Desktop Mashups: Closing the Gap Between the Web and the Desktop

June 9, 2009

Google and other web-centered businesses try to establish the web as a platform similar to an operating system. Many new web services use interactive GUIs to offer email clients, messaging, word processing and other services that are in their behavior similar to traditional desktop applications. Client-side scripting, well defined APIs, and asynchronous data exchange are the key elements of the interactive web interfaces that rapidly spread under the Web 2.0 label. Developers of mashups go even one step further and create new web applications by combining data from different sources. Numerous examples of excellent mashups show that the remix of data in mashups can have significant advantages compared to the individual web applications still providing the data. Unfortunately, the Web 2.0 has also some major disadvantages and creates new problems.

The most obvious technical risk of storing data on web servers or computer ‘clouds’ is the catastrophic failure of a web service that can create major problems especially when no backups or alternative programs exist. Besides such technical issues there are also very serious problems related to the code and the user’s data as Richard Stallman recently pointed out. He argued that the common business models of cloud computing get users over time into a trap of proprietary systems including the lose of control over their data, and paying eventually more than for an equivalent desktop application. Other unsolved issues include privacy and ownership questions related to user generated content. Stallman criticized also that the Javascript code of websites, although freely distributed, is in many cases so obfuscated and unreadable that it can not be considered to be Free Software.

A white paper published by MAYA Design questioned the technical reliability of the centralized cloud structures. The authors describe today’s cloud computing as “radical experiments to be performed by gigantic, non-redundant entities.” Their suggestion is that “until we understand a domain of endeavor very, very well, we should insist on decentralized, massively-parallel venues for dubious experiments. In the information economy, it means net equality, information liquidity and radically distributed services (and that’s pretty much the opposite of “cloud computing” as described today).” Also, they argue that short-term profits, and insufficient thinking about architecture, security, and resilience may guide developments in the wrong direction. Serious warnings about the risks of cloud computing come now even from more main stream journals such as The Economist saying that the economic advantages are associated with the customers’ risk of “losing control once again, in particular over their data, as they migrate into the cloud.” As a specific example they mention that it may be more difficult to change the service provider than migrating from one computer application to another: “For a foretaste of this problem, try moving your MySpace profile to Facebook without manually retyping everything.” Another report entitled “AppEngine doesn’t fit the needs of startups on the runway” described how limited data access made impossible to run a statistical analysis of the user data. Without the option to download the data buried in Google’s database the analysis has to run as a job on AppEngine but once the database grows bigger and the job takes longer it will hit AppEngine’s 5 second timeout and cannot be completed anymore.

The big challenges of keeping the freedom and independence from individual web services, and supporting the principles of free software on the web including full control over our data show that we need to develop alternatives to the risks of purely web-centric applications. We have to ask for example: Is it possible that we use mashups and other new developments of Web 2.0 without loosing the freedom to control our data and the application’s code? Mashups provide very interesting new concepts for integrating different applications. In fact, Google CEO Eric Schmidt expects that Web 3.0 will be “applications that are pieced together.” I agree that small scripts can be simple but powerful tools for redirecting and manipulating data on the web as well as on the desktop. This concept of integration has the potential to change our desktop paradigms. I first realized this when I started writing small Bash and Python scripts. However, I think the more important step was adding a tiny CGI server. This user-centered server running on my desktop allowed me to integrate both web services and desktop applications with some small but very efficient applications I needed.

Mashups require essentially only two things: 1) client-side or server-side processing, and, 2) proper APIs supporting efficient data exchange between client and server. Most mashups are relative small applications that improve usability by linking and remixing the data of more than one dedicated web application such as Google Maps, photo sharing, social bookmarking, social networking, web mail, and chatting. The data exchange among web applications is achieved by a simple old-fashioned text channel. In fact, this type of communication is very similar to the traditional data exchange between concatenated text processing tools in Linux/Unix, also referred to as pipes. The only important advantage of the text channel used for web applications is the standardized structure for sending data, e.g. “key1=value1&key2=value2”, and of character encoding for data to protect the control characters, e.g. equal sign. Unfortunately, these key concepts of web APIs that are essential for mashups are largely missing in our desktop applications.

The few programs with useful APIs that I know include Krunner and Korganizer both part of the Linux/KDE environment. Krunner has since KDE 4.2 an exciting new concept that allows not only starting programs similar to the command line interface (CLI) but shows also dynamic search results while the user is typing. Krunners numerous plugins include also calculations, searching for contacts, opening profiles, and other interesting links to different programs. Korganized includes both a GUI and a simple CLI for a calendar. The CLI allows to add, list, and delete calendar events. So, it can be used as API for scripting and data exchange with other applications. Most applications, however, do not support such data exchange although they would greatly benefit from higher integration at the desktop. Thunderbird, for example, would be much more useful if I would be able to place links in calendar event, ToDo items, notes, etc. pointing to specific emails with additional information. Ideally, Thunderbird would support both opening of the email and returning of their content for displaying it in a desktop mashup. Thunderbird would also benefit from a “command line” similar to Firefox’s URL supporting smart keywords and bookmarklets. Combining the desktop CGI server with Firefox’s smart keywords allowed me to send, receive, and remix data similar to mashups. Ubiquity on Firefox also promotes mashups and is a very interesting new concept except that without a mechanism like a CGI server it is unable to communicate with other desktop programs. I think an integrated solution for a CGI interface would promote integration with other desktop applications and mashups including the powerful ubiquity. It would allow us to develop very efficient user-centered desktop mashups that can basically integrate any desktop application that has a sufficient API. I think the common web standards for URLs provide an efficient and robust text channel for the exchange of structured data at the CLI level.

In summary, running a tiny CGI server on the desktop allows to build applications integrating different data very similar to mashups on the web. Developing such desktop mashups doesn’t mean that we need to abandon web applications but that we will have more freedom to choose and that we can integrate both desktop applications and web applications. I think the last point is especially important because it is a major limitation of web operating systems.

Running Windows 2000 on SuSE 10.2 with VMware

October 14, 2007

This article consists of notes, comment, and links supporting the installation of VMware Workstation and VMware Player. My primary motivation was to run programs in Windows 2000 on a SuSE 10.2 Linux host. Read more …

Securing SSH with DenyHosts

October 14, 2007

The frequency of brute force attacks on SSH is increasing but can be stoped with DenyHosts for example. A simple hack starts the program only to check during a login.

Some month ago, I noticed in the log files of a server an increasing frequency of SSH attacks. The attackers obviously used scripts to try long lists of logins such as common names and standard accounts. Fortunately, I had already changed /etc/ssh/sshd_config to deny access as “root”, and to permit access only for a selected group of users:

PermitRootLogin no
AllowUsers Dave Jack Scott

This was an initial step to minimize the risk of brute force attacks. To slow down the speed of an attack I also limited the number SSH daemons:

MaxStartups 1

Unfortunately, OpenSSH does not have any mechanism build-in to stop unsuccessful login attempts that come from the same IP address. So, I started looking for other tools and found DenyHosts and SSHDFILTER interesting. I preferred DenyHosts because it uses /etc/hosts.deny to block the access from attacking IP address and did not require IPTABLES.

DenyHosts can be automatically started as cron job or can run in daemon mode. Both methods can be used to check in defined time intervals the log file for failed logins and deny access from an IP address with more failed logins than the defined threshold. The configuration file for DenyHost offers very fine-grained settings. For example, the threshold for known users can be higher than for unknown users so that the tolerance for password typos can be higher than the maximum number of unknown login attempts. Another very interesting feature is the purge option that can be used to reset the number of failed logins after a defined time interval.

Running DenyHosts in daemon mode or as cron job leaves the time interval between two scans of the log files as time window for an attacker. The log files showed that the frequency of SSH connections during an attack was about one per second. To keep the time window short it would be necessary to run DenyHosts very frequently just to secure the SSH port while SSH access is not in permanent use. I thought running DenyHost once per login to check the log file would minimize the effort and the time window for attacks.

Eventually, I found a hack to realize it. It is quit simple. I configured DenyHosts to write all blocked IP addresses in /etc/host.blocked and use it in the first line of /etc/hosts.deny. This avoids that IP addresses would be added to the end and gives /etc/hosts.blocked a high priority because TCPWRAPPER stops scanning /etc/host.allow and /etc/hosts.deny when it finds a matching IP address. All logins from not blocked IP addresses get eventually to the last line where the SPAWN option is used to start the Python script DenyHosts:

sshd:ALL:spawn python2.4 /usr/bin/ –purge -c /etc/denyhosts.cfg: allow

Since then the log files have shown that the access from attacking computers was denied after the defined maximum number of failed logins was reached. DenyHosts is an excellent program that even allows running an external program or script as a plug-in. This could be the next hack to keep a record of denied IP addresses for statistics because it would be interesting to know if the attackers return after their IP address is removed from /etc/host.blocked. But using DenyHosts with a threshold of five failed logins for unknown users does not allow them to run a brute force attack anymore.

Related bookmarks — absolutely

The OpenSSH project recently asked for donation so that the development can continue. PLEASE CONSIDER TO DONATE!

The Evolution of Bookmarking — Bookmarks, Firefox, and

October 14, 2007

Saving bookmarks for websites has dramatically changed during the last years. The current rapid development looks even like an evolutionary jump.

When I started saving bookmarks for websites many years ago I had only about 200 that I could easily organize in a directory with not more then three levels of folders. This served well the purpose at the time. Later I started using AltaVista to search the web. Eventually I switched to Google as primary search engine because Google’s format of the search results made it more successful to choose websites that provided the information I was searching for. The big difference was that Google’s two lines of sample text containing the search terms showed the context while other search engines at the time were lacking this critical piece of information.

When Google became popular most people saw it as a new way of getting to websites and bookmarking became a less interesting feature of browsers. All the search engines were very successful when the information I was looking for were sufficiently characterized by a few words. So, it was always very easy to find the latest version of a popular program for download. But in many cases it was impossible to get an article back that I had read a few weeks earlier if there was no specific combination of words that could be used to filter it out.

The bookmarking eventually returned but the meaning had changed. Before, bookmarks were primarily used to revisit websites but at this point in time they became for many people the permanent history of evaluated websites. It was possible to go back to an article and get the information whenever it was needed. This is different from the common history of a browser that contains just any website and a large history can make it very difficult to find a particular website.

The number of my bookmarks started growing faster than before. The effort of organizing them in a directory became eventually too much, so that the bookmarks became just a list that was lacking any kind of context. All conventional bookmark managers still try to attack this problem by reducing the effort for organizing the bookmarks in a directory. But besides the effort there was another problem of directories. It was almost always possible to place a bookmark in multiple places of the directory tree but not in a unique place. When I saved it in one directory it could be that I would search for it a few months later in a different place. This is not only true for bookmarks but also for files or any information organized in a directory. A few commercial bookmark managers addressed this problem by using a database and keywords for each bookmark. But adding bookmarks was not very practical because it required to scroll through the long list of keywords in order add a few to the bookmark.

A few years ago Mozilla introduced a search bar in the bookmarks manager. Well, it was searching the titles only and titles do not necessarily contain the context that was important to me. For example, a website with the title “The Evolution of Bookmarks” would show up in the search results when I searched for “bookmark” but it would not in a search for “Firefox”. So, the hack was to bookmark the website and to add quickly the context that was important to me, e.g. “The Evolution of Bookmarks – Firefox GUI search annotation”. This changed dramatically the way I used bookmarks. Eventually, I spent a few hours and converted most directories into this pattern. The few remaining directories are exclusively used for the bookmark sidebar that contains frequently visited websites. A directory with main bookmarks at the top level and one or two sublevels provide a visual reminder and quick access.

Saving bookmarks and adding context information helps to bring back the websites quickly when the information is needed. I started thinking about an interface that would provide the functionality needed to make adding words as context easier, less time consuming, less vulnerable to spelling errors, and remind me of words I had used before. A long list with keywords with a scrollbar was not very appealing but multiple lines of space separated words that I could click at in order to add them or to remove would improve usability I thought. Also, the bookmark should show up at the sidebar or at the bottom of the screen as soon as I load the website so that it reminds me of previous visits, provides the context I had added, and any notes I made about the website.

Well, it turned out that this wish list was partially realized in, an online service for social bookmarking. Joshua Schechter has designed an excellent interface to add bookmarks and enter tags as context. There is no unnecessary link or text on the interface. It provides all tags that were entered before as links so that entering by keyboard and by the mouse can be combined. has the very best interface I have ever seen for a social bookmarking service. I’m sure some people will argue about that but for me the clear design of follows the wonderful principles of simplicity that are also used by Apple or Google. Using one of the famous bookmarklets for adding bookmarks to makes it very similar to saving it in the local bookmark manager.

The major difference between a tree hierarchy and tagging seems to me that looking at the hierarchy while selecting items provides some guidance in choosing the next item. This reinforces the relationships within the directory tree to some degree while the current tagging systems misses any relationship between tags – a critical issue that was also addressed in a whitepaper by Hans Reiser because it is important for the next generation of ReiserFS file system on Linux. Potential solutions to this problem include the definition of main tags and the support of tag combinations. It would be helpful if the user could define the main tags that show up in an extra line on top to reinforce the use of main categories. A second line could show the tags that have been used in combination with the selected main tag(s).

In the future, browsers will hopefully adopt the tagging of bookmarks and provide a faster response time than any web service can. Synchronizing public bookmarks with a web service such as could combine the advantages of local bookmark managers with social bookmarking and guaranty the privacy of bookmarks that are not public. Maybe we will see in one of the next Firefox versions a bookmark manager with an interface similar to the interface or even more advanced. The bookmarks in the sidebar fit more in the concept of a web panel than into the flat list of tagged bookmarks and could be separated if necessary. The integration of annotations and citations would be a big step forward. This does not mean that bookmarks and annotations need to share a database but bookmarks, tags, annotations, and website management icons for allowing software updates, allowing popup window, etc. should appear together in the user interface to show everything that is associated with a website. To show any saved information about a website as soon as the website is opened would make a big difference because it would provide reminders and annotations that can guide the browsing. What once was a bookmark manager may become the information management system that keeps together links, annotations, tags, existing copies of websites, and other meta-data. This looks very promising to me and could become the next generation of Firefox to “rediscover the web”.

Related blogs:
Rethinking the User Interface For Bookmarks — Two Different Purposes May Require Two Different Interfaces
Advanced Tagging — Hierarchical And Ordered Tags
Firefox Hacks :: Tagging Bookmarks
Firefox Hacks :: Search Simultaneously in Firefox,, and Archived Bookmarks
Is Yahoo going to be

External articles and blogs:
Security and Privacy Risks of Google’s new Firefox Extention “Safe Browsing for Firefox”

My related bookmarks at

Google Print and the Society

October 14, 2007

The problem with Google Print is not the technology but that Google privately owns the digital copies of the books and is running an advertising business using these copies.

Google is working to complete the Google Print service that will make the content of many books searchable and show short citations from the books as results. There are suggestions that Google should add micropayments to Google Print and make everyone happy: users, authors, and publishers.

The problem with Google Print is not the technology but that Google privately owns the digital copies of the books and is running an advertising business using these copies. May be Google will not claim an exclusive right of ownership for the digital copies but nobody else will be able access it without Google’s permission. So, the issue is a bit different from “fair use” or “Creative Commons”. The EU got aware of the private ownership and pushes now for a public project to make the books of European Libraries available on the web. Google has already become a monopoly on the market for ads and the dependency of the society on Google search as a private service is not unlike the problems with the privatization of water. There are three independent search engines left on the market and I would not consider that a stable and sustainable system.

Amazon became an empire as well. Anybody who buys something through Amazon pays the TAX (10%). The micropayment model for Google Print would include paying the TAX to a monopoly as well. And I really mean TAX and not fee for the service because their market dominance gives them the freedom to set the price. This is not unlike the price for music downloads where less than 10% of the revenues go to the musicians. Recently, a computer journal made a transparent calculation with fair fees for the download service and kept the payment for the musicians the same. They ended up with 50 cent per download as a “fair offer”. I think the rest is a TAX to a protected business that was successful lobbying politicians to extent copyright protection more and more.

A few days ago ArsTechnica had an article about a new project: Google Base. In the report it says: “The last sentence there really speaks to what they’re after: eBay, Craigslist, and classified ads. Users will be able upload all kinds of items for sale, and you’ll be able to geo-locate them, compare them, and search them via Google.” I think it even includes Amazon’s service for used books and other items from extern sources.

It seems people start to realize the hidden price tags of monopolies like Walmart and Microsoft. Why not for Google? Is it really to early to realize the risks of Google’s market dominance? Is Google’s PR motto “do no evil” still unquestioned? I think it is time to look for alternatives beyond Google that can generate a sustainable development and without the threads of monopolies. The announcement of the Open Content Alliance to make books available on the web hosted by the Internet Archive, a non-profit organization, comes just in time. Hopefully the Open Source ideas of collaboration rather dominance will spark even more such projects in the future and the neoliberal changes in copyright will be cutback to a balance between creators, distributors, and consumers.

Update: Google Patent for User Targeted Search Results
Google has filed a patent for user targeted, or attention targeted, search results which will change the ranking of Google’s organic results per each individual user based upon that user’s search behavior, location, sites visited, and even ‘typing behavior’. Read more …

Related external articles & blogs:
Security and Privacy Risks of Google’s new Firefox Extention “Safe Browsing for Firefox”
Reining in Google — The Washington Times

My related bookmarks — absolutely

Google Problems and Critics

October 14, 2007

Google’s corporate PR image Do No Evil does not fit reality.

Following the news about Google it is surprising how persistent Google’s corporate PR image Do No Evil works. Many reports show that the problems are growing and that there is something wrong with Google. Read more …

Advanced Tagging — Hierarchical And Ordered Tags

October 14, 2007

Could the order of tags be a general solution for hierarchical tagging? It would be similar to relations between words within sentences or to the order of folders in a directory and without enforcing a structure.
Tagging and meta-data in general are new and very interesting approaches to handle data by adding semantic data or relations. We see currently the first generation of online services that support tagging for bookmarking, photo and other services. If you are interested to read more about tagging I highly recommend an article about and Flickr and a comparison between tagging (folksonomies) and hierarchies (directories).
Common for the first generation of tagging systems seems the use of flat tags that do not have a hierarchy or any relationships to each other. I guess everyone who has used tagging for a while has noticed that the lag of relationships between tags is a limiting factor. For example, the meaning of “social bookmarking software” and “bookmarking social software” is very different. In one case “social” is associated with “bookmarks” and in the other case with “software”. So, the problem is that a search for “social+bookmarking” shows both items.
Hans Reiser has discussed this problem of relationships between tags in a very interesting whitepaper about name spaces because it is important for the next generation of file systems that will provide mechanisms for tagging or other meta-data. He proposes a solution that uses a special character to define hierarchical tags, e.g. “subject/strike to/elves from/santa”. But is this solution consistent with the general concept? I mean would a bookmark with the hierarchical tags “computer/hardware” be listed under hardware? I’m not sure what to expect because is doesn’t look like a familiar metaphor.
I wonder if ORDERED TAGS would be a more general and simple solution for hierarchical tagging. It would be very intuitive because it is very similar to writing sentences in a string of tags, e.g. “development software social bookmarking.” All tags in this order could be searched as flat tags. But I could also search for ordered tags: “social bookmarks”. The tags “social” and “bookmarks” needs to be in this order but can be at any position within the string of tags. In Hans Reiser’s example I could search for “to elves”+”subject strike”. A disadvantage of not using an operator that explicitly defines a relationship is the risks of getting items with unintended tag orders. But I think, examples of search results from Google, Yahoo, etc. show that unintended word orders have a low practical relevance.
A more complex search system could even support wildcards to search for ordered tags beyond neighbors. For example, I could search for a bookmark with the ordered tags “computer hardware video card” with “computer * video”. Or I could use a small set of tags as main categories and place one, two, or more tags of the main categories in the most left positions within the string of tags. The search term for the first main tag could be “^computer”. Well, I could go on with examples for regular expressions to search for computer and computers at the same time because tagging systems (folksonomies) tend to be inconsistent or different users have used different tags for the same thing, e.g. socialbookmarks, social_bookmarks, social-bookmarks.
A well ordered set of tags could even be equivalent to a well-defined directory structure, e.g. “computer hardware CPU shopping budget”. But the big advantage is that I can use ordered tags as well as flat tags to search for items. A special character to explicitly define a hierarchy or relationship between tags could be useful if necessary but should be optional and not enforced nor should it hide one or more tags of the hierarchy.
Update: Recently, I posted a script to search bookmarks that can search for ordered tags when regular expressions are used.
Related bookmarks – absolutely

Firefox Hacks :: Tagging Bookmarks

October 14, 2007

The hack is a simple but powerful guideline that makes use of bookmarks search in Firefox.

Many Firefox users hope that tagging of bookmarks will be implemented very soon. But very few seem to realize that bookmarks search in Firefox can be used as simple tagging system:

  • Bookmark a website,
  • Go in the opened dialog window to the end of the tittle field, and
  • Add a separator code plus the tags.

For example, when I would tag this blog I would expend the title “Firefox Hacks :: Tagging Bookmarks” to “Firefox Hacks :: Tagging Bookmarks – tips tricks howto”. The tags Firefox, hacks, and tagging are not necessary because they are already in the title. Typing “tips” in the search field of the bookmarks manager lists all bookmarks with “tips” in the title including the bookmark for the Firefox hack.

I’m using this simple hack since the Mozilla browser has the function to search bookmarks because I quickly realized that the titles of websites are frequently not very descriptive or that I need to associate the website with projects I was working on. The free association with projects and ideas is actually an important issue that will always require some manual editing. It is not necessary to organize the bookmarks in folders. They can just remain in a flat list because I retrieve the bookmarks with the search function.
Related blogs: Advanced Tagging — Hierarchical And Ordered Tags
Firefox Hacks :: Search Simultaneously in Firefox,, and Archived Bookmarks
The Evolution of Bookmarking — Bookmarks, Firefox, and

My related bookmarks at …

Firefox Hacks :: Improve the Usability of the Bookmarks Manager

October 14, 2007

Opening the bookmarks manager with keywords avoids problems with big bookmarks lists and Flat Bookmark Editing makes the bookmarks manager to an active workspace.

The handling of bookmarks in the current versions of Firefox results in some annoyances when the number of bookmarks reaches several hundred or thousand bookmarks. For example, Firefox does not respond for several seconds after the mouse pointer came across the bookmarks menu because it is putting all bookmarks into the menu.

Fortunately, the user interface of Firfox and other Mozilla-based browsers can be opened from the URL bar:


I saved it as a bookmark, opened the bookmark properties, and entered as keyword “bm”. Since then I have never used the menu again because it is much fast to press Control-L (to activate the URL bar), enter “bm”, and hit the Enter key. A double click on a bookmark opens the website in the same tab. But it can be opened in a new tab when the Control key is pressed while double clicking. This keeps the bookmarks manager open and saves time when the bookmarks manager is used more frequently.

The second major improvement in usability was to install the Firefox extention Flat Bookmark Editing (screenshot). It works like a tool in Photoshop that stays open instead of being opened and closed with every change. This type of permanent dialog window is always very useful for any continuing editing processes. Saving bookmarks without changing anything has been the primary use in the past. But adding and changing tags or meta-data requires a better editing process. The lower effort for editing is very beneficial for the active use of bookmarks. Flat Bookmark Editing makes changing tags or short notes much quicker. I hope Firefox will soon have a sidebar like Flat Bookmark Editing that loads all information simultaneously with the website and would be a very effective annotation tool.

Related blogs:
Firefox Hacks :: Search Simultaneously in Firefox,, and Archived Bookmarks
Firefox Hacks :: Tagging Bookmarks
The Evolution of Bookmarking — Bookmarks, Firefox, and

My related bookmarks at …

Rethinking the User Interface For Bookmarks — Two Different Purposes May Require Two Different Interfaces

October 14, 2007

Bookmarks are useful as FAVORITES in a menu for quick access to websites and as a large INDEX of marked websites with valuable content.
The reasons for saving bookmarks have evolved with the development of the web. Currently, there is a lot of discussion on the web about tagging of bookmarks. The discussion has two primary reasons: 1) frustrating experiences with organizing large amounts of bookmarks in directories, and 2) new online services for sharing tagged bookmarks within a community. It seems all users of online services for tagging of bookmarks, photos, etc. would like to have a new bookmark manager that supports tagging and can synchronize the local bookmarks with the online service. However, I think the necessary rethinking of the interface for bookmarking is not only a decision between hierarchy and tagging paradigms.
The redesign of graphical interfaces for bookmarks must address two different purposes. One is a menu for quick access to websites and the other is a personal index of marked websites. Both concepts are important for the interaction with the web and have been kept in one interface over time. But it may be better to have two separate interfaces.
The menu’s purpose is fast access to frequently visited websites. In principle, its function is similar to program menus or to sidebars with collapsible groups of items or submenus. Both have a very simple directory structure that provides not only the items but serves also as an visual reminder for the items and their shortcut keys. Menus and sidebars with the most frequently used bookmarks on the top level and a few more grouped in submenus or folders are good examples for this concept. The name FAVORITES emphasizes the concept very well that has been very beneficial for organizing a limited number of bookmarks in my sidebar.
Bookmarks without the high visibility in the sidebar or in the menu are a “permanent history” of marked websites. This personal INDEX contains bookmarks that have been saved because they were interesting or valuable. Searching this INDEX can provide pre-selected information very quickly. Tagging seems a perfect tool to categorize these bookmarks for the INDEX in a very flexible, quick, and simple way. Important for the interface of the INDEX are quick adding of bookmarks, simplified editing (e.g. Flat Bookmark Editing), and fast search for tags and text (title and description).
My own bookmarks file has the FAVORITES at the beginning of the bookmarks file to be visible in the sidebar and the INDEX as long flat list of tagged bookmarks below the FAVORITES. Unfortunately, this structure has affected the response time of Firefox (details) and it is not useful at all to have a large file of bookmarks listed in the menu nor in the sidebar. To solve this problem and to tailor the interfaces for FAVORITES and the INDEX more specifically to their purpose it seems reasonable to have separate user interfaces for the FAVORITS and the large INDEX. I’m sure if their data should be separated but it may be easier to have a file for the FAVORITES that just contains titles and URLs in the order of the menu while the content of the bookmarks is saved in the INDEX.
Important for the FAVORITES is that bookmarks can be manually ordered and that folders can be used to organize and collapse groups of bookmarks. The FAVORITES could contain dynamic or “LiveFolders” for dynamic search results of the INDEX, e.g. a selection by tags using a link structure chrome://browser/content/bookmarks/bookmarksPanel.xul?search=favorites+firefox. But the primary purpose needs to be visual feedback and quick access to websites. The manual order of bookmarks can be as important as in regular menus.
Important for the INDEX is the excellent support for tagging and Flat Bookmark Editing (screenshot). A new but very useful concept for entering tags are “tag clouds” combined with dynamic recommendations as the POST interface (screenshot) demonstrates. I hope very much that some time soon we will have for the sidebar a form like Flat Bookmark Editing and that opening a website loads simultaneously the bookmark when available so that I can change tags and make notes and annotations. This type of tools that do not require to open and close a dialog became very common for graphical editing, e.g. Photoshop, Gimp. Firefox’s Find in websites and in bookmarks as well as Flat Bookmark Editing demonstrate how this concept of avoiding dialog windows can improve the usability of web browsers too.
A redesign of the user interfaces to add, edit, and use bookmarks is a great opportunity to introduce new concepts and increase usability. Improved user interfaces for bookmarks including “dialog-free” annotations of websites could result in an interesting tool for information management and may pioneer the way we will use tagging and other meta-data in the future.
Related blogs:
The Evolution of Bookmarking — Bookmarks, Firefox, and
Firefox Hacks :: Improve the Usability of the Bookmarks Manager
Firefox Hacks :: Tagging Bookmarks
Firefox Hacks :: Search Simultaneously in Firefox,, and Archived Bookmarks

My related bookmarks at …