tag:blogger.com,1999:blog-6569681.post116543106651807593..comments2024-01-15T13:17:33.771-08:00Comments on Geeking with Greg: Spam is ruining DiggGreg Lindenhttp://www.blogger.com/profile/09216403000599463072noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-6569681.post-1166835935799232922006-12-22T17:05:00.000-08:002006-12-22T17:05:00.000-08:00well, i stopped visiting digg half year ago, exact...well, i stopped visiting digg half year ago, exactly because of the spam. and it was such a nice site...Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6569681.post-1165605176939954782006-12-08T11:12:00.000-08:002006-12-08T11:12:00.000-08:00That's a good point, Jeremy. It is hard to know h...That's a good point, Jeremy. It is hard to know how "good" Digg is without comparing it side-by-side for clicks or user satisfaction with some other content.<BR/><BR/>Kevin might argue that, as long as people are using the site, it is useful. That probably is true, but it doesn't help determine how much more useful it could be with changes.<BR/><BR/>My post is more trying to argue that spam is getting worse and worse on Digg because of the profit motive for spammers. I am predicting (speculating?) that this will eventually overwhelm Digg, make it useless, and cause most people to abandon the site.Greg Lindenhttps://www.blogger.com/profile/09216403000599463072noreply@blogger.comtag:blogger.com,1999:blog-6569681.post-1165543625912879732006-12-07T18:07:00.000-08:002006-12-07T18:07:00.000-08:00Hey Elias - Servus! - how're things in Japan with ...Hey Elias - Servus! - how're things in Japan with Masataka? :-)<BR/><BR/>I guess my question would be, either to you or to Greg, what does "spam" mean in this context, anyway? If the goal (as Greg says in his <A HREF="http://glinden.blogspot.com/2006/12/youtube-cries-out-for-item-authority.html" REL="nofollow">next post</A>) is to help people find the information they need, and the stories that make it to the top of Digg are stories that people need, are they really spam?<BR/><BR/>In other words, if the story is good, does it matter how it got to the top? I am not arguing one way or the other; I'm just asking the question from the evaluation perspective. If the goal or the objective function is satisfying consumer demand, what is the role of algorithmic purity? <BR/><BR/>Actually, from a true wisdom-of-crowds standpoint, a truely algorithmically pure algorithm would not let any user see any other user's diggs until the voting period was done. Independence among actors is one of the key, crucial aspects of wisdom-of-crowds. And so it is already clear that Digg isn't completely algorithmically pure; it already does not correct for non-independence biases. So how much purity does it take to be pure?<BR/><BR/>About a year ago, at an open-forum panel discussion @ Yahoo, I asked Kevin Rose a related question: "How do you know if you are good?" He stammered for a while, then tried to explain to me the notion behind Cranfield-style ad hoc retrieval evaluation (i.e. standard search evaluation). My response was, "yes, yes, I know that's how you do it for search. You take two systems, and that system with more relevant docs at the top of the list is the better. But how do you do it for Digg? When the modus operandi is people voting for stories, and displaying those stories with the most votes.. how do you know if that story is good or relevant or satisfying of user needs? How do you know if Digg is good?<BR/><BR/>He looked at me like I was a bit crazy, like I was silly to even ask a question like that. It was obvious, he felt, that Digg is good, because it shows stories that gets lots of diggs. But to me that sounded like circular reasoning. It sounded like tyranny of the majority, with no consideration for the larger number of people in the splintered, long tail minorities. <BR/><BR/>So it seems to me that before we can figure out whether spam is killing Digg, we have to be able to distinguish between (1) when spam actually gives people what they want, and (2) when spamless Digg doesn't. In other words, is Digg good?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6569681.post-1165477858140783202006-12-06T23:50:00.000-08:002006-12-06T23:50:00.000-08:00I think you are way too negative. There will alway...I think you are way too negative. There will always be some form of spam, blogspam, ... but overall it's much too useful to surrender because of some Spam, just like emails. <BR/><BR/>It would be interesting to know what kind of companies are successfully spamming Digg. Do you have some examples? (At least on the front pages for the different categories I've never seen any.)<BR/><BR/>You might have linked to this in one of your early posts... anyway, here is a comment arguing why <A HREF="http://www.seomoz.org/blogdetail.php?ID=1436" REL="nofollow">why spam doesn't really work on digg</A>.Eliashttps://www.blogger.com/profile/06133166115774380054noreply@blogger.comtag:blogger.com,1999:blog-6569681.post-1165467082164617822006-12-06T20:51:00.000-08:002006-12-06T20:51:00.000-08:00That's cool, Greg. Sorry if my Dude comment came a...That's cool, Greg. Sorry if <BR/>my Dude comment came across<BR/>as rude or ad hominem. I'm<BR/>glad that Findory and this<BR/>blog are around. Didn't want<BR/>you to adopt Bogatin style,<BR/>'at's all.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6569681.post-1165465340783227892006-12-06T20:22:00.000-08:002006-12-06T20:22:00.000-08:00Sorry about that, Anonymous. An alternative is to...Sorry about that, Anonymous. An alternative is to copy-and-paste (or slightly rephrase) things I wrote in previous posts, but I thought it was both more honest and more useful to do excerpts.<BR/><BR/>Jeremy's comment seems to indicate that at least some prefer the references to previous posts over pulling content from previous posts without the reference, but I'll think about if there is a way to do it that doesn't look so repetitive. Thanks, Anonymous.Greg Lindenhttps://www.blogger.com/profile/09216403000599463072noreply@blogger.comtag:blogger.com,1999:blog-6569681.post-1165455732433546742006-12-06T17:42:00.000-08:002006-12-06T17:42:00.000-08:00The "see also my previous posts" is one of my favo...The "see also my previous posts" is one of my favorite parts of this blog. What it says to me is that Greg isn't just knee-jerk reacting to the blogosphere. It says that he has got a larger context in which he is discussing many of these issues. And, if we are interested, we can follow that context.<BR/><BR/>I wish the regular news were so considerate. I wish, when you tuned in to CBS or CNN or even CNET, they tried to provide a bit more context and perspective instead of just dumping stories at the viewers.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-6569681.post-1165437662290217672006-12-06T12:41:00.000-08:002006-12-06T12:41:00.000-08:00Dude, you need to go a little easyon the "See also...Dude, you need to go a little easy<BR/>on the "See also my previous posts"<BR/>routine. You sound a little like<BR/>Donna Bogatin with her<BR/>content-free self-referential<BR/>posts on ZDNet. Though, unlike her,<BR/>you do have meaningful things<BR/>to say most of the time.Anonymousnoreply@blogger.com