Forum Moderators: open
Dmoz really seems to be a great project, even in France; but..should the average surfer accept not to find a pertinent information in Dmoz when Yahoo is updated daily?
The commercial link would introduce the editor.
<div id="Layer2" style="position:absolute; width:50px; height:50px; z-index:45; left: -55px; top: 91px">
<a href="Googlegalproblems.htm"
<img src="100billswithGoogleguyasprez.gif" width="48" height="48" border="0" alt="sherman anti trust suit sherman anti trust suit sherman anti trust suit</a>
</div> ;)
ps As yet I don't know of any 100billswithGoogleguyasprez.gif but maybe soon after this posts...
One thing is for sure: a directory feeding Google could take nearly any submssion fee.
Should Google start their own they would be a big step closer to controlling webpublishing.
In a way the Google directory tab says, if our normal Google-machine-algo generated results were not good enough, check out the human edited results in this tab.
Its finally time for the Altavista-like Prisma categories or options on top of, or next to the normal search results.
Its finally time for the Altavista-like Prisma categories or options on top of, or next to the normal search results.
If you're referring to something like Teoma's "Refine (Suggestions to narrow your search)" list of topics in the right column of search results, I hope Google never includes that feature which I find quite confusing and hardly of any use. "Quality results" means including the very best, most relevant resources in the top 10 results even for the most generic queries, not asking users to refine their search IMO. In other words, with a 2.5-billion page index, precision is what matters most.
there are many instances for which such suggestions would help:
- UK versus US spelling [webmasterworld.com]
- Offering plurals to single search queries [uk.altavista.com]
- offering suggested synonyms to search term queries [webmasterworld.com]
- Search for Jaguar [google.com]
A sub-listing of: Apple Software - The Cat - The Car, would be nice? (same for Ford (president-actor-car))
Altavista I thought even argued that in the long event, such a "pre-categorisation" would teach searchers to use more words in their search queries.
It could make the results page look cluttered, but it could be an opt-in, or only show when there are many Google results in those varying suggested categories similar to the search query.
Wouldn't having some kind of paid Google dirctory go against everything about no payment for any kind of placement within the search results?
I came across a forum that is made up of current DMOZ editors, and I personally found their audacity, arrogance, and attitude very alarming. They seem to actually believe they truly are 'holier than thou', and much, much smarter than you, the average bear.
If Google does continue with them, they should launch some sort of probe into possible corruption (sorry for being overly cautious; I can't afford a lawsuit against their deep pockets). Invite public input; I think I could contribute a thing or two.
I came across a forum that is made up of current DMOZ editors, and I personally found their audacity, arrogance, and attitude very alarming.
I know which forum you are talking about.... I have one site still waiting for over 3 months and yet it still hasn't been added - despite me asking twice on there...It's in the queue they say... Tsk...
But you know, despite it's many problems - DMOZ is free and MOST of the time it proves to be an invaluable way to get Pagerank for people who cannot afford Yahoo! ($299 a year per site, give me a break...)
So please don't drop it Google... Pretty please, with sugar on top. And a cherry.
I would venture to guess, anyone who has shown support for DMOZ in this thread, is listed in DMOZ. I don't blame you for wanting to protect that, but I don't think you can argue that the system is seriously flawed and corrupt in a kindergarten kinda way.
If Google is aware of this (which I'm sure they are), I can't imagine them not making DMOZ irrelevant in calculating rakings, in the near future.
DMOZ as a concept is still sound. Many categories are run efficiently and honestly by a team of hard-working editors. Unfortunately, there are a number of corrupt editors who, not only reject valid applications, but also falsify reasons for rejection. And yes, I do know this for a fact and have seen the evidence. The majority of the corrupt editors appear to work within the most competitive areas, as you'd probably expect.
Are DMOZ aware of the corruption? Yes. Have they done anything about it? No.
Unfortunately, DMOZ's inability/unwillingness to deal with their own inside-corruption will be its downfall. Which is a shame for the majority of categories which are operated with a good level of quality.
To answer the question posted in the thread: I don't really know if Google will drop DMOZ. I really, really hope so. IMHO, using search results which are filtered by unethical editors for their own benefit is not a good reflection on Google.
They must know submissions for keyword-keyword-keyword.com or exclamation-1-keyword-keyword.com sites are all spam
I don't think we should just lable a keyword-keyword.com domain name as spam, in many cases they deliver relevant results. It makes it difficult to brand a site but at the end of the day we're all after online customers not creating pretty websites. (Although I think achieving both is possible).
Does anyone actually know why the ODP needs human editors, why not just humans checking for spammed sites?
I beg to differ, computers do it better. It just takes a team of 50 PhDs to make the software. I bet there is a way to make DMOZ history and to also decrease the directory features reliance on single human editors.
Why not allow people to list their site in any category. But only one category. To get listed in a higher level you must meet higher and higher criteria. Higher PR, 7/10 editors that approve, random sampling, etc...
ok, let's say I visited a site from Google. bam a pop-up appears and asks me how would I categorize the site. The options they give me are 1) the same category the author submitted under. 2) similar categories. 3) false categories.
Now, if 5/6 people identify the same category as the original submitter then the thing sticks.
Or there could be a help Google organize site, when you go there Google randomly picks some sites for you to categorize. If you pick the same category as other people then this too re-affirms the selects. Google can design a system to randomly poll unique people. They could also have super volunteers (people that have higher then normal rating efficiency) that can spot bad apples, if multiple unique super volunteers point out bad apples then let the system re-organize. But never give the keys to just one human. Humans do it better, but humans can also be corrupt and biased.
Overall, I think random sampling of people/volunteers plus a few PhDs would really help eliminate the need for DMOZ.
Just this evening I tracked down a ultra bad editor, it was clear he owned an affiliate site which was placed very high in the directory tree under one of his categories. Then the organization he resold for was buried deep, very deep. It is crazy that DMOZ is so corrupt.
PS, The apple is only a little rotten, most of it is very good. But when you are aiming to create the mind of God a little rotten gray matter is a very bad thing.
Why? Well quite apart from the fact that I view it is a vastly superior resource from a users perspective, I think its downside is far less problematic than that of Yahoo.
Yes, ODP has a fair number of lazy/corrupt editors. These are a problem which should be addressed. But at least they CAN be addressed.
However, the flip side of Yahoo takes a lot more getting around.... good old $.
As a commercial entity they are driven by profit - they are driven to get more and more out of you for your listing. No problem if you are a big enterprize... but if you are small?
In the long term, as they up their margins, this problem will increase, forcing more and more truly excellent sites out. Smaller players will be marginized and the quality of their directory will deteriorate as a result. I see this as inevitable, as short term greed always wins in the end with this type of outfit.
The choice Google makes will therefore be interesting, and will be a good indicator of the long term likelihood of them retaining their perceived 'purity'.
Lisa, please inform a meta editor of this and it will be handled. Not the same day, but it *will* be handled (investing abuse takes time, as firing someone requires solid proof). You can also sticky me the category and I'll see to inform a meta who is actually interested and competent in the area, which may speed things up a bit.
Then the organization he resold for was buried deep, very deep. It is crazy that DMOZ is so corrupt.
PS, The apple is only a little rotten, most of it is very good. But when you are aiming to create the mind of God a little rotten gray matter is a very bad thing.
Thanks for the last paragraph. Among several thousand active editors, there are bound to be a few dozen bad apples. And they are bound to appear in competitive areas. Why is anyone surprised about that? It is good to know that on average, more than three thousand good sites are added to the ODP per day. While the abusive editors may stick out to someone if they operate in their favourite category, they are still an extremely small minority.
Looking at those proportions, saying that "DMOZ is so corrupt" is really ridiculous. Luckily, Google tends to look at the broader picture, along with hundreds of other sites who use the ODP data.
In that case, there is no machine that can compete with the humans, simply because the abstractions and the subject matter in the web pages are far too complex for any pattern recognition or AI software to handle. The day when we can build a machine that will come close to the categorization perfomance of a skilled editor then we will probably also be able to build robots that can do all the things that a human can do.
However, if the number of web pages that have to be organized is large (and growing), and if those pages do change over time, then the categorization performance of the small editorial staff will begin to decrease.
History has shown that in a situation where manual labor becomes a limitation for a system's performance, the only way to increase the capacity is to increase the level of automation and machinery. There are many examples of this, e.g. farming, manufacturing industry, and routing telephone calls on a switchboard. To me this suggests that the future of directories must also go in this direction. And there are in fact already today examples of 2nd generation directories which are founded on these principles. In one of those systems, a web site owner becomes the editor of his own page by the use of a meta tag which refers to a taxonomy specifying the subject category, geographical location, and language of the page.
Google may or may not drop ODP in favor of Yahoo in the short term. However in the longer term I predict that Google and everyone else will be forced to embrace new directory paradigms building on automation and distribution of the task of organizing the web to the Internet users themselves.
yup. they could do this very very easily. and they could create a bigger directory than the ODP and better quality control than Yahoo.
but then why are we asking "will google drop the ODP"? with FAST having so many more features and looking increasingly better than google, surely we should be looking at whether FAST will take on the ODP or Yahoo or maybe create their own directory?
>>If Google was smart, they would blow both of them off and start
>>their own directoryyup. they could do this very very easily. and they could create a bigger directory than the ODP and better quality control than Yahoo.
Have yout two guys talked with their human resources department recently? Looks like their 50 resident geniuses have troubles seeing the simplicitly in this that is so obvious to you. Or do you have any other explanation why they haven't already replaced the ODP with an automatically generated high quality directory some time earlier this year?
I also agree with mat about FAST (which is, first of all, really SLOOOW when compared to Google: not exactly good for a SE company with such a name).
Oh yes, you tell me AllTheWeb's got the "nicest advanced search options page on the Web", which is very cool for us power users, but then how many regular users even care about that?
I'll repeat: if a SE needs to ask its users to "refine their search" (that's Teoma) or use the "advanced search" feature (that's AllTheWeb) in order to provide them with quality, relevant resources in the top 10 results, then that search engine has missed its target.
(Sorry for the "me too" --and slightly off-topic-- post, I just had to agree/disagree).
Have yout two guys talked with their human resources department recently? Looks like their 50 resident geniuses have troubles seeing the simplicitly in this that is so obvious to you.
How is it that the fact that they haven't done it yet is an indication that they have never considered it, or that they aren't planning to do it in the future?
I know that there are many loyal ODP editors who will defend the directory to the death, but I think the notion that Google creating their own directory is something that is too tough for them to do is a bit silly.
Building a system that allows people to submit sites for human review is childs play compared to what they have already accomplished.
The reality is that the only real value a DMOZ or Yahoo listing has anymore is the authoritative PR
bump you get from it. If Google is going to continue to use a human reviewed directory to help determine which sites on the web are most important, then it makes perfect sense to eventually control the directory being used to make those decisions.
Looking at those proportions, saying that "DMOZ is so corrupt" is really ridiculous. Luckily, Google tends to look at the broader picture, along with hundreds of other sites who use the ODP data.
Bird- I agree with you that the majority of editors are ethical and hard working. "One bad apple spoils the bunch!" This is particularly true in this case. Because it's obvious, in which categories this occurs and nothing is done about it. Last time I looked ODP does not stand for "Opinionated Directory Project".
The argument that
hundreds of other sites use ODP datacan be construed as a bit of a stretch. Isn't the ODP data "free", and could that explain why so many sites use the data?
My only wish is that ODP was truly open, as advertised.
