a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment by mk
mk  ·  4660 days ago  ·  link  ·    ·  parent  ·  post: Can we have a feature to ignore particular hubs?
I see. Ok, bear with me here.

I don't follow thenewgreen. You'd think I would, given our relationship, but he follows a ton of people and is a very active sharer. I like some of the stuff he does, but not all of it. At the same time, I do see a good number of his posts, because people I do follow either follow him, or follow people that do. And, as it works out, thenewgreen I see, is more-or-less the kind of thenewgreen I dig. For example, I follow cgod and he follows thenewgreen. When cgod shares his stuff, it's almost always the slice of thenewgreen I am interested in. I can see that your example with derek is a test case where this might not work perfectly, but whack-a-mole with tags likely won't be perfect either. I could see thenewgreen constantly surprising me with new tags on content I'm not keen on (#thebeatles4evar) Sorry tng! :).

That said, I was thinking on the ability to ignore external posts based on tags, and that is something that I am pretty sure I'd like to implement. If the Hubski crowd loves #nastystuff, it shouldn't be a constant impediment to me finding new content.

Humor me and see how things feel after that implementation. Like the ignored users, I am going to make ignored tags red. That might help someone determine if there are better ways to structure their follow list. i.e. Maybe there is somebody sharing a lot of thenewgreen posts that you like that you aren't following.

I've got to drive to work, but I'm happy to continue this discussion. There are a few things I haven't mentioned. I do appreciate the feedback, btw.





mk  ·  4660 days ago  ·  link  ·  
At work. Just one more thing to add regarding ignoring posts by tag on users you follow: it does have some technical baggage. If a very large of users had a very large number of ignored tags that needed to cross-referenced for every post in their feed, it could actually create a pretty significant demand. Of course, there are ways to deal with this, but all baggage brings more baggage. Ideally, posts should be served without too much overhead (individual requirements for seeing a post ideally have a light data demand). IMO the best way to think about it is: What questions does Hubski need to answer before I see this post? and how much digging is required to answer those questions? Those types of considerations always come into play. They aren't the primary factor in my decision here (they don't need to be yet), but I always have to think about that end too.
kleinbl00  ·  4660 days ago  ·  link  ·  
You're describing "trickle-down" social economics.

On the one hand, you're arguing "the system will pick up the slack." Okay, that's easy for you to say - you designed the system. It will undoubtedly give you what you need, or you'll tweak it so that it does. It's not so easy for the rest of us to hear - after all, most of us are here because the "system" of Reddit failed us. Your vision of the way this works is "if something is popular enough, you'll see it because your friends like it." The problem is that the way you envision the system, if my friends consistently like things that I don't, they're going to cease to be my friends. Your example with thenewgreen is perfect - you know him best, but you don't follow him because he shares too much. So if he finds something cool that you might actually like, you'll be the last to know. That's unacceptable. You're basically saying "Hubski tracks influence as far as who follows who, but eventually the whole thing crumbles because once someone starts indulging interests that don't appeal to you, your best bet is to dump him entirely, good with bad."

Your "baggage" complaint is a red herring, related to your unwillingness to address "tags" the same way every other institution on the planet addresses them. If there were consensus on tags, it wouldn't be that big a deal - your problem right now is that "tags" are fundamentally "editorialization" with a mix of categorization thrown in for sport. it's the #media/#foxnews/#murdoched dilemma - these things should, systematically speaking, be one thing, not three.

Finally, you're extending dislikes in the same direction as likes. Most people "dislike" far fewer things than they "like" - while most people like ice cream, some don't like rocky road, some don't like raspberry swirl. Likewise, not a lot of people are going to block #music but some of them might very well block #dubstep.

If it really worries you, limit us to ten "blocks." You're worried about problems down the road, and I think those are things worth worrying about... but your users are asking for things now. It's scalable now. See how it works. If it gets unwieldy, deal with it when you start to see the first inklings of problems, not when they're theoretically over the hump. Again, I think that a proper approach to the "tags" issue clarifies a lot of this.

b_b  ·  4660 days ago  ·  link  ·  
"...after all, most of us are here because the "system" of Reddit failed us."

I wonder if this is a source of some of the disagreement. I'm not going to speak for mk, but I think he may be worried about creating a lesser version of reddit, instead of a new and different platform. I never was a redditer, so I'm naive about the subject, but its more subject-centric, rather than user-centric, no? I wonder what your thoughts are on how to maintain some website identity if tags become the focal source of content filtering on hubski (or maybe I'm missing what you're saying). It seems that following people is the essence of hubski, and that following tags is just a convenient way to find new, possibly like-minded, users. Thoughts?

fr33lunch  ·  4660 days ago  ·  link  ·  
I, too, was not a reddit user nor really any other social news site. I have been fairly active in subject-specific forums that interest me, woodworking, home theater, off-roading, etc. While neither of those topics are directly linked, the vBulletin software that they all use is the same. The format makes it easy to slip into the conversations without thinking to much about the process.

My company uses a credit card terminal to process all electronic types--like most retailers. However, because of our OS, we went with a different type of terminal than you might find at most retailers like Walmart or Home Depot. So the process of verifying the purchase amount, swiping, and authorizing payment are a bit different too. The terminals display step-by-step directives so that the customer can know how to proceed. Instead of actually reading the directions many turn into Neanderthals--mashing in the keypad, tugging on the power cord in search of a stylus, swiping their card over and over and over, all while muttering the phrase "why can't they make these damn things all the same?!" or the occasional "they're all a little different".

I think we all become Neanderthals when we are accustomed to something and then are introduced change. We like what we like--at least what we think we like. http://hubski.com/pub?id=18703

From a business standpoint, I know the perils of failing to make a product to fit the consumers needs. As a craftsman, I want to make something that I am proud of. I know that the right person will find my product as awesome as I do. However, I am not a craftsman by profession. Anything I make is part of a hobby. I don't rely on "customers" for payment. If I make something for someone it's usually as a gift. If they don't appreciate it they can go fuck themselves and I'll refrain from making anything else for them.

kleinbl00  ·  4660 days ago  ·  link  ·  
>I, too, was not a reddit user nor really any other social news site.

"I've never driven a car before, so I don't see why you should have a steering wheel rather than a steering tiller."

A little video history, if you will, that's not necessarily en pointe but still entertaining:

http://www.youtube.com/watch?v=_amZsf8A1Lo

"Tags" are not an invention of social media. They are not an invention of Reddit. They are not an invention of Del.I.Cious, they are not an invention of Facebook. They aren't even an invention of Getty or Corbis, who have far more claim to the title than anyone else. They're an invention of Henriette Avram for the Library of Congress, who have been using them since 1968:

http://en.wikipedia.org/wiki/MARC_standards

This isn't a "I'm a neanderthal mashing the buttons" because I'm not used to change. This is a "taxonomy without meaning is useless." It's like your terminals - what sort of pompous, entitled, self-important retailer thinks that I should be required to read directions in order to give them money? I'm the customer. If standards have developed, ad-hoc or otherwise, those who deviate from them are the ones with explaining to do, not those who do not choose to adhere to your whims.

Every single person using this site learned the Dewey Decimal System in 3rd grade. Every single person using this site has been involved in some form of categorization or other. And every single person using this site has two things they can follow: users and tags.

I can block users. This was suggested, implemented and celebrated. I can't block tags. This has been suggested and the argument against it is "well, everybody else does that, but we want to be different."

If you're going to alienate your customers, you need to give them an experience better than what they'd get elsewhere. If you're going to deviate from the norm, you had better be able to provide substantial benefit to put up with the alienation.

Putting the brake pedal on the right because it seems like a neat idea and let's see how it works out is not visionary thinking.

fr33lunch  ·  4659 days ago  ·  link  ·  
I'm in agreeance about the need for a more universal form of tagging. It is actually one of the first things I was looking for when I first got to the site. Since the current process is completely subjective, I don't even bother using tags to follow content. So for me, bI'm already ignoring tags - just not in the sense that you guys are asking for. I believe it was said in a previous post, but the ignoring of tags would be worthless until a stable process for tagging was adopted.

In an effort to illustrate further what I'm talking about in regard to the terminals, I'll share with you this candid ditty. Everyday I hear my customer's frustrations. Everyday I also am frustrated with the process of explaining to them the simple process of waiting for the total, agreeing to the amount, and swiping their card....would you believe that even after instructing customers countless times a day how to use them. If I'm making a purchase through them, I will often swipe first and then wait for a total even though I know how to use them.

My company isn't as much pompous as it is cheap. The OS that we use was written in the 80's. It infuriates me that they haven't upgraded it or many of the other processes that would be very cost beneficial to fix. But, I'm low on the totem...

kleinbl00  ·  4659 days ago  ·  link  ·  
I hear ya. I didn't mean to pick on you with the terminals - I just didn't like the argument you were making.

Tags, whatever you want to call them, are useful. That's why they've been adopted. This shouldn't be a controversial point. I'm willing to spend a lot of time exploring the best way to implement them, but I'm pretty firmly convinced that a feature that has become commonplace elsewhere shouldn't be treated cavalierly just because we're "trying something new."

fr33lunch  ·  4658 days ago  ·  link  ·  
Not a big deal. Not my best argument I suppose.
mk  ·  4659 days ago  ·  link  ·  
I can block users. This was suggested, implemented and celebrated. I can't block tags.

To the extent that you can block users, you will be able to block tags with the next update.

mk  ·  4660 days ago  ·  link  ·  
Not much time to write yet. There is some truth to what b_b said as far as motivations are concerned.

It seems that following people is the essence of hubski, and that following tags is just a convenient way to find new, possibly like-minded, users.

Also this, which you said:

Hubski tracks influence as far as who follows who, but eventually the whole thing crumbles because once someone starts indulging interests that don't appeal to you, your best bet is to dump him entirely, good with bad.

Seems to me to hold particularly true for tags. But it's not one user, it's a crowd creep. -That's one of the failings of subreddits, IMO.

I did code up the ignore-by-tag for external posts. I'm going to test it more, and maybe push tonight. I have been working on PMs, and I think that's something everyone can agree should be done. (I think you'll dig the way PMs work). Anyway, I have to pass some cells, drive home, and cook for and entertain some company. I'll revisit.

kleinbl00  ·  4660 days ago  ·  link  ·  
>It seems that following people is the essence of hubski, and that following tags is just a convenient way to find new, possibly like-minded, users.

Fine. Right now, though, any given submission has four variables (title, URL, text and tag) and two ways to subscribe to something (user and tag). Like it or not, tags are a big part of the way content gets distributed.

Frankly, I'd rather have no tags than bad tags. Right now, Hubski has bad tags. You'd be better off using any of the tag-clouding plugins, have them parse the titles of my posts for statistically-significant nouns, then present those clouds in some sort of flower gallery for users to browse. Tags as they are currently implemented are either so vague as to be useless or so specific as to be commentary rather than categorization.

If you're going to have two dimensions of affinity, those dimensions need to be equally useful. You have a very firm idea as to what "following users" is, and it shows. You seem to have more nebulous ideas about what tags aren't and it also shows. If "tags" are something that you don't want to support, drop them and spend your efforts on something you believe in. Because as it is now, "tags" are nigh onto meaningless.

mk  ·  4659 days ago  ·  link  ·  
I actually wish that there were a better name for tags. I tried to come up with one for a long time.

Categorizing content by topic is something that has been implemented in many places, with mixed results. In the case of Reddit, topics are as much (and sometimes more) definition of a community, than of subject matter, and the content and its quality is dependent upon the number of contributors. Community content usually has a lifecycle, and eventually the forum drops in value. It's not something I wanted to mirror. Its not terrible, but it has been done and done.

Hubski is in large part designed with the purpose of gathering around personalities and interests, rather than topics. That's the experiment. The jury out on whether this is an improvement when compared to something like Reddit. But that's not my yardstick. I'm more focused on whether or not Hubski is interesting and rewarding, rather than whether or not it is an improvement on Reddit. In fact, it might not be easy to make the comparison when all is said and done. i.e. Are Twitter or Google+ better than Reddit? Depends on the expectations.

I implemented tags to provide a dimension of user-based grouping, whatever it might be. If a user wants to create a tag to subcategorize their posts, that's fine. If a transient event like OWS pops up, and users want to group their posts around it, that's fine. If users want to editorialize with them, that's ok too. Tags are also likely to be useful when I implement search and can improve recommendations. The tag is an optional color to be added to the post. Part of my reasoning for tags is due to the organic creation of hashtags on twitter. In the user/subscriber model of content curation, people wanted a way to group content, so they created one. Hubski doesn't have a search mechanism yet. But if it started with one, I suppose organic hashtags could have arisen to turn 'Hubski search' into a filter. One of the biggest problems with Hubski tags IMO, is not what they do or don't do, but the expectations that they bring.

As a means of subscribing to topics, Hubski tags are not very effective. If I wanted to support subscription to topics, there are obviously better ways to go about it. But I have spent a lot of time in topic-based communities, it's not an unsuccessful way to build forums. But I set out to try something different. Perhaps you are right, and having unequal dimensions of affinity is deleterious to this effort. Maybe we should eventually drop tags altogether. That's an idea I am willing to entertain. Probably more than expanding upon them. In fact, perhaps the best way to go about this is just to make an effective and interesting search/recommendation engine which users can leverage in their own way.

At this moment, I am going to roll out this ignore-by-tag for external posts, which mirrors the ignore-by-user for external posts. I'd like to see how that works, and to what extent it is used. I'm working on PMs, and that is my current short-term goal. I want to roll that out, and let it set in. Barring some unforeseen demand, I'd like to decide the future (and clarify the goals) of tags then.

kleinbl00  ·  4659 days ago  ·  link  ·  
There isn't a better name because there's a history and lineage to tags. They are what they are. The reason they aren't "tags" on Reddit is because tags are uncurated. Subreddits are curated. someone is ultimately responsible for what gets assigned to that topic.

It definitely makes tags more ephemeral, more dynamic and more fluid. That's a good thing.

The problem is that in hubski, you have built a social network based on friendship and affinity without recognizing that "common interests" are one of the fundamental basics of friendship. People meet their friends through church, through hobbies, through shared addiction, through all sorts of commonality. If I were to meet Gary Cobain on the street I'd probably want to hit him. But since I'm a big, big fan of the Future Sound of London, I'm likely to extend him a lot of rope as far as his opinions and viewpoints. Likewise, I really like hearing what David Bowie has to say about stuff, which makes me more interested in his music - I stopped listening to Laibach and Echo and the Bunnymen cold having mixed them and interacted with them in person. They're assholes, but you wouldn't know that from listening to their music. (well, Laibach maybe)

You see "tag" and think "subject" or "category." I see "tag" and I think "interest." Do you really think members of a model railroad club would hang out together if they didn't have the HO vs N rivalry to tie them together? This expectation isn't related to "tags" it's related to human nature.

Both eHarmony and Match.com use the Meyers-Briggs personality test to get their matches. Here's the interesting part: eHarmony matches people with similar profiles. Match.com matches people with opposing profiles. They both seem to work. Personality only takes you so far, though. What do you have in common?

This stuff matters. Hate the word "tag" all you like but recognize that you didn't include it out of obligation, but out of societal necessity.

mk  ·  4659 days ago  ·  link  ·  
someone is ultimately responsible for what gets assigned to that topic.

I suppose that is the other deliberate experiment here: How much can we avoid top-down curation?

I see what you are saying about interests. Just thinking out loud, perhaps some sort of option is rather than followers slicing up a user's posts, maybe posters can purposefully create subchannels for partial subscription. A user might be allowed a finite number, they can name them, and there you go... Following would give the default option of the total hub, but you could follow a subhub instead. Hm. Something to think on.

kleinbl00  ·  4659 days ago  ·  link  ·  
>I suppose that is the other deliberate experiment here: How much can we avoid top-down curation?

I would aim for 100% and consider it a failure if it's any less than 99.999%. Curation belongs on blogs.

>I see what you are saying about interests. Just thinking out loud, perhaps some sort of option is rather than followers slicing up a user's posts, maybe posters can purposefully create subchannels for partial subscription.

This really is a semantic issue, not a philosophical one. I just don't see being able to ignore one aspect of a person's interests to be "slicing up a user's posts." I don't much care to discuss politics with my father-in-law, but I eagerly listen to anything he has to say about orchids. That doesn't mean I'm not 100% committed to him as a "friend" it just means he has interests other than my own. Likewise, I'm interested to hear what anyone has to say about orchids, regardless of how well I know them - and maybe, through affinity to orchids, I'll discover we have other things in common.

You see "fragment." I see "aspect."

b_b  ·  4660 days ago  ·  link  ·  
I actually agree with you on this point. I'm not really a fan of the tag thing generally. I don't think they're too value added.
thenewgreen  ·  4660 days ago  ·  link  ·  
I see them like I would signs in a book store "self help", "travel" etc. They are tools to discover. They are also a tool to archive. I recently started using #tng to tag songs I've written. This is so I have an easily sourced catalog that I can reference. I realize that nobody else will ever use it as such but it's strictly for me.

Beyond that, I will occasionally go to the "discover" link and read posts from a specific tag. I tend to use the "video, best posts, active and badges" links on the bottom of the screen VERY often. I think they get often overlooked.

I guess what I'm getting at is that I agree, Im not a big "tag" fan. I could really take them or leave them for "following" purposes.

blimpy  ·  4660 days ago  ·  link  ·  
being able to control the "outside posts" seems a good solution to me. I think I may be following too many people. i need to be more picky. thanks mk.