These are my thoughts on building a multi-directory search portal.

Introduction:

At the most basic level this would mean that with one search box and one search, you would search the indexes of 2 or more similar web directories.

Such a portal could exist on either it’s own domain or on a subdomain. Domain is probably better should other directories come along.  (Random thought: Could portal be part of Indieweb.xyz?)

Minimalist or Portal?

Minimalist example would be search.yahoo.com which is just an uncluttered search box vs. yahoo.com which is a full blown portal considered cluttered by some.

Unifying the Search

My suggestion, and it is just a suggestion, is to beta test with a free open source crawling site search that can be set to index more than one domain like either phinde or sphider.  One of these might work perfectly or will show it’s limitations but the learning experience of trying these first will be useful prior to coding a custom search.  (What would be icing on the cake is if Chrome based browsers would recognize this search form and allow you to add the search to the browser.  Search installed on a person’s browser gets higher use.)

Optional Portal Features:

  1. Cross Index of Categories:  I already touched upon my idea under “categorization systems” here.  This could be on separate pages from the search form above.  I kinda like this idea: no scripts just a page of plain HTML so low overheard and cost, it might encourage people to browse rather than search deeper into both directories.
  2. Web Search Form: Example: Searxscript.) Not sure what i think of this.  On some level you are competing against yourself. OTOH you are giving the user more reasons to use your portal.  I have done this before with both dropdown menu and radio buttons on the search form and very few people used it.
  3. Leverage RSS Feeds:  Use headline feeds from our blogs, and newest listings to keep portal fresh and not just a searchbox.  Free.

I’m not married to any of this.  Just throwing out ideas as they flit through my brain-pan.

This was also posted to
/en/linking.

Liked this post? Follow this blog to get more. Follow

Q.: Does your directory have a crawler/spider?

 

Why yes we do.  Although he’s getting kinda old and is a grumpy old git and he hobbles more than crawls.  When he’s not mopping the floors and washing the bottles he can be found adding URL’s to our directory.  Generally he can add about 20 URL’s a day, sometimes 25 if we juice him up with caffeine.  After about 20 site reviews he starts going all cross-eyed.  We tried naming him Bradbot but it never really took.  🙂

Seriously:

Not really.  Indieseek does have a meta tag fetcher.  When I put a URL in the field the meta tag fetcher tries to grab that page’s Title, Description and Keyword meta tags and fill in those fields.  It does the same thing when the public submits a URL and it saves some time.

Other than that, everything is manual.

Liked this post? Follow this blog to get more. Follow

For context you should read my post about decentralized search.

If you are using self hosted WordPress you can have a small links directory running in no time.  This uses the “Links” feature in WordPress that was never removed, just papered over.

  1. Add the “Links Shortcode” plugin from WordPress.  What this plugin does is reactivate the Links function in WordPress and lets you display the links categories you add on any page using a short code.
  2. Once you create a page and place the short code any links you add will appear on that page.  Follow the instructions on the plugin site on how to place and configure the short code.
  3. Create categories and add links in your WP admin under “Links”

It’s really just that simple.  And it’s free.

You can use this to make a blogroll and/or a links directory.  IMHO every blog should have both even if the “directory” or links page is only 20 links to start it will grow over time. This lets people surf from site to site based upon your recommendation which is a powerful thing.

You can divide things up over several pages.  The link listings have the option of Ratings, Title, Description URL and more.

 

Tips and Advice:

Start out making this for you.  This is a great way to keep your permanent bookmarks and share them with visitors.

If your blog is about one topic (ie cooking, hunting) your little links directory should probably match your theme.  If your blog is about anything and everything save whatever you want.

My suggestion is to name your links page “Directory” in your navigation menu – we need to get people used to seeing that word.

You can use this for anything: links to sites you like, links to your friend’s site, links to causes or charities that are important to you, links to sites or pages you check all the time.

Consider this part of the guerrilla war against the Google search silo. If a thousand bloggers all do this it starts making a difference.  Maybe we will teach people how to surf the web again.

This was also posted to
/en/linking.

 

Liked this post? Follow this blog to get more. Follow

And why it matters.

It matters because the first 5 listings are going to be clicked on most.  Just like in a search engine, where the top 3 listings for a query get the most clicks. (Thus spawning the whole SEO industry.)

Alphabetical: The oldest way to rank listings in a category is alphabetically.  Which is great if your website’s name starts with the letter “A”.  The alphabetical thing started a whole trend back in the Yellow Page phone book days of naming businesses “Acme”, “Ace”, “AAA Exterminators” etc. to game the listings and be at the top.  That is the whole downside of alphabetical.

X Rank: Another way to rank listings is using a third party ranking factor like: Page Rank, Alexa Rank, etc.  The problem with this is the rankings are all based upon popularity.  So if you put the most popular sites first they stay the most popular.  That too does not seem fair, especially for new sites that have no rank.

Click Rank: Another factor that used to be popular was Click Rank.  Sites that got clicked on the most rose up in the ranks.  Again the problem was once they got to the top they tended to stay there forever.

Rating and Comments:  Many directories have a formula that will rank sites by user rating (usually stars) and/or how many comments the listing gets.  This never really worked out.  Webmasters always tried to game the system.  Also, in search, most people don’t bother to rate or comment, they just want to find answers to their query as quick as they can.

Editor’s Rating:  This is a subjective rating given by editors of the websites within a given category.  This can work well when you have expert editors taking care of subject that they are experts in.  But it falls apart quickly when you have one guy who is a Generalist, trying to judge sites on subjects he/she knows little about.  However, if you do have expert editors, this might be the best of the lot.

Here at Indieseek.xyz I use Alphabetical ranking in categories.  I could use the others but alphabetical keeps it simple and does not add to server load.

 

This was also posted to
/en/linking.

 

Liked this post? Follow this blog to get more. Follow