you make other hacks to this code, or have other suggestions for me, please let me know.
Twitter #fixreplies fiasco is telling
I’ve seen many people try to explain (or argue with someone) about why Twitter matters, what they use it for, why it’s cool. These conversations are often interesting in how they demonstrate that even people that love Twitter often can’t explain it to anybody. Sometimes I think these people aren’t even sure themselves what they get out of Twitter, or why they find it compelling.
And I think the #fixreplies fiasco shows that even the Twitter brain trust may not understand the value of their own service very well. We certainly know they don’t seem to have any idea how to make money with it.
For those that haven’t caught wind of this yet, recently Twitter stopped displaying public tweets that are deemed to be “replies” between your friends and someone who you do not follow. In the past, such tweets would be seen as sort of one side of a conversation. If you follow @Fred for instance, you might have seen something like the following in your Twitter stream:
Fred: @Mary I liked your idea about http:…
In this case, let’s say you don’t know who @Mary is. Since you like Fred, you might want to know who this Mary character is and what she’s about. A tweet like the above was a perfect opportunity to be exposed to new people that your friends engaged with. In other words, it was a great way to discover new potential friends. it was also a great way to see what your friends are talking about – what topics were so interesting to them that they were compelled to respond. So they helped us stay more in touch with our friends, both in terms of who else they talk to and what they like talking about.
In addition to the tweet itself, it might also have said something like “in reply to Mary” which is a link to the original tweet by Mary that Bob is responding to.
By removing these tweets from my stream, Twitter has instantly made itself less valuable as a social tool. On Facebook, I can see when a friend comments on one of their friends’ items, even if that second person is not my friend. Friends of Friends is a major component of any social network and part of what separates a “social” site from just another web site.
Twitter argued that they had to do this for technical reasons. That makes no sense to me. What’s really special about this tweet by my friend Fred, as far as Twitter is concerned? Why are they treating it specially? In terms of the relationship to my stream, as a follower of Fred, how is it different than other tweets by Fred? Of course it’s not. So much so, that people have talked about simply inventing a new format for replies and not using @Mary anymore so that the message will appear in the stream. UPDATE: Twitter claims the “technical problem” is having to check every account’s preference every time a reply was posted was causing a huge strain on their database – so they removed the option and gave everyone the “most common” choice.
It also represents another step toward violating the end-to-end prinicipal , something that has gone toward greatly increasing the value of the Twitter network and separating Twitter from others. As I have said, dropping the end-to-end advantage is a big deal:
This end-to-end model permits myriad uses of the underlying network and fosters incredible innovation.
…
At first look, Twitter becoming more “intelligent” about the contents of tweets may appear to be a good thing. But it is a slipery slope. Once they start adding meaning to the 140 characters, the network becomes more restrictive. The more Twitter remains a “stupid network” the better, IMHO.
I think that fact that even Twitter themselves don’t understand what makes their service valuable tells us something. I hope they figure it out and don’t destoy the special thing they have created.
Update: an interesting counter-perspective: my thoughts on #fixreplies and the ‘dot fix by @HelloKit.
Also, see the whatumissed hack for a way to see these missed replies again: http://mrblog.org/whatumissed.html
Sheep throwing on Twitter, redux
UPDATE: Edited to reflect the fact that Luca is not the developer of buytter.com
UPDATE 4/26/2009: Since posting this, it’s interesting to note that “how to block buytter” is among the top incoming Google searches sending users to Mr Blog – so apparently I’m not the only one considering these questions.
Yesterday, I spoke briefly about Tweefight, an App/Game that effects tweets that look a lot like the kind we often see on Facebook:
I just won a fight against @mrblog on Tweefight. I won! Do you want to fight? Try now …
Of course Tweefight was not the end, but the beginning. Today we find buytter.com where you can “purchase” your Twitter friends – sound familiar?
filos bought @mrblog for $ 200 http://buytter.com/filos
This led to a discussion about the implications of such applications. As I’ve noted before, Facebook, apps operate in the Facebook “walled garden” and, as a result, Facebook can offer users the option to “Block this application” (to opt-out) to avoid messages of the type above from appearing in their stream (what Facebook calls the “news feed”).
In the case of Twitter, because it is a mostly a “Stupid Network” (the good kind) Twitter doesn’t know which messages are bots or app-induced and which are genuine hand-crafted works of human imagination, so there is no “Block this Application” option.
I suggested that we should define a convention whereby app-generated messages like those above should be flagged with a hash tag (such as #fromapp or similar). Luca (developer of Tweefight) argues that Twitter should automatically handle such “app messages”:
“if a service opens the web twitter interface pre-filling it, twitter should add a tag”
As I said the other day, this might be a slippery slope. (To be fair, Luca is really proposing a Twitter UI (end applciation) change, not really a “network” change, so perhaps it’s not that risky.)
Anyway, these apps raise some serious questions for Twitter. Let me explain why these app messages received so much attention from me.
At one time, I didn’t expect to do much with Twitter. Now it has evolved into one of my high-attention-demanding data streams. Let me explain. What I mean is sort of a combination of “attention worthy” and “pain” (or scarcity). By that metric, Twitter is also a “high-cost” network. Email is lower cost – it doesn’t demand as much attention in terms of being an interruption – email can be done time-shifted. At the other end of the high-attention-demanding spectrum is a ringing telephone. Not only does it have to be dealt with in real-time, immediately, but it also is an interruption – I have to stop what I was doing to deal with it. This means an incoming phone call better be worth my time – my attention is the scare resource here, not the dollar-value cost of the call.
At first, Twitter was pretty low on this scale – and I didn’t expect it to move up. I was wrong (not the first time, and won’t be the last). Over time, I’ve tweaked and tuned my Twitter setup, trimmed who I follow, device settings, etc. such that Goldilocks would be proud – it’s “just right”. I even set things up so Twitter hooks into one of my more precious (high-attention-demanding) channels: SMS. I receive direct messages on my mobile. I also have scripts in place so I receive @mrblog mentions via SMS too. This has been “just right” – that is until yesterday, when I started receiving Facebook-App-like drivel messages like those above on my cell phone. It’s not fun being interrupted by an annoying SMS tone because someone threw a shoe at you. Suddenly my “just right” Twitter data flow became “too hot”.
So just as Twitter was getting useful, and I could trust it with high-attention-demanding respect, blam! Now I have to rethink all that. This is why the Tweefight and buytter apps are important. Such apps have the potential to change Twitter in very fundamental ways. For me, this happened within a few SMS messages, but for others it will happen too, eventually, one way or another – and they may not even realize it – they will just never use Twitter in a more “attention-demanding” mode, thus changing the value, if not the very nature, of Twitter.
Today, Twitter holds something over on Facebook. Twitter has access to my SMS channel. There’s no way I’d let Facebook send me SMS text messages – they’ve already proven they’re not worthy (changing the value of the Facebook news feed). So Twitter’s stream today is higher value than Facebook’s. But that can change, and apps like Tweefight and buytter provide an example of a possible shift in direction of the Twitter user profile. For some people, this maybe increases their value of the Twitter info stream – I’m not one of those people.
Where Twitter goes from here is really up to all of us.
Tweefight – the next way to throw shoes
It’s gone way beyond snowballs and shoes. Not only are there at least three separate “shoe throwing” applications on Facebook, now you can throw just about anything at your friends, from watermelons to sheep.
That’s fine. It’s not my cup of tea, but I know a lot of people enjoy such things. But, thankfully, Facebook lets me block these applications to prevent flooding my Facebook newsfeed stream with messages from these apps, like: “Bob threw a show at you. Throw one back!”
Yesterday, I got the following in my Twitter stream:
I just won a fight against @mrblog on Tweefight. I won! Do you want to fight? Try now http://tweefight.com/?opp=filos
It turns out Tweefight is a new service/game launched by buddy Luca Filigheddu. Clearly, a lot of people love this kind of thing. As noted above, I’m not one of those people.
When someone fights you on Tweefight, a “tweet” link is generated that posts a tweet of the type shown above. I have scripts in place that send all @mrblog references to my cell phone. The last thing I want to see in that stream is a “so and so threw a shoe at you” message. Unfortunately, Twitter doesn’t have a mechanism to block such tweets.
So I asked Luca if he could add an “opt-out” option to the Tweefight site, so if someone engages me in a “battle” on there, instead of getting a “tweet” link, they will see a notice that I’m not interested in receiving such messages. He agreed to implement an opt-out, which I think sets an excellent precedent for all future such “entertainment” applications. Since twitter doesn’t have the equivalent of “block this application”, such sites are going to have to police themselves.
Apparently the site is getting a lot of usage (so much so it was off-line much of the day yesterday) so it’s nice that Luca is so receptive to suggestions. It looks like Tweefight will do just fine without me – which is terrific – to each his own.
Twitter is a “Stupid Network” (the good kind)
I’m sure I’m not the first to point out this analogy, but one of the things that makes Twitter interesting is how it parallels the “stupid network” model, as prescribed by David Isenberg in The Rise of the Stupid Network.
Twitter embraces the end-to-end principle which is a central design principle of the Internet itself. A “stupid network” is one that merely passes packets independent of the what application uses the data in those packets. In other words, the network has little knowledge about the contents of the packets it handles and exerts little influence over them.
Twitter passes 140 character “packets” (tweets), acting as a “stupid network” for the most part. Twitter does not interpret the contents of a tweet – it passes whatever data the tweet contains. Some applications “overload” tweets with special content recognized by that application. This end-to-end model permits myriad uses of the underlying network and fosters incredible innovation.
Over time, Twitter has become slightly more “inteligent” interpreting “hashtags”, url-shortened links etc., but it still permits all kinds of application-specific data.
At first look, Twitter becoming more “intelligent” about the contents of tweets may appear to be a good thing. But it is a slipery slope. Once they start adding meaning to the 140 characters, the network becomes more restrictive. The more Twitter remains a “stupid network” the better, IMHO. By deciding to be “stupid” and supporting the end-to-end prinicple, Twitter is inviting creative minds to use their “network” for new applications, which, it turns out, is a very smart thing to do.
Twitter in the news
With Twitter Inc’s recent raise of another $35 million, bringing their total to $55 million, with an estimated valuation of $200 million to $250 million the company has been making the rounds on the wire. This AP story today, Can all that Twitters turn to gold amid the gloom? says that “Twitter intends to start testing ways to make money this spring.”
[CEO Ev Williams] hinted the company is exploring ways to charge for expanded commercial access to Twitter, but emphasized that all personal accounts will remain free.
Well, I guess that’s as long as Twitter is still in business, that is. At some point the VC money will dry up and one of these revenue strategies is going to have to pay off.
[Twitter usage] was seen in just 5 percent of respondents between 45 and 54.
This matters for Twitter’s financial future because most younger people don’t make a lot of money, which could make it more difficult for the company to appeal to advertisers.
The aritcle suggests that Twitter may be challenged to start charging for things users have been getting for free, as has happened with other sites like Yahoo!, Facebook and MySpace:
Its prospects are clouded even further by the resistance that Facebook and MySpace have faced as they have tried to inject ads into forums where people primarily goof off or fraternize.
Twitter’s revenue plans remain secret for the moment, but they are under a lot of pressure to fish or cut bait in terms of revenue generation.
Introducing taglets.org
A little over a month ago, my good friend and colleague, Mark Petrovic, had an idea for a project/service. That idea evolved into taglets.org, which we introduced today.
At first, to be honest, it was not obvious to me how it would be used or by whom. But Mark is smart and I trust him. So that was enough for me. It was also a perfect opportunity to get some hands-on experience with the Google App Engine platform.
Now that the idea is actually operational and I’ve had some time to play with it, I’m very intrigued by the potential for Taglets. It’s got some of the traits of the biggest ideas, being simple, general, and flexible. As Mark says:
The service is first a platform, with a reference implementation layered over it via the www.taglets.org web site, with the platform defined by a public API that developers can use to create and manage users, tags, and comments.
While it may look at first glance that this is a twitter clone of some kind, the fact is, Taglets is not intended to replace anything else. It’s fundamentally different. For marketing people, that spells disaster; but in this case, that’s fine, because many of the best ideas never had a business plan, target market, or clear demand. If you ask “what problem does this solve” the answer will be, “you tell us.” It’s open and flexible, and like many things before it, we have no idea how Taglets.org might get used, ultimately.
Here are some examples of how Taglets.org is different:
- The followers of a tag are anonymous. A tag is the target (to-address) of a message (comment) but unlike Twitter et al, that tag is not necessarily associated with a particular recipient/person/entity. Think of this like the street number on a house. Anybody outside that house can learn that location’s address (tag) “123 cherry lane” by simply observing it and anyone can send a message to that “tag” by postal service. But that doesn’t mean they actually know who lives there or, in other words, who they are sending to. One can learn the address of the house, it’s “tag”, simply by driving by. They can’t learn who lives in the house by name, yet they can still send a message to that “tag” – the people inside the house determine who actually receives that message, without the sender being aware of how their message was ultimately distributed.
- The sender of a message (comment) is anonymous. The sender may elect to include a “return address” or sender identification of some kind within the content of the message, but there is no “from-address” associated with a comment.
- Comments are transitory. There is no history (timeline). Comments posted to a tag are distributed to the followers of that tag (via email, Twitter DM, or HTTP POST) and discarded by Taglets. Tags can be searched. Comments cannot be. Obviously, if you receive Taglets.org notifications via Twitter DM, they are archived at Twitter, but they are not archived at Taglets.org itself.
- Tags are public. Anyone can search a tag, by name or description (keywords). Anyone can create a tag. Anyone can follow any tag. No follower knows who else follows the same tag.
- Tag names and descriptions are self-describing and have no formal meaning to Taglets.org. The creator of a tag describes what that tag refers to with a description field, but this is opaque to the taglets.org service itself (simple free-form text). Users will decide on the conventions they want to use in tag names and tag descriptions, similar to the way that twitter status messages can be overloaded with links, photos, hashtags, and @user constructs.
Taglets.org allows a person to send a message to “something” even if that “something” is not yet tagged and has no recipients. They can simply create a tag for it and send a comment, even if it goes to nobody.
A “tag” can be something as finite as, say, a device’s MAC ethernet address, or as abstract as an idea. It can be a person, a category, a blog name, a company, or a product.
When you follow a tag, you can have updates delivered via email or Twitter account (Security note: Taglets.org never requires or stores your Twitter password). You can can also tell Taglets.org to POST to a specific web page when a comment arrives for the tag. This might be used for a Wordpress hook, for instance, or to trigger some other action via PHP script, or similar.
You can learn more at the Overview page on the Taglets.org site. There’s also a Google Group at: http://groups.google.com/group/taglets
