Once again I'm just sad to see the continued erosion from the awesomeness that was Twitter. Twitter are killing off the good and magnifying the bad, all in the pursuit of goals that seem to be forcing a continuous decrease in the quality of end user experiences.
The dumbest thing about this is that monetizing Twitter is trivial, without changing anything else. Just put occassional and obvious ads in our streams, and charge us to upgrade to see an ad-free steam.
Some other obvious approaches mentioned down the thread: Charge for API access, charge for exceeding API limits, charge for exceeding certain follower numbers, charge (at various levels) for analytics, charge for feature sets inside things like Tweetdeck, charge for access to certain API functions and so on.
It's hard to see $9 billion worth in these ideas, but that's only if we think in the short term.
Twitter could be genuinely worth that in 5 or 10 years, but only if it focussed on customers, and that's only going to happen if the founders vision trumps the very short term investor greed.
Saying it's trivial is kind of ridiculous; it's not. It's worth considering that rolling back the API like this will bring a lot of users back to the site to actually see those ads.
Regardless, as a developer it sucks to see companies take away fun tools, but as an entrepreneur I like seeing them at least try and execute an actual business plan.
I suspect if they made it a terms of service violation to block adds they could fairly effectively police smartphone apps. If nothing else it would be interesting to see them try.
I am also of the opinion that a vary cheap add blocking option say 1$/month would prevent most people from jumping though many hoops for add-block software.
I agree with what you're saying, but you really don't have to "jump through hoops" to get ad-block software. It takes about 30 seconds to install an ad-blocking extension.
I think the better argument is that most people don't know/care for ad-blocking software. I think AdBlock and its alternatives are prevalent in the tech community, but we're just a small part of total users.
>It takes about 30 seconds to install an ad-blocking extension.
I'd be willing wager that those 30 seconds of barrier would deter like 70-80% of the computer using population. Very few people as a whole know or are comfortable with browser extensions.
If it was only a dollar or two per month and not a -complete- pain in the ass to sign up for, I'd probably pay anyway.
I wonder if you could encourage the people who hate ads but would happily take a legitimate option by detecting ad blockers and instead of haranguing them for it, put up a polite advert for how to legitimise their ad free usage? (thinking in general, not just for twitter)
That's a good idea. Just the other day I turned off Ad-Block on the Let's Play Archive (http://lparchive.org/Dwarf-Fortress-Boatmurdered/) because they had a little banner message that asked me nicely if I would consider it.
Is it wrong that I blame IE for a lot of this? I feel like IE was a major factor in ramming the "DON'T INSTALL TOOLBARS / ACTIVEX ADD-ONS NOT NO WAY NOT NO HOW" knowledge down consumers' throats (and that still remains good advice).
Most people aren't even aware that they could be experiencing the web ad-free. It just never even occurs to them that should a thing could be possible.
This seems to me to be the problem. They took too much money to monetize in this way. A few bucks a month from a few million users, plus whatever they can make by occasionally inserting ads into the stream (which can't be that much considering how many people would never even see an ad because their streams are too noisy) just doesn't seem like enough to justify all the VC money...
Agreed, that's why I said "a few bucks a month from a few million users". I can't imagine that more than a couple million of their hundreds of millions of users would pay to remove the ads from their feeds.
Perhaps they can charge for certain items that can be considered premium, such as verified accounts. I still doubt there'll be a couple million of users willing to pay though.
I have a hypothesis that any company reaching a certain growth threshold becomes evil :) If there's already a fancy named law on this matter, let me know. Oh, btw, the very top level is reserved for companies that are too big to fail :)
It seems pretty obvious that Twitter is trying to become a business and these alienating changes are part of the strategy but the unfortunate part is that they seemingly have picked the completely wrong strategy around what their business should be.
Twitter really from a tech prospective is a protocol and it seemed that Twitter as a business was in line with what they were providing. Twitter.com acted as the centralized hub and the developers were invited to build upon and enrich the Twitter protocol. The goal was the more tweets and users the better no matter where they were coming from.
Then they realized they need to start making money.
It seems that from the business folks the reaction has been to strong arm users back to using Twitter controlled products with the assumption that they can start to serve ads. I don't blame them necessarily in that this is the standard business model for a social network but I think it will fail specifically for Twitter. In my opinion advertising will cannabalize their business - they should be looking to sell analytics and insights - thats where their value is. Yes its harder, less proven but they are just destroying themselves with their current strategy. Im sure nobody wants to see ads on Twitter and I wouldn't be surprised if nobody really wants to serve ads in that Twitter is exceptionally noisy and the functionality is very streamlined. On the other hand the value of real-time trends or the licensing to do so is exceptionally valuable to just about every company. Perhaps its not as easy of a business model as ads but at least its seemingly more scalable with their ecosystem.
To me it seems like the larger problem is that Twitter doesn't just have to be profitable. They also have to justify their $9 billion valuation. Which I suspect will be impossible.
"they should be looking to sell analytics and insights - thats where their value is"
This seems incredibly obvious to me, and is what I meant elsewhere in the thread where I mentioned knowing what everyone in the world is thinking right now.
Yes, it is hard. But they've got the profile and the money to round up the talent for it.
Making money from ads is cute and all, but there's a whole other level they could be working on.
> Then they realized they need to start making money.
The VC mantra is: 'We acquire a large amount of users for free our services. Someone (the buyer) will monetize them later'. Users are the currency of internet startups.
On the other hand the value of real-time trends or the licensing to do so is exceptionally valuable to just about every company. Perhaps its not as easy of a business model as ads but at least its seemingly more scalable with their ecosystem.
Maybe there is no serious money on that type of business model. Remember, most employees, investors and execs are trying to dump Twitter to investors, who cares what happens 2-3 years from now? The revenue needs to be shown now, ideally before advertisers get to see any hard stats on effectiveness.
Facebook has been adding more ads, Google is virtually all ads so why not Twitter. Facebook will tire people with ads, Twitter too, Google will lose their reputation little by little considering that fairness is the core of it's business. But many will have cashed out and others will hold the bag
What I find frustrating is that twitter has ignored /really obvious/ opportunities to monetize in a natural way, in favour of this strategy that involves backtracking on the things that makes twitter /twitter/. Opportunities that I see:
-- Charge celebrities and companies for the capability of having huge numbers of followers. Remember Ashton Kutcher's race to be the first twitter user with 1 million followers? Yeah. Having that many followers is valuable to the person with the followers, and caused actual strain on twitter's infrastructure. Why didn't they charge money for that?
-- Charge third party developers for increased API capacity. It's obvious that this is a pain point. Why not just provide the option to pay to remove this pain point? Maybe they are planning to? who knows
-- Charge users for special account abilities- like, perhaps, many of the features offered by tweet deck.
And yet we have them pursuing this artificial advertising strategy. sigh.
I can only think that the strategy they are pursuing is that they see twitter as a passive medium for celebrities to transmit messages to their fans. Like it's a text based TV, that they can run ads next to. Perhaps they have the user demographics to back this up. Us weirdo geeks that use it as a a many-to-many communications medium are the vast minority.
> Charge celebrities and companies for the capability of having huge numbers of followers.
You can't. Servies like these grow because of celebrities and their fans. Taxing the celebs will stop their own growth. Also, there are lots of would-be celebs who don't make a lot of money and are trying to use twitter to get the word out there.
> Charge third party developers for increased API capacity.
For the same reason as above, you can't. A lot of people come on facebook and stay there because of apps, charging devs for API access will kill innovation. Also, if something is popular and growing, taxing it will be bad for both the users and the dev if the dev can't pay. Also keep in mind that most popular apps may not have a monetization strategy yet.
> Charge users for special account abilities- like, perhaps, many of the features offered by tweet deck.
This is a good approach and is the LinkedIn model. I'm sure they're thinking of doing this with talk of 'lightweight analytics' etc so long as they don't make the paid users too powerful and disenfranchise existing users.
> You can't. Servies like these grow because of celebrities and their fans. Taxing the celebs will stop their own growth. Also, there are lots of would-be celebs who don't make a lot of money and are trying to use twitter to get the word out there.
The celebrities in this case, are using twitter as an ADVERTISING platform. you said it yourself- would-be celebrities are trying to use twitter to "get the word out". is this now not an opportunity worth paying for? Would the fans, noticing they are unable to follow Ashton, blame twitter, for charging a reasonable price for that huge megaphone, or Ashton, for being unwilling to pay for this enormously powerful broadcast medium? I don't think, in this case, limiting growth is a serious issue.
> for the same reason as above, you can't. A lot of people come on facebook and stay there because of apps, charging devs for API access will kill innovation. Also, if something is popular and growing, taxing it will be bad for both the users and the dev if the dev can't pay. Also keep in mind that most popular apps may not have a monetization strategy yet.
Apple charges $99 for an SDK. straight up flat fee, annual charge. Are you saying this has seriously harmed innovation for iOS?
What I'm proposing isn't even that. It's charging for say, having an app that has more than 10,000 simultaneous users. If you are charging money for this app (and many twitter clients do), is asking for a fee really as harmful as you claim? I don't think so.
I think this makes sense. Even besides the celebrities on twitter, there are many party promoters, news outlets, bloggers, merchandisers, etc. that use twitter to advertise to their followers, and it provides a far more efficient advertising medium than the alternatives. The service charge for having a huge following would probably be well worth it (perhaps even negligible) to certain major account holders.
Sure you can charge celebrities, you couldn't have done it early on, why not now though? Facebook is making a ton of pages now through the pay to see seen stuff.
I would keep the API free and open, sure charge for the larger feeds and enforce 3rd parties to show sponsored tweets where appropriate.
>> Charge third party developers for increased API capacity.
> For the same reason as above, you can't.
Twitter is already charging developers $infinity for > 100k API tokens; lowering that to $1/per token or something isn't going to harm the app ecosystem.
I genuinely think that twitter fails to think about quite a lot of things, and is actively hostile to those within its ranks who do. That is just my opinion though. My genuine opinion, based on their public behaviour.
I dislike this new trend of linking to 'meme' like images within a blog post. It really detracts from the post and gives it a childish feel. I clicked those links expecting hard evidence about Twitter's bad API, instead I get irritating comics and stop signs.
What gets me is that they aren't even re-hosted: they're hotlinks to other websites' imagery. Besides being incredibly lazy, leaching off of other people's bandwidth without giving them credit or placing the imagery in their original context by linking to the pages they appear on is just a low move.
Yeah, you're both right. I'll move the images local. Not intending to leach. Just trying to lighten up an otherwise kinda depressing post. Seeing four years of work come crashing down.
As pg says, running a startup is like being repeatedly hit in the face.
Thanks, I appreciate it: I'm just not a fan of people not getting the credit for or benefits of their creative works. While I don't have an opinion on the use of the images, I can definitely understand where you're coming from and what you were trying to do.
Leeching bandwidth? Really? This is not 1999 anymore, no one cares if their images are hot linked anymore. Be ready to be surprised, people found ways of making money from hot linked images.
I would care. There's no excuse for your site affecting my server in any way. If it's a site that gets enough traffic the number of connections being made to my web server for a single resource could be enough to adversely affect my site. Besides why are you going to take the chance that the image stays online and in its current location. Just rehost it, it takes such a minuscule amount of effort.
The asshole aspect of deeplinking, I thought the atrociousness of incessant linking had been explained perfectly almost a decade ago: http://maddox.xmission.com/c.cgi?u=banish
Bad companies start getting really obsessive about being the only ones who end up profiting through their platform. They view another company getting traffic and making even a little money through their platform in the same backwards way that music executives consider pirated albums to be lost sales.
They always end up lashing out at the wrong people - in this case smart developers who are giving them free promotion and creating value for their users.
It's sad because Twitter had by far the best platform to build innovative software off of, but now I wouldn't touch Twitter's API with a 10 foot pole. There's too much risk that I'll spend lots of time putting something together and then getting shut down.
The worst part about it is that spammers still easily use the Twitter API but legitimate software gets rate limited or shut down. It's really sad given the potential that Twitter had.
It would depend on app.net's data policies - how much (and how fast) we could get information out of them.
Of course, they own the field, so they could change the rules at any time, thus (once again) killing the project.
From a technical basis, 90% of the effort was building a robust system atop Twitter's. The actual core behavioural categorisation engine was relatively trivial.
Ie, it would require a major rebuild, without lessening the risks at all.
Do social networks tend to cannibalize themselves over time?
Initially, growth is the most important part of a social network, so it makes sense to prioritize usability. Over time, pressures towards monetization mount (either internally or externally, among spammers). Monetization tends to degrade the user experience, creating opportunities for new entrants.
Network effects can keep you going, but these aren't Ma Bell's network effects. Users can install competing social apps next to each other using almost no time and less money. They get to keep the more established network as a fallback, then spend most of their time using the one with better features or more "exclusivity." Even if this doesn't lubricate transitions enough for established users, network effects are basically nonexistent among new cohorts (read: the young, emerging international markets).
If this is correct, it would explain why no one under 20 actually uses Facebook (or email, for the matter), and why every move from Twitter smacks of desperation.
So, I don't know, all hail snapchat? (At least until something displaces that in five years...)
They saying goes that if you are not paying you are the product (or will become eventually).
Wonder how much people are willing to pay? And what it would cost?
But personally I fear the biggest road-bump will the paying itself.
It's not fun anymore for developers. Of course it isn't, but do normal users care? I can't say I've enjoyed using twitter any less since they introduced the API limit.
I can say I've all but stopped using twitter altogether since their purchase/bastardization of tweetdeck. That client was nearly perfect.. multiple networks support, persistent searches...
Twitter is not important enough for me to have to run a separate application for it.
Echofon for OSX is no longer functional since API 1.0 was deprecated.
I've moved to another native client, but as a user who likes to see multiple avenues for engineers to subsist (i.e. not just working for bigco), it's saddening to see.
The author is talking about application/rate_limit_status which is also rate limited. The headers are great if you keep track of them from the response on your other calls, but if you want to see your rate limit status across all endpoints you can use:
I'm sad to see the Twitter ecosystem collapsing. There are a couple of projects that I'd like to do involving Twitter, but I've backburnered them because it doesn't seem like a safe place to play.
Almost every single feature of Twitter (including its logo!) was created by third party devs or users. Not surprising that when the API is severely restricted, the fun stops.
This is misleading-to-false, particularly regarding the logo.
The logo was created by Simon Oxley. It's misleading because while Oxley is a Twitter user, he didn't create it for Twitter: he created as stock imagery, and Twitter purchased the rights to use the image via iStockphoto: http://www.wired.com/business/2009/03/twitter-designe/
It gets into false territory because subsequent versions of the logo were developed by designers either Twitter employed or contracted with.
Even if you're talking about the "T" logo variant, it's also not true: the "T" logo was simply the "T" from Twitter's original wordmark, which was simply the Pico Alphabet font designed in 2001 by a Japanese design firm, Maniacker's Design: http://www2.wind.ne.jp/maniackers/pico.html
buying doesn't equal creation, so webwielder would seem to be correct. Derived versions of the logo are new creations, see all the arguments or derived works and software licensing. Also, "original artwork" and "simply the Pico Alphabet font designed..." doesn't match either.
To go for functionality, the hashtag and retweet were developed by users not Twitter.
The sense I meant "original wordmark" (not artwork) was that it was the first wordmark, not that it was novel. It was not created by a Twitter user or dev: it's just the name "Twitter" in the Pico font, which predates Twitter by several years.
Moreover, it's misleading to claim the bird logo was created by third-party Twitter users (or devs): the fact that Simon Oxley has an account on Twitter played no part in the acquisition of the rights to use his work. They purchased the rights to use stock imagery.
That is to say, its association with Twitter is due to Twitter buying the rights to the image and using it in their branding, not due to Oxley's involvement in the Twitter community or his offering it up as a community take on Twitter branding (which he did not). It's not in the same category as borrowing from or being inspired by the Twitter community as was the case with hashtags and retweets.
The subsequent logos were created by internal Twitter designers or contractors. While I'm sure they are Twitter users, they are definitely not third party Twitter users and their usage of Twitter is incidental to their design work.
I've suggested before that we should explore an independant organization that rates companies and their treatment of devs who use their API. If a company has a history of constantly trimming usage limits, it should be documented. At some threshold, recommendations could be made to avoid or be weary of certain APIs.
We are returning back to the 90s where API access came from BD deals and the rest of us did dirty parsing to make things work.
For a while we got the illusion those days may be gone. And why not, it sure felt like that. But now we are getting to a stage it's almost worse for services to have public or free APIs at all.
I'm looking at the features and the future features this app has and wondering who the heck it's for. I've never gotten DM spam, I've never had to think about using a machine to unfollow junk accounts.
I use Twitter to talk to some of my friends, share things I think are cool links, follow a few people whose work I admire, and occasionally tweet links to my own work. I don't follow back everyone who follows me; I basically regard Twitter as a sort of IRC channel where I get to choose who's in it.
How do people who actually need this kind of tool use Twitter?
Marketers, and people building an account for their business. Follow anyone who looks like they might be interested in what you offer, remove the spammers (TwitCleaner's main use for me), and unfollow people who don't follow back after some time has passed.
You can build a Twitter account with real followers pretty quickly that way.
TFA is about the API, but when I read the title prior to clicking my immediate reaction was "it was never fun"
I'm in my 30s so I don't feel that "get off my lawn" old, but I just never "got it" I guess. Any tech person I want to follow normally has a blog because you are limited to the character limit.
I wish twitter had built in filters, what drove me away was all the pictures, the "lolz", and food/drink shots in exchange for a few tech links that aren't on the persons blog and if I were not following them I might not have known.
Twitter feels like reddit to me while RSS is akin to HN. HN is super focused, and while some funny OT stories make it to the front page, for the most part it is solid tech news. reddit you get _everything_
If twitter could somehow provide "channels" or something like professional/personal I'd really see the value. Not everyone is going to hashtag their tech only stuff or personal only stuff
> HN is super focused, and while some funny OT stories
> make it to the front page, for the most part it is solid
> tech news.
Interesting: My perception is the complete opposite, and I've actually found myself using Twitter (with Tweetbot) for tech news much more than HN lately. I think this is because Twitter uses the people I choose to follow to decide what to show me, whereas HN uses everybody, and to make a story popular with everybody it generally has to make people angry/mad/happy. That means that HN seems to have more and more polarizing stories reaching the front page, whereas Twitter seems more focused.
I'm the same demographic and also never really got into Twitter, but I definitely see the appeal.
Comparing Twitter to RSS is definitely way off as RSS has really nothing to do with social, and Twitter has everything to do with social. HN/Reddit, by comparison, are almost exactly the same thing with just different community standards.
Twitter is really about a streamlined and open conversation. It's a more stripped-down social experience compared to Facebook or Tumblr or whatever other social networks came before, but yet it captures the essence of online interaction in an elegantly minimalist fashion. I think this leads some people to find it utterly vapid because online interaction is simply not the same as in-person, and Twitter exacerbates these characteristics by lowering the time and location requirements down to near zero, so you can participate anywhere, anytime, on any device without ever having to take more than a minute because you're limited to 140 characters anyway. There is definitely a purity and beauty to it, but it always gives me the sense that I'm wasting my time.
The images are gone now. As per recommendations by other hn users above. The only links remaining are to directly pertinent or explanatory information.
It's a shame that Twitter themselves don't seem to be doing anything to clean up the spam. I use to enjoy following twitter during sporting events but these days every trending topic is immediately swamped by automated spammers.
No. Off hand, it'd require batching the users into groups of 100 and using an API call (/users/lookup) for each group. That gives you a user object each (which you don't care about), but also the latest tweet.
The limit is 180/15 mins, ie a max of 18k in any 15 minute window.
Twitter is trying to go public and everyone holding shares is thinking of the possible riches. After that, they can leave to start their own thing, stay or possibly retire.
What we/others say is a distant second. Right now they need to show growing revenues and that's it.
Not surprised at all. One could argue it was the API that attracted so many people to Twitter in the first place. It seemed really smart to me at first to create such a great ecosystem, but maybe there's more behind that scenes that I don't know about?
Either way it's sad to see this service go. The guy seems nice and provided a good service, but clearly he's not going to spend a bunch of his own time and money if it's not rewarding anymore.
The dumbest thing about this is that monetizing Twitter is trivial, without changing anything else. Just put occassional and obvious ads in our streams, and charge us to upgrade to see an ad-free steam.