How will the Web of today evolve into the Web of tomorrow? The way we think of web development -- as HTML pages on a server being accessed by desktop browsers -- is about to change. The way we'll access the Web, the way we replicate and accept data, even the formats in which the data is passed, are all changing.


I'm not a betting man, so don't put all your chips down on one of my predictions, but this article details how I see the Web trending in the coming year. Much of this technology is available now, but it's used among small pockets of smart, revolutionary people. As they cut their teeth on these new ideas and products, the technology will make its way to the masses. It's my prediction for 2008 that these innovations will become understandable, both commercially and socially, and easy enough for anyone to use and access.


Some of these predictions already exist today for the alpha-geeks who are riding that cutting edge of technology. There are many ideas on that leading crest, and though some will fall away, others will prosper. This list represents the nine items I see emerging into the mainstream and becoming more mature in 2008. They'll move from the hands of the alpha-geeks to the lives of everyday people.


1. Facebook Plateaus, and No One Cares



Social networking sites aren't going to disappear, but I think the most popular, Facebook, won't continue on its current trajectory. If you look at the graph of MySpace vs. Facebook popularity below
If we compare several popular social network sites -- Friendster and Hi5, Facebook and MySpace -- on the graph below (or online), we see that trend again: a slow start, substantial growth, and then a plateau. And 2008 will see either a new competitor enter into this crowded field, or the social networks will begin to chip away at each other's customers. Facebook seems to the biggest on the block so, everyone's championing the under-dogs. With recent privacy blunders, a general lack of trust in Facebook, and campaigns run to encourage users to close their Facebook pages, many users are beginning to realize that the best social network is just the Web itself.
Alexa Traffic Report for various popular social networks (click to view image)
In this last graph (which you can also view online), we see a comparison of the various social networks with Blogger.com, a free blogging platform.
Alexa Traffic Report for various social networks compared to Blogger.com (click to view image)

The Blogger.com platform alone is bigger than the biggest social network. If you add all the other free blogging platforms -- Wordpress, LiveJournal, and so on -- to the mix, you can see that the hype of social networks just doesn't compare to that of individuals with their own blogs.
As people get tired of fighting with corporate advertising, and money-hungry companies who are all too happy to put their own interests ahead of their customers, users will move to their own blogs and web sites. Social networks might be good for finding old friends, but so are search engines. If you want to get in contact with an old friend, blog his or her name, or try searching for your own name -- you might just find that an old friend is looking for you.


2. Data, Data Everywhere ... and Not a Drop to Drink



O is the new i is the new e. Years ago, when traditional technology started to move online, it be came e-everything -- e for electronic. Email, Ecommerce, eMachines -- you've heard 'em all. Then Apple started to make iMacs, and i was the new e. I stood for Internet, then the iPod appeared and i didn't really stand for anything. 2008 is the year of the O. O is for open. Open will be the hot buzzword: OpenID, OAuth, OpenSocial, and more.
In this new era of openness, data ownership will become an issue. Most people are unaware of the rights they have -- or lose -- when dealing with social network sites. Think about the comments you make, and the images you upload, who owns those? You spend months and years adding all that data. It should be yours, but this is not always the case.
Data portability is another key issue associated with open data. When you join a site, you should ask yourself, "how much of my data can I get back out?" It's your data, your words, and your property. You should have the right to take all that with you if you leave. You spend valuable time and effort building it up, but guess what? Often, it isn't yours -- if you quit the site, you can't take it with you. This situation is known as vendor lock-in, and 2008 will see more of these walls begin to come down.
Before you join yet another site, demanded that you can quit at any time and take everything with you. Sites that honor your rights will excel, and sites that don't are doomed to collapse under the weight of their own unhappy customers.


3. Pull is Dead! It Was Killed by the Push


Services like Twitter, Jaiku, and others make it easier and easier for us to generate data almost in real time. That's both a blessing and a crutch. We want that data to be open -- to be able to freely move around -- but at the same time, we don't want various services requesting our RSS feed every 10 minutes and killing our bandwidth.
When a service requests your HTML, RSS, or any other file, that action's called a pull -- the service is pulling the data from your server to theirs. Pulling is pretty easy and straightforward, but the requester has to keep asking the source to see if the file has changed. This alone doesn't require a lot of bandwidth in cases involving a small group of clients and vendors, and caching is not important. But pull doesn't scale to the size of the Internet when you're dealing with near-real-time data. Can you image what would happen if a site like Facebook pulled your Twitter feed every 10 minutes? Now imagine millions of Facebook customers pulling millions of Twitter feeds every 10 minutes. Even if nothing's been updated, the service needs to make the pull request to find out, and that amounts to tremendous bandwidth usage!


The solution to this problem is to PUSH data between federated systems. A push system allows the customer to inform other sites only when an update is available. So instead of Facebook polling Twitter every 10 minutes, Twitter pushes an update to Facebook only when something new is added. The by-product of this approach is a huge bandwidth savings and a reduction in the delay between services. Previously, Facebook could be as much as 10 minutes behind your Twitter account, but with a PUSH system, Facebook is informed of the presence of new data at the same moment that the post is made to Twitter. Another advantage of this approach is that you can PUSH to multiple services at once. So when you update your Twitter feed, you could inform your Facebook/MySpace page, your Jaiku, lifestrea.ms, and other accounts, thus saving bandwidth exponentially.


The other byproduct of a push system is that it requires us to federate the systems and agree on a standard end-point and format. This promotes data portability and the strange side-effect of mixing and matching systems. I personally like Flickr for my photos, but some of my friends use Picasa. In a federated system Picasa could push an alert to Flickr to indicate that my friends have added images. Then, in my Flickr stream of friends' photos, I'd see images that were hosted not by Flickr, but by Picasa -- and vice versa. This scenario allows each person to use the system they're familiar with, and at the same time, interact with friends on other systems.
One group working to solve this problem is the team creating Instant Syndicating Standards. In 2008, more groups like this will emerge to tackle this problem.


4. It's a Bird, it's a Plane ... What is That Thing?


Be on the lookout for the introduction of more nontraditional connected devices. Cell phones or expensive PCMCIA cellular cards used to be the only truly connected mobile devices, but now, more and more devices will use cellular technology, but they won't be phones that we recognize. The new Amazon Kindle has a cellular modem built in, which allows it to phone the Amazon mothership so you can download more ebooks. It uses a traditional cellular connection like a regular phone, but it isn't a phone you could use to make a regular call.
The cellular world is morphing into something new to facilitate this evolution. We can see already that the iPhone is not just a phone -- it's a small computer with a cellular/WiFi "always on" connection. Nokia has been building similar dedicated WiFi devices for awhile, with its N770, N800, and N810 Internet tablets.


Even telecoms are getting into this new market. In the US, Verizon agreed to open device and software access to let any device that meets its specs onto the company's network. As such, any home-brew hardware hacker can now begin to build something that connects to the Verizon network and passes data around. This facility allows for even more non-traditional devices to benefit from the existing cellular infrastructure.
Companies like Nabaztag have been working along these lines for awhile. In 2008 we'll see more hybrid devices hit the mass market. People will become aware and rethink what they use to "be connected".
Thanks to