Fractals of Change

Nerd Help Needed – WiFi at JetBlue JFK

There’s supposed to be free WiFi here at the JetBlue JFK terminal. Last time I was here no one could get it to work so I didn’t feel bad that it didn’t work for me either; was even a little smug that I could limp by on EVDO (which is usually fine in US terminals but has bad reception here). But now everyone else is connected.

I can see the hotspot: everyone who is on is telling me to use the one called “default” rather that the one called “Jet Blue hot spot” – that one is an ad hoc network so I don’t really want to use it. Default is infrastructure even though it has that ugly name.

I can connect to default. It gives me an IP address so my system thinks it has an Internet connection. But, whenever I try to access the Internet, I get a message that the page isn’t found as if didn’t have a connection.

Those who are successfully logged on (I hate them!) are telling me I have to open a browser to agree to the terms of service. I know that; but, when I open the browser, it tells me I’m probably not connected to the Internet rather than giving me the JetBlue logon screen.

I had no problem logging onto WiFi in Burlington airport this morning (free and terminal wide!). So what’s going on?

When I look at ipconfig (this is obviously a nerd post), I see my connection but no dns server. Hmmm… Why? Did I change some sensitive parameter and forget to change it back.

Logging on with EVDO works – just slow.

Just for the heck of it, I delete the profile for default in Intel Proset Wireless. Now I tell WiFi to look for networks and it finds default and presumably sets up a profile for it. I connect. And I can still get to the Internet. But I’m also connected to EVDO; maybe my packets are going through there. Logoff EVDO; still connected and can get web pages starting with JetBlue signon.

But did the WiFi connection inherit something from the EVDO connection? I power off the EVDO card and am going to try disconnecting and reconnecting the WiFi. Pause. Breathe deeply. Reconnected without incident (my flite’s still scheduled on time, too).

This does NOT convince me I know what’s going on or will get connected next time Do know (and this post will help me remember) to use default rather than Jet Blue hotspot (not good implementation by JetBlue). Do know that I have to start the browser to logon before getting any other connectivity (usually the case but something that has hindered the deployment of WiFi phones).

Would welcome a comment from someone who really knows what’s going on. BTW, am running Windows XP with latest mods on Toughbook with internal WiFi.

Verizon OPEN Wireless

Very surprising and welcome announcement from Verizon Wireless yesterday:

“Verizon Wireless today announced that it will provide customers the option to use, on its nationwide wireless network, wireless devices, software and applications not offered by the company. Verizon Wireless plans to have this new choice available to customers throughout the country by the end of 2008…

“ ‘This is a transformation point in the 20-year history of mass market wireless devices – one which we believe will set the table for the next level of innovation and growth,’ said Lowell McAdam, Verizon Wireless president and chief executive officer.”

Lowell’s right. And Verizon Wireless is right to open up. There’s plenty of room to be cynical about this; after all, Verizon Wireless is trying to STOP the FCC from putting an openness requirement on the 700Mhz spectrum to be auctioned. Very well-informed Om Malik posts: “Do we really believe that Verizon is going to be happy being Pipes-R-Us?” He point out many ways this there may be less to this announcement than meets the eye.

As an optimist, I think Verizon’s acting in enlightened self-interest which is just great. Here are some of the factors which may have influenced their decision:

1. They DO have the best network coverage in the US – especially for data coverage. Best way to capitalize on that is to have developers build appliances and apps which run on their service.
2. Because they are CDMA (a protocol NOT used in most of the rest of the world), they run the risk that no new stuff will be developed for their network, especially if it remains a walled garden. The US is a big market but the rest of the world is even bigger.
3. The next time somebody develops an iPhone-like breakthrough – and somebody will – they want it run on their network, not be locked to AT&T.
4. Amazon’s Kindle, which could but doesn’t run on Verizon’s network, is a clear example of how usage may be sold bundled with a device. They’re not going to be a better bookstore than Amazon. They need these innovations to be on their network.
5. They know that the next year or so will bring huge device innovation including (I think) wireless connectivity in almost every GPS and associated services. They know that their network and their data service – EVDO – which can do handoffs at 80mph is well positioned to benefit greatly from this
6. As voice minutes turn to VoIP minutes and WiFi minutes, they’re better off keeping some of that traffic onnet even as simple bits rather than losing it all. Note that, to their credit, they DID change their terms of service to allow VoIP over their data service.
7. The FCC has said that the huge chunks of 700Mhz spectrum going up for auction have to be used “openly” by the winner. It would be hard for Verizon to operate a network which is half open and half closed. Maybe they DO want that spectrum, don’t think they can change the rules, and want to be ready for it.

If you’re torn between my optimism and Om’s pessimism, Galeal Zino suggests a test:

“Let's see which comes first:

“1. Articles about the millions of dollars of deep packet inspection and payload-based billing systems that Verizon is going to incorporate in order to "manage" their soon to be "open" network.

“2. Articles about third-party mobile devices that interop with Verizon's network, using CDMA for voice when necessary, and using unrestricted IP to place VoIP calls using third-party solutions when more appropriate.” 

What’s twitter?

If you’re a hermit, you’ll hate twitter. In fact, if you aren’t a gregarious super node, you probably won’t like twitter or see much use for it. But – with you or without you – it’s growing quickly and has the attention of very communicative and well-read bloggers. As part of my new technical gig, I have to understand sites like this even though I’m more a reclusive type. This post is about what I’ve learned: you may want to skip it if you already twitter but you’re also welcome to stay and join in the discussion.

Best thing about twitter is it has a simple purpose: it provides you with a way to find out in 140 characters or less what you’re twittering friends are doing right now and to tell them what you’re doing. You can get the stream of 140 character messages, which are called tweets, as text messages, as instant messages, or in a simple web application. You can create them on a cell phone, in the simple web application, and in a couple of other ways which begin to get interesting and which I’ll get to below.

You are a “follower” of those people whose tweets you want to see; you only get tweets from those you follow unless you elect to look at the “public timeline” which has a gush of every public tweet – can’t imagine why you’d want to do that unless you’re in sociology or marketing.

Unless you make your tweets private, they’re both part of the public timeline AND visible to anyone who wants to follow you. Of course I immediately made my tweets private meaning I had to OK would-be followers (or insult them by not OKing).  That’s not the true twitter experience so I gave in and made them public. I do get the screen names of new followers in email and a chance to become followers of theirs as well.

If you are a twitterer, you can follow my infrequent tweets under my screen name of tevslin. My last tweet is from 3:45AM this morning complaining that I was up for an early flight. Don’t know why you want to know this; don’t know why I wrote it; but that’s what social people do.

One use of twitter I do understand is facilitating spontaneous meetings of people who’d like to see each other and now have a way to know when they’re both in the same place at the same time. Lots of nice stories about friends waiting for delayed flights at different gates of the same airport and getting a chance because of twitter to have a drink or a meal together.  Two weeks ago in my pre-twitter days, I was in NYC and missed a chance to see a friend who was also in town; maybe we would have gotten together if we twittered.

Some people – Jeff Pulver is a good example – use twitter to say that they’ll be somewhere and happy to meet people with certain interests. Jeff has lots of followers (and follows lots of people); this works for him.

You can send direct tweets to someone else; I don’t think the facility is used much but it does have an advantage over texting: the tweet will get to the recipient (assuming the recipient twitters) on his or her cell phone or computer depending on which they happen to be using at the moment for messages – you don’t have to guess what device they’ll be on.

Twitter spam is, in a sense, impossible. No one can make you follow them. You can stop following anyone who is boring.

Applications are growing up around twitter which take advantage of the fact that it’s a good way to notify people of things they may want to know – especially expanded versions of the answer to the question “what are you doing now?”

An application called seesmic (still in closed alpha) makes it easy to take short videos of yourself with your screencam or webcam and post them to the web. An option is to have a short description of the video and the link automagically appear in twitter so your followers can see and hear your latest status with a single click.

Dave Winer has created twittergram which makes it easy to post audio from your cellphone or PC and have a tweet with a description and link appear to your followers. Twittergram can also do pictures posted to flickr. (had tip to Fred Wilson for his post on twittergram).

Twitter’s happening, even if it isn’t your cup of tea.

Quick note to nerds: twitter streams are xml. Makes it easy to write web apps both to feed them and to view them.

Full disclosure: I’m a minor indirect investor in twitter through Union Square Ventures.

Kindle – Reader Questions and Comments

In a recent post, I opined that Kindle may be more important as a crude opening wedge for free (sponsored) Internet access than as a change-agent for the way people read books.

Reader ellen has a good question:

“The free internet service after the $400 buy price sounds too good to be true. I would certainly buy one for that reason alone.

“I am going to ask a really dumb question. Does getting wireless access to the internet mean every time you are near any wireless you will be able to log on or does it mean you have to go through the same towers as cell phone access? Having had so much trouble with at and t cellular for my phone would sprint have to be well covered in my state? No one around here uses sprint for cell access. Everyone uses verizon, because for some reason it is best in my area of massachusetts. After many years and 4000 leftover rollover minutes I a dumping at and t for verizon.”

Ellen, that’s a good question. You shouldn’t buy a Kindle if you aren’t usually in a place where there is a good signal from Sprint. EVDO uses the same towers as voice service; so, if Sprint voice isn’t good where you are, EVDO won’t be either. Verizon also supports EVDO but that won’t do you any good with Kindle because it’s tied to Sprint’s network. Moreover, it is possible to be in a place where voice service is good and data not. If you know someone with a reasonably modern cellphone who has Sprint service, ask him or her to look at it near your house and see whether it says “EV” or “1X” where it shows the data connection (different phones display this differently). 1X is a slow data network and Kindle won’t work on that; it is rapidly being replaced by EVDO by both Sprint and Verizon.

Reader Aswath, who is a smart guy, is confused:

“I am a bit confused about the "free" access to the Internet. If I am right in interpreting this, then why do I have to pay a monthly fee to read FOC? That is why I thought that I could access only those sites to which I have subscribed (signified by a bookmark in the browser?). But then both you and Pogue [nb. NY Times writer David Pogue in this article] say it differently. Hence the confusion.”

Internet access IS free through the builtin browser on Kindle. You can, for example, read Fractals of Change in the browser – free – just as you do in the browser on your PC or Mac. However, you can also elect to pay 99 cents/month to subscribe to FOC. That is confusing; why would you pay for something you can have free?

Well, two possible reasons. One, the version you pay for is presumably (I don’t have a Kindle yet) better formatted for the Kindle screen. It has no ads and doesn’t have the sidebars that the browser version does. The Kindle browser is described by Amazon as “basic” so that may make FOC pretty ugly when viewed that way.

More important in the case of a blog or other periodicals which you can subscribe to is that they are downloaded to the Kindle automagically rather than being fetched from the web. That makes a difference if you are going to want to read this content when you’re offline – on a plane, for example, or traveling through country where Sprint EVDO (and therefore Amazon Whispernet) is not available – assuming you were online at some time so that the download could take place.

Both of these reasons for buying content rather than consuming it free are even more important for books. You can read my novel hackoff.com: an historic murder mystery set in the Internet bubble and rubble free at www.hackoff.com. Many people, however, don’t want to read a whole book online. You can also download and print PDFs free.(but not on Kindle). Or you can get it nicely formatted so that you turn pages as in a book rather than scrolling and read it on Kindle offline – but that costs $4.76. If you do that, BTW, Amazon stores it on your virtual bookshelf forever even if you delete it from Kindle to make room for more books. Probably the real comparison here is with the hardcover edition available from Amazon for $18.96.

Reader Marc Orchant did try hackoff.com on Kindle: “hackoff.com looks great on the Kindle. The formatting is clean (e.g. Q&A in the first chapter) and I'm looking forward to reading it as I count myself among the survivors of those perilous bubble days.” Nice to hear. Thanks, Marc.

Reader Terry Gold also likes Kindle (wish I had mine): “I've had a Kindle all weekend, and I'm already planning on how I can get rid of most of my paper books. I'll keep the ones that mean the most to me, but for everything else, this is the way to read. I just had Amazon send me a sample of your book even though I have the hard copy and I can read it on the web. The Kindle changes reading for me.”

On a somber note, reader Dylan Salisbury asks: “What happens after someone hacks into a Kindle and tethers their PC to the internet connection? If they had really "bought" that internet connection, it wouldn't be a problem (you can run any traffic you want over your own connection, right?) -- but I have a feeling such uses will be shut down or denied.”

I think Dylan’s right. If such hacking becomes noticeable, it probably would be shut down in some way. And that raises an interesting question: is it a violation of Net Neutrality to shut down some uses of “free” Internet access? I’d say no in this case since Amazon hasn’t been using the free access to the web as a selling point and it’s pretty clear what it’s for; but certainly there may be plenty of argument over this.

The Snows of Whiterock Mountain

Near the top of Whiterock Mountain in Vermont.

Puzzle: why is Tucker wearing a red blanket and Bruiser wearing a bandana (hard to see)? Vermonters please don't answer; too easy a question for you.

Kindle – Shape of the Web to Come?

Forget the argument about whether or not you want to read books on a device instead of on comfortable paper. Forget whether you want to pay a subscription price for blogs like Fractals of Change that you can also read free in a variety of different ways. The Kindle points to a possible future for the WorldWideWeb in a way that has nothing to do with books.

Kindle comes with free browsing! Free! You pay once to buy the device and then you browse free anywhere that Sprint EVDO reaches, apparently for as long as you have the Kindle. Remember all that stuff I blogged about buy vs. rent? Here’s a chance to BUY your roaming Internet access for a onetime fee.

After the mandatory questions about e-books and e-book readers that we’ve agreed to ignore in today’s post, David Pogue gives some attention to the importance of free browsing albeit buried well down in his piece in The New York Times:

“But the part that will really rock your world is the Kindle’s free wireless cellular broadband service.

“Now, if you just splurted your coffee, you’re forgiven; “free” and “wireless broadband” have rarely been used in the same sentence before. The Kindle goes online using Sprint’s 3G cellular data network — the same service that costs $60 a month for corporate laptop luggers. The Kindle’s price tag stings less when you realize that Amazon is going to pay your entire wireless tab.

“So the Kindle can get online almost anywhere — not just in little coffee-shop hot spots, but in cabs, in lines, in doctor’s offices.

“There’s even a crude Web browser. It’s fine for text and graphics, lousy for Web layouts and useless for streaming audio or video. But with some effort, you can use it to get news, rebook a flight, monitor blogs and even check Web e-mail (like Gmail).”

In a previous post I glibly said the Kindle couldn’t be used for email. That’s because I use Outlook, which doesn’t run on Kindle, as a client and forgot how many people use a web interface to read their mail. Duh!

Now, as David says, this is hardly the be all and end all in browsing. It’s not good for porn videos and doesn’t even do color. The browser is “clunky” and should at least work sideways, which it doesn’t, for good web viewing. But the browsing is free!

Will this model spread? That’s the important question. There’s actually a good chance that it will.

Don’t know what the deal is between Amazon and Sprint which supplies the underlying network. Sprint has had a disappointing experience recruiting and keeping customers lately so probably had network assets to spare. But, once a network is built, extra bits don’t cost anything substantial until you have to buy more network.

If a part of the sales price of each device provides capital for network expansion, that’s a scalable model. (Don’t worry about depreciation; the device’ll get obsolete at least as fast as the network assets and have to be replaced so more capital’ll come in). If Amazon pays Sprint on a per use basis instead of a per device basis, that pays the debt on network expansion; the model works either way (or both).

And what about Amazon? Well, they sell the device and presumably make money on that; they’re sold out at the moment. But they also “sponsor” the access. They chose the bookmarks for the browser. The device comes linked to the Amazon account you bought it through. With this device, it is even easier to buy real books from Amazon. In fact, you can download the first chapter free and then decide whether you want the paper or electronic edition. They’ll have a recurring revenue stream as well as the initial sale – and can help pay for the cost of network operation of that’s part of the deal.

Kimble may well be the crude leading edge for “free” sponsored web access. If so, there are profound implications for web development (hint: develop apps that run in a browser), the Internet access business, and even Net Neutrality. There may even be a tie in with Amazon’s hosting business for such apps. Hmmm…

My prediction: we’ll see “free” sponsored access on car GPSes before another year has passed: the opportunity for roadside advertising is huge and the screen is already paid for.

Full disclosure: Both this blog, Fractals of Change, and my novel hackoff.com: an historic murder mystery set in the Internet bubble and rubble are available for Kindle. You can pay to download them from Amazon OR you can use the browser to read them free online. Choice is good.

Kindle – Free Internet Browsing for Just $400

Both this blog, Fractals of Change, and my novel hackoff.com: an historic murder mystery set in the Internet bubble and rubble are available for Kindle, Amazon’s new e-book reader. These two reasons are enough for me to buy one even at the $400 pre-Christmas price; but there’s a third reason that might convince even non-authors: free Internet browsing.

Update: I should’ve been clear. You have to pay to subscribe to FOC ($.99/month) or to buy hackoff.com ($4.76) on Kindle in Kindle format although both are free on the web and even through the Kindle browser (see below). Fred Wilson hates this but it doesn’t bother me because the connectivity is free (once you buy the device). Usually you pay connect to the Internet and get the content free. This is another choice for readers.

In a New York Times article this morning, Saul Hansell quotes Amazon founder Jeff Bezos: “If you go back in time, the landscape is littered with the bodies of dead e-book readers.” Presumably Jeff means the devices and not the people who used them.

Obviously, Amazon means to succeed with this device despite the fact that other e-book devices have hardly taken the world by storm. It is different in at least one important way from its predecessors: it comes with a free wireless Internet connection via Sprint EVDO service (which Amazon calls Whispernet). Clearly, this connectivity is meant to make the device easy to use and written material easy to purchase. For comparison, an unlimited EVDO plan from Sprint costs $60/month but you don’t have to have an account with Sprint to use EVDO.

You can do more than just buy e-books or order from Amazon over this connection. Kindle includes a web browser. Unlimited use of this browser over Whispernet is free. Note: This isn’t absolutely clear from the Kindle documentation so I called Kindle support. “Yes,” the CSR said, “free,” and “yes, unlimited.”

From an author’s point of view, the inclusion of a browser is a breakthrough. My blog as well as the online editions of hackoff.com and The Interpreter’s Tale all include links which I think added to the stories but get lost in the paper editions. People already read blogs online, both because of timeliness and links. I’ll start reading books online when they are richer than paper books – that means links that work!

This isn’t full Internet access. There is limited e-mail available through which you can receive attachments which Amazon converts to Kindle format at $.10 for each conversion (or free if you email them to your non-Kindle email account). Other people you authorize (remember, you’re paying for the conversion) can also send you attachments. But this isn’t a Blackberry; you can’t do your regular email through it you can only do email in the browser.

Other than downloading Kindle-compatible content and products from Audible.com, it doesn’t appear that you can do any other kinds of file transfer over the Internet connection. You can use the included USB cable, however, to transfer photos and music from your PC directly.

Even the browser isn’t fully featured. According to the User’s Guide: “Your Kindle comes with an Experimental application called Basic Web which is a Web browser that is optimized to read text-centric Web sites. It supports JavaScript, SSL and cookies but does not support media plug-ins (Flash, Shockwave, etc.) or Java applets.” That means no YouTube on your Kindle. Note: TechCrunch says that Kindle DOESN’T support JavaScript. I’m assuming they mean Java since web access these days is almost useless without JavaScript but don’t have a device so can’t be sure.

Depending on Sprint EVDO has its plusses and minuses: there is no searching for a hotspot as there would have been with WiFi and no worry about signing on to a WiFi service. On the other hand, Sprint EVDO isn’t everywhere in the US and is hardly anywhere outside the US. Amazon marketing says: “With Whispernet, you can be anywhere, think of a book, and get it in one minute. Similarly, your content automatically comes to you, wherever you are. Newspaper subscriptions are delivered wirelessly each morning. Most magazines arrive before they hit newsstands.” I buy the one minute; but “anywhere” is quite a stretch.

My bet, WiFi will be added soon. If people are going to use the live links, it won’t be satisfying to download where you have EVDO connectivity and then read offline. But WiFi is becoming pervasive in homes and hotels and’ll soon be in planes (I hope).

Maybe Kindle is the wave of the future for free web access. See this post.

My Friends the Super Nodes

The super nodes are what makes undirected networks efficient. They’re responsible for everything from the apparent fact that there are no more than six degrees of separation between any two people on earth to the rapid initial spread of AIDS. Memes become fads as the super nodes propagate them. Hits don’t happen without super node attention.

A lot of the discussion Matt Blumberg, Brad Feld, Phil Hollows, Jeff Pulver, Fred Wilson and I had about the future of messaging was actually about social networks (see my post here and Fred’s here). Social graphs area fancy name for the lines which connect nodes in a depiction of a social network – a network of people. If you draw the simplest kind of social graph where there is a line between any two people who know each other, you’ll find that most people have less than the average number of connections while there are a few people who each have a huge number of connections: these are the super nodes.

You know who they are in real life: they’re the people who always know someone for you to contact whether it’s a job or a soul mate you’re looking for. They know everybody you know and they know lots of people in the place you’re about to visit as well as some people in the place you’ve just been. They’re gregarious (it’s easy for them because, when they do meet someone new, they always have friends in common).

Us average joe nodes tend to cluster: we and the circle of people we know all know each other and not many people outside the cluster. Turns out in networking theory that one of the reasons the super nodes are so important is that they have friends all over the place. So an idea – or a disease – jumps from cluster to cluster via super nodes. You need these long connections if you’re looking for a job because you already know most of the people your friends know and you wouldn’t still be looking if they had a good opening for you – but your friend the super node knows people in other clusters who may be looking for someone just like you.

In cyber-life they have zillions of friends on Facebook, their blogs are well-read, they have good Google-juice, and they are followed by hordes on twitter (and do a fair amount of twitter-following themselves). Supernodedom is a kind of celebrity but it is two way. Celebrity is watched but doesn’t necessarily see; super nodes really see and communicate with lots of other people. Otherwise they couldn’t help you make contacts.

Turns out Fred and Jeff are super nodes (not that I didn’t already know this). When Jeff met Facebook founder Mark Zuckerberg at a conference, he immediately asked Mark for help with what for him is a pressing issue – Facebook only allows you to have 5000 friends. Terrible problem if you’re a super node.

Facebook plugins and twitter apps that work fine for you and me break when Fred or Jeff try them. No one allows for the number of friends that have to get loaded for these guys. They need special tools to let them rank and prioritize the volumes of communications they get from all these friends – and the ever increasing number of new friend requests.

If you’re writing a network application, you have to make it simple enough for the average user (who has a below-average number of connections) and robust and scalable enough for the super nodes. You DON’T want to forget about the super nodes because they’re the ones who are going to make or not make your product spread virally across the web. Moreover, if you have a network service, you always have a problem getting enough initial users to make the service valuable to new users; each super node is worth hundreds of us ordinary users in network value to a social network of any kind because of the number of connections they have and the number of people who want to connect with them.

I suspect that some social networking products may only appeal to super users. I’m trying twitter, for example, but haven’t yet found it compelling (after twelve hours). I signed up with a closed twitter account so the only people who can follow me are those I permit – that’s not the way super nodes act at all, I quickly found out. Fred and Jeff twitter all the time and have many ideas for making the service even more useful to super nodes.

Although you can have a successful network service serving mainly super nodes, you CAN’T succeed if your service is only fit for us ordinary nodes – networks don’t work right without the super nodes; the rest of us don’t have enough connections.

And The Answer Is…

The answer is “we don’t know.” The question is “what did you guys decide in your discussion yesterday about the future of email?” But we learned a lot and hypothesized even more. Of course, being blogger types, we’ll share and also hope you can add to the discussion.

Is email dead? Fred Wilson has a good post today pointing out that email has become just part of an array of messaging options. His list of the others:

“Instant Messaging
Blogging (each post is a message)
Skyping (text+voice)
Voice Mail Transcription (voice to text)
Twittering/FB status update
Web mail
Web site messaging (FB messages)
Comments on social media
Social gestures (actions in the news feed)
Text messaging (sms)”

I’d add to Fred’s assertion that each blog post is a message the interesting analogy that each comment is a “reply all” but not as annoying as reply alls in email which tend to fill inboxes with unwanted messages.

We spent a lot of time talking about social networks both because Jeff Pulver gave an impassioned description of life in his social media living room and because all of these messaging options above and many others rely on and create a “social graph”. One way to think about an application like Facebook or MySpace is as an explicit way to manage, use, extend and restrict your social graph.

So what’s a social graph? (Mary says I need to point out that I’m using graph in its mathematical meaning of the lines that connect points and not in the vernacular meaning of a line that looks like a mountain range and is used to represent or misrepresent some trend.) You and all the people you communicate with are the points on a social graph. One way to draw that graph is with all the lines radiating from you; but, of course, the people you communicate also communicate with each other so there are even more lines. Once you start to draw lines out from the other nodes (people), you start to create a larger social graph which contains your own graph; this is the friends-of-friends graph. Etc. Etc.

Way before there was an Internet, people made good use of the information in their social graph to decide, among other things, how much attention to pay to strangers. “I would like to introduce you to…”; “I’m a friend of…”; “…said I should look you up”; “you’re from Vermont; do you know…”

Arguably the Internet and all the ways of messaging on it have increased the need for us to use our social graph to filter wanted communication from spam, information from noise. It’s become much easier for more people to reach us but we don’t have any more hours in the day to absorb or respond to messages. We don’t want to be cut off from making new friends but we do want a way to be selective – especially when making unseen e-friends.

A simple application of the social graph is that some spam filters block mail from anyone not in your address book. Facebook uses a positive approach to do filtering: you only befriend and open yourself up for communication with people whose invitations of friendship you’ve accepted or people who’ve accepted yours. But note an important aspect of befriending on Facebook: when you get a “friend request”, you are able to see which friends you and the stranger have in common. A simple spam filter is indiscriminate compared to using “friends of friends”.

You can imagine an inbox which arranges your messages according to how well-connected the sender is on your own social graph; well-connected could be interpreted in number of ways including gross number of connections; volume of communication on each connection; connection strength weighted by how close you are in terms of volume of communications with each of the connectees of the sender; the quality of connection as measured by messages in rather than messages out (someone whom people write to a lot is putatively more interesting to hear from than someone who sends a lot and gets no replies).

This gets interesting when you realize that email systems like gmail, Hotmail, and Yahoo Mail already have the information necessary to construct and mine these graphs (hopefully with your permission) as do Microsoft Exchange servers. Bloggers, the links between blogs, and the comments on blogs are also social graphs AND they can be connected to the social graphs of other messaging forms. So far each messaging form like email, IM, phone calling etc. has had its own social graph in the form of an addressbook. But we nodes are using all modes.

Hmmm..

BTW, Fred said that he is about to write about social graphs as well so look for his post.

Thinking Aloud

This morning Matt Blumberg, Brad Feld, Phil Hollows, Jeff Pulver, Fred Wilson and I are getting together to think about email in the larger context of social networks, instant messaging, RSS, SPAM, and other interesting stuff that’s going on. Brad outed the fact that three of the group – Fred, Jeff, and Brad, himself – are all preparing for this discussion by thinking aloud on their respective blogs. Somehow that fact, too, is relevant to what we plan to talk about; I’m just not sure how.

Jeff is the evangelist of the Social Media Living Room, a place where he can (and does) communicate with an extraordinary number of people in a number of different ways, most of them less real-time than a phone call and more immediate than an email. He explained this vision in his keynote at VON and, to his disappointment, it went over the heads of most of the gathered telco executives. He also told the execs to hire 13 year old product managers and fire them when they reach 18 and get stale in their thinking. Jeff may be optimistic on the pace of change but I think he’s right about both the direction and the importance of social networks like Facebook to the way we communicate – even though my social media living room is much smaller and quieter than Jeff’s.

Fred expands on a post by Saul Hansell and points out that Microsoft Hotmail and Yahoo Mail each have social graphs with more then twice as many nodes (sorry, I mean networks with over twice as many people – slipped into nerdtalk) as mySpace and Facebook while AIM and gmail have about the same number of users as the big social networks. Some spam filters do take advantage of this by filtering mail which is NOT from someone in your address book. Come to think of it, cell phones use our personal address books to find the names of callers and assign them the appropriate ringtones.

In Saul’s post Yahoo and Google talk a little – not very specifically – about their plans to leverage both the address books and the pattern of who gets mail and messages from whom how often to compete with Facebook. But Facebook is the competitor de jour so take that with a grain of salt. Both companies give due deference to the privacy of our information, of course.

Brad – who, like me, has a tendency to think about servers – responds to the talk about Hotmail and Yahoo Mail by writing: “So what.  Seriously.  The real data lives in the gazillions of Microsoft Exchange servers that are distributed around the world and connected to this magical thing called the Internet.  Don't think about your inbox (or your Outlook PST file) - think about "the server."…

“The amount of "social information" - especially in a business context - is staggering….  As far as I can tell, everyone is focused on the client side (Outlook) rather than the server side (Exchange).  This confuses me since the information, distribution, and the leverage (especially with regard to selling stuff) is on the server side.

“But the bigger and more mysterious question is "where is Microsoft?"  This is their world and their domain.  Over 15 years they demolished IBM/Lotus (and everyone else) in "email" only to be ready to fumble the next wave of this.  I don't get it.”

What do I think? Thanks for asking.

email hasn’t changed much in the twelve years since all the corporate email systems and all the private networks got connected to each other through the Internet. Our use of email has exploded; the messages have gotten bigger and richer as access bandwidth has increased; new devices like the Blackberry have been designed around email; but email is pretty much what it was a dozen years ago. For kids, however, email is often not as useful as text messaging. And spam has gotten to be a problem.

And there hasn’t been a significant advance in landline phone calling since the invention of the push button phone. But cellphone calling is often directory based and the device is used for text messaging, pictures and video as well as voice.

Following Moore’s law, the chips which are in almost every communication device are 256 times more powerful than they were twelve years ago. Graphics are fantastic compared to what they once were. Voice and other sounds can be carried much more richly than on the old phone networks.

The incremental cost of communication is near zero and getting closer all the time.

We think aloud on our blogs before going to a meeting (but we do still like to meet face-to-face!)

I think we’re overdue for huge change. I’m sure it’ll be fun and disruptive. I just don’t know what we’re changing to. When I do, I’ll blog about it.