Some people say that the bubble's going to take a downturn in the next year or two - that huge numbers of copycat startups are going to shut down, people are going to be out of work and Web 2.0 cheerleaders are going to eat their (our) words.
While startup churn is inevitable in any industry (thank goodness we're not restaurant founders!) I think this forecast is selling the future short. There are some big trends I'm really excited about for the web in 2008. Whatever happens to the economy, there's at least a whole lot of innovation to be inspired by right now. Ultimately, I think that will end up brightening the picture for all of us around the world.
For each of the 5 big topical trends described below, I've assembled some resources I think will be useful for anyone who wants to keep up with cutting edge developments in these fields in the next year.
These resources include:
* An OPML file of top blogs on each subject. This is a bundle of feeds you can import into your reader.
* A filtered RSS feed of just the most popular items regarding each topic (using AideRSS). Remember, whenever you subscribe to new RSS feeds - some of the magic won't be visible until you mark all the initial items as read and new ones come in again.
* A Custom Search Engine that you can bookmark and use to search inside the top news and reference sites regarding each topic.
If you'd like to recommend any additional feeds or sites to add to these resources just let me know and I'll check them out. Please do, in fact. I hardly have the energy to make these lists exhaustive by myself. That's kind of the point of this whole web endeavor, isn't it?
All of these resources are dynamic; sites added at a later date will be automatically delivered to everyone who subscribes to these OPML files today.
So let's get to some trends that are shaping up to make a big impact on 2008...
Data silos and walled gardens are a huge loss of opportunity and more people are figuring that out every day. The developments in this direction seen just in the last half of 2007 are too numerous to list here but some of the subthemes include the following.
Open Data Resources:
*RWW Open Data Feed Favorites OPML file (save link)
*RWW Open Data Feeds - Best of Feed (copy and paste to your reader)
*Preview the above feeds before subscribing (pop-up window)
*RWW Open Data Sites Search (Visit and Bookmark)
As is aptly demonstrated from the category above, the future is likely to be even more swamped in data, social and content options than the web is today. From Google Reader's recent incorporation of both feed recommendations and shared items in Reader from your contacts in GMail to the ascendancy of services like Last.fm, Pandora and StumbleUpon - recommendation is beginning to make a big splash already.
Dr. Rick Hangartner, Chief Scientist at recommendation engine MyStrands, posited the following about the relationship between search engines and vendor-specific recommendation engines in a recent guest post on mobile search blog MSearchGroove:
In the near term, search engines will increasingly incorporate simple recommender technologies to handle approximate queries (e.g. “You asked for this, and based on similar queries/behavior by others, you might be looking for this.”). But in the long term, the recommender industry will be larger, and recommender technologies will be more pervasive than the search industry and search technology as we know it. [Because there will be recommendation going on all over the web.]
Recommendation Resources:
*RWW Recommendation Industry Feed Favorites OPML file (save link)
*RWW Recommendation Industry Feeds - Best of Feed (copy and paste to your reader)
*Click to preview the above feeds before subscribing (pop-up window)
*RWW Open Data Sites Search (Visit and Bookmark)
A Semantic Web has been in the works for a long time but is just starting to hit the scene for real. The idea is that semantic web technologies are able to derive meaning from online content and determine connections where none have been made explicitly.
If I'm looking at a web page about Assata Shakur, for example, then SemWeb tech should be able to determine that she's the subject of the page I'm looking at and that it's a biographical page. Once that's been determined, semantic technology can leverage the two trends discussed above (data openness and recommendation) to do all kinds of interesting things.
As I wrote in coverage of an excellent interview with Semantic Web scientist Yihong Ding - once our software is capable of deriving meaning from web pages it looks at for us, then there's a whole lot of work that's already been done, allowing our creative human minds to reach new heights. By pre-processing online content for us, Semantic Web technology lets us start from a point of higher abstraction.
Richard MacManus called Semantic Web application Twine possibly the first mainstream consumer semantic web app, but there's a whole lot of innovation going on in this space. Major companies are starting to leverage Semantic Web technology under the covers of existing websites as well.
Semantic Web Resources:
*RWW Semantic Web Feed Favorites OPML file (save link)
*RWW SemWeb Feeds - Best of Feed (copy and paste to your reader)
*Click to preview the above feeds before subscribing (pop-up window)
*RWW SemWeb Sites Search (Visit and Bookmark)
While Michael Arrington says the release of the iPhone relieved him of any pressure to build a mobile version of TechCrunch, I think there's still going to be a whole lot of innovation in the mobile space well into the future.
Most of the people online in this world access the web through a tiny little computer they carry in their pocket and also use as a phone.
Mobile means more than just small, though. It also means portable, fast, location-aware and tied to voice, media and the meat-space.
Mobile is already a great analogy for data portability in general - people are thrilled in the US that we can now switch cell phone carriers and keep the same phone number. Imagine if we lost our contacts when we switched phones. The same type of expectations are totally reasonable for services online.
Once mobile really gets tied into open data on the web, to recommendation engines and to the semantic web - then we'll be cooking with gas.
Mobile Industry Resources:
*RWW Mobile Feed Favorites OPML file (save link)
*RWW Mobile Feeds - Best of Feed (copy and paste to your reader)
*Click to preview the above feeds before subscribing (pop-up window)
*RWW Top Mobile Sites Search Engine (Visit and Bookmark)
We're only going to get so far if we just tell the world, "trust us, all this ephemeral crap is going to change your life!." A big part of why there isn't widespread consumer demand for OpenID is that the benefits of it haven't been clearly communicated. The concept is gaining steam almost in spite of the communication of its advocates, many people believe. The future of OpenID and many other key technical innovations, lies in communicating with people about what they can do with the tools. That is not easy to do with things that are complicated or new.
Just as video has changed the web forever because visual communication is infinitely more evocative than text - so too do I expect the perceived value of visualization to grow by leaps and bounds in 2008.
I wrote a post about 3 methods of visualizing best practices in social software design over Thanksgiving, highlighting the work of Thomas Vander Wal, Chris Messina and the Google OpenSocial team. To that list I'd like to add Dave McClure's SlideShare archive, where you'll find images like the one below. This stuff is pure gold. Powerpoint is the future? Well, effective visual communication of complex data-based concepts is going to be an invaluable part of the future.
Visualization Resources:
*RWW Visualization Feed Favorites OPML file (save link)
*RWW Visualization Feeds - Best of Feed (copy and paste to your reader)
*Click to preview the above feeds before subscribing
*RWW Top Mobile Sites Search Engine (Visit and Bookmark)
I hope these resources will prove useful for our readers in the coming year. Please let us know about any sites that ought to be included here - or let us know if you think I'm barking up the wrong tree and these won't in fact be hugely influential trends in 2008. Thanks for getting all the way through this long post!
Comments
Subscribe to comments for this post OR Subscribe to comments for all Read/WriteWeb posts
There will be a huge growth and interest in nich social networking in 2008. Niche sites offer advertisers highly targeted demographics.
Users also want to connect with like-minded communites of people rather than get lost in the huge populations of facebook, myspace et al.
Search engines like findasocialnetwork.com help users and advertisers locate niche social networking sites thus promoting their use and visibility.
Posted by: Fason | January 2, 2008 2:54 AMhttp://www.readwriteweb.com/RWWOpenData2.opml
seems to be invalid
Posted by: Heimetli | January 2, 2008 3:10 AMGreat post Marshall, thanks.
What do you think about the likes of Qik allowing live streaming from mobile devices? Do you see this having any impact in the year ahead?
Posted by: Tom Raftery | January 2, 2008 3:13 AMMarshall, your OPML file is valid according to my validator.
http://validator.opml.org/?url=http%3A%2F%2Fwww.readwriteweb.com%2FRWWOpenData2.opml
Dave
Posted by: Dave Winer | January 2, 2008 6:18 AMDr. Rick Hangartner said...
In the near term, search engines will increasingly incorporate simple recommender technologies to handle approximate queries (e.g. “You asked for this, and based on similar queries/behavior by others, you might be looking for this.”).
This is a monumental task, it is doable, but I am not sure when. There are new algorithms which have been published in the computing literatures over recent years that can do these sort of thing more efficiently but I haven't seen wide adoption of those techniques by commercial vendors. These new algorithms (a few them and also there are many variants of these few) are from the domain of Multilinear Algebra and Tensor Calculus (these concepts are the same - but slightly different). They are not new, since Einstein developed his General Theory of Relativity using Tensor Calculus almost a 100 years ago (1916). Even though they've been around for a long time, the data-analysis community & researchers recognized their potential use in intensive task as search engine, recommendation engine, speech processing, image processing, etc, ... over recent years.
Current search engines & recommendation engines, etc, do use linear algebra algorithms, where the input data to the algorithm is a 2-dimensional matrix (rows & columns) only. In Google's PageRank, the input is a matrix of document by document links (outwards & inwards) frequency. For textual search engine, the input is a 2-dimensional matrix of words by documents frequency. For a recommendation engine such as Amazon, the input data is a 2D matrix of users (customers) by items frequency.
Currently, Google computes both PageRank and text content (latent semantic indexing - LSI) individually. I don't know how Google combines those 2 indexes together as one retrieval during a live search query. With the advent of multi-linear algebra, the input data becomes multi-dimensional (more than a 2D matrix). The data could be 4-dimensional matrix such as links of documents by documents by words by documents. See, this one multi-linear algebra algorithm achieves in one computation, the equivalent of 2 different algorithms that Google are currently using (PageRank which uses a document by document 2D matrix and LSI which uses a words by document 2D matrix). But in multi-linear algebra, any dimension is allowed. So, for the task that Dr. Rick Hangartner described, where recommendation is incorporated into web-search engine such as Google, the input data is a 6-dimensional matrix such as links of documents by documents by words by documents by user by query (word/s). The beauty is , it is just one algorithm, rather than a combination of PageRank, LSI, recommendation, which is doable, but it would be very difficult. Another hypothetical situation, of the power of multi-linear algebra is that it has no limits in theory (perhaps only by computer memory) in the dimensions it can crunch, eg: Say, that a search vendor who uses a multi-linear algebra algorithm for its system, decides to localize the recommended search (for a specific state or a country, etc) and in a specific season of the year. Such input data is an 8-dimensional matrix such as links of documents by documents by words by documents by user by query (word/s) by location by season (eg - winter, summer, spring, autumn). In fact these data exists today (search query terms that are related to seasons & locations), but they're pretty difficult to compute & index. If they're being done today, it is done via SQL filtering (for locations only) first, which are then narrow down to the document indexing.
Multilinear algebra algorithms not only apply to search but also to recommendation. For, Amazon, it could be something like the following (3-dimensional matrix):
Customers who bought item A also bought item B and also look (browse) at item C
Vendors such as Google, Yahoo, Microsoft , IBM, Amazon are already aware of recent publications in multi-linear algebra. Either they're working & researching on their own version (which the public would never see, since their work is commercial, except Microsoft & IBM who frequently publish their researches in publicly available peer review computing journals) or contemplate on developing the algorithms that are publicly available in the literatures. All these search engine vendors were present at the Workshop on Algorithms for Modern Massive Data Sets at Stanford in 2006. The presentations are freely downloadable from this site and that includes all slides for presentations in multi-linear algebra & tensor calculus.
I believe that such search capability is just around the corner.
Posted by: Falafulu Fisi | January 2, 2008 8:02 AMHi Marshall,
Nice post. I think you comments on visualization are spot on. I find Information Aesthetics (http://infosthetics.com/) has some great coverage of innovations in representing data online.
I'm much more skeptical of mobile making the big break-through. I've always thought it lacks a compelling user need plus advertisers that I have worked with aren't nearly ready to target users at a GPS level.
Added to that I just received my first location based mobile ad delivered by Bluetooth from a shop I walked past. Now I'm adding bloody annoying to my list of reasons. The shop has a window, I don't need a mobile ad to tell me what great stuff they have inside!
Posted by: Nigel Eccles | January 2, 2008 9:18 AMMarshall this is an outstanding post!
Thank you for putting all these resources together so well. I have loaded all the feeds and bookmarked the search engines.
Posted by: Neil Hinrichsen | January 2, 2008 9:46 AMGreat post, thank you!
re this quote:
"Just as video has changed the web forever because visual communication is infinitely more evocative than text - so too do I expect the perceived value of visualization to grow by leaps and bounds in 2008."
I don't think it's necessarily true that visual communication is infinitely more evocative than text. What is true, however, is that many people find it much easier to create compelling video than text. Video has broadened the pool of web contributors enormously.
Ben
Posted by: Ben Nevile | January 2, 2008 12:17 PMGood post. Very informative. Thanks a lot for this. Digged.
Posted by: Villa Orlando | January 2, 2008 1:59 PMI think we are in the era of visual communication, visual communication has changed the way we browse the web. This post is very well put together.
Posted by: Post On Fire | January 2, 2008 2:16 PMHi Marshall,
Thank you for the post and mentioning my work again. Semantic Web is about to have some big breakthrough on creating real-world applications in 2008 or the least 2009, I believe. But we may not wait for W3C to lead the practices (anyway we have waited for many years). Industrial companies such as Radar Networks and some new academic efforts will push the practical side of Semantic Web research more aggressively. In person, I will graduate this year and look for an academic offer at present. I expect to conduct more productive research projects and have more collaboration with the industrial realm in future.
Happy New Year!
Yihong
BTW: For RWW readers, if you are interested in investing or sponsoring research on building real-world Semantic Web applications, I would more than happy be contacted and discuss my projects with you. Thank you, Marshall, for allowing me prompting myself a little bit here.
Posted by: Yihong Ding | January 2, 2008 2:51 PMThanks Yihong Ding, yours is the most relevant self-promotion in comments that I've seen all week. You rock on with your bad self, as they say. Thanks for commenting.
Posted by: Marshall Kirkpatrick@Fason,
Posted by: Khurt L Williams | January 2, 2008 3:46 PMWhen you say niche social networks are you talking about things like Ning.com and KickApps?
dude, you are on some serious hustle with your blog posts. you've taken so much work off of my plate! thanks
Posted by: Baratunde Thurston | January 2, 2008 4:22 PMNew year, new life? Time will tell... In the meantime, I invite you to read my latest post on my blog about a new definition of Web 2.0 and the proposal for an exciting Web 2.0 world-wide project: World 2.0. That article is in English language, even if the rest of blog is in Italian language.
Posted by: Dario de Judicibus | January 2, 2008 4:50 PMps. For those who think there is another .com bubble burst comming, have a look at these links. You'll change your tune.
-mind blowing demo at a recent Microsoft tech show
http://www.ted.com/index.php/talks/view/id/129
-The new face of the web, called Silverlight. In 2 years, the web will never look the same.
Posted by: JB | January 2, 2008 5:39 PMhttp://silverlight.net/showcase/
all I have to say is that 2008 is going to be an exciting year.
Posted by: spuds | January 3, 2008 12:26 AMGreat work Marshall.
For those who use Netvibes I've created a tab for each of these which you can access from the link below (or click for a specific one).
http://eco.netvibes.com/search?q=ReadWriteWeb
Subscribe to Recommendation feeds
http://eco.netvibes.com/subscribe/229677
Subscribe to Semantic Web feeds
http://eco.netvibes.com/subscribe/229678
Subscribe to Mobile Data feeds
http://eco.netvibes.com/subscribe/229679
Subscribe to Open Data feeds
http://eco.netvibes.com/subscribe/229676
Subscribe to Visualisation feeds
Posted by: Neil Richards | January 3, 2008 3:54 AMhttp://eco.netvibes.com/subscribe/229680
@Khurt, I'm referring to niche social networks created using things like ning, kickapps, snappville et al.
Posted by: Fason | January 3, 2008 8:20 AMhttp://findasocialnetwork.com is a search engine to help find and promote niche social networks.
I've done something very silly! I've made available an (almost) complete list of the RSS feeds to which I subscribe in my main Google Reader account. It's not a huge number, just a couple of hundred, but Sciencebase regulars might like to check it out here:
http://www.sciencebase.com/david-bradley-rss-subscriptions.html
I deleted a few of the more embarrassing ones that were in there, like Ricky Gervais, Russell Brand, Rush is a Band etc, so you're left with the ubergeek stuff.
Dave B
PS There is an OPML version available but it's not quite as current as this page.
Posted by: David Bradley | January 3, 2008 11:11 AMNot only very informative, but educational in the most effective way by delivering information using the resources about which you're teaching us! This is surely how information will be passed in the future, as access to filtered streams rather than specific facts....(meta-blogs???)
Posted by: ian fenwick | January 6, 2008 1:29 AMI am taken aback by the variety of social search and bookmarking sites that have popped up. It's only going to grow bigger. http://www.grandcanyonquotes.com
Posted by: Gary | January 7, 2008 8:57 PMhttp://www.slideshare.net/josanku/web20-trends-updated/
Posted by: John | January 11, 2008 10:33 PM