Good question, webworld!
Sadly, there seems to be some confusion regarding the main question a (potential) directory owner has to ask himself/herself before embarking on the journey to world domination:
WHO is my (main) CLIENT?
I am assuming that there could be only two valid answers:
The implications from the answer you give are very significant. Let's start with the USER. For a USER, a directory has to be a rich source of relevant information. That would indicate, for instance, that starting with an empty directory is not a good idea. People are spoiled, and if you managed to persuade them somehow to TRY the directory, they will not be happy with one that gave them zero or hardly any results. As you know, you never get a second chance to make a first impression :)
One can, of course, start by loading up all the DMOZ data as a 'starting point', But then you are not much different from DMOZ, are you, but certianly far less KNOWN than DMOZ - so why should anyone come and use your directory?
Alternatively, you can start building it under wraps, until you are ready to go live with some meaty and useful information. That takes time, money and quite a lot of effort - not for the faint hearted. Even if you have just 50,000 categories (DMOZ has 590,000) and you wish to have 10 sites per category, you're looking at 500,000 sites! Hack, even 10% of that is really hard work!
Last but not least - what's the Unique Selling Point? There are literaly HUNDREDS of directories that are only distinguished by the colour scheme and the logo. They all use the same category structure that was pioneered by Yahoo and then copied and adopted by DMOZ. When looking for a particular SUBJECT, say film producers, it is very hard to beat either DMOZ (under Google)http://directory.google.com/Top/Arts/Movies/Filmmaking/Producing/Producers/ or Yahoo [dir.yahoo.com...] (note the similarity!) so having 'quality links and no spam' will hardly cut it with users.
Well, then, you may ask, is there a way to bring users to a directory WITHOUT reliance on Search Engines? Sure - advertisement is a good start, but only a start. There must be something UNIQUE about the directory that will make them bookmark it and entice them to come again and again. If that doesn't happen - only a bottomless coffer will do.
However, life can be so much simpler if your client is a webmaster. Most of them (wrongly in my experience) do not expect directories to deliver traffic - or just a trickle. They do want to be listed mainly for the PR, link popularity, being found by spiders etc. In other words - you are a means for ANOTHER end. That can be a very lucrative business, if you play it right. Here is the recipe:
- Use a simple script and inexpensive hosting
- Copy the directory structure from DMOZ. It is the most comprehensive, it's free, and thousands of hours were spent developing it.
- Charge a reasonable amount (don't be too greedy)
- Don't reject anyone (find some excuse to accept every site)
- Ask the SUBMITTER to do most (if not all) of the work. Tell them to write their own title, description and to find the right category for their site themselves. As you will have the same category structure as everyone else - they will know how to do it!
-Provide a good servioe! Be very fast to approve sites (24 hours is good - one hour is outstanding), and offer to refund them if they are not happy. (unlikely if you accepted them anyway)
OK - so you will not get to dominate the world, but you will have some bobs on the side - not too bad.
Best of luck whichever way you go!