The new interface is streamlined and more social with an updated relationship system. A focus on consistency (e.g., limiting user control of visual elements) and removal of clutter (e.g., presenting tags in a drop-down menu rather than a cloud) characterize the design changes made. Also, a few tweaks to group sharing were made to help reduce share-spam.

The "Favorites" tab has been renamed "All" and includes every site a user ever liked, including multimedia content. Similarly, the "What's New" page, which displays sites a user's friends and subscriptions reviewed, has been renamed "Recent Activity."

In the new interface, shares are found in the main Discover tab and display the list of sites that were sent to a user by mutual subscribers with direct-share permission. Soon, the site will display sent as well as received shares.

The once-retired Similarity Meter, a diagram showing a users' Favorites similarity with other users, is being resurrected. And by popular demand, group shares will no longer be available in bulk; users will have to select other users within groups to share items with them.

Design changes that limit user choice in favor of overall site consistency are threefold: List view rather than a grid view is the default for all profiles. In contrast to the 22 background colors once available, a mere seven of the most popular color themes are now being offered for profile customization. And the visual editor has been scrapped. "Rest assured," the site reads, "we are still working on making sure all your past blog entries look good in the new interface. Some text customization will still be allowed.

"Our goal is not to limit your creativity, but to place emphasis on content and ensure a consistent user experience." An admirable aim, indeed; we think StumbleUpon is working in a good direction with these changes.

We also admire the new drop-down menu for tags:

Users are able to toggle back and forth between the beta redesign and the old site design over the next month. "To switch back to the old look," the site says, "simply click on the Beta link located on the main navbar (see image below) and on the following screen click the Leave Beta button. To return to the new interface, use the link Click here and try the new StumbleUpon! located at the top of all pages in the old view."

Today, Faviki is releasing an upgrade to their service which will give you even better control over the tagging process, making bookmarking even easier than before. They're also announcing support for OpenID.

The biggest upgrade today is Faviki's enhanced tagging interface. In the past, Faviki struggled with some of the tag suggestions pulled out of Wikipedia because they were too long and too hard to enter for practical use. Plus, users wanted to use tags of their own creation, not the tag suggestions.

For example, if someone is tagging an article about the soccer player "Filippo Inzaghi," they may want to tag it by the player's nickname "Pippo." Before, this was not possible. But now, if Faviki doesn't understand a tag, it will pull in possible matches and ask you "What exactly do you mean by ______?" After you pick your selection, Faviki will remember your choice.

This is an important change for the service because it means users can tag web pages any which way they want, but they're still linked to the structured data on the back-end. That way, when someone searches through Faviki's community tags, all the web pages for that particular item or concept will appear, even if people tagged them using their own personal keywords.

Beyond Wikipedia

Another change in Faviki's service is the ability to define new tags. Prior to today, the service was limited to searching Wikipedia for tag suggestions, but now it has the whole web at its fingertips. If a tag is entered which doesn't match anything from Wikipedia, Faviki will search Google for relevant URLs and then ask if the links presented represent the same tag. As multiple users go through this process, Faviki learns what URLs best represent that concept and adds the new tags created by the users to its database.

API, OpenID, and More

Faviki has also just launched a Save/Edit API that provides a way to save and edit bookmarks from other applications. In addition, they've introduced support for OpenID. Other new features arriving today include a smarter autocomplete list, the ability to convert tags, spam control, the ability to export/backup your bookmarks, and a new tag description tooltip.

The only issue we have with Faviki is the same one we had before: there's still no import function available. That means you'll have to leave your extensive bookmark collection behind if you want to use this service. We suppose that it could be difficult to properly tag and match all of our old bookmarks, but without this feature, Faviki doesn't have the best shot at attracting the heaviest users of social bookmarking services.

But now, that's all changing...or so they say. The company has basically scrapped their original concept in a revamp that's more of a "makeunder" than it is a "makeover." The new Secondbrain focuses on making bookmarking simpler while ditching most of the service's other features.

What's Gone from Secondbrain

Regular Secondbrain users will have to deal with the most dramatic fallout from this switch, starting with the fact that some of their content has gone missing. Imported content like pictures, videos, bookmarks, etc. from other social media services no longer exists in the new Secondbrain. Only content imported manually or with the Secondbrain bookmarklet remains.

The new Secondbrain also no longer does social media synchronization or importing of Delicious bookmarks. These are temporary limitations as the company decides on how to reintroduce these features in a more user-friendly way. For now, Delicious bookmarks can only be imported manually by browsing for and selecting your exported bookmark file generated by Delicious for import into the service.

Also gone is the lifestreaming-like feature which let you follow other users and all the content they were sharing. Perhaps realizing that sites like Facebook and FriendFeed dominate in this area, Secondbrain has decided to switch this option off, now allowing you to follow specific collections only. These collections are sets of aggregated content (blog posts, photos, videos) on a particular subject. Here are some popular collections to give you an idea. All content added to Secondbrain has to go into a collection now, but it no longer has to be tagged - that has become an optional feature.

If you choose to use this part of the service, Secondbrain almost becomes an alternative RSS reader of sorts, pulling in filtered lists of "best of" content on topics you care about. Even better, you can just grab the RSS feeds for the collections themselves (a feature added last week) and pull them into your preferred feed reader instead.

Is It Worth Revisiting?

Overall, the new service may appeal to those who are still actively using social bookmarking and unlike similar sites like Delicious or Diigo, you don't follow people, you follow specific sets of content people create. (Diigo's "groups" feature would be the best comparison).

Still, the bigger question about the new Secondbrain isn't whether or not the new simplified service will appeal, but whether or not social bookmarking is even all that hot of a service anymore. It almost seems as if social bookmarking was just a pre-cursor to the social media sharing types of services we use today, like Twitter or Facebook. Because, really, if you want to casually share a link with your friends, what services do you turn to these days?

Social bookmarking still makes sense in some cases - like organizing research, sharing all the links discussed in a podcast, or compiling topic-based resource guides. In those niches, Secondbrain could still have a shot at staying afloat, but their real shot at glory may have already come and gone.

On the infrequently-updated company blog, the sole remaining employee Mike Koss shared the news about the additional funding, saying it was a "small round" but enough to enable Faves.com to operate for 2+ years, even without increasing the site's revenue.

He also shared some of the plans he has for the new Faves.com, a site that will no longer be solely focused on social bookmarking in the traditional sense, but will try to increase its relevancy in today's era of the social web by allowing for integration with Twitter, Facebook, and other services.

Koss also talked about the Ma.gnolia disaster and, in the spirit of full disclosure, wanted to make sure Faves users understood the status of the service. Specifically, Koss noted that Faves exists in a fairly large-scale data center with four front-end web servers and six back-end database servers. Nightly incremental backups are made as well as weekly complete data snapshots, but no recent full-scale data recovery test has been done. He also said that improving site performance and reliability going forward would be one of the challenges he has to face.

On the one hand, it was refreshing to hear Koss speak honestly about where Faves is now and where it's going, but on the other, it can easily make one feel a little wary of using a service that was only recently gasping for air. That said, as long as Faves.com isn't your primary service for storing your bookmarks, it's probably going to be OK to just enjoy using it as the new social utility it aims to become. Now the question is: will you?

LookSmart acquired Furl in September 2004. At the time, the company announced that it wanted to leverage the data it gathered about users' online behavior through Furl in order to create better search algorithms for web search. Over time, however LookSmart has withdrawn from improving search and the company sold off its FindArticles search engine to CNET in late 2007. Given that Furl is not part of LookSmart's core business anymore, it only makes sense for the company to offload responsibility for the service. We can't help but wonder if LookSmart actually tried to sell Furl but, given the current economic climate, wasn't able to find a partner that wanted to acquire the service directly.

A Good Fit for Diigo

Diigo looks like a good fit for Furl, as it also provides social bookmarking services at the core of its product. Furl at one point looked like it had a bright future ahead of it, but it never really took off. In February 2008, Furl launched a major redesign of its site, which included interesting new features like cached-copy archiving, export in bibliographic formats, and recommendations. According to Diigo, Furl has about one million users - though, as usual, it is hard to say how many of these are active users.

It is not clear how (or if) Diigo will integrate Furl into its own product, but in the announcement, Diigo stresses that it will release version 4.0 of its service soon. Whether this will already feature some integration of Furl's features remains to be seen. For now, Diigo has released a tool that allows Furl users to easily transfer their bookmarks over to Diigo.

When we first reviewed Qitera, we already liked it a lot. Qitera's main emphasis is on creating a search engine for your bookmarks, no matter whether the bookmarked sites were on the public Internet or hidden behind the paywall of a for-profit database. Qitera is a very capable social bookmarking site with all the standard features you would expect, but what really sets it apart from its closest competitors is its search functionality.

Search Engine Integration

Whenever you do a search on Google or Yahoo after installing or updating the Qitera plugin, the top three search results from your bookmarks will automatically appear at the top of the search results. These results will include your own bookmarks, as well as bookmarks from your friends and colleagues on Qitera.

As Qitera's Joerg Lamprecht points out in the announcement, we often search for the same things over and over again (and so do our friends). By first displaying search results from sites that you have already bookmarked before showing results from the Internet, you can break this cycle and get faster access to a set of search results that you have already marked as relevant by bookmarking them in the first place.

From a business perspective, this integration with Google and Yahoo also makes perfect sense for the company. By seeing the Qitera bookmarks every time you search, you are far more likely to return to Qitera and make it the default app for your bookmarks.

Last month, Qitera also released a better integration with Yahoo's delicious. You can now import your delicious bookmarks to Qitera, though sadly, this is only a one-way street so far. We would also like to see the possibility to sync your Qitera bookmarks back to delicious.

In the last two months, Qitera has also addressed a few of the other problems that we discovered when we first tested the site. You can now, for example, get an RSS feed with the latest updates from your social network on Qitera, and the service now also finally recognizes duplicates.

Coming Soon: Qitera for the Enterprise

Qitera has also announced that it is developing an enterprise version of its service, which it will be launching at Cebit in March.

First of Its Kind?

We think this integration with Google and Yahoo is a great step forward for Qitera, and we are not aware of any other social bookmarking service that has created a similar plugin. If you haven't tried Qitera, we definitely think it is worth a try.

For a more in-depth look at all of Qitera's features, please refer to our review of the service from last December.

To mark its big day, we offer below two videos. The first an introduction to the tool for readers still unfamiliar and the second a screencast demonstrating just how easy and useful it is to make 5 changes to your Delicious experience. Those changes took us under 5 minutes.

We also want to take this opportunity to thank the Delicious team and especially now post-Yahoo founder Joshua Schachter, for making this awesome service what it is. We really appreciate it.

First, an Introduction

Thanks to CommonCraft for another great video.

And Now for Something New

The following video demonstrated five of our favorite ways to use Firefox plug-in Greasemonkey to radically change the Delicious experience. This is really easy to do, as you'll see, and we've included all the links below the video. With just a handful of clicks you can integrate Delicious into sites like Google Reader and Digg, you can sort and view Delicious in brand new ways, and make a number of other changes.

Note that there's no audio in this video, we just went through the steps. We hope that's ok for readers but if you'd prefer it be narrated, let us know.

Links shown in the screencast:

Greasemonkey

Delicious for google reader

Sort by popularity or other

digg.licio.us

subscribe in delicious

Favicious

Autopagerize

The extension lives on the right side of the browser window and can be expanded and closed at any time. Comments can be kept private, or shared with individuals and groups, as well as with the whole ReframeIt community. Once you share a comment, other users can rate it, bookmark it, or share it with other users. On pages with many comments, you can also filter annotations by ratings, number of replies, and shares.

Lots of Social Features

ReframeIt also provides its users with extensive social features on its web site. As has become standard for social sites, you can also 'follow' other users and be notified when they leave comments.

ReframeIt's community is mostly organized around groups. Currently, most of the active groups on the service are clustered around political topics, but the company also envisions school or college classes to use the tool for reading and annotating texts online.

Competition

ReframeIt's closest competitor is clearly Diigo, which has a more extensive feature set and also puts more emphasis on social bookmarking and discovery.

If you are mostly interested in sharing annotations, Reframe It is a perfectly good and easy to use solution, especially if you just want to share those annotations with a relatively small group of friends, colleagues, or classmates. If you need more advanced features, however, Diigo represents a better solution, especially if you consider that ReframeIt only represents a sub-set of Diigo's functionality.

Our main complaint with the service is that the browser extension does not allow you to access your social network and past comments without having to go to ReframeIt's web site.

Overall, however, we were impressed with the user interface design and functionality of both ReframeIt's site and browser extension.

For a quick look at how ReframeIt's interface works, here is the company's introduction video:

For those who already have established communities on reddit, being able to point your own domain to reddit without having to host the open source version of the service yourself is a great step forward. Also, being able to customize your CSS stylesheets gives you the opportunity to make reddit fit into the design of your own brand. This makes using reddit a lot more palatable for those with established names who would like to experiment with social news sites, but shied away from it so far.

Reddit definitely doesn't seem to be afraid of giving up control. In the end, though, this move is only going to help it grow its audience - and while the audience might sometimes not even be aware that they are looking at a custom reddit site, reddit itself will surely run advertising on those sites, so its bottom line is only going to benefit from this.

One thing reddit doesn't allow you to do, though, is to create your own voting algorithm - though judging from the direction the developers have been moving in lately, this is probably only a question of time.

Contest

Also, reddit has announced a contest for those who want to start their own community on reddit. Whoever manages to create the largest sub-reddit within the next month can take away a Macbook Air and a reddit bobblehead.

Reddit Keeps Growing

Reddit also announced that they have seen a 300% increase in subscribers and subscriptions since unveiling their latest redesign in May.

While reddit is still much smaller than Digg or Yahoo Buzz, it is definitely driving the development of its site forward a lot faster, and, at the same time, pushing its competitors to become more open and creative as well. While Digg is trying to keep very tight control over its service, reddit is moving in the opposite direction and judging from the numbers cited in this announcement, it is working out quite well for them.

Ma.gnolia, one of the most popular second tier social bookmarking services on the web, announced today at the Gnomedex conference in Seattle that the company has thrown itself to the mercy of the development community and is going to make its code available in open source.

Ma.gnolia is tiny compared to Yahoo's Delicious, but in every way other than network effects, it's more interesting. Unfortunately, that's all pretty academic. When interesting competes with the powerful network effects that come from the huge number of Delicious users - Delicious wins. Today's announcement may help grow Ma.gnolia quickly; at the very least it's a daring move.

There aren't a whole lot of people using Ma.gnolia, but a lot of them are people who place great emphasis on web standards, data portability and other forward looking ideas.

The coolest features of Ma.gnolia?

• OpenID support, so much support in fact that Open ID is the only way you can get a Ma.gnolia account now.


• Attention Data - Ma.gnolia may be the biggest service provider that gives meaningful support to the APML attention data standard, letting you port your taste data from their service over to any other service that will import it.


• OAuth support. You can give 3rd party apps access to your Ma.gnolia data without giving them your password, because Ma.gnolia supports the open user authentication standard OAuth. That's great, now we'll see how many 3rd party apps care about the relatively tiny set of Magnolia user data.


• Customer service by IM. It's great. The company uses Twitter and Get Satisfaction but it also offers support by instant messenger! That's really nice. Not just any IM, either, but the OpenID savvy Pibb. Pibb has RSS feeds for chats, which means you can drop a question into the help chat, subscribe to the feed, then get notification by RSS when someone replies and jump into the chat room to IM with them. It's great.


• Social features. Delicious is trying to be more social than it used to be, but Ma.gnolia does a much better job of it. Sharing with groups is really easy and you can "thank" anyone for bookmarking a particular URL. That's cool.


• It's pretty. Ma.gnolia looks nice and is relatively accessible (maddening site architecture aside). There's screen cap thumbnails all over the place, including in the daily auto blog posts populated by bookmarks.

Why an Open Source Federation is Cool

The value proposition is really not clear here but I think this is it. You get a local social bookmarking system with your additional features and your branding, but with the network effect of access to all other Ma.gnolia users' bookmarks for research and sharing. If enough people go for it, it will be cool. That's the classic problem for social media tools and it's not going to be an easy one to over come. Open sourcing the service is a bold step that could be a big win. We do hope it will work.

A Million Ma.gnolias Will Bloom?

Will communities all over the web download, customize and participate in a federated Ma.gnolia? Maybe. It's hard to know. Unfortunately, Ma.gnolia founder Larry Halff's presentation announcing the open sourcing of Ma.gnolia here at Gnomedex illustrates the problems the company will continue to face. Just like the service Halff created, the man himself seems like a brilliant guy who you know has great ideas but communicates them poorly enough that it frustrates people pretty quickly. The value proposition is unclear, the site architecture is frustrating - right now it's a service for standards true believers. This author uses it personally, though almost every time I do I grumble and ask whether I should go back to using Delicious.

If a federated Ma.gnolia thrives and can capture the network effects so important to social bookmarking, it will be because of the value of open technology and in spite of Ma.gnolia's own struggles. Network effects are key to social bookmarking because a large number of users makes these services a place where you can discover cool things quickly, where popular items for a given tag have risen out of a large number of candidates and where the things you bookmark can be seen by a lot of people. If you want to do research, you want to do it at Delicious right now, not Ma.gnolia.

We hope that this strategy serves Ma.gnolia well. It does offer a business model in charging companies for custom implementation, but we'll see how many companies go for it. Companies may be better served by a social bookmarking service aimed at enterprise use, with solid granular privacy controls, like ConnectBeam.

Will you consider implementing a local, federated version of Ma.gnolia? The code should hit the web in December, so there's plenty of time to contact the company and make sure it gets open sourced in a way that works for developers. Let us know your thoughts in comments.

How do website readers prefer to share stories they find with friends? According to the company behind the widely used sharing widget ShareThis, after emailing a link, the most popular method of sharing is now Facebook. The numbers are interesting - but there are also some big caveats to keep in mind.

In our enthusiasm for Web 2.0 style tools, many blog publishers may forget just how popular sharing by email is. It's clearly the favorite method. Email sharing does tend to be one to one however, having items shared on Digg or Facebook has the potential to reach many, many more people.

The big surprise here, though, is that Facebook and MySpace have emerged as hugely popular ways to share items from off-site. Have they found greater mainstream success in the relatively short time these sites have supported item sharing than dedicated social bookmarking sites have in the years they have been online? It appears that may be the case.

We found these numbers via Amit Agarwal's blog, which is always a great place to discover new things about the web.

Why This is Important

When publishers add the ShareThis system to their websites, they can choose which services to include buttons for. It's an important detail to take into consideration and knowing which services are most popular can help make this decision. Here at RWW we don't use ShareThis, we use another service called AddThis. Looking at the numbers from ShareThis, though, would lead us to believe that sharing by email needs to be added and sharing by Facebook needs to be given higher billing in our widget. Other sites might make other decisions based on this data. GigaOm, for example, doesn't offer sharing by Facebook at all - something our friend Om might want to change.

Caveats

A few things to take into consideration, however, include the following:

• Your site's audience may vary. Different communities around different content topics probably have different trends in the sharing tools they use. We assume, for example, that there aren't a lot of people sharing ReadWriteWeb stories on MySpace - but maybe we're wrong!
• Some of these services use bookmarklets. These numbers aren't for all sharing, just sharing that goes on through the ShareThis widget. Delicious users, for example, don't necessarily think of what they are doing as sharing (it's often bookmarking for personal use) and that service has its own bookmarklet.

None the less, the take away here for us is this: email, Facebook and MySpace are very popular ways for people to share things online. Publishers neglect them at our own risk.

Vi.sualize.us is a great social bookmarking service for images. You can think of it as the Delicious for images. A place where the cool photos hang out. If you head to their homepage, you'll see a ton of great images that users of the service have recently added. From fashion to architecture, you'll have a blast going through the archives of users on vi.sualize.us. We recommend using their Firefox plugin to bookmark images whenever you'd like by simply right-clicking on an image to save it. You can also embed your latest bookmarked images on your blog or subscribe to a friend's RSS feed for vi.sualize.us.

FFFFOUND!

Ffffound! was one of the first image bookmarking services to hit the scene. With a fantastic selection of images to browse through, the only downside to this service is that it's invitation only. A bookmarklet and IE extension are provided to make it easier for users to bookmark images to their Ffffound account. A neat twist that this service has is that the more images you bookmark, the more the service will recommend personalized images for your viewing pleasure for a more dynamic experience.

We Heart It

We Heart It appears to be more of a Ffffound! clone with a better user interface. Still, the service is pretty appealing with great images, social tagging, and commenting, which Ffffound! happens to lacks. We Heart It provides a bookmarklet for saving images and videos from Vimeo or Youtube. However, this bookmarklet does not work in IE.

Bookmarking Taken to the Next Level

Bookmarking is going further than we might have originally expected. We can now bookmark images, links, save webpages just the way we found them and tons more. I'd love to see these sites incorporate the ability to upload photos that we take from our cellphones. After all, those are images too. Let us know what you think the next step for bookmarking will be. With We Heart It's ability to bookmark videos, what will we see next?

The foundation of Socialmedian is its user created 'news networks.' Right now, there are over 1000 different networks on the site. Given the nature of the service, it is no surprise that most of them deal with technology in some form or another. Users can add links to these networks, but Socialmedian also automatically suggests stories based on keywords users can add when creating a new network. This is a very smart idea, as it allows even small networks to have a constant stream of updated news.

Vote Till You Drop

To share items on the site, users can either use a bookmarklet or 'clip' any story already on Socialmedian and share it on another network. One of Socialmedian's most interesting aspects is that it allows its users to vote on almost everything on the site. Users can, for example, vote on the ranking of the keywords and sources that are used to seed the networks.

With all this voting, it would be easy to consider Socialmedia as just another Digg, reddit, or Newsvine clone, but its personalized networks and extensive social networking features set it apart from most of its competitors. While both Digg and reddit are moving towards more personalized experiences, Socialmedian has made those the foundation of its product right from the beginning.

Keep Following

Out of all its core features, Socialmedian's social networking features have probably seen the most extensive remodel. Most importantly, users can now 'follow' each other. This is, of course, a similar model to what other social networks have done in the past as well, but this feature also allows you to see what topics and posts are most popular just among the people you follow, which puts a nice twist on the following/follower scheme.

Among some of the other new features are the ability to directly post to Twitter, as well as Socialmedian's use of Google's Social Graph API to help seeding new networks and recommending stories to its users.

Socialmedian is a very nicely designed social news network with a feature set far too extensive to cover in this short post. Thanks to its extensive alpha test and close interaction with its early users, almost everything on the site feels very well thought out and we did not come across any major bugs. Overall, we would be surprised if Socialmedian couldn't carve out a very nice niche for itself in the social news market.

Browzmi was founded by Travis Parsons in 2006 and is currently being developed by a five person team. Browzmi has not taken any venture funding yet. Registration for Browzmi is open.

Browzmi allows you to surf the web with your friends. The core of Browzmi is made up of its bookmarking and sharing function, as well as the built-in IM. The chat function, as well as the real-time updates on the site, are enabled through an XMPP and Jabber back-end.

It's important to note right away that Browzmi is not a screen sharing application. While you can follow your friends around the web, you are not sharing the exact same experience with them and you don't see their mouse cursors moving around etc. Browzmi instead focuses on real-time sharing and discussion.

This concept, of course, is somewhat similar to that of Me.dium or Yoono. However, the interesting aspect of Browzmi is that it does not rely on extentions at all.

Just Like a 'Real' Browser

The Browzmi interface mimics a regular browser, with a location bar and search form at the top, as well as sidebars on the left and right. Thanks to this, new users should feel right at home when opening up the application. There are also back and forth buttons, as well as a home button that takes you back to the Browzmi 'Everyone' page.

Once you surf to a site, you can comment on a page, bookmark it, and make your opinions about it known by voting up or down on it. You can also 'clip' pictures from a site, making Browzmi a photo sharing application as well. All your actions on the site are saved in a stream that you and your friends can access and which looks somewhat like a stream in Friendfeed.

Sidebars

The sidebar on the left side of the screen ("My Stuff") is mostly reserved for seeing where your friends on Browzmi are right now. Thanks to the "Surf with Friends" feature, you can follow them around the web and chat with them. This sidebar is also where you can access your favorites (which basically function as bookmarks), see your activity stream, and see your friends' online status.

The right sidebar is called "Explore More" and it allows you to see where other users are on the service right now, as well as related content on Flickr, YouTube. Here, you can also see what other users who are not in your network have said about any given site.

Thanks to the XMPP back-end, all of your friends actions are being pushed to you instantaneously .

Privacy

When we first reviewed Me.dium, which offers a number of features similar to Browzmi, we were worried about the privacy implications of the service. Browzmi allows you to make your stream public or private, but any of your online activity through Browzmi does, of course, show on Browzmi and is available for viewing by your friends or (if you don't set the privacy features right) anybody. Given that users are probably quite aware that what they are doing is public - that is, after all, the whole point of the service - we are not too concerned about this, but it is still worth pointing out.

Coming Soon

Browzmi is currently working on adding number of additional features to the site, including allowing users to sign into the site with their AOL, Yahoo, or MSN credentials, which would allow everybody to chat with their IM friends right from Browzmi and also to invite them to surf the web with them. Integration with AOL's IM system will be announced in the next few weeks and the others should follow relatively soon after that.

Browzmi is also planning on allowing its users to comment on specific items on a site. This is something a few other bookmarking services already allow, but without allowing for real-time sharing.

The Browzmi team is also going to release an API in the future, which will allow developers to add widgets and customize the Browzmi experience.

Verdict

We were skeptical of Browzmi at first. Why, after all, would you need a browser within my browser? However, with the IM integration, its various RSS feeds, and straightforward surfing experience, Browzmi is actually a lot of fun to use. If you used it with Fluid or Mozilla's Prism, you could even make it your standard browser interface if you were so inclined. While that is probably not even Browzmi's intention, it does open up some interesting possibilities for the future.

In our tests, Browzmi worked very well and we didn't come across any bugs. For some sites, where Browzmi might be problematic, you can download a Firefox extension, but in our tests, we never needed it.

Browzmi, like most similar services based on social interactions, is going to get more useful as more users start joining and as your list of friends growths. In many ways, it is complementary to other services like Friendfeed, in that it gives you an option for real-time sharing and discussion, while you can also import most of your Browzmi actions to other services through Browzmi's RSS feeds.

Once Browzmi adds more IM services, we think that it can grow quite rapidly and organically, as there will be zero friction for new sign-ups.

Features

The basic premise of Evernote is that you can throw all kinds of files at it and then search for particular words in those files later. Full or partial screenshots are easy to add as are photos emailed from your phone and text entered directly into the application. The company says you can take photos of signs on the street and labels on bottles of wine, then search for text in those images to recall them later. There's all kinds of other features like annotation, sharing, a widget etc.

Storage space is limited and Evernote is now announcing a premium account at $5 per month for 500MB per month of uploading, free users get 40MB each month. That 40MB will go fast if you're uploading full screenshots.

There are a number of services similar to Evernote, but few are as lightweight, inexpensive and multi-platform as it is. Last100 writer Dan Langendorf is excited about Evernote and says it compares well in terms of features to more heavyweight competing services like Yojimbo, Soho Notes and DevonThink. Another marginally related service is Iterasi (disclosure: a consulting client of mine) though that service and Evernote are not as similar as I suspected.

Evernote has a whole lot of promise that's well articulated in the company's demo video. Check it out and then read on to learn about the problems we had in trying to use the service.

The Problems

The OCR search in Evernote is far enough from perfect that it's a real disappointment. False positives are annoying, but missing what should be readable text means that Evernote fails to recall documents that it promised to find. We saw a high percentage of false positives and too many cases of failure to capture text even on screen captures of web pages. The webcam capture wasn't good for much as small images are too fuzzy, but we're told by other users that a good point and shoot camera can get business cards into the system recognizably.

We ran into other little problems like an inability to login to the mobile web interface, the desktop app freezing up and a Mac desktop interface that was not as intuitive as we would have liked.

Recommendations

If Evernote's OCR could improve then we'd love to see additional features like:

• the ability to capture the current contents of a particular application instead of just the screen or a portion of it.
• search from the desktop application.
• adding items by RSS feed.

We'll keep trying Evernote for non-essential uses, but we were really looking forward to it and are disappointed with the performance issues it faces right now.