Hacker News new | past | comments | ask | show | jobs | submit login
Twitter is forcing us to drop ability to flattr creators by favoriting tweets (flattr.net)
148 points by chinmoy on Apr 16, 2013 | hide | past | favorite | 80 comments

It's extremely clear that Twitter is turning its back on the developer community who helped the company reach success. We were the first adopters of Twitter, encouraged our friends to join, and made it mainstream. Now, when we want to build apps on their platform, Twitter tells us they're closed - that they don't want to return the favor.

So, yeah, I can see how Flattr, a service that revolves completely around a community of people who actually appreciate others for their work, is upset at the company whose notion of "thanks" is a Cease and Desist.

As developers, we should be alarmed - we shouldn't be blaming people for 'not reading the terms of service' or saying that 'Twitter has the right to do what it wants'. These answers might be technically right, but they don't capture the true feeling of what is right, which is that developers should build apps that enhance Twitter's ecosystem.

So, that's it for Twitter for me. I only expect Twitter to get worse about their developer policies in the future, and I don't want to be a part of a community that treats developers poorly.

Is it worse if Twitter lets flattr do this and then shuts them down later?


In another era, maybe 5+ years ago, any API provider that had such draconian service terms such as Twitter would have been promptly dropped by most API developers.

Why do people continue to develop to Twitter's API? It's obvious that Twitter doesn't want people to use their API. In my years of developing on top of 3rd party APIs from Microsoft, Oracle, etc, I've never seen such contempt for developers as I see from Twitter and Facebook. Developers should abandon Twitter en masse.

But they don't. Developers keep using the Twitter API and they keep getting shut down after spending significant time developing their app. Fool me once, shame on them, fool me twice, shame on me.

I think developers keep doing it because they are still hoping for that one app that will get them acqu-hired by Twitter or Facebook, so that they can become fabulously rich. It's the one big difference between now and 5+ years ago, and especially before the Instagram acquisition.

Your comment is so filter bubbled that it hurts.

People keep developing against the Twitter and Facebook APIs because that's where the users are.

Why does it matter how many users there are if you aren't allowed to make money off of them?

Twitter is far worse than FB in this regard. Unless you can figure out a way to generate revenue by putting content into the stream (eg content publishers), Twitter is not a good place to build a business.

Unless you expect Twitter to sue you, "not allowed to" is pure ideology. In reality, you make money until they notice you, then walk away.

For a side-project, that is great. But if you are building a company where you have multiple people looking to you to keep supplying the paycheck that pays their mortgage, that just seems irresponsible.

Notwithstanding that, why would you want to build something that could have the rug pulled out from under it at any point? I build stuff because I enjoy building thing. If it wasn't working in the market and fails, it is my fault, and I can live with that. If it is successful in the market and fails through somebody else's actions (that I should have been cognizant of), that just sucks.

Because the alternative is to build CRUD app startups like Basecamp, with no dependencies on any API or 3rd party data. Useful, no doubt, but a saturated market.

Saying "CRUD application" is like saying "Java application". It's an implementation detail and not an actual feature.

Tumblr is a CRUD application, yet it shares almost nothing with Basecamp.

Focusing on the technology does not make great products.

FWIW, I think there are a lot of potentially interesting applications that could be built on Twitter, but they've proven themselves hostile to the idea. I don't even trust them as an OAuth provider anymore.

Wait did you mean bubble-filtered?

We're not talking about users here; if you're a Twitter developer you are by definition someone who should at least pay passing attention to which way the wind is blowing in tech. If you want to try some ideas out by building against Twitter that's great, but if you are trying to build a business and you don't realize that Twitter is trying very hard to make sure no one makes any money from their platform but them, then you deserve to have your legs cut off when it inevitably happens after months or years of hard work building traction.

With Facebook or Google APIs you can make the argument that's it's worth the risk, but idealogies and diatribes aside, with Twitter you have to go in with your eyes open.

> We're not talking about users here

Yes, we are. Flattr works fine without Twitter. The Twitter connection is "just" a way to get more users. Flattr is not going to go bankrupt because Twitter blocks them (or, at least not as a direct result).

Therefore, in a thread about Flattr (or any other business that uses the Twitter API but does not vitally depend on it), a "devs that still talk to Twitter are stupid" rant is completely misplaced.

Poor choice of words on my part. I meant we're talking about developers making decisions, not users making decisions.

It's a gamble that will pay off if the powers to be deem it to not be a threat. But then the conundrum is why would twitter allow any other company to profit off it's platform?

This might look like a failure for Flattr, but if you compare them with the flood of micro-payment services that sprang up 5 or 6 years ago, 99% of which vanished without a trace, then clearly it marks a kind of success for Flattr that people are still talking about them, and still thinking about their business model. Of all the micro-payment and/or donation sites that tried to make something like this work, Flattr has been the most successful so far. As a point of comparison, look at Kachingle, which so far has failed to gain any traction, and which has attracted more derision (and lawsuits) than praise: http://kachingle.com/

Kachingle made a big bet on getting a business method patent and (last year) they lost that bet. They also bet big on getting the New York Times to sign up and, when it didn't happen, took an ... unhelpful ... tone about the subject afterwards.

Yes, I know. I worked with Kachingle very briefly. The amount of internal chaos and fighting was amazing, much worse than what I've seen at other startups.

Without commenting on the internal stuff, I think Flattr's slow-growth strategy has been much more effective. As you point out, the base rates for failure is very high.

I've been watching since about 2008. I remember thinking that Contenture and Tipjoy would be serious competitors.

I brought this up in the previous twitter api thread, but it is interesting to get more thoughts.

Perhaps the answer is to stop using apis.

I developed a reddit extension, and when reddit cut off my api access I was able to crowdsource a datafeed through an inbrowser extension. I'm not saying it applies here, but with some creativity I think some apps could be refactored to use a similar approach.

It is not a clean or easy technological solution, but it seems like the api route isn't clean or easy politically.

Imagine if Google tried to get off the ground by using apis to crawl or if Facebook had used approved apis to populate its initial database. They'd be quickly killed, just like what is happening api innovators right now.

Facebook has shown a willingness to send Cease and Desist letters to people who do this[1]. I'm sure they would be more than willing to follow up with a lawsuit. Most sites (that I've actually looked at :) with APIs have clauses against the automated gathering of data from their site. This would fall directly within that.

I suppose if you are located somewhere where you feel like you are outside of the reach of their legal arm, you might be able to get away with it. Otherwise, you are in a legally shaky position, at best.

1. http://www.zdnet.com/blog/facebook/facebook-bans-browser-plu...

Can anyone remember a few years ago, where it felt like almost twice a week there was a new thing leveraging from Twitter in some way, doing something useful or something novel or something completely terrible? That was great, it created genuine value and drew more users in.

If you're a developer and you build on Facebook, or Twitter, or any other API, you now need to be ready for them to pull the T&Cs out and ban your ass and kill your project because damnit, we can't extract value from your additions that'll help us effectively monetize our customer base.

(p.s. flattr looks like a great service and I'm sure they'll find a work around, but they should have known this was on the horizon)

Is network effect so strong that Twitter can go so long at to try to destroy its own platform? Or is it the lack of an alternative? Makes me think of Oracle doing the same with Java by trying to extend copyright to APIs.

Is Twitter even a platform any more? They are going out of their way to limit and prevent 3rd party app's so they can distil down what Twitter is.. a status update service.

Beyond using Twitter for authentication I don't see why you would want to risk anything else on the Twitter API.

They don't share the user's email address. Kinda rules out using them for authentication, IMO.

Why? It protects the user's privacy, unlike Facebook. Ask them for their email address after they authenticate and make it an act of volition.

So far it seems yes. There's a reasonably established alternative at identi.ca, but so far it mainly has users who're there for "ideological" reasons, so it's a good place to follow e.g. Debian developers, but hasn't succeeded in attracting the broad range of people who use Twitter.

How long until twitter rolls out it's own tweet-directed payment service? Maybe they finally found a market-tested monetization strategy.

Yeah it's called kill Flattr and take their idea, oldest game in the book.

Cleaner fish http://en.wikipedia.org/wiki/Cleaner_fish sometimes become food when they're no longer wanted.

Twitter is a culture, and bringing money into the culture of favoriting and retweeting could change it, possibly for the worse. Twitter has a right to stop that. I'm not saying their intentions are necessarily so pure (probably not), but it's also not out of the question.

It's also not out of the question that I'm just a brain in a jar and all of you are an illusion. So maybe let's first talk about likely things, and then about the infinite amount of things that cannot be disproven... ?

What about Favstar and its pro features? The service makes money of people's favourites, albeit less directly.

I think it's different. As I understand it, with favstar you can't actually give someone money. You can give them a trophy (or other virtual rewards) which you yourself buy with money. The key difference is that the system is still fundamentally about nothing more than status. You make content and receive thanks and status VS you make content and get paid for it. Innovations to the former system are far less likely to cause a big cultural shift than a switch to the latter system is.

Twitters war on developers continues.

Ugh, I guess Twitter wants to be the next MySpace. These companies need to ADAPT if they want to stay relevant.

This isn't Twitter's lack of adapting, this is them actively retracting. They're not losing relevance through inaction. They're making decisions to purposefully reduce extra featuresets.

Interesting... They seem to crack down heavy-handed on companies that offer payment related to twitter. Is this about user experience or is this a harbinger of their upcoming monetization strategy?

Same reason Apple doesn't allow in-app Bitcoin transactions, they can't monetize it. It's anticompetative- but that's their right, in this case.

Flattr was also hurt by Apple too remember...

More and more Twitter reminds me a fat and ugly beast.

Flattr's blog post seems to have a bit of that "not fair!" tone to it, which they - for better or for worse - don't actually have the right to say.

I can definitely see the argument from Twitter's side, even if I don't agree with it. But more than that: these are Twitter's terms! You can't roll in and start arguing for benefit-of-the-doubt with the people that wrote the terms for their own platform.

If you build your app on another company's platform to monetize their platform and they decide you can't, you're just falling into a trap you set for yourself. Even if it would be cooler if they allowed it.

I helped write that post a bit. What is sad is that the typical story is the big fish always push out or eat the smaller fish. We would love to offer Twitter a 50/50 split of our fee (or better) as a rev share, but it is doubtful they will take that offer. It seems like everyday now there is another story like this and that is very discouraging for the open web, something that Flattr was built to help support. If you believe in the open web, support it.

> If you believe in the open web, support it.

The World Wide Web is dead. Welcome to the Internet, where you can inter-network between your favorite walled gardens.

PS. If you're unhappy about this change you can tweet to the offical @twitter account with the hash #bringbackflattr, please be polite, thanks all.

I honestly don't know if they're doing it because of some undisclosed master plan, or just because someone is being pigheaded.

Either way, the publicity via HN should shake loose a better answer and I -- as a potential future competitor, hello, nice to meet you -- will be interested to see which one it is.

We love competition, especially US based, have you seen Copper.is and Centup :)

I'm in Australia, actually. Thanks for the names, I periodically lose track of everyone who comes and goes.

I haven't used Flattr in a few years so I'm not sure exactly how this feature works. However company's features or businesses being destroyed through failure to read the rules/guidelines set out by the company they are working via is becoming more and more common. If you are developing for iOS, Twitter, Facebook etc. read the rules first. If there is anything that could be interpreted in any way to affect you get clarification.

> read the rules first. If there is anything that could be interpreted in any way to affect you get clarification.

And how do you get clarification, if I may ask?

I've been asking for over 2 years about a clarification on the iOS terms of service, no answer yet.

Those rules are now longer than the US constitution ...

The rules are arbitrary and capricious. I'd say, "don't rely on a single platform for your business especially if it has a track record of draconian policy."

or make sure that your core business is in no way dependent on any free api that you cannot sign a legally binding agreement of privision (which is essentially a formal clarification of what you can and cannot do).

e.g., if you are writing an iOS app, make sure you also have a desktop (or web) and/or android version which could survive a takedown from the appstore.

> I haven't used Flattr in a few years so I'm not sure exactly how this feature works.

Flattr is the best of the microtipping services that has so far emerged. They removed one classic source of angst by capping the total monthly fee, so tippers don't get a little sting of worry every time they click the but button.

Compare fixed-rate mobile phone plans with pay-as-you-go. Even though PAYG plans are often cheaper, many people will pay a premium simply to avoid having to think about what they might or might not owe at the end of the month. Flattr get that part right.

> If there is anything that could be interpreted in any way to affect you get clarification.

I disagree, for two reasons. Firstly, it's easier to get forgiveness than permission. Second, even if they're kicked off Twitter, they got exposure during their time there and more since getting the boot.

>> "it's easier to get forgiveness than permission"

This is true in many cases (and is fine for Flattr) but in other cases I've seen (AppGratis for example) their business is in serious trouble. If your entire business relies on an interpretation of the rules it's better to ask for permission because if you aren't forgiven the consequences could be serious.

Does anyone have a clear view of Twitter's Terms on commercial use and can clarify it?

For example, Flattr and Ribbon.co have recently run into trouble, but Chirpify, Amex and Dwolla look like they might have Twitter's blessing.

Can anyone give a good explanation of how it works?

Man this was what made flattr fun to use. O well, at least I can flattr youtube videos.

It's even worse for them than that. If they are violating the TOS it is a violation of the Computer Fraud and Abuse Act 18 U.S.C. ยง 1030. Federal felonies for all involved. :-/

I went and had a quick look and I think you are over-stating the situation. 1030 appears to deal with either going around authorisations for USG computers (a)(3), or going around authorisations for other computers (a)(4) in order to commit what would be in other settings ordinary crimes and misdemeanours: intentional damage (5)(A), reckless damage (5)(B), damage and loss (5)(C) and various species of fraud (6),(7).

I'm pretty sure Flattr's actions qualify for zero of these.


I believe he is referring to the case of Weev.

Who is not a direct analogy. In law, your motives matter when determining whether an act was criminal. The fancy term is mens rea, literally "guilty mind".

Unfortunately this high school civics version of the U.S. criminal justice system is not accurate. "Mens rea" comes from the traditional common law, which is now increasingly divorced from actual federal criminal statutory law. An increasingly number of federal crimes say unlawful intent may be presumed or is even irrelevant.

Note I'm not taking a position on whether any possible TOS violations rise to the level of a CFAA violation, or whether DOJ would care, just that waving around terms like "mens rea" does not an actual rebuttal make.

Or, if you prefer a high school civics sur-rebuttal, "ignorance of the law is no defense." :)

I'm not American, but I do resemble one on the internet.

It does vary from jurisdiction to jurisdiction, but strict liability is still rare and still largely constrained to summary offences, not indictable offences.

Regardless of whether you are in a purely code jurisdiction or purely common law jurisdiction or some hybrid, there usually has to be additional effort by a legislator to indicate that strict liability applies.


that's it , I can finally stop wasting my time on writing an app which obviously will be pulled down due to violation of twitters terms.

They are going to pivot to compete with Spotify next :D

time to get an app.net invite and abandon twitter i guess. At least for me.

twitter doesn't make any money if you use their apis. in the same way google wound down google reader, i think you'll see the same neglect hit the apis. twitter needs developers focusing most of their attention on products that make money.

I am currently in the process of patenting a technology for a business which I suppose will indirectly compete with Flattr.

To be quite honest, I admire how far they've gotten. Of the various micropayment and microtipping ventures, they've been able to steadily plug away. I gave them about as much odds of success as the many other companies that have come and gone. (How this bodes for my own ambitions is left as an exercise for the reader).

Naturally, I think that their model has critical economic and technical flaws that put me in a better position. But they're in the market making money and I'm on HN being all hand-wavy and mysterious (sorry: pre-patent secrecy).

If I could give them one piece of advice -- and it goes for everyone who's had a crack at this -- 10% isn't enough. The incidence of the micropayment/tipping service's share doesn't fall on the user, it falls on the websites. So you might as well pick a sustainable rate, rather than hobble your cashflow.

Is a patent really necessary?

It's neither necessary nor sufficient for success. But I believe it will deter some competitors into choosing an easier alternative. Writing iOS apps for sharing pictures of your lunch with people nearby, for example. And some investors -- particularly investors in low-risk cultures like Australia -- would prefer to invest in something with IP.

I am not in a strong position to start a fence-swinging business.

I have no startup pedigree, no contacts in SV, no famous friends, no rich parents, didn't study at MIT/Stanford/UCB or work at Google/Facebook/whoever and I cannot move to SV for the foreseeable future.

I'll take every legal advantage I can.

And judging by the downvotes, applying for a patent is unpopular. I feel that it's more inventive than slide-to-unlock, but folk will have to take my word for now.

It's endemic of a broken system based on completely asinine patents on things like (as you mentioned) slide-to-unlock, rounded rectangles, "[X that has been around for thirty-five years] but on a computer!" and as one of my clients wanted to patent, "chat inside a Facebook page."

Applying for a patent is unpopular because (software|technology) patents are unpopular, wrong-headed, and generally used by companies like Lodsys to extort money from independent iOS developers who don't have the resources for legal staff.

I don't know anything about you other than vaguely recognizing your handle from around HN. It's possible your idea is extremely novel, even groundbreaking, and it completely deserving of legal protection. It's very unlikely. Chances are you're just going to waste time and money.

I agree with your general position and I just want you to know I take none of your criticism or remarks personally.

I suspect that if it ever gets a mention on HN, someone will say something like "Jacques Chester tries to patent logins" or "Jacques Chester tries to patent apache weblogs" or any number of things that don't at all describe what I developed.

Part of the trouble is that patentese is hard to read. While developing the technology to be submitted I wrote and refined what amounts to an RFC. It's a few thousand words with about a hundred lines of pseudocode near the middle.

This in turn blew out to about ten thousand words and 5 diagrams with ~200 reference numbers.

When my first examiner's report came back he'd found and rejected two candidates for prior art; in both cases it took me about an hour to make heads or tails of what was described.

I don't feel sorry for Flattr. The creator, Peter Sunde, one of the creators of The Pirate Bay, not only facilitates the sharing of content against the wishes of many of the owners, but charges 10% to use Flattr.

If Flattr was really all about helping others, the fee would be much less.

I still remember when he spoke at a conference and he talked about how everything should be shared/free. It seems he's not being very honest with us.

With the amount of advertising on TPB (and traffic), I know he was making a profit (and paying his salary).

Why should he get to profit on the backs of hard-working developers and musicians and at the same time saying they shouldn't be able to earn a living?

Come on, you can't compare TPB with Flattr. If your gripe is with the 10% fee, fine. I also think it's too high, and Sunde himself said when the economics allow for it, the fee may be reduced.

Why should he get to profit on the backs of hard-working developers and musicians and at the same time saying they shouldn't be able to earn a living?

When did he ever say the developers / musicians shouldn't be able to earn a living? Just because someone thinks culture should be shared for free with the world, doesn't imply that the creators (more like "remixxers") shouldn't be able to make a living from their work.

Find me a developer/musician who had to close up shop because TPB ran them out of business and I will eat my words.

>If Flattr was really all about helping others, the fee would be much less.

How much less? Five percent? One percent? The only other micropayment processor I'm aware of that has better rates than Flattr is Paypal's micropayment service at five percent plus five cents. Given Paypal's massive size advantage I'd say Flattr's doing well so far.

>...talked about how everything should be shared/free.

I'd think founding a startup aimed at micropayments for all kinds of artists and other creators would indicate that Peter's ideas about distribution of art and creator compensation are more nuanced than simply "everything should be shared/free." He says as much in TPB AFK. He certainly doesn't believe that copyright as it stands has any place in a digital world and Flattr seems to be his manner of investigating ways to share content while still giving back to creators.

>With the amount of advertising on TPB (and traffic), I know he was making a profit (and paying his salary).

I've found no numbers on just how much has been made via adverts or what it was spent on, though I'd imagine the costs of bandwidth for coordinating the majority of the planet's bit torrent traffic could easily reach thousands per month. The recently released documentary has one of the prosecutors arguing that there were 64 unique ads on the site at any given time, each costing 500 dollars per week, coming out to about 1.7 million in revenue each year. Gottfried says in the documentary that there were four unique ads, and that even 110k annual revenue was bigger than they had typically gotten but a more realistic figure. I'd love to see the actual balance sheets, though I doubt anyone from the defense or prosecution has any to speak of.

>...at the same time saying they shouldn't be able to earn a living?

90% of flattrs that cost next to nothing for the creator to take advantage of constitutes an incapacity to earn a living? Sounds better than what an artist would get via itunes or an old-fashioned record deal. Of course, this is all operating on the poor assumption that most artists would ever be able to earn a living off their work, no matter how draconian copyright law became or how perfectly DRM was designed and implemented.

We can get a hint about how much of the advertising went to the founders, by comparing it with an other high alexa ranked site: Wikipedia. Their budget is public.

At the time when Peter Sunde operated it, Wikipedia had about about 3 times the traffic of TBP. Wikipedia also had about 10 times higher of the operational budget (not counting Salaries and wages). We get this by looking at the TPB advertising money as reported by the police, and comparing it to the WP budget.

One can then either assume that Wikipedia administrators are swimming in money and have all bought small island in the Caribbeans, or that running a services on that scale is actually quite expensive.

Link to 2006-2007 WP budget: https://upload.wikimedia.org/wikipedia/foundation/4/49/Wikim... Link to TPB budget for 2006: https://en.wikipedia.org/wiki/The_pirate_bay#Advertising

> One can then either assume that Wikipedia administrators are swimming in money and have all bought small island in the Caribbeans, or that running a services on that scale is actually quite expensive.

Running services on the WMF may be quite expensive but that tells us little about TPB - the WMF is not really comparable. They're running hundreds of projects, not just the English Wikipedia; their staff are living in one of the most expensive places in the world, San Francisco; they're developing & supporting an infrastructure much more complicated than 'upload a torrent file and a textual description box', due to editing pages by all sorts of users with different privilege levels and rich media and exploiting HTML5 and working with new MediaWiki extensions like Semantic and of course the truly enormous size of Commons' media database of millions of photos and images and videos; they work under many more legal pressures than TPB (which takes joy at thumbing their nose at any legal issues); coordinate dozens of chapter organizations worldwide (some with significant amounts of revenue like the de chapter); and do other things like the floating Wikimania conference or the DVD encylopedias aimed at Africa or surveys or editor-retention projects or third-world article writing contests etc.

What TPB does is impressive in its own way, but simply is not on the same scale of complexity or variety.

While some of that is true, I am talking 2006.

At that time, Semantic was developed and funded by Karlsruhe Institute of Technology, Commons was just a year old and was of 1/16 of todays size. I am also not sure how many chapter organizations they had, or if the work towards any DVD encylopedias was in the works then.

I also intentionally excluded travel for the above calculations.

An other telling part can be seen in the Wikimeda fundation 2006 budget under the heading of "Internet hosting". To my understanding, that number is exclusive bandwidth costs.

The lowest I have seen is WebMoney with a 0.8% flat fee on all transfers. I have no idea though what the fees for depositing and withdrawing money from the system are though.

Card processors take 3% and have a ready made solution they can sell and sell again with zero modification. I don't see why a custom app for a specific purpose which will have limited use can't say they're taking 2x the card processors, 6% (adding up approximately to the 10% total cut).

edit: fixed a typo that inverted my whole point :)

10% seems reasonable to me but I agree the Pirate Bay connection is probably hurting their business in all kinds of ways.

On my anti-Flattr checklist, you forgot to complain about how Flattr uses EUR to count money instead of USD.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact