Forum Moderators: open
note: no bashing or flaming of the ODP or Google.
Just your view of how the world of Google and SERPs in your sector of the industry would change?
In mine's it would be massive, especially for companies who have enjoyed an extended relationship within the ODP only. There is thousands of them.
?
It sucks.
I totally empathize with the website owners who cannot figure out which category best fits their site.
And I feel for the Editors, who, thanks to the submitter's confusion, face a large proportion of submissions that belong elsewhere.
The problem is that ODP bases its management scheme on the taxonomy. But any taxonomy that is rigid is not effective for a body of listings that are as dynamic as website descriptions. Rich-content websites are not static. They can and do add pages that would, in a dynamic scheme, qualify them for reference in another topic.
Further, the notion that topics have fixed boundaries and relationships to each other is a carryover from the age of the Dewey decimal system. Faced with a deluge of information, it was necessary for librarians (who were unaided by computers, remember) to invent some way of ordering the chaotic mess.
Fast-forward to today, and the idea that a website MUST be described as being in one category or another. Or, in some directories, maybe in two categories, at most.
The idea that the category must be absolutely fixed was and is thought to be the only way one can assign "expert" volunteer Editors, since making one in charge of a well defined and bounded topical subset makes it easy to say who is in charge.
Add to this the simplistic drill-down processes created by system designers who mimic this rigid notion in their hierarchy layouts.
So here we are: two constituencies at each other's throat, thanks in no small part to a useless taxonomy scheme.
One result is that in this and similar threads, there is much concern for the poor webmaster’s quandary - where does my site best fit?
I am sympathetic, but I am actually much more concerned with the dilemma faced by the search *users* - namely, how does a normal, non-web savvy person find anything quickly, when facing a taxonomy with maybe a hundred thousand topics?
Gang, the whole notion of a rigid taxonomy is absurd. Information changes, both with respect to how it is described, and its relationships with other information.
More importantly in the specific case of the ODP, most sites with any degree of content can logically and linguistically be associated contextually with *many* "categories", especially if the categorization scheme in question has 3 or more levels of drill-down depth.
So, even before dealing with all the operational problems of the ODP, we ought to give the Editors - *and* the submitters - a fighting chance, taxonomically speaking.
It starts with the fundamental that the website creator *must* be allowed to say what topics her site is related to. It is then the obligation of the ODP to design an editorial management scheme that can accommodate the reality of a site quite properly "belonging" in numerous categories.
In our version of the DMOZ directory, we use dynamic categories. (Not because we are trying to "save' DMOZ, although it would be great if it turns out we can play some minor role in that effort. The fact is that we *had* to invent a better scheme of managing topics, simply so we can do a better job for our paying website clients.)
Under our methods, a given site may be shown in DMOZ as "belonging" to a hard-coded category, but in our version we have the flexibility to associate the site with dozens of categories, which we call topics.
A website owner in our system has the right to ask to be associated with any number of topics, just as she can with any number of keywords. If one stops to consider, this makes sense, since "topics" or categories are actually just keyword phrases, right?
How could the ODP employ such a flexible scheme, and still avoid expert Editors from falling all over each other? That question can be answered in a more appropriate forum. But as a long-time specialist in massive automated workflow systems, I assure you it can be accomplished quickly, and fairly.
So, should Google drop the DMOZ database? No. They should apply some of their expensive talents to the effort of making it better. That’s what we are doing, anyway.
That's just one reason why it won't work to have the submitter having the final say on where their site lives. Another reason is that some submitters will want the most appropriate category, e.g. Food/Fruit/Apples/Granny_Smith, whilst others will want the highest category available for a similar site, e.g. Food/Fruit. You'll end up with mixed up categories.
Similarly, the PFI model breaks down by creating parts of the directory filled with affiliate sites all pointing to the same core site. Taking a look at the contents of a link farm is a good measure of the kind of sites out there that dmoz aims not to include.
Independent human editing is necessary to assess submitted and otherwise obtained sites on the basis of content to put them in the most appropriate location or exclude them according to the guidelines. In the vast majority of cases, dmoz editors are editing sites with which they have no affiliation or bias towards or against.
The problem with that is that, as a site owner, I would take my bog-standard online pharmaceutical sales affiliate site and add one page on each of Napoleon, the banana, the Golden Gate Bridge, the war in Iraq and how to tie five types of knot, then request and expect links in all of those categories, simply to increase the ranking of my core site.
I could never ever picture any of the complainers here offer a second of their time to support a community driven project ;)
The Dmoz sucks because there are too few editors, and too many cheating webmasters, more editors are always welcome. But if you become an editor you'd have to treat your completetion the same way you treat your own sites. I don't think most of the complainers are capable of that, but feel free to prove me wrong.
The value of a Dmoz link is heavily overrated currently, it doesn't matter so much as many think. Google is not treating Dmoz links any different than other high PR links.
It's just that it's much easier to get a Dmoz link rather than other high PR links.
That sounds like a good idea for your pay-for-inclusion directory. I'm glad the ODP data made a good resource for you in creating that directory. It's probably helpful to many people.
I'm strongly in favor of there being multiple directories out there with multiple different focuses. I think that the more that occurs, the more chance a surfer has of finding what he's looking for; and if all the directories are spidered, the better search results are too. Win-win for everybody. I don't really understand the endless debate over which-is-better-ODP-or-Zeal; I'm glad they're both there, and the Yahoo directory too, and ibizwiz's business directory, and my own little webpage indexing niche academic sites. They're all good for the Internet.
It would be a shame if they all merged into one muddled, grey mass. The strength of the ODP as a directory is that the sites are independently reviewed for inclusion... webmasters don't get to write their own descriptions and place their site all over the directory and so on, so our users get more objectivity and less advertising. On the other hand, sometimes I'm *looking* for advertising, so it's a good thing there are sources for that, too. And sometimes I'm actually looking for a *review* review... i.e. "This company is really good but their stuff takes forever to get there." The ODP doesn't provide that either, but there are sites that do.
I think diversity is a good thing. Trying to turn a cow into a horse is just going to lose you a lot of milk and still give you less speed than the real horses.
The review process then rules on if each category association is sufficient or not. Then some Editor has to determine if the site has worthy content, or is just an affiliate front, or whatever.
Anyone who runs a bona fide search site wants to eliminate SPAMmed listings. We have found a method to do that (using humans, you betcha) that still allows the site owner to be the one to say what topics they wish to be associated with. But this doesn't mean they can associate willy-nilly.
We then invented a drill-down method that lets topics be defined much more flexibly. Not rocket science, just taking advantage of database functionality in ways a static hierarchy cannot.
The same methods could be applied to DMOZ, and the same tools for assisting the Editors in the ODP could be employed - IF there was management direction and a bit of funding. That remains the main problem, as posted eleswhere.
The problem with that is that, as a site owner, I would take my bog-standard online pharmaceutical sales affiliate site and add one page on each of Napoleon, the banana, the Golden Gate Bridge, the war in Iraq and how to tie five types of knot, then request and expect links in all of those categories, simply to increase the ranking of my core site.So what exactly do you do with sites covering more than a subject in different subdirectories that are each one worth an inclusion? Reducing to evil affiliate sites does not cover the entire problem.
A site with genuinely significant and useful content on different topics in different fields (e.g. bananas and Napoleon but not apples and oranges) is already eligible for listing in several ODP categories. The point is that under the current system it is the editor who chooses whether that content is sufficient and appropriate for multiple listing. The deeplinking of the BBC website is a good example - News in the news category, Health in the health category, History in the history category etc.
If the submitter got to make that decision instead of the editor then you'd have more instances of submitters trying to get round the system by creating token content in multiple areas and expecting a listing in all of them.
The submitter already has 'first say' in where their site is reviewed because they can submit their site to any category. The general rule of 'submit to the one most suitable category' is the general rule because generally websites do not contain content in multiple diverse areas.
This is a good idea: if you're serving the site owner, and if you're depending on the site owner for submittals and/or finances. And if the site owner is getting honest, knowledgeable, professional keywording advice. [Professional journals have people dedicated to this kind of analysis.]
In the ODP context (where none of the above is happening) it would be a non-starter. And the fact of experience is: over three-fourths of the submittals I've looked at weren't even in arguably the right place. And that's not even counting the usual suspects (malicious spammers).
Several people have alluded to the underlying problem already: lots of people don't have a clue about hierarchical taxonomies; some people that have a clue aren't willing to work with someone else's taxonomy (and, of course, all taxonomies are not alike); some decisions are arbitrary, some require knowledge of the subject matter, etc., etc., etc.
The fact is, we can't depend on either editors or submitters fully getting the concept, and as editors we spend a great deal of time in internal forums trying to come up with a taxonomy -- not one that's ideal, but one we can nearly all live with. It is a difficult problem.
The result, of course, is that many people can't, and many more won't, use a directory. They want a search engine instead. Hence the popularity of Google, etc., and the necessary emphasis on keyword-picking (which is an art in itself) and keyword-targeted ads even in directories.
I'll echo the thought: people wanting to play with other indexing schemes (including "webmaster pay for keywords") are cordially invited to use the ODP data as a base or a testbed. If you publish such a site, be sure to submit it to our "Sites_Using_ODP_Data category."
I would mention in passing that the ODP taxonomy is nothing like as rigid as careless non-editors have supposed: if you are planning to use the ODP data, you need to be prepared for literally thousands of taxonomy changes per week.
Well, and IF that was the kind of directory we wanted to have. It isn't. But as I said before, there's room for many types of directories on the web. I'm glad you found a system that works well for what you do. No one directory is going to be all things to all people. So you keep giving the site owners first say in the listings in exchange for money, and we will keep giving independent review the first say on a volunteer basis, and there will be that much more information available for surfers. Win-win, isn't it? *shrug*
So, should Google drop the DMOZ database? No. They should apply some of their expensive talents to the effort of making it better. That’s what we are doing, anyway.
How Google arranges ITS directory, I'm far less qualified to comment on, since I don't work for Google. However, given how hard Google works to keep webmasters and SEOs from manipulating its search engine results, I'm personally dubious that they would be very interested in letting webmasters have the first say as to where and how many times they should be listed, and with which keywords. Google seems to me to be supremely focused on the surfer first, webmaster second (the reason I usually look to them first for a search engine), so I'm not sure why they'd want to expend all that time and money bringing on editors of their own to review webmasters' opinions about why they should have a second, third, or twelfth listing in Google's directory.
If they did, though, I doubt it would hurt their search engine results very much, or make me less likely to use Google. The more aggressive submitters would get at most one or two additional links from the Google directory, and as all their competitors did the same thing, the effect would be basically erased.
Ah-hah... I had gotten the impression from some of the posts here that the issue was Google 'getting' the ODP to do things differently.As I am sure it is/was. I just meant I thought in that post ibizwhiz was referring to the Google directory.
If I had to choose, I'd take a pr4 link from the odp before I'd take a pr7 link from anyone else.
/odp is one of the most useful sites on the net today - don't change a thing.
Why should older sites have the advantage?
isn't that the case in any business? You see opportunities earlier, you benefit the most.
Why should it be possible for me to register a domain, demand my free lunch served at once and feel entitled to blow away all the others that worked on their stuff for years? That's not the way live works.
(But - once your in, I agree it's not very fair to get only half the glory because Google just won't update its directory.)
I think if you have a commercial site, you're going to have a much tougher time. This summer I did up some unique content pages in my hobby site because I felt like it. On a lark I submitted them to various odp cats and they went up within 2 months.
Also, my hits from odp listings are non-negligible; actually, they're about in line with other static links I have, such as from my Yahoo listing, from a specific directory within my hobby area, and the organic links I get from other sites in my category. Sure, I get way more hits from Google searches, but I wouldn't call those odp listings junk.
And I do like the Google directory. Sure, I don't use it the way I used to use the Yahoo directory (back when it didn't suck) but I do end up in there from time to time. And the cat that a site is placed in by editors can sometimes be informative.
The neatest thing about Google's odp mirror is its PR sort, which I think is the best value-added odp use I've seen. Though it is sometimes amusing to find sites that haven't existed for years in directory.google.com with apparent PR of 4.
