As of today digg has category check boxes for customizing digg stories. You can un-check topics you do not wish to see. This is a great step in the direction of making digg hit more target audiences. But I believe it can be taken farther.

I propose a more advanced customization system which will allow for more functionality in the future. What I’m thinking about is a system very similar to Smart Folders on Macs. Basically users will be able to create smart topics of their own which will show up in the topic bar.

How would this work?

To conceptualize this you’ll need to step outside of the box for a minute. Right now, when a box is checked it shows everything in that category. If I don’t want to hear about Election or Political stories currently, I would need to un-check those categories on the customize page. Now lets say I don’t care about all the stories as a whole, but I want to hear about a specific candidate (Joe Smith). With the smart filter feature I would create a Smart Feature topic with the following options:

Filter Name: Joe Smith

Rules
Where Title contains ‘Joe Smith’ and
Category is ‘US Elections 2008′ and
Media Type is News

This will give me another topic in my category bar with the ‘Joe Smith’ topic under Custom. Then only Election stories about Joe Smith will show up on my front page and I can navigate to the custom category page at will.

This will be great and all but it raises the question about what if I have ‘US Elections 2008′ checked also. In this case, all election stories will show up on the All page, but I will still have my Custom filter to view only those topics specifically.

This functionality leads to a broader user base because suddenly the news that the users see can be manipulated to topics of their interest.  This also gives the topics functionality back to the users, as users can create topics that interest them. Instead of the topics that Digg employees think they want.

It can be annoying to upload the same user icon to n* sites. Which makes most users not even bother. The normal steps to upload an Image are as follows.

1. Login

2. Navigate to Profile

3. Select Edit User Icon

4. Browse to the file on the Computer

5. Upload

6. Crop Image

7. Finish

In the best case this can be 7 steps. A simple solution to this would be to allow the user to give a URL or RSS feed.

URL

With the URL, the image could be located on another server, then once the user enters the URL the website server could create a local copy to avoid hot-linking.

In the best case scenario this could get rid of one step (Browse to the file on the Computer). This is a good solution but not a great one. It still requires me to login and change my photo.

RSS

My favorite solution is the RSS solution. Many photo sharing sites (picasa, flickr, zoomr, smugmug,…) provide RSS feeds for a users images. This is an ideal solution for the user icon, because I can login once provide the RSS feed for my photo stream, and that’s it. When I upload a new photo to flickr, then the website can pick that up automatically and display it for me. I no longer need to think about changing my icon.

What’s even better about this solution is if many websites adopt this technique, then I will only have to change my user icon in one place. What a time saver!

Many people online have been yelling for a Web Based OS. I claim that one already exists. There exists the tools for an underlying OS. There has yet to be a platform to bring all of the already excellent web services together. Some of the potential services that I’m thinking about are specialty sites like gmail, plaxo, delicious, … .

Mail

It is obvious how the mail app would work, but what are the advantages of having email in the OS? Can control mailto; When a link is clicked in any onther app then the WebOS can handle it.

Maybe the WebOS could be a unified platform that makes everything pretty launching the web services in iframes?

Or the WebOS could be something else. It could just be a platform that allows the separate web programs to plug-in and communicate.

What would make a WebOS stand out from other platforms is the ease of letting other programs to connect and communicate. There are a few essential technologies that need to line up for this to work.

1. File Management
2. Open API
3. Program Management
4. Platform Upgrading

File Management

What makes a WebOS so desirable in File Management is its ability to have files anywhere, The ability to have a revision history of files, Automatic Backups, and search-ability with file indexing.

A WebOS would start with just a basic filesystem. Everything on the web is data-based, but a webapp is not usefull unless the data stored can be interacted with.

Web API

This is essential to the success of a WebOS. If the user is restricted to Google services, then it will fail. But it’s more than allowing the creation of gadgets/widgets. The third party apps should have full screen space as well as much control as other systems.

If on web app (plaxo) created some files or data points on the user’s site. Then another app that the user has should be able to access the data points. Once given permission the second app could call something like:

data.plaxo.contacts —–> XML

This will return xml to the second app and it will be able to access the contacts.

The permissions for access to the data would be similar to unix, there would be a group/user/guest access. When an app tries to access a file the first time the user will have to grant it access.

Platform Version

This is one advantage of having an online OS. Version updates are automatic. However, some may see this as a disadvantage. For them they can set an option that allows them to keep their current version. The options could be as follows.

• automatic update
• notify me
• do not update

Program Management

Program management is a little more tricky. The model for most OS like platforms today is to have the program code base hosted on the program’s own server. Facebook has recently taken the steps to provide developers with space.

If the OS is going to regulate the upgrade of each program then, the program should be hosted on the OS’s server.

The mechinism for installing a program will be as follows.

1. The user finds the program, clicks the program’s link.
2. The link will link to a codebase witch, after the user approves, will install on the OS.

The url may look something like this. http://webos.com/install?u=http://osprog.com/j.webp

Today there are many web services available but there has not been a gathering place for all of these services. The WebOS’s that exist do not allow for full functionality, and this is their downfall. A WebOS must allow for total control over files, user customization, and program interaction.

Digg is very good at giving me the most popular content on the web. Since its inception this content included news, flash games, images, videos, podcast episodes, lists, how to, and articles.

The idea behind digg is that the news is user approved (not editor approved). But there are some aspects of digg that remain under single user control. Here are the list of aspects that are under multi-user control and single-user control.

multi-user control

• podcasts
• stories
• comments

single-user control

• story category
• story title
• story link
• story description

All of the single-user controlled items can be converted to multi-user controlled. Here’s how…

Story Category:

When a story is submitted it’s up to the submitter to pick the category. This is problematic. If the submitter misunderstood the article or simply picked a bad category. This can be solved by getting the user’s feedback when they select bury>wrong topic. Digg can ask the user which topic the story belongs in.

Once the Wrong Topic bury is clicked, the suggest category box could look like this

Then if the system gets enough votes, the stories topic will be changed.

Story Title:

The story title on digg is often misspelled, misleading, incorrect grammar, or just incorrect. In digg’s case the power of the people should be used again. The story can have an edit icon next to each title. When the edit button is clicked the title is replaced with a text box that has suggest features.

The user clicks on the small pencil to the right of the title.

Story Links:

Story links are sometimes direct links that go straight to the story described. Most of the time links are broken, down, or blog spam.

The best link can be suggested just like the best title. When you click the edit link button a text box with an suggest feature will appear.

This will cut back on people posting the links in the comments.

Story Description:

Now the story description is a little bit tougher. Because of the length of the description it would be impracticable to do the same as the link and the title. So my suggestion is this.

When a story is submitted the description that the user puts in will become the first comment of the story. So this comment will remain the story’s description as long as it is the most dugg comment. When another user submits a comment and gets dugg up, then that comment will become the description

Summary:

Digg became a fun and unique site because of the control it gave to all of the users. Not all aspects are up to multiple users.

The category, title, link, and description can be updated to benefit from the masses.