My current project: Soulmix

This will be my 139th blog post. That means that I’ve published 138 times while barely talking about my actual work! That isn’t so cool. In an effort to be more open, I plan to start writing more about what I am actually doing.


Those who have been following my blog have probably seen me link to my current project Soulmix.

Soulmix beta 1.0.

I started working on Soulmix last June, and initially, it was conceived as a niche community site for sharing content related to living a good life. People would share content related to life (personal development, relationships, psychology, lifehacks, etc.), upvote the posts they liked, collect their favorite posts, and discuss within the comments.

After building the site, I beta tested it for a few months. It seemed to have potential. Return traffic was growing steadily, but slowly. I sent a weekly email newsletter curating the best posts on living a good life, and it was well received. People sent emails thanking me for it, and I could track email clicks throughout the entire week. I think if I kept pushing for another year, it could have grown into a great little community for life-related content.

Making the decision to pivot.

A few months in the beta testing, I came to two realizations:

  1. I was getting bored of it. I enjoy reading great content about life, but after a while, it gets old. I have many other interests. I found myself wanting to discover and share content related to all of my interests; not just on living a good life. In short, creating a niche content site felt too restricting for me. I strongly believe in finding founder-market fit, and that means creating a site that I personally love and want to use everyday.
  2. I realized I was building interaction mechanisms. Soulmix was conceived as a niche site, but what I was building was general interaction mechanisms. Upvoting allows masses of users to choose great content. Collections allow users to save the stuff they like. Comments let users talk about stuff. Nothing about these interactions is specific to a niche. They are general, and can be used for general content.

These realizations made the next step obvious for me. I had to broaden the scope of Soulmix, and focus on the general interaction mechanisms.

Focusing on interaction mechanisms.

Furthermore, I’ve begun to strongly believe in keeping products simple. Soulmix had content shares, upvotes, collections, and comments. That felt like too much. I decided that I could go two routes:

  1. The discussion forum route. This would include content shares, upvotes, and comments. The problem was that I couldn’t think of anything much better that Reddit. I would like the Reddit community to be slightly more friendly, but at scale, I understand why it is difficult (if not impossible) to control the Internet masses. The other problem was that historically, I haven’t been a discussion forum kind of person. I lurk, but I’ve never been a big commenter. That means I would never be the power user of a discussion forum site.
  2. The collection/curation route. This would include content shares and collections. This route seemed much more appealing for two reasons. First, the large sites in this space (Pinterest/imgfave/weheartit/etc.) are mostly image collection sites. As much as I like images, much of the content that I consume are links, not images. This leaves me space for building something that I might love. Second, I love the passive social model behind these types of sites. It lets you interact with others through shared interests, but I don’t need to actively think about (or inject myself into) conversations.

So, about a month ago, I decided to pivot. The next version of Soulmix would be a general site for collecting stuff I liked online.

Soulmix beta 2.0.

This brings me to what I’m working on now: the new and improved Soulmix!

Soulmix is now an online tool for organizing and sharing the stuff you love on the web.

Your favorite images, links, and videos are typically scattered all across the web. Soulmix lets you take them all, and organize them into remixes (or collections). It gives you your corner of the web where you can collect and share the stuff you like. You can also follow others users (or their remixes) to discover new great stuff.

Sound good?

Soulmix recently became functional again and is in private beta. If you’d like to check it out, just request an invite *nudge nudge* 🙂

I’ll be letting people into the beta with time, and if you join, I would love to hear your thoughts!

Curation as the third frontier of the web

One of my recent big interests has been related to our relationship with content on the web. I’ve written a few posts on this about the importance of old content, and on finding the best content on the web. I’ve been beginning to believe that curation will play a huge role in our relationship with content.

Today I stumbled across a great blog post that helped me crystallize more of my thinking on this topic by Patrice Lamothe (founder of Pearltrees) called The Web’s Third Frontier.

The most interesting part to me was a section on the founding principles of the world wide web:

The founding principles

These principles are simply the initial objectives that Tim Berners-Lee and Robert Caillau stated for their project. Eliminating technical jargon, these objectives can be broken down into three general, universally applicable propositions:

1-    Allow anyone to access any type of document

2-    Allow everyone to disseminate their own documents

3-    Allow everyone to organize the entire collection of documents

When I came across this, things began to make sense.

  1. Information access: The Internet allows anyone immediate access any type of document via a URL.
  2. Information dissemination: Blogs enabled simple creation of content, and timely dissemination has been enabled via mailing lists and RSS readers.
  3. Information organization: This is where we are stuck with limited options.

This is a huge insight. Access and dissemination have been solved. Organization hasn’t been solved, and it could be a game changer on the Internet.

As far as organization goes, we are limited to (1) the time-based nature of blogs and news sites, or (2) the retrieval process from automated search.

Curation provides a much-needed third option, and it sounds good in theory, but has been difficult to implement in practice. Outside of Pinterest, which is mainly about images, there hasn’t been another huge winner in this space (and many have tried!).

What will to take to crack the curation space? I can’t say I have an answer, but am starting to come up with some interesting thoughts. If you have any thoughts or opinions on current and/or future curation solutions, I would love to hear from you!

P.S. This is post number #99 in a 100 day blogging challenge. See you tomorrow!

Follow me on Twitter @alexshye.

Or, check out my current project Soulmix.

How do I find the best content?

Last week I wrote a post asking: what happens to old content on the web? This post continues from the thought process in that prior post.

When a person searches for information on the web, they only care about one question: how do I find the best content?

Because most of the best content is old content, the question often becomes: how do I find the best old content?

It gets more complicated. What does ‘best’ mean? What is best is often subjective. Suppose I am looking for relationship tips. You can’t really find a best set of tips. It may depend on many factors including my age, sex, cultural background, maturity, outlook on life, etc.

So the question really is, how do I find the best old content for myself?


As of now, search is probably the best option. Search relies on the fact that that over time, the structure of the web points towards the best pieces of content. That is, the best content has the most and best incoming links.

Search looks for the one best set of results across the web. As I said above, for many queries, there is no one best set of results. We are all different people, and the best set of results will differ between people.

This must be one of the big reasons Google cares about social. Personal information enables personalized search.

How good can personalized search get? Who knows. Even if you have a lot of information, as Google does with Gmail and G+, it must be tough to develop the algorithms to automatically determine the best results.

Recommendation engines.

Many startups are working on being the best recommendation engines. Usually, the challenge is the find the most relevant new content to present a user. But, it doesn’t have to be that way. For example, Pandora is great at finding old music that you might like.

I’m not well versed in the recommendation engines that exist, so it may mean that there is no clear winner for general content yet. But, it could be interesting for an intelligent recommendation engine to suggest the best old content.


It is possible to discover great old content via social feeds and social networks. Usually people don’t search for content on social networks though. Instead they stumble across good content. Most of the time, this content is new content. However, people occasionally post great out stuff. And if you were so inclined, you could ask your social network a question and hope for pointers to the best content.

A big problem with finding great content on social is that not all content is shareable. People share what they are proud about, but won’t share what they are more ashamed of. For example, if someone is searching for the best data on sexually-transmitted diseases, or birth control, they most likely wouldn’t broadcast this out on their social networks.


Aside from social, there are great link aggregators/communities that are largely anonymous. The largest that springs to mind is Reddit. Through anonymous aggregators, you could find great content on almost every niche of the web. On Reddit, simply search all of the subreddits and you’ll find communities on all kinds of obscure and dark niches on the Internet.

Similar to social feeds, you tend to stumble across information on these aggregators. And most of the links are new links. But, if there is a social discussion component, things may work out. If one was searching for embarrassing information, the best bet would be to find the right subreddit, and ask. Because you are anonymous, the people won’t know you, but if you ask the right way, you may find the best old content for your query.


Recently, curation sites such as Pinterest have popped up. On these sites, people manually curate their favorite content. A big plus is that if you can find a person with your tastes, you may find the best content on the web specific to your liking. The downside is that you need to find the right set of people to follow. This takes upfront investment.

Also, with curation sites, you aren’t really asking a question. Instead, you follow people and stumble across what they have curated for you. The one X-factor here is that large curation sites provide a great data set for search. For example, I’ve recently started using Pinterest search for finding recipes. It is surprisingly good. Of course, Pinterest search doesn’t cover all verticals, but it is interesting that can be useful as a search solution.

So, how do you find the best personalized old content?

There are many ways to start going about it, and there are a bunch of startups tackling parts of this. Still, as a consumer, I don’t have a great solution.

The opportunity seems large enough that solving a sliver of the problem would make a great startup. And solving more than just a sliver? That would be huge.

If you have some thoughts, I would love to hear them!

(Photo credit: Mark Probst/flickr)

P.S. This is post number #54 in a 100 day blogging challenge. See you tomorrow!

Follow me on Twitter @alexshye.

Or, check out my current project Soulmix, a site for sharing the best of the web.

What happens to old content on the web?

As I perused the recent Startup Edition on writing, I came across a great post by Kevin Dewalt. The premise of the post is that blog posts lose their value over time. This is very true. Blogs just aren’t designed for browsing past posts. Kevin proposes an awesome strategy for dealing with this by turning old blog posts into an email course.

Beyond blogs, there is a larger problem here: over time, all content on the web loses value.

This makes sense. In the daily (or even hourly) news cycle, new content gets churned out quickly. As readers, we have a limited attention span, and the old content usually disappears from our attention to make room for the new.

The problem is that not all content should lose its value. Some content is evergreen. It will never go out of date, and it is so good that it should always be easily accessible.

How should we deal with this evergreen content? How do we find it, or resurface it, when it may be useful?

There are countless news sites, social feeds, and other distribution platforms for new content. How many distribution mechanisms are there for old content?

For most, the main mechanism for discovering old content is the search engine. If an epic piece of content has the right keywords for the query, page structure, and incoming links, search will work just fine. But, this isn’t always the case.

For the more dedicated, they may search their private stash of bookmarks, or their saved posts on Digg, reddit, StumbleUpon, etc. This may help the individual, but isn’t a general solution to the problem.

The even more dedicated will organize their bookmarks, and publish curated lists on their own websites or blogs. These are more helpful, but the curation is scattered all across the Internet.

More recently, curation platforms such as Pinterest have popped up for people to organize and publicly share boards containing their favorite pieces of content. This provides a great way to discover old content based upon tastemakers. But, searching through a sea of pins can be time consuming, and isn’t always easy.

I don’t think there is a good solution here yet. Still, content on the Internet continues to be created at a rapid pace. Finding the best stuff can be difficult, and with time, will probably get more difficult. It seems like there is a big opportunity here.

P.S. This is post number #47 in a 100 day blogging challenge. See you tomorrow!

Follow me on Twitter @alexshye.

BTW: I’ve been thinking about this problem a lot for current project: Soulmix.