Two folks blogged about my thoughts on having a corporate ombudsman for Mahalo, and the “publication that prints lies constantly, and will probably be sued for libel soon” (you can guess their name) dinged us for having a link to my blog (gosh!) on one of Mahalo’s pages (the iPhone Problems page). Some thoughts on what folks are saying:
- Certainly a challenging job. I don’t know if it’s even a job for one person, or if that person could survive it for very long. The “you should have this in your results”, “this should be the first result” criticisms will probably come in hot and heavy. Myself, I know that I’m not capable of this kind of job. Sure I’d try to be fair, but I know it would be so hard.
Will a Mahalo ombudsman silence all of Jason’s critics? Hell no. However if Jason does put this into place and picks someone who it truly respected and can be honestly impartial, I think it could be a model other companies could replicate for themselves (Flickr for controversial photos, fan fiction, etc).
I agree, it’s not an easy gig. However, the small stuff around “I should be above this person” is going to be easier to navigate in my mind. At this point we have the discussion boards available on Mahalo and if folks want to debate a SERP we will listen all day long and make the decision based on the communities input. We have 50 full-time folks, 800 part time, and tens of thousands of users already. My feeling is that intelligent discourse is the best way to come to a decision. Is it perfect? No, but it’s a lot better than the stance that machine powered search engines take: “talk to the algorithm.”
The bigger issues? Those are the ones I’m thinking about, like:
- Who falls into the first seven links on the abortion SERP?
- What about hate speech? What should the top seven links for a hate group or racial slur be?
- Should we have racial slurs in our index or just dump those search queries to machine search (i.e. should we pun”).
- Should we link to adult content, how should we do that, and how do we protect kids? Right now Google has a safe search setting–should we model our service after that? Should we dump adult searches over to machine search engines and keep Mahalo as a “safe search” engine?
- Should take a position that the we have a result for every, single search term? Should that include “how to build a bomb” or “how to INSERT ILLEGAL ACTIVITY HERE.”
- Do we have a higher ethical and moral benchmark since we are human edited?
I don’t have the exact answers for any of those issues. In fact, I’ve been avoiding dealing with those issues for the first six months while we get the top 25,000 search terms completed. My hope is to first build a community of individuals who respect and trust each other, and then let them debate it.
An outside perspective, even if it didn’t provide answers, I think could help guide a discussion. It would also help make sure that anyone could have input without feeling it would have a negative impact on their relationship with Mahalo (i.e. what if one our team disagrees with a position, but does’t want to bring it up with the group).
Robert Seidman brings up some issues at 2AM:
- First of all, I agree with Faulkner, the link to Jason’s own twitter-esque posting about Jason’s frustration with the lack of cut & paste on the iPhone should not be on a site intended to provide “useful information”. But that wasn’t what raised the biggest question for me.
When I saw that the page had been authored by Conrad Quilty-Harper who writes for Engadget, with links to stuff Conrad himself had written on Engadget it set off all kinds of flashing red lights about bias (which Jason’s Ombudsman post was about).
I confess to being an Engadget junky (hey, I read Valleywag religiously, it’s not like I have any shame), but the thing is, aside from a stray self-destructive tendency here and there, I don’t believe in the notion of “negative asymmetrical risk”. That’s just a fancy way of saying I don’t believe in things with very little upside and huge downside. The downside here, is that people will question, and rightfully so in this case, Mahalo’s bias. Certainly this sort of thing will be mocked on Valleywag, but it would be a legitimate question/concern by anyone who raised it.
My advice: don’t use Conrad for technology pieces or at the very least, don’t let him link to his own pieces because even if they are the best ones, people will wonder about…bias. It’s not worth it. Also, because of Jason’s history, Mahalo needs a company stance about linking back to stuff in the Weblogs Inc. empire or risk a lot of flack.
Very good catch Robert. We do, in fact, have a rule that folks can not link to themselves or do SERPs that they are involved in. So, Conrad, who is an intern at Mahalo, made a mistake in linking to his own work. It’s actually my fault because I asked him to do a serp on all the iPhone problems because I thought it would help people. Of course, I didn’t realize that Conrad did not actually go through our Guide training problem.
This one is 100% my fault. I’ve done two things since reading your post Robert:
1. I told the team please never link to me unless that have to because it would look like a silly omission to not have me linked (i.e. on a page about Weblogs, Inc. or Silicon Alley).
2. I asked our editorial director CK Sample to have someone take over the “iPhone Problems” SERP and make sure we are not linking to anything that might look like bias.
I agree 100% that perception is reality and that you’re better off avoiding all this. In fact, our radical transparency–that is built into the system–makes debates like this possible. The second we do something that has a hint of bias we will get our butts kicked… AND THAT’S A GOOD THING!
We want to get called out when we make a mistake because it will make the product better. In fact, I’m trying to figure out a way to REWARD people for finding bias or problems on the site.