|Google and DMOZ|
Why doesn't Google build their own?
After having a site left waiting for three years to be listed in DMOZ, and having much more respect for Google's technical prowess, I'm left wondering why Google is still dependent on an outside group after so long. There has been no shortage of discussion over the years regarding the problems with DMOZ, not the least of which being the lack of transparaency. Many of the operating practices and behavior seen in DMOZ appear to be counter to Google's.
Past experience indicates that Google could come up with their own directory which would outshine DMOZ. Key to this has to be transparency. Let people know where their site stands in regards to listing. Let site owners control their listing. Google already has facilities in place to let people demonstrate that they own a site. Once that ownership is demonstrated, let the owner see where their site stands for being listed and let them update their own listing. Many of the complaints of DMOZ would go away if it wasn't such a black hole of information.
Has anyone seen indications from Google as to why they haven't gone this route?
I think Google won't do something like this because overall the value of directory these days is very very low, though being listed in DMOZ is pretty good thing.
You don't know that being in a directory is of low value to google, Google has made no such comment,
Furthermore, Google is currently updating its own dmoz based directory regularly, or so it seems, virtually every month now, some one notes changes
Theoretically, dmoz is perfect for google, it categorises websites manually, with human review an oversight, for free
Google has indicated that automation an scalable solutions are their thing, so running a personnel intensive directory is probably not likely for them
From what I've read here an elsewhere, Google know exactly what 'category' (area/niche) a website belongs to, by analysing a website's external links and overall content -- If not it wouldn't be possible for Google to decide whether or not a webpage is relevant for a given search query.
Google could make their own, more accurate and dynamic (compared to DMOZ and others), directory and place websites in categories based upon information that Google allready has about the websites in their index.
I believe that it can be done by using the data that Google allready have. But the question isn't whether or not it's possible for Google to make such a directory. The question is more like why should they do it?
[edited by: OutdoorMan at 2:18 pm (utc) on Jan. 23, 2008]
I don’t think people really appreciate the size of DMOZ, although it seems to have been shrinking for a while now, it is still a significant independent resource.
If your site meets the criteria and is better than some of the others listed in your preferred category then it can be very frustrating.
The thing to remember is DMOZ is not uniform. Some areas have editors which are interested in building the best resource for people who share their passion, other areas are short of editors and in some areas editors seem to have no interest in added new sites.
In an area we keep an eye on I see some shenanigans have started happening again (Like removing a Company with 50 or so offices that’s been listed for years in the same National category to a County category (i.e. several layers further from the root)
If your site can not get listed for 3 years or your site gets “moved” then as a matter of principle it can be a big issue to you.
What Google et al. have to assess with any resource “Is this skewing of the data insignificant enough not to affect their algorithms?” If it is significant then there are two things they can do - Place less weight on the value the source delivers or build your own better model.
Rebuilding DMOZ would not be a small undertaking. Not using DMOZ data would mean the lost of huge and on the whole clean resource. The answer I would guess lies probably somewhere in between, as a result DMOZ is devalued and as a result good editors leave and the cycle goes on.
|You don't know that being in a directory is of low value to google, Google has made no such comment, |
Value of directories for general web search is about zero. The whole history of Google is beating directory of Yahoo using algorithmic search. Why would they want to invest money into creating directory which is anti-thesis to what they did so far? There was some use for DMOZ and they took advantage of it (it was free afterall), it also helps Google identify DMOZ clones, but to actually put efforts beyond this passive use is not something that Google appears to want to do. This does not mean they will tell you that, but I am inclined to think that actions speak much louder than words.
Why would Google want to build their own when they get the DMOZ one for free.
I'm sure that Google does a cost-benefit analysis. The overhead of updating the directory is surely pizzley at the most. The overhead of one more link in their list of dozens of subsites -- again, somewhere below pizzley.
The value of the directory is presumably greater than pizzley, and of course it doesn't have to be MUCH greater than pizzley, in order to make keeping it a no-brainer decision.
Replacing it would be very expensive -- I'd estimate in the 7 or 8 digits. Would the INCREASED value of their own version make that kind of investment worthwhile? I suspect not. But I'm not the business manager making that call (or, for that matter, any other kind of Google employee.)
Google doesn't mind posting content that is absolutely useless to most users (patent applications, books, etc.) so long as they think there is some class of users who WOULD find it of use. After all, they have a LOT of disk space.
no, im not sure i agree
Google COULD build a directory and if they charged a review fee like Yahoo it would be self funding however, they probably think that the same effort and investment into search would be better spent.
Over the last few years the google directory has slid from being a link on the google home page (1 click away), to the list of services tab(two clicks away), to the "more" tab (three clicks away). That tells me that the directory use is falling in interest and its also hard to moneytise a directory effectively.
Yahoos is going the same way and a number of other directory sites have changed to PPC type listings to make them more viable finacially.
I know some of the dmoz editors here are still pro the directory but search is so much more convenient to use, that a directory now is boardering on being obsolete. The net is also now so large that any directory would struggle to keep up even if charging review fees, let alone a free one.
Dmoz just fills a stop gap for google imo, it costs little so may as well leave it as it is for now. If google directory had a lot of users it would have adwords on it imo
Thats my take on it anyway and why google dont bother building their own directory and its probably a short matter of time before they just disband with it all together
[edited by: RichTC at 12:02 am (utc) on Jan. 26, 2008]
RichTC, good analysis!
I think your last statement says it all. It would probably look better for Google if they did away with Google Directory all together. Relying on an outside data provider which has a lot of baggage (controversy) doesn't help Google's image and if the service is not helping Google's business model there is no use keeping it around.
Have you considered that the very reason they use it and keep it around is that it IS helping their business model.
|if the service is not helping Google's business model there is no use keeping it around. |