5 ms·
I don't understand how this isn't a settled issue. If clickstream data is 1 of 1000 signals, and you create clickstream data for a specialized query that will n
by boredguy8 16y ago
I don't understand how this isn't a settled issue. If clickstream data is 1 of 1000 signals, and you create clickstream data for a specialized query that will never trigger off another signal, then your created data will be reflected. That sounds exactly like what happened.
You'd have to make the argument that using this data is wrong, somehow. But to make that argument, you'd basically have to argue that users shouldn't be able to share their habits with whomever they want to. I doubt that argument can be made in a compelling way.
I'm surprised Google is pushing this further, and a little disappointed.
- ddkrone 16y agoYou should be able to share your results with whoever you want but that's not what happens. You share your data with just bing and google. You don't really own any of the data you are generating so saying you should be able to share it with whoever you want is saying hop before you've jumped.
- boredguy8 16y agoWhere I click on a screen, and what I click on, is not my data?
- ddkrone 16y agoOf course not. Try inquiring google or bing about obtaining all the information they have gathered from your toolbar and see what response you get. Actually I'll tell you: "The gathered data is anonymous and we have no way of verifying who sent it to us". So how exactly do you own information that you have absolutely no access to and with no way to transfer to competing search companies?
- ugh 16y agoThat’s decidedly not what owning your clickstream data means. The argument is that users should be able to decide to give away their data, whether they can later access that data is irrelevant. Access to the collected clickstream data is simply not part of the agreement and as long as users are not coerced or tricked into agreeing that’ certainly unproblematic. (I can similarly agree to give away my photo collection without getting any right to access it ever again. That doesn’t mean that I don’t own my photo collection.)
- ddkrone 16y agoYou just said you can give away your photos by agreeing to not have any access rights to it and still own those same photos. It seems like you are using a different definition of "ownership" or maybe "click stream" than I am so this argument is pointless unless you define exactly what you mean by those words. By "click stream" I mean the aggregate history of clicks along with contextual information related to those clicks as one coherent, single entity and not individual clicks looked at one by one each of which was owned by the person that generated the click and who transfered ownership of that single click to somebody else. If you meant giving away a copy of your collection then that's a little different but that's not what's happening with user generated click streams. The only click stream copy is the one at google, microsoft headquarters and you have no access to it so you don't technically own it.
- ugh 16y agoNo. You can give something you own away. You obviously don’t own it anymore after you gave it away. Being able to give what you own away is part of ownership. (You can’t give something away you don’t own.) Since you can still make a local copy of your clickstream data (every browser has some sort of history function) you don’t even really give it away, you keep it and hand over a copy (maybe not technically but it is still the exact same data) to Bing. Here is an analogy: Imagine a scale which, after you agreed to it, automatically sends your weight to the manufacturer every time you use it. The manufacturer made it clear when you agreed that you will not be able to access your weight history. While slightly creepy, I don’t see any ethical or legal problems with that. You own the weight displayed on the scale, you can therefore also agree to send this weight to whomever you want, even if they don’t let you access your weight history. You are also free to make your own local copy of your weight history (for example by writing you weight on a piece of paper every time you use the scale).
- thezilch 16y agoAccept in the "original" [torsoraphy] search. Forget Bing having to compete with Google's "spell correction team;" Bing need only use Google [copy] as a high frequency signal on tailing queries. That sounds like exactly what happened, and it's wrong.
- ugh 16y agoWhy exactly is it wrong?
- DenisM 16y agoLook at it this way: Google is setting trends on tail terms due to its massive market share. Once the trend is set, Bing captures the trend by monitoring user's behavior and provides that to the other users. By reflecting user's behavior in its own index, Bing seems to replicate the Google's index structure which created that behavior. Had there been no Google, there would have been different trends and Bing would capture those instead. It's not Bing's fault that Google is so influential, and certainly not a reason for them to stop dedcuing relevance frm user behavior. Just because Google creates trends, does not mean they own trends. The users own their own behavior and are free to share it.
- hackinthebochs 16y agoThis comment really gets to the meat of the issue here. Search is about discovering and then predicting user trends. Google in fact is creating user trends through its position in the market. Should Bing be locked out of the market because Google is currently in the dominant position? If Bing is allowed to discover user trends then it will necessarily end up being a replica of Google's results in some cases. There's just no way around it.
- rgarcia 16y agoI think you can make the argument that in order to have a "free market" of ideas (when it comes to building the best search engine) you shouldn't cannibalize the ideas of your competitor. If Bing's signal consisted of some generic method for mining clickstream data that just happened to pick up on clicks originating from Google, then this wouldn't be a problem. The fact that they're deliberately deconstructing Google URLs (per the Microsoft research paper that Matt cites) makes this questionable.