The # Discover Tab
I've been really impressed by the quality and especially the diversity of news I am getting from the #Discover Tab on Twitter lately.
This exchange between Hunter Walk and me from last week is a good example of how I feel about it (sorry that I could not figure out how to embed a Twitter conversation):
I actually don't love the visual treatment of the #Discover tab, particularly on mobile, but also on web. I wish it was just like the timeline. I don't need all those "blobs" of images that pop up sort of randomly and jarringly.
But the data I am getting is super relevant to my interests and incredibly diverse (sports, entertainment, friends, tech, NYC, etc, etc).
I am curious how all of you are finding the #Discover tab these days.
Comments (Archived):
strangely, I clicked on the discovery tab this morning (which I rarely do) and was pleasantly surprised by the results.Have to say I think Twitter is killing it lately the rate of adoption by media here in the UK has really ramped up. Still think it’s more of a broadcast medium than conversational though.
Its broadcast with the ability to be conversational (like Hunter and I did). It certainly is not first and foremost conversational
I’m seeing more and more spontaneous conversations spin out of Tweets on Twitter lately. A lot more than before. And they are getting better at threading them.
I’m not experiencing this.Big fan and user. I just don’t see conversations as their nature.
selfcast
noisecast mostly!
Wow. I’m surprised too.Would love to see if Twitter could separate content by topics and present it that format. Its very tweet focused at this moment (half of which I ready already on timeline).
I like it comingled. I have a very disorganized mind. I like search and hate folders.
A whole new twist on the punk mantra of ‘search and destroy’ 😉
Rarely use it – just had a look – nothing of relevance/interest, I’m afraid to say – I wonder if it ‘learns’ from repeated use – ie, what you click on to signifies a hit and thus relevance and the curation engine duly ‘learns’ and strives to ever-improve its results for you. Well, that’s what we did in ensembli – and look where that got us, lol… 😉
It should be based on who you follow, what you tweet about, reply, retweet, etc
@carl_rahn:disqus and then move into importing (customer choice) what you read, search and so on from the web. Don’t know about ensembli, Carl, but it may have been #years/months ahead of its time (if your tag is meant negative)… at this point, going after keyword/NL/DM/ML leading to personalized curation is worth it. There are ways to do it that are not that expensive getting the platform together. With the fragmentation happening in developing a true Virtual Assistant going on, the opportunity of doing it right becomes bigger.
I had started noticing that recently as well. The vast difference between now and when they first released it shows that Twitter is really going all in on their media strategy.
native
I’ve been optimistic but not as wowed initially. My interests are so wackily broad that I should be an easy implicit one to provide news to but not proving so oddly.Two things though:-want to get discovered, add photos or graphics. Twitter embedded photos drag me to new sources all the time. In fact, using Twitter more than Tumblr recently to share graphics.-with all the sources out there, I still don’t find that much new stuff outside of my conversational networks each day. Need is still there and strong.
You want to see some new stuff, follow some folks from new places. I don’t remember how it happened, but for some reason I and someone in India ended up following each other, and now I’ve got a couple of Indian correspondents on Twitter. One of them retweeted this from one of his followers last week:http://www.youtube.com/watc…Which has to be seen to be believed. Just an insane act from India’s version of America’s Got Talent. That’s something I would have never seen if I hadn’t been following a couple of these Indian folks.
sound advice.what’s the old saying….keep doing the same thing and nothing will change.
Sometimes you have to give serendipity a nudge. Don’t stop watching the video before the first minute, btw. Goes in a completely different direction than you might expect.
I love tweeting with people in other cultures. I asked one guy why he was following me when all his tweets were in Dutch and he explained that while he tweeted in his native language, he still enjoyed reading tweets in English. Makes sense. I followed him back (and a few others tweeting in other languages) — it adds flavor to my timeline.I also enjoy following @TheMime. His tweets look like this: …
Yikes! Peeked at the video. Gives new meaning to “Mad Men.” I saw 3 minutes’ worth.
I love the expressions the judges make while watching them.
So that’s why your Tumblr has been more bare. I miss your photos showing up there — since there is less to scroll through than on Twitter to see them. Anyway, can’t you just have your Tumblr posts automatically go to Twitter as well?
Hi Donna.Been careful about my time lately.Bunch of travel. Heads down on client and my own businesses. Focusing.Busy doesn’t mean productive is my new working orders to myself lately.People throw themselves into social nets a lot cause they are there, and fun, but they the easiest and the most obvious are not the best avenues for productivity in my opinion.
I hear you. Glad you’re busy…er, productive.
Everything is really good actually. You had best let me know when you are in my neighborhood though so we can go out, chat, be unproductive and enjoy ourselves!
I seem to have this finely tuned at this point. I have a tight group of people I follow. There are about 50 prolific tweeters I unfollowed because the 10% that is their good stuff gets retweeted by other people in my network anyway. Not talking about anyone here on AVC…though I won’t name names. ;)This remaining people are my discovery network…and at least right now, I discover more than I have the time to read!
You’ve inspired me to give this a chance.I need a healthy dose of random daily inspiration. Need more than I’m getting actually.Tuning my streams today!Thnx.
Made the cut. Phew!
But of course. 🙂
I’m not sure I’ve ever clicked on it, but did just now out of curiosity. It’s *way* off. Stories about pawning my wine cellar to buy an Aston Martin (I don’t have a wine cellar, or think I’ve ever clicked on any links relating to wine or Aston Martins), a libel damage case about a z-list celebrity (totally irrelevant, not the sort of thing I’d read) etc.
Just tried it with other account, seemed much more accurate. That one is new, and only following a handful of people, so that suggests to me that the discovery is being done on who you follow, not what you click on.
It’s both…if you have the option checked to use your web surfing to personalize your twitter experience (under settings).
This is something that will improve the more you use Twitter. I’ve not been a very heavy user lately so I think it’s having some problems with relevance but that should improve if I start to tweet and interact more.I’ve been saying for almost a year that I can envisage #discover becoming the default view for Twitter as we continue to use an increasingly visual web. Twitter Cards make a lot of sense in this scenario.
I do check the #Discover tab sometimes.. but i look at it with an attitude similar to the one when i look at my gmail spam folder.
Yes, it;s a tough hurtle to overcome.
Frankly, I didn’t know it existed. #UndiscoveryTabThanks for the heads up. Will check it out.
Me too, I will chk it out later in the day
I usually type something in the search field when I use the #Discover tab, but the tweets it highlights are often relevant (albeit ones I’ve seen already).BTW, I tried advertising on Twitter again this week, with a promoted tweet, offering a Cyber Monday sale on the Portfolio Armor iOS app. For $100, I got 5,018 impressions, 130 clicks (2.59%) and two app sales. Probably not enough to be profitable (unless the two buyers both subscribe for a year’s worth of the new feature coming with Portfolio Armor v.2.0), but better results than the first time I tried advertising on Twitter, as part of the promotion they ran with Amex.What I think would make these results even better is if advertisers could pick a few key words, and then have their promoted tweets only appear in the streams of Twitter users who have used those words in a tweet, or have them in their profile. In my case, if I could have had my promoted tweet only appear in the streams of users that had tweeted about “risk”, “investing”, etc., I probably would have had better results.
I just sent this comment to Adam Bain who runs revenue for Twitter
Great, thanks.
there is no targeting system in twitter? Targeting is really critical. Terms + timing (Only if they used x terms in y hours) should really increase CPM as clickthroughs would be much more relevant. Wow
No targeting that I can see. Agreed having it should improve results.
Bad…very bad….
You’re exactly right. Better targeting is key.Question: I heard since my last time doing an ad test on Twitter that they let you select a specific tweet to promote? It used to be auto selected which was a huge turnoff for me. I want to pay to promote a landing page link, not a blog post.
That’s a good question. It says they let you manually select a tweet to promote, which is the option I choose, but when I looked at my stats periodically, it had the last five tweets checked off.
They let you pick now…by default they pick them all, but you can say you want to ‘manually’ select the tweets to promote, then you have to go in and unselect the ones they picked by default…once you do that, you can pick the ones you really do want to promote going forward (and remember to manually stop promoting the older ones each time you start promoting a new one)
Cool.
i dislike twitter almost as much as i dislike apple (as i view both companies as deeply untrustworthy), so i have a strong bias here which may be tainting my views. however i have trouble seeing how twitter will be good at direct conversion ads. it seems more like a possible branding play, especially as a way of marketing content, increasing follower count, and building organic links. have you tried google adwords? how did it fare?in my advertising experience, which is admittedly rather limited, nothing has beat google adwords. the closest, and the one that could beat it if the right factors are in place, is youtube.
May have to pick your brain about this offline, but these Twitter results are better than the ones I’ve gotten with Adwords the few times I’ve tried it: better click rate, lower cost per click, and better conversion rate.
oh wow i’m surprised to hear that. perhaps it is related to the fact that you have an inexpensive app, which is more conducive to impulse buys and thus less dependent upon the customer already having a shopping intent in his/her mind.i am about half way done that book you recommended, the social networking business plan by david silver. OUTSTANDING. at $13.72 it is a full $0.73 more than i prefer to pay for kindle prices but well worth it. one of the best internet business books i’ve ever read. it’s already set off a chain of new ideas in my mind. thanks for the tip!
I should note that I advertised the PA website, and not the iOS app, when using Adwords. Maybe the results would have been better otherwise. Edit: Giving that a shot now, with an Adwords search campaign limited to iOS 4.0 and above devices (i.e., devices that can download the Portfolio Armor iOS app). PA is actually fairly expensive for an iOS app. Most apps are free, and the majority of paid apps are $4.99 or less, I think.Glad you’re liking the book so far. I got it in hardcover and found it worthwhile.
“i dislike twitter almost as much as i dislike apple”That would make great adwords copy;”Find out why I dislike twitter as much as I dislike apple andwhy I like gold”.
HAHAHAHA i should run a test ad campaign with that copy! 🙂
That sounds a bit like AdWord applied to Tweets. It makes a lot of sense.
Agreed on both counts.
“and two app sales.”One of the things I first learned about buying advertising was the yellow pages. You had to commit to an entire year, because the book came out once per year. It was expensive (and that was good because most competitors didn’t want to spend the money). Evaluated on any given 1 week or daily period it could have (and was) a total bust. No sales. But over the course of a year it definitely paid. No question about that. In year two of course we picked up a big account that lasted for 6 years. They simply choose three display ads. Once again, that was year two. Otoh, I ran an ad in the Wharton Alumni Magazine and got no response. So I ran it in a years worth of issues (still no response) and then pulled the plug. (Kinda knew the circulation wasn’t great but the cost was low and bla bla … support the school etc.)Important note to startups: If you are selling advertising keep this in mind. Try at all costs not to allow people to sign up for short periods of time unless you know that the type of ads you are offering will work that way and provide feedback.That said there are many reasons why twitter might not have worked for you. And it might very well not make sense to try it for longer given the economics of your offering.Final note: One of the best ways to decide if you should spend money in a particular way is to observe what either competitors do, or similar businesses, do. If a business repeatedly spends money a certain way, in general (re-read that) you can bet they are getting something for their money. (Especially with smaller type businesses that don’t have “budgets” to be spent). One of the things I noticed and tracked in the yellow pages (before I ran my ad) was that competitors ran either the same ad year to year (I pulled up old issues of the YP for this) or increased their ad size. After seeing that the decision practically made itself.
Interesting lessons from your Yellow Pages experience. It might be partially relevant to a new channel I’m considering, where there’s a hefty annual commitment to participate. That has its upside as well as its downside, as you note.
I love the discovery tab and have been getting super relevant information from it with respect to certain LP related topics, such as private equity, infrastructure investing, and sovereign wealth fund activity–almost surpassing any single publication.I have also used the tactic of having a separate Twitter account, where I only follow the certain type of folks for instance LP’s instead of GP’s –folks that tend to tweet much less often and whose signals sometimes get lost in the noise of the prolific.I have tried it with lists but with less success.The email aggregation of news is great but seems to be inconsistently delivered. What up with that?All in all, Twitter is reminiscent of Craigslist–hugely effective but sometimes a little hard to navigate and somewhat messy.
We discussed messy here the other day. We concluded that messy is good mostly
yes…of course. 🙂 Messy is good.
i like to think the mess is great on some level, but it is great on microcosm levels – we in tech like our mess, sports fanatics like their flavor of mess, politics, etc – all centered around organic, instantaneous conversations; but no one has really figured out how to make ‘magic’ out of that madness. Discover is still just ok, and I actually am betting the better innovation will come from outside of twitter, as even as they focus in on engagement/ user value – they do it in to improve their advertising initiatives.
Twitter Discovery has become very good, and it’s complemented by their equally relevant daily Email digest. But Discovery is already getting fragmented. We discover stuff based on who we follow on Twitter, Facebook, LinkedIn, Disqus, etc. I’m not sure that one discovery will rule them all, but Twitter’s is definitely top of mind & innovative. It would be useful if they revealed specifically how they do it. What are the key triggers behind the method?
When I’ve remembered to look at it, I’ve picked up some good follows from the highlighted RTs/favorites/follows. The stories at the top, eh.
They need to put it on TweetDeck.
Yessss
They need to bring back the trending column on tweetdeck. And I need to more creatively use lists on tweetdeck 🙂
They need to put a lot on TweetDeck — others (including Hootsuite) are starting to leapfrog.
“sorry that I could not figure out how to embed a Twitter conversation”The embedding feature has gotten worse recently, unfortunately. If memory serves, you used to be able to drag the corner of the box with the HTML code in it, so you could expand the box and cut & paste all the code. Now you can’t (or at least I can’t), so it’s difficult to copy and paste it correctly.
Such an important feature and so poorly done!
Storify FTW!checked it out… it’s not useless but I think not quite good enough yet. No way I should be getting ‘Jessica Simpson is pregnant gain.’ At least I hope not!people discover is a major pet peeve, in past it showed me abandoned accounts that hadn’t tweeted in forever, people who tweeted every 5 minutes, and followed me and 10,000 other people and had only 15 bots following them.(edited)
On the iPad, the LinkedIn news feed and Flipboard offer a much better integrated newsreader experience to flip through the stories, go to the original site, come back, I think a list of stories w/preview no longer meets the bar.
“so poorly done!”People who design products know them to well and aren’t in the best position to see something that is obvious to a newbie or someone with less skills. Sometimes this is actually intentional (the mystery and discovery adds to the allure) but mostly it’s just the lack of understanding of having to have the opinions of ordinary people considered in the product design.I’ve made multiple comments about disqus, some have been received and acted upon. Some have not.Up in the upper right corner of your comment if you hover your cursor two things appear. One is “collapse” and the other you don’t know until you click (it’s “flag”) Not only is this poorly done (and easily fixable) but “flag” means different things to different people. Things like this could easily be identified with just a small bit of research obviously.
BTW, congrats on the Bloomberg West exposure this week — you’ll have your pick of video clips for Saturday’s post.
Was it any good? I missed it
Yeah, I think you’ll like it. I’m guessing they’re doing one segment per day this week, on Bloomberg West, and I caught two of them, one on Skillshare, and one on Etsy. I think even someone unfamiliar with either company would have groked the key points from those segments.
I don’t think I’ve clicked on #discover since the feature was first introduced — so decided to do so in response to this post. My first thought was that it was way off — which I can understand since my interests are varied and lately my tweets have been less focused — and conversations have been more personal than themed — but as I continued to scroll down, the relevancy increased so I wonder what this means?Of the first 20 recommended tweets:5 were of great interest (RT’d one)5 moderately of interest10 low interestSummary — Those 5 of great interest were worth scrolling through the 20, but I’m not sure how often I’d think to come back to do this. The 5 of moderate interest made sense as to why Twitter recommended them, so, I’d want to see more along those lines. Of the 10 tweets of low interest, a few were from tweeters that I’d typically find of interest including at least two that I already follow.Conclusion — I’m impressed that Twitter has picked up on the instability (or is that diversity?) of my interests. It will be interesting to see what is recommended when I’m back in my game in terms of tweeting.Recommendation — While some users might hate this, I wouldn’t mind if Twitter occasionally prompted me when I’m on the site to click on #Discover for recommended tweets — sort of like Amazon and Netflix. I have been receiving emails from Twitter to this effect and am impressed by the attempt, but I’m not as inclined to leave my workflow to click over from email.
that is 25% accuracy. Definitely could get better. 50% at things would get interesting.
That’s such a more succinct way to say it.
@ShanaC:disqus follows the KISS strategy. We are not at a level where it is possible to do 20 listings. 5 is a more realistic number following the top interests of customer. After the accuracy bar is achieved, spread the number to 8. At that point look at doing the 8 plus a condensed curation of an additional 5 or so that can be summarized based on you asking another person, “What’s up?”.
Generally, the sample size doesn’t alter the result, but in this case, if I had used 10 tweets as my sample, the % of spot on tweets would have been lower. The larger sample produced better results — probably because Twitter perceives my interests as vast. If they struck 50% accuracy, this would be enough incentive for me to return frequently to the discover feature, possibly without prompting.And I’ve never been called “S” by a nicer guy.
I was referring to sample based more on regular customer base and the time element reflected by larger group. Do remember ‘S’ reflects wisdom knowing how much isn’t known, so it is a compliment. 😉
Somewhat. I recognize simple things may actually be very complex. And need time to figure out all the numbers.It makes me fun to work with though – I like to distill down just to the essentials.
Oh I know, kinda like that discussion this past year where recent news validates my design (shhh).Per my response in this forum, it is reminding everyone to think of the customer base. Media interests always affirm eveyone’s poor me regarding time. OTOH, there is real pressure regarding time. So moving into getting what you want, when you want it (media/gossip) most will not be looking for 1,000 links. The KISS method opens up interaction/interface and will drive up hits leading to upward ad revenue…..Otherwise, you know I love hearing/reading your discussion of numbers looking outward 😉
I was actually thinking about it at the time that you could morph the discovery engine into the ad targeting engine. but at 25% you would need to be way more accurate.
heck if it was 10% but that one discovery was brilliant it’d be valuable. So it’s not as simple to judge as percent accuracy, it’s a weighted combination of relevancy/value * accuracy.
Yes, even 10% has some value, but is it compelling? Will this motivate me to click on the #discover tab when I could use that same time reading my already large feed?
Not if you think about the targeting as a form or could be spun out as ad targeting (I hope twitter is reading this) . 70% hit rate should be the goal. not 25%
If 1/10 tweets on the discovery tab made you $1000 you’d repeatedly visit it. So relevance is not as simple as % accuracy. You could have 10/10 mostly relevant suggestions, but not get any value in clicking the tab over visiting your normal twitter inbound. (also pertains to @donnawhite:disqus )
From your lips… funny at my last check the results were at 50% and very interesting, actually. I feel like I got a promotion. LeanStartups, StockTwits, WSJ and McKinsey showed up. Not sure why NFLMemes keeps showing up.
Okay, credit where credit is due. I looked again 9 hours later and 50% of the sample was of strong interest and only about 25% was way off. Guess I should have waited until Twitter had its caffeine. Or maybe the fact that I’ve tweeted more in the past couple of days is starting to influence the results. Obviously you get what you put in…can algorithms produce much more than that?
#Discover is not providing much that is relevant to me.I use Favorites as my “read it later” tool, usually for tweets that involve links with articles that I cannot get to now. I wish twitter would directly give me that functionality so Favorites would be a more relevant “like” indicator.In addition to better meeting my use case, twitter should then have a better signal to surface things that are interesting to me in #discover.
i also use the star as a bookmark. i think a star and a later icon would be awesome.
I find it interesting how many people don’t even know about the discover tab. It’s also interesting how you two (and myself) use the favorite button as a readitlater like tool. Begs the argument that users don’t fully understand buttons and that they often don’t use them for what they’re meant for…
Yes!
I use the Pocket app as my *later icon*.
Same here.
later, yes!
Like many, I use Instapaper on mobile and it works great.But agree entirely that Twitter needs to build this function into their web app.
Thanks, will give that a shot
GetPocket is another one that works well with twitter to save articles.
signal, yes!
I’m pleasantly surprised (eg, this article: http://www.nytimes.com/2012… ) – but I tend to use tweetdeck. So I do’t get lots of discovery there. I think interfacing discovery as part of an app might be a bit better. 🙂
I use twitter because I find it to be the best place to discover news and conversations. So I don’t understand the discovery tab. I’ve never found the stories to be better than what’s in my main timeline. Feels superflous to me.Agree with Fred on the design. Let me expand the tweet if I want to see the image. I don’t want images showing up automatically, especially on mobile.
i have a tightly curated timeline. i only follow 738 people. so i find #Discover gives me the best of the rest of Twitter
haha.Fred, there has to be a better way for those of us with lesser attention spans than you to follow that many people in our Timeline, but somehow curate it so the firehouse is manageable.Lists does not do it for me, and my upper bound on the follow list seems to be approximately 125.
I cannot resist suggesting you try Kuratur for this 🙂
I follow about the same number of people as Fred and find it works pretty well, particularly since most aren’t power tweeters.
I come back to it after each update to see if it’s gotten better. I’d like to be able to see the “best of the rest”.
To me, this is how Twitter should build its advantage, not API limitations. The people who want an alternative client are power users who have special needs. I’m a HUGE fan of Tweetbot because basic Twitter just doesn’t cut it for me on mobile.But a great #Discover tab would drive me to use the official client. Those are the kinds of interesting but non core features Twitter should build and not expose to the API to promote the use of their core user experience.
I think thats what they are going for, but I’m not sold on them being able to pull it off. My #discover is polluted with mainstream news ranked by high # retweets, favorites, and those I follow that have high follower #’s… Still very much Hive mind curation
I think they should open up the API a little bit more, but you should have to really prove that you’re not a bot. It would be terrible to have Twitter receive useless API requests that does nothing but slow down the service.
I’m talking about the shutdown of alternative clients.It’s become too important of a service to not have power user tools, and they refuse to update TweetDeck for Mobile.
Ah, I see what you are talking about now. I use a built in Ubuntu client, Gwibber I believe. Does this affect other devices using the Twitter API?I know that I use a Twitter bot that pulls my tweets into my IRC chatroom. It’s a nice feature and I’d hate for that to go…
That should be fine. They are destroying the client ecosystem by capping each one at 200,000 users. I understand from a competitive reason why they’re doing that. I just wish they’d use other features and restrictions to do it and let basic publishing and consumption continue unabated.
Well for one the mobile client for Android requires too many permissions. Know a trusted third party app that is both light and doesn’t require that many permissions?
This might help for embedding tweets and conversations (not that I’ve done it lately)https://support.twitter.com…
Here’s an example of how to do so for that tweet (ymmv aka not sure if it’ll automatically snag the conversation)
There’s a bug in Twitter’s embed html code cc @dpinsen Dave Pinsenyou have to manually change//platform.twitter.com/widget…tohttp://platform.twitter.com…
thanks!
It’s not your bad, they had a bug in their share code (identified it below). Those types of bugs are critical because they limit threading the app throughout the web.
I didn’t find my Discover tab particularly relevant. Is there something I can do (besides tweet) to help tailor it to my needs?
Sorry, still not really useful for me, it probably depends on how narrow/well-defined the people you follow are. Also all the embedded content requires you to scroll down a lot.
I agree with both Hunter and Fred in that the content has become relevant and highly targeted to me, and I like that it changes / updates very often.But Fred I also agree with you on the mobile #discover’s design / feel. Jarring is the right word – it feels like the stories are right on top of one another and blend together. Would like to see a better and clearer division between stories. Would also like to “favorite” stories and have them automatically join my #Discover tab (a la pocket, read later, etc.).
of course i’m feature blind and didnt even notice that tab til you mentioned it. at a glance, not useful to me. the stuff thats relevant is from sources i follow anyway. not a huge fan of algorithmic news and information curation generally, maybe bring that bias to my casual review
I follow 2,000 people on Twitter. I have found that #Discover is used a little too liberally to be helpful.
The content is getting better while the experience is getting worse. Especially so, as you say, on mobile.I also like the content of the Activity subsection of Discover but it’s so buried that I rarely remember to visit.
Related to this, there’s a pretty good recent video of Dick Costolo at the University of Michigan via a GigaOm article: http://gigaom.com/2012/11/2…DC sees Twitter “as the pulse of the planet”.
“I actually don’t love the visual treatment of the #Discover tab, particularly on mobile, but also on web. I wish it was just like the timeline. I don’t need all those “blobs” of images that pop up sort of randomly and jarringly.”a) is this a design problem – can you imagine the same content being displayed less jarringly?b) are you habituated to the twitter feed in a way that makes it hard for you – in this context – to ‘absorb’ another style of presentation?c) am I wrong in thinking that the #discover feed is somewhat similar to the tumblr feed? Do you dislike that – especially on mobile?
The biggest problem with Discover is how far away from top-of-mind it is.Twitter broadcast is in the main stream for 99% of the population.Discover is supposed to be like a stream – but the whole point of short form broadcast is the context in which it is delivered:This is uninteresting:”I love hotdogs” – strangerThis is interesting”I love hotdogs” – ZuckerbergDiscover doesn’t really tell me enough about who is sending me this content. So I’m left to guess what’s behind the link / quote / etc.Personally the biggest problem with Discover on mobile is that stupid layout. You’ve got a weird discombobulated hierarchy of data that is displayed so confusingly that it’s more stressful than helpful.Here’s all the shit I need to parse in Discover, content block by content block:=========First content============[avatar] name @usernamecontent of tweet and then htttp://link_________________________________headline of link | |[source avatar] name | photo |====Second Content===============GIANT IMAGEoverlayed:[avatar] name @usernamecontent of tweet and then htttp://link
I had not paid much attention to the Discover tab until reading this. I have looked at it before but found it to be uninteresting. I was more interested today simply because I’ve had a personal gripe for a while that despite the many interesting people I follow, the same 10 or so keep on appearing over and over again in my timeline – and it’s 10 people who I find particularly dull and would prefer not to see in my feed but who I don’t want, for various reasons, to unfollow. It’s annoying me enough to make me stop wanting to engage with Twitter at all. So, Discover enabled me to see some others with different content. I realize that I can also create a list but I’d really prefer if it Twitter gave us the ability to choose who shows up in our twitter feed and that we were able to amplify some voices and mute others.
I’ve been using twitter like I do google for some time. Twitter search is awesome. ( I’m not a big fan of their push search though.) The following would help twitter search: 1) remove retweets2) collapse similar tweets via a classification algorithm. 3) Include a geo location field.
I have never clicked on it. I shall try now.
OK, so an article on “Dancing with the Stars” is the first link? I doubt I will be back.
Lol. Maybe you are following the Kardashians?
Ha. No.
I wouldn’t waste my time following them…
Brutal.
I find it mildly interesting, but not compelling. I see a lot of stuff in there that I understand why Twitter thinks I’m interested in it, but in reality I am not interested in it.For example, I’m getting news about a Rhode Island murderer because I follow a lot of people from Rhode Island and retweet items about Rhode Island startups. I have less than no interest in the Rhode Island murderer though.I loved Summify and now love the Twitter emails that I assume are powered by Summify. But if I understand correctly, the emails are about Tweets within my Twitterverse and the Discover tab is about relevant Tweets outside my Twitterverse?As it is, I wouldn’t be able to sacrifice any time using the Discover tab because the payoff isn’t big enough. I read the Twitter email updates religiously.
And I haven’t been reading the emails, but this makes me think I should give it a try. Oy — one more reason to stay up late.
Same here: jarred me on the iPad, but find it useful. Will figure out at some point how to make the app come up on my stream, rather than “discover”.
This is probably off subject but I think Twitter is probably the most over hyped, lied about internet company on the face of the earth. I keep my FaceBook account mostly to see what people are doing and for random trends in advertising.With random friends I find it a pretty good barometer of things going on. For instance, about 10% or so of those ‘friends’ are Instagram Users, FourSquare has dropped to almost zero and so on. Social Cam which started strong has been a no show for over 3 or 4 months now.The sharing of videos on YouTube has dropped from the beginning (high) everyone sharing videos of their past to less than 3% still doing so.Which has lead me to this question? Who actually uses Twitter. I can understand celebrities, semi-celebrities or anyone who thinks they have something to sell, mainly themselves. But for the rest? In my opinion it’s a very limited primitive form of communication. As with FaceBook it’s a time filler.The irony is that it seems as EVERY TV show uses someone using this feature. Here is my question? They honestly have to pay for this.On another note and it’s not me but my wife who watches Reality Shows. Has anyone noticed how for a company that is supposed to be going bankrupt, Blackberry is still the predominate phone..check for yourselves.What I’m saying is ‘you can’t believe the hype’, not on a virtual product whose stats can be manipulated any way possible…
“Twitter is probably the most over hyped, lied about internet company on the face of the earth.”First, I’m not a twitter user. I normally don’t like to make comments about something that I don’t even use (I tried it but like golf found it not my thing). But since you raised the issues you raised I will chime in. (I’m eating lunch.)”the most over hyped”By over hyped I think you are referring to the mainstream media as well as the tech or new media. The latter is self explanatory as far as the interest (both users and it fills a need). The mainstream media (ex. the nightly news with Brian Williams) it makes total sense that they would mention twitter. It provides a constant stream of content, all at no cost. Free. What’s there not to like about that? It’s not a wire service which they have to pay for. Or getty images. Or archive videos. And they don’t have to send reporters out and stand on street corners. They just reach out and grab something interesting from the firehose. Makes total sense that they would be fanboys and wish it success.” In my opinion it’s a very limited primitive form of communication.”People tend to think that it’s great that they find out information practically in real time. I think this is way overdone. Why in the world would I give a shit and need to know right away when something happens in the world or even in my home town? Would it really matter if I found out the next day or at night? This is almost purely infotainment. It’s only value (unless you are trading on the info or perhaps have some other personal or business reason to want to know) is in it’s distraction value. I’m not saying that there aren’t reasons that can be given that refute my statement. But I don’t believe that the majority of people have a real need to know the news in real time as if it’s going to alter something in their lives other than provide another source of joy. Which is fine as long as it fits into their time allotment for entertainment.”On another note and it’s not me but my wife who watches Reality Shows.”Both me and my wife watch Reality Shows. (RH series – everything but Atlanta). We morsel over every little nuance in behavior on the show.”Blackberry is still”. I noticed that also. See, you learned something from a reality show right there. Product placement. Did you notice the cars they drive, or the way their houses are decorated, or the way they dress? All no accident. The emotions (and interactions) on those shows are genuine, even if the plots are somewhat directed. And you can learn from that if you pay attention.
@LE…I’ve come to think that I’m better responding to Fred in the afternoon after I’ve had a chance to think about what to say, lol. Totally agree that from a marketing standpoint it’s brilliant. All I’m saying is that I just don’t believe that it will become a product that goes mainstream.On the other hand, if twitter would allow more characters to have a deeper discussion then I think it could have the possibility of being a substitute for chat.As far as the cars, the worse placement I saw of this was The Glades. The conversation between the two main characters about the car she was driving made even me cringe…or is that fringe, which has just about the same amount only a little more cleaver..I say long live Kayne West and the Kim K. relationship…boy was that made in commercialism heaven!
Of interest, regarding twitter usage per Pew:http://pewinternet.org/Repo…
Diego Basch has a much better quantitative study than pew/princeton’s archaic and useless data-yielding format here:http://diegobasch.com/42-of…Conclusions can only be as good as the data sample!
Twitter was at the center of a major shift in direction for me. It is a powerful tool. I am not as dependent on it as I once was, but I wouldn’t want to live without it. I am quite passionate about this. If they started charging usage fees tomorrow, I would pay.But I will say that you get out of it what you put into it. At the same time, you can pretty much make it what you want it to be. That’s the beauty.
Hear Hear on your last paragraph!I look at it as an effective way to curate news and opinion and it works for me.
I thought it was for me too, but they’re not acting like the kind of company I want to be associated with and so I deleted as many tweets as they’d give me access to and stopped using the service. I don’t miss it one iota.
Hmmm. Come to think of it, I have missed you on Twitter.Guess I’m behind on the latest developments re Twitter corporate. I was troubled by some issues a couple of years or so back but then they seemed to redeem themselves — for instance their IPA. Am I missing something — besides you?
Very sweet of you to say.Twitter was pretty half-arsed until various 3rd parties did their innovating for them. Twitter could have been the custodians of the Next-Web-Protocol (generating an enormous amount of wealth), but lacking sufficient vision to embrace the symbiosis they declared war, thus relegating themselves to a mere footnote to the history of the internet.
depending on the friend, some are really active foursquare users.And I do train checks every once in a while of “phone” predominance (i go home on the LIRR, and I will take a look around me at what phones people are pulling out. Everyone’s non-work phone (and many work phones) are not blackberries.
Twitter is the defacto standard for sharing media publicly.Facebook is the defacto standard for sharing with a group of friends. (although it has had a habit of trying to sneakily share everything publicly)If you’re an early adopter, you share stuff publicly and want to know what other people share publicly. You might say they’re a small number, if you don’t market to them. But you don’t reach late adopters unless you succeed with early adopters first.Blackberry for a long time was the low-cost smartphone which might account for popularity on reality shows, in addition to product placement.
testing
I checked out the discover tab today. I hardly follow anyone, so it was pretty general. It included tweets from people I follow.Recently I wrote an application that automatically summarizes 10q and 10k filings with the SEC. As an experiment I tried tweeting links to them with the $ticker convention, but I got no traffic. After doing dozens of searches over a couple of months, not once did one of those tweets turn up in the search via the $ticker convention. They did show up in the Topsy search results.My conclusion is that Twitter has a long way to go as method for finding information.
THERE A DISCOVER TAB? WHERE?
I don’t use the discovery tab that much but I do use it when there is very high profile stuff going on, like the presidential debates and other stuff like that. It allows you to see trending conversations as well as those of the people you are following.
Not in my muscle memory yet. I use #discover every now and then… but the vast majority of the people I follow on Twitter are through lists, and it’s unfortunate Twitter doesn’t take into account lists in suggesting people to follow and for articles in the #discover tab (but there are still plenty of good reads to be found there).
test
you passed!
Discover has never worked for me, and I find it totally useless. It may be that my friends use public Twitter to discuss, “open letter style” issues with security and startup, and use Facebook (locked to friends or at most friends of friends) for anything social or personal, sharing news articles, etc.The only content I see in discover is stupid articles from people I “have to follow” but don’t actually like.
Hi @facebook-517182859:disqus ! I’ve never seen you around here. Twitter’s Discover rarely shows me anything I am not aware of already, but it doesn’t give me irrelevant content either. I wonder if your user ID causes anomalies in Twitter’s recommendation algorithms: octal is compelling, lots of gravitas, like a sentinel value (or something… there’s a better analogy, no doubt, but you get the idea).