Chat with me.

Tuesday, August 12, 2008

Twitter won't bill for followers. At least for now.

   GigaOM posted today that Twitter is may be going "fremium" asking for a payment if your followers count go over 2000. However Evan Williams (one of Twitter's co-founders) replied that this is not true.

ev-status

In various interviews Twitter co-founders shared that they explore the possibilities of taking the money for twitting from corporate users, however I am sure that it would aply more to some specific brands and some extra services.

Sunday, August 10, 2008

Twitts of people who twitt a lot are less valuable?

3 days ago I started using our alfa-version of twitter client. I personally prefer to use the web-based version, as I don't really need to be alerted about new messages instantly. It doesn't have a skin (design) on yet and lacks some features but the alerts and groups work and show how many new twitts each person has.

updates
I have updated the page about the alfa-version. If you are already following me, d-msg me and I'll give you a link to try it.

I now follow 158 people on Twitter and some time ago I started to notice that I use Twitter less and less frequently 'cause I get tons of Twitts that are really worthless to me. But after I've set alerts for the 14 people which twitts I don't want to miss twitter got it's permanent place in one of my tabs.

What I also noticed is that people that twitt less usually twitt about some important things in their life or about the world around them and these twitts are really interesting. Of course there are some exeptions. So before pushing that send button next time take a second and think if your Twitt is interesting for the people that follow you, newsworthy or funny.

That is my personal notice, I would love to hear your thoughts about that.

Wednesday, July 23, 2008

Links to Twitter services, that I find useful.

This is a very short list :)
It's just the tools I use personally.

  • Twitter Karma - let's you check your followers/friends and mass follow/un-follow/block them.
  • Quotably - shows threaded conversations, so you don't miss any replies.
  • Hurl.ws - my service of choice to shorten URLs, it's hosted by google and lets you easily track clicks by adding /hits/ at the end of your url.

Friday, July 18, 2008

Twitter gives it's full feed to Gnip.

In a techcrunch post Michael Arrington breaks the news that Gnip now gets the full Twitter PubSub Feed and all developers can access it via Gnip's API.

Twitter's top managers stated earlier that they are worried about if it's safe to give away the full feed and in the Twitter Development Talk group they Alex mentioned that if the XMPP feed would be given to anybody it would require an additional contract that would prohibit resyndicatíon of this content. Currently only a few sites were given access to to this feed including FriendFeed that is actually building a competitor to Twitter on top of this feed.

I really think it's a great move as Twitter actually keeps the control over the feed as it can still cut any site from using it and it can cut Gnip in case it wouldn't be able to block the sites that abuse the feed usage.

For all the developers it's simply great, because that would allow us to provide updates to users in real time handling all the communication with the end-user from own servers without pushing the API calls to their limits. The Twitter API would serve to get the additional protected information.

I still didn't see a company that did the work with 3d party developers right. Google's i-home is a tutorial on how the work with developers shouldn't be done, Facebook lets crappy apps win the popularity race without supporting the apps that create user value and Twitter is navigating the waters that has high storm risk. And I really think Twitter should help some small apps and bots get some popularity advertising them inside Twitter. I know some absoultelly great Twitter aps that will never get very popular, but are really valuable and need some help in promotion.

Tuesday, July 15, 2008

Twitter acquires Summize and extends the official API.

N8807858_4739_normalstevekinney: Twitter – who makes no money – just paid for a service that made no money (Summize). Maybe everything I thought is wrong.

   Great news for all Twitter developers: Twitter will extend the possibilities of the current API with search functionality from summize. Summize.com now redirects to search.twitter.com and as I understand we will soon see a change in the Twitter UI also.

If you need detailed info Alex Payne answers some questions about the acquisition in this post

Why is this important?
   Some applications were already using Summize API, but it's really tricky to count on a 3d party API as it can shut doors or change rules any time it wants. As it is now the part of Twitter developers can count on it and use it, so expect a lot of interesting search and tracking apps to appear in the future.

Threaded @conversations?
Summize can currently locate and display dialogs. The real question that I think is important for most of the Twitter community is if summize will offer this feature to 3d party developers via API. Normal conversation tracking is the #1 feature that Twitter really needs. After Twitter has shut the "PubSub" for most of developers it is impossible to make something like quotably.com without a special "yes" for Twitter, so the search API is the only hope to get this done.

Michael Arrington's interview with Evan Williams:

you can read the full post at techcrunch.

Thursday, July 10, 2008

Twitter should really force it's development rates.

The news about a man getting out from jail helped twitter to get to it's peak of 0.14% reach on the 3d of June. However it has fallen back to 0.06 right now and that's even lower then it was before the news. The fact is that Twitter can't keep the acquired mainstream traffic because it lacks some features that are vital for such a service. Even a simple HTML page with tips whom to follow could make a difference. There are tons of interesting bots and bloggers on twitter but it's really hard to find them.

As for other competitors - they won't see any light anyway.

You can see the Alexa's traffic graph here.

Tuesday, July 8, 2008

Requests to Twitter API are getting back to 70 per hour.

As Alex Payne stated here:

We're attempting to get back to 70 requests per hour this week.  We've just now raised the limit to 40 per hour and things are holding steady.  If all continues to go well, we should be back to 70 today or tomorrow.  We'll try to keep it there barring special events that are announced ahead of time.

Wednesday, June 11, 2008

New site listing lots of Twitter tools.

Everthing Twitter is a new site that lists lots of interesting tools for twitter. It's a really nice collection with additional descriptions.

You can also find a nice small collection at TwitterHolics

Tuesday, June 10, 2008

Why will all Twitter-Clones fail?

As twitter started to gain traction lots of services popped up that were cloning it. The latest and one of the most successful was plurk but all of them don't really have a chance until they have the start traffic that would beat twitter in user-base amounts right upstart.

Why?

Twitter doesn't have a viral loop strong enough to grow by itself.  It wasn't designed the way to fully exploit this. The dynamics of jaiku and pownce clearly show that too. However twitter grows... but why? The answer is really simple. Media loves twitter getting news out from it. If media loves you, they will write about you even if you are not newsworthy, but they definitely won't write about a smaller service as it won't be important to them. And as I already wrote twitters viral loop isn't strong enough to grow the service all by itself without the help of the media. So all microblogging services that appear that would attract less users then twitter from start are doomed as they won't get any press coverage.

Competition.

Well google actually thought it's a great idea to buy jaiku that's in fact is just a twitter clone and to be "clean" and launch a service they have stolen bought. However it can take years till they finally go out from their closed beta as it always happen with google's acquisitions, but it could also be earlier leaving twitter with less time to prepare for real competition. If google launches it inside it's gmail using the chat status messages twitter would be dead.

Stronger viral loop.

Earlier in this post i wrote that twitter doesn't have a viral loop that is strong enough to grow the service by itself. It's actually half true... If microblogging ever reaches the point where 20% of Internet users actually use it actively or passively a madness will start. When you have one friend using some service you won't join, if 5 of your friends tell you about it - you would.

Friday, June 6, 2008

KillWithMe.com ? Untraceable - the movie. Viral distribution.

   Just watched the movie "Untraceable". Well a bit late for a movie review of course, but I want to post about would it possible to repeat something similar? Let's leave the killing thing aside, my post is about marketing and not about committing crimes :-)

So let's try to repeat the creation of the bot-net and make the site intended to kill people go viral.

The guy that has built the site in the film is a true genius. You can be good in marketing or tech or at killing people, but being perfect at everything is a rear composition. You need a A-class team to repeat what he did.

1. Tech-possibilities.

   The movie is actually well done in tech-terms. It seems that the authors consulted some professionals and know how Internet works in general.
   Hosting a site on a bot-net is possible, however no one in the world now controls a bot-net of this size. The most successful hackers get around 100.000 installations a day, but that are already teams of at least 5 people. Creating installs cost money. Yes, hacking is not only illegal but also requires money. To create a bot-net of this size the guy would have needed around $1,5M in raw tech costs, 6 Month of time and a great team.

   Taking the site down. Well it's a movie :-) The register of the domain would have blocked it instantly. But let's assume it was some .island domain and FBI couldn't reach the register. Cable providers would block the site really fast and there are also the updates of browsers that could have been used for that. The possibilities of how to take a site down are endless, no site can survive if anybody with money or influence wants it down.

2. Content and viral promotion.

   Some people tend to think that it is enough to have great content to make your site go viral. However it's not. A site that kills victims in real time has good viral potential, but there are tons of sites that list similar recorded media. Each viral promotional loop has it's borders. Starting with one person it usually stops somewhere hitting the users that don't tend to share everything with their friends. Imagine that viral growths is a tree - some branches have lots of branches and leaves, but the leaves are the dead-end where the viral promotion stops. So you have to seed a lot of trees to get the scale.

3. Starting viral promotion.

   How could this guy seed some trees from his cellar? On the second video he has hit the 6.000.000 visitors within an hour. As we know he had a huge bot-net and he had access to user's social network accounts. He could use them to post the link to the site. However passing on usually requires entering the captcha images and it also costs money. Theoretically having HUGE technical possibilities and money it is possible to make it happen, but I haven't heard of anybody capable of reaching that amounts of users in such a short time.

Could you repeat it?

I know there are some really smart people reading this post. Is any of you capable to reach 6.000.000 visitors in 2 hours of time? Due to my calculations making this happen would require something like $2M investments and around 8 month of preparation and I am still not sure if I could repeat it. Well... may be I am just not smart enough.

Sunday, June 1, 2008

The Gillmor Gang - FriendFeed interview.

You can listen to the interview here.
Still can't understand the point why is there so much noise about FriendFeed, taking in considiration their lack of traffic. They did get good funding, but right now I don't see FriendFeed as a tool that can be interesting mainstream users. Right now it's pushing the ammount of noise up without the possibility to filter it anyhow.
The guys from gilmor gang were actually pushing FriendFeed to implement the same features Twitter has right now, while FriendFeed was telling "we are not about conversations". I don't really see a future for a site like FriendFeed without being a communication tool. It can get some fans, but unless you get the scale which is not possible without being a comunication tool it just won't get enough people using it on a daily basis.
The tech problems at twitter force hardcore users search for other options, but I think 95% of twitter fans don't really care about that and will patientlly wait till the problems are over which i beleive will happen very soon.

Sunday, May 25, 2008

How to force friends use Twitter?

When a new person gets inside Twitter he usually faces a problem that he doesn't actually know anybody out there. That's actually a huge problem for Twitter going mainstream. When you join Facebook you actually can just upload your photos and tell your friends you have a new album there or if somebody is already registered you can write a mail and the person will be notified, or you can just join some groups, see the photos of people you don't quite know add some apps. The point is that a person joining Twitter feels lonely with 0 followers and doesn't really have much to do. Due to some stats Twitter has something like 10.000.000 people registered but less then 10% ever come back.

What is vital for Twitter is to make use of your own contacts so your friends can be notified on what you write. Plaxo made exactly the same thing, the only difference is they were sending too much emails without people actually knowing all of their contacts get spammed. If it's a weekly or daily e-mail - it's a service, if you get a notification on everything without a chance to easily change settings it's spam.

But lets get back to the question of getting your friends use Twitter. Even if you make them register it won't help. A couple of services are still needed to make this work and I'll add the needed ones. At first give them your RSS feed to integrate as a live-bookmark or in their news-reader. Then explain that they can generate such a feed and can just update Twitter via IM, e-mail, SMS. The right way to get real friends inside Twitter is actually not to force them, but let them watch you and reply to you without changing their habits and the way of life they are used to.

Friday, May 23, 2008

Facebook's new design. One small step...

TechCrunch published an article with a screenshot of the future facebook's design.
Removing the left menu could be seen as a small step, however it isn't.

I actually expected this redesign since facebook started it’s app-platform. Keeping the left menu bar eats the space that could be used by apps. Now you have the chat with you and you are free to go anywhere. That's exactly the layout I use for my current project (mine is different of course, but i also keep all of the page free).

It can grow really big if facebook makes an effort to promote good apps made for it. Right now it’s not that interesting for serious developers, as promoting an app on facebook is almost as hard and expencive as promoting a standalone app. So if you don’t have a product that is designed to spam the contacts on facebook making an app for facebook or go standalone is a question. That leads to the fact that good apps are burried in tons of crappy apss making it hard for users to find them and moving serious developers away from making facebook a “platform of choice”. The very same mistake that google made launching it's i-google homepage. Services should be pre-selected and the user should have the possibility to switch it if they like the other one better. For example the micro-blogging button leads by default to twitter, but can be changed by the user to jaiku or pownce. Why force each user to check all the garbage to find a dimond he needs?

The web operational system.
As more and more apps are developed for the browser the importance of which PC you are using is going down. That's a real threat for microsoft and the current opportunity for the next Tech Star.
Though i really admire Yahoo I think Facebook is the only company right now that makes everything right. If microsoft wanted to ensure it's future business it should really buy facebook but with the team included. Without a team facebook is just a site that doesn't make huge profits now. Well I guess the search button will lead to live-search ) But that's not enough. Techcrunch is absolutelly right that google has the tech to make it's operational system, but they really mess the thing making everything a standalone project. You just can't drag-drop a note made in google-notebook to google-docs for example and a comfortable client to work offline is missing. My bet is that Facebook will be the most important web destingation soon if it actually improves it's messaging system.

The "walled garden".
I won't say Facebook should open up ) But the possibility to contact a person that is using Facebook from outside is vital. The thing is there are lots of people that don't want to be that public and we still write our e-mails on our business cards instead of the facebook/linkedin profiles. A business card could have an additional profile information, but without the possibility to contact a person that is registered in Facebook without having a profile there and a descent mail interface it just won't become a standart de-facto of internet messaging which is vital for the web operational system. You will be where all your contacts are.

Sunday, May 18, 2008

The future in the world of thousends social networks :)

Some History
When people speak about social networks names like Facebook, Myspace, Hi5 and Friendster come in mind. However the idea isn't that new. Long before that, a site appeared on the web with a simple idea and a great domain name: classmates.com. The difference from the actual networks was is they actually charged money for their services and it was a very sucesfull site. Well... why was :) It is still very sucessfull and actually buys lots of advertising meaning it is profitable for them to advertise. In 2002 I was discussing with one friend a possibility to open a social network in a country that wasn't covered by classmates.com and we came upon a $1,5 Millions in advertising expences per country to gain enough momentum to become widely used. Unfortunate for us we couldn't get more then $750k together and we don't have a lot of venture funds nearby (3 in europe). Well it surely sucks doesn't feel great to know about the next big coming and watch it happen, but i got used to the fact that after 10 years in web I am better at predicting things and developing projects rather then pitching VCs and that's not the point of my post ) The point is that there are lots of companies that actually have the money and each of them wants an own social network. The calculation is simple who owns the means of communication, that owns the traffic that is then distributed to all the other projects. Let's leave alone Google with it's entry point strategy and huge media support making it near to impossible for any other search engine to establish itself regardless of the technology used. So what we are seeing now is that finding and contacting people gets tricky as they spread between social networks.

Why the hell do people use social networks?
An easy question for social network addicts is a hard thing for internet professionals. Exchanging photos and arranging meetings is great but it could be easily done via e-mail or IM. The thing all of the big companies missed in the battle over the best e-mail with the hugest mailbox size is actually that people don't know the e-mails of each other. Search engines meant to find people didn't establish and couldn't fullfill a basic need as it is impossible to advertise them - people rearly come back and don't bring the money spent on ads back. Right now Google, Microsoft and Yahoo are desperatelly searching for a way to gain control over the contacts that previusly lived in theirs hotmails, y-mails and gmails while the solution is just in their hands.

What's next
As the number of social networks will grow more people will be trying to "break free" and that's exactly what Facebook and others don't want. Well... GMail is really better then facebook's internal crappy messaging system and other companies like ZenBe roll out great solutions too. I won't even write about the usability of the home of thousends ugly pages where i spend 2 hours once a month (it is in my ToDo calender) trying to understand what people are doing there and keep up with the web. Last week Google, MySpace and Facebook all rolled out some solutions named simular to "DataPortability" and making exactly the other thing - send your visitors our way to login (or sign up) and you could have a fancy iframe on your site. Thanks for a generous offer, but i humbly reject ;) Many newspapers and blogs are atacking social networks to open the data and talking about it's the only way for survival however opening data is a straight way to the grave. This time social networks just made a move that's not too sweet for web developers, they have to roll out something better.

Will the "walled gardens" survive?
I guarantee not :) It's a del.ico.us piece of cake lots of companies and web-developers including me want to cost and work hard to make it happen. And we will ) Leading to a decentralised social web. Whoever loses, users win.

no © you can keep all the mistakes found in this text

Sunday, May 11, 2008

What features do twitter users want?

In this post I'll gather links to posts and info about twitter feature requests.
10 Features I Wish Twitter Would Implement
Twitter Tools Wishlist

HashTags and Groups on Twitter.

Thinking about grouping and tagging on Twitter I have stumbled upon two interesting posts about this:
- Twitter Hashtags and Groups
- Why I Unfollow People Who Use Hashtags On Twitter

Theoretically HashTags could be a solution, but given that "#" signs actually eat up space in short Twitter messages I think that simply "not displaying" them by clients and twitter is not the best option. And acutally often you need to tag the message with a word that doesn't appear in the text.

My idea is to make an open database where Tags are stored.
So you can just press TAG on any message and assign a TAG to it.
Later on you can check posts of a special person, your friends or just everybody TAGed by you and the comunity with a special tag. So you can even navigate your own posts based on the tags assigened by your friends.

I would love to hear if you think that tagging is usefull for Twitter.
Fell free to drop me an email, reply here or make a blog post and notify me to link to it.

Saturday, May 10, 2008

How do people update Twitter?

In this post i want to focus on the stats I gathered about the distribution of twitter updates. The data simply represents % of the posts made with each service. Twitter-addicts usually use clients and update more frequently so if we would want to track unique users instead of posts ammounts I expect the weight of web-updates would rise dramatically. I'll digg more inside the stats as i move on with my project.
Please note, that "TwitterFeed" actually updates Twitter automatically without user interaction. So as i move on in this post, I'll break data into smaller sections.

update methodpopularity
web46,83%
im5,83%
txt5,17%
twitterfeed6,67%
twit6,50%
twhirl6,33%
twiterrific5,17%
P3:PeraPeraPrv2,33%
twitterfox1,83%
twinkle1,67%
movattwitter1,50%
other10,17%




Let's see how it looks like if we compare the services that twitter owns vs. 3d party clients. RSS updates are excluded from the calculation.
update methodTwitter + TXT + IM3d party clients
popularity62%38%


And the most popular Twitter clients in May were:
update methodpopularity
twhirl17,84%
twiterrific14,55%
twit18,3%
twitterfox5,16%
P3:PeraPeraPrv6,57%
movattwitter4,22%
twinkle4,69%
other28,63


Japanise language Twitter clients got an impressive 10,33% share.