Forum Moderators: open
It's hard to imagine any other site of similar significance that is so willing to both schedule outages during prime usage time and tolerate (apparently) unplanned ones. I realize we aren't talking about a site like eBay here, but just about every site of any significance attempts to avoid downtime for any reason. If an outage must be scheduled, typically it is done in the wee hours on a weekend to avoid interrupting most users. Normally, too, major changes may be tested using redundant systems so that if unexpected problems are encountered it's not difficult to roll back to the functional setup.
DMOZ may not be taking orders online, processing travel reservations, etc., but letting these kind of protracted outages happen risks alienating the volunteer editors that are so critical to its survival as a viable resource. Legitimate site owners will also be discouraged if they keep trying and can't get through. (The spammers will just try later...) Overall, the whole experience certainly suggests a feeling that DMOZ isn't important enough to maintain and upgrade properly.
If they can't maintain it properly they should hand it over to someone who will (Google?). I get the feeling though that they would rather kill it than give it someone who knows how to use its prestige positively.
Did I miss something?
Why manage all of that information for the sake of the editors? If that was the case then why would anyone who wasn't an editor care if the editors saw their site?
I think the main thing is that the editors care about what they are doing - that way the end product is of a high quality.
Lets face it - all us volunteer editors arent doing it for the glamour and recognition, are we?
JOAT
The database backend management system currently in use was never designed as user friendly.
With the upgrade comes better, faster more efficient service from volunteers that like doing what they do.
I for one will be much happy to wait -- (since I do wait when DMOZ is running -- the old way).
The recommended way to use that data is per the RDF dumps, which makes the end users completely independent from the dmoz.org uptime. Those sites that "leach" directly from the site are currently tolerated, but face the risk of going down with it during each maintenance term. If you want to search the ODP data in an efficient way, use the Google directory. It is both faster and offers much better search features.
I agree with the comment about dmoz.org traffic up to a point, but I don't think that justifies the lengthy outages we are seeing. I've got sites that do far less traffic than dmoz (and are of infinitesmal importance by comparison), but I'd be horribly embarrassed if planned upgrades (or any kind of failure, for that matter) caused this sort of downtime. These outages suggest a lack of testing and/or a lack of a backup plan if problems occurred.
But I stress this is rumor ;)
dmoz.org site itself is used by relatively few end users
Well this is somewhat inaccurate. In September, my site recorded over a 1,000 visitors directly from DMOZ.org.
This maybe pale traffic compared to Yahoo, and market and research dependent, but alot better than Looksmart and Fast (at least for me).
While most heartily sympathizing with the frustration of all of you who can't find your favorite site: a couple of points: For non-editors, it is just adding a day to the immensurable amount of time it will take to get a site listed, that's all. You aren't going to lose any visitors.
And, as to the proposal: I hereby nominate this the "King Canute Memorial Thread."
When I was unable to reaching dmoz.org [now it works], in fact the server was not down, it was firewalled on the 80. [->strange thing instead, 25 was not filtered :)]
So, seems logical, if this happened from 1 October [I don't know, don't tested before today], to assume that Googlebot has spidered dmoz in the last days, also if we've seen it down so often.
Seems logical to assume also some other things IMHO, but who knows.. ;)
cminblues
OK, I'll bite... :)
