Twitpic is shutting down | Hacker News Reader

Twitpic is shutting down

blog.twitpic.com

477 pointsuptown153 comments

So, with three weeks notice, all tweets that have ever been posted that included a picture using Twitpic (which is a remarkable number) will no longer have their picture available? In fact, do we even have three weeks? I just did a search for "saurik twitpic" on Twitter (https://twitter.com/search?f=realtime&q=saurik%20twitpic) to find people who had posted pictures of me, of them for me, or of my products to show to others, and I'm seeing just a bunch of broken images (the few that work, as in the few that still show images on Twitter's website, such as https://twitter.com/MuscleNerd/status/296068187353661440, are simply a cache from Twitter's twimg.com: clicking through to Twitpic doesn't work). It seems like twitpic is already offline :/. (So like, I was going to go through and try to frantically download images I'd find relevant in the next week, but I guess it is already too late.)

(edit:) Even the broken previews are now disappearing (making this both a little less obvious that there is missing content and also a little more barren-looking). I guess Twitter notices the original image is gone and stops trying to render it through twimg.com, so most of the tweets in my posted search result are now just "imageless" (before they looked like this one, which is older enough that I hadn't rendered it yet to get its cache to clear: http://cache.saurik.com/tinyimg/twitpicoff.png).

The upload functionality seems broken right now. The site says my upload went fine, but I can't see the photo.

Are their image file servers getting hammered? Weird.

The Twitpic homepage[1] is also 404'ing big time, perhaps it's just a devops issue more than them pulling the plug before the sunset date?

It almost seems like they're playing a little hardball and showing everyone what things will be like if they disappeared from the web, which seems to be what Twitter is threatening them with since they're unable to secure the use of their name and yet millions of images are linked based on that (domain) name.

Still, we've seen this story before, Twitter doesn't give a fuck about developers now that they have sufficiently used them to build their platform and brand.

But, what's funny is, it was the use of "tweet" that Twitter frowned upon for 3-rd party names -- "twit" was considered fine...

[1] http://imgur.com/YLOEKEs

> Twitter doesn't give a fuck about developers now that they have sufficiently used them to build their platform and brand.

I really hate that this is the standard model for building startups these days. Make a free service, grow your userbase using investment money and by encouraging developers to build on your platform, then once users and developers are dependent on the platform, start littering it with ads, selling users' data, and shutting down all your API and all the developers' work. We've seen this happen over and over. Will it ever stop? One thing is for sure, developers will wizen up to this if it continues and stop building on top of platforms, which just sucks for everyone.

There is fresh meat out of school every day and they are replacing the old ones already and continuing the same way.

I completely agree with you, but for the use of the word "dependent". What user is dependent on Twitter? Seriously? Can't one stop tweeting in an instant without any ill effect on her life?

For developers it's a different story -- but they should know better.

Aside from people with huge audiences and Twitter-related businesses, anything you choose to use and incorporate into your life is at least somewhat of a dependency, even in the presence of substitutes.

For example, I love Chick-fil-A sandwiches. When I moved, the closest Chick-fil-A is now far away. I was dependent on Chick-fil-A, and I'm less happy without it, even though I have missed no meals and probably eaten healthier in its absence.

All kinds of things are like this - if emacs went away, I could learn vim. If my apartment building went condo, I could find another place to live. If my favorite TV show got cancelled, I would find something else to watch, read a book, etc. But those are all things I chose to use, and taking any of them away would reduce my welfare by definition, even if only a little bit.

If words have meaning and if we try to stay away from hyperbole, you are not dependent on Chick-fil-A, never have been, never will be.

What would you suggest (more for e.g. Twitter users than my example)? Reliant? Used to? The whole concept of dependence is that substitutes aren't perfect. Windows/Mac, Android/iOS, emacs/vim - no one is dependent on those, but a lot of people would be pissed if any one of them went away.

I would say that users have a vested interest Twitter's continued service.

Also, I don't think substitution has anything to do with dependence. If "dependence" signified inferior substitutes, I'd expect "independence" to signify fungibility rather than autonomy.

The issue is obviously one of semantics but your concept of dependence is not the widely accepted concept. Dependence means completely reliant on - no substitutions would do. The thing you're dependent on is a requirement, not a preference. For example, you are dependent on food. Remove food from the equation and you will perish. You are not dependent on Chick-fil-A.

Twitpic is dependent on twitter. There are no substitutions that they can make and still survive. They can't port to a service comparable to twitter as there are none.

EDIT: Whether or not users are dependent on twitter is up for date. I imagine it depends on their use case.

> Dependence means completely reliant on - no substitutions would do.

I disagree with that. Package managers have concepts like optional dependencies or alternatives (only 1 of multiple options required). Anyway, you seem to be speaking of dependencies in a technical sense - build dependencies, package dependencies, etc. In a more general sense it is quite normal to say something like "I depend on my bike to get to work" even if I could take the subway or use Uber, or "I depend on YouTube for my daily dose of cat videos" even though you wouldn't really die without them.

Arguably, Twitpic isn't dependent on Twitter per se, just a large microblogging website.

Twitpic can possibly open up a Chinese branch that runs on Weibo instead of Twitter, for example.

You just made a great argument for why you aren't dependent on Chick-fil-A.

You could find out how many people were dependent by taking it away. What would happen if tomorrow Twitter was completely gone?

I guess society would move on. But probably one way they/we would move on is by building another Twitter.

I would split a hair here: It's not the users as people who are dependent, necessarily, it's the content they've spent time contributing that is dependent on the existence of the 3rd party service and on Twitter itself.

I have many, many twitter acquaintances I have literally no other way to contact (many I don't even know their actual names). So yes, dependent in that it's an independent, non-duplicated social graph, completely disjoint from other social graphs. Leaving it would, in fact, cause an ill effect in my life (multiple ill effects, actually), just as withdrawing from any social group.

True in this case, but I don't see this being a trend and definitely not the "standard model for building startups". What are some other examples of companies that have done this?

Legally speaking, most terms of service include a term which allows the provider to change everything in the future, without any limitation. Maybe developers are advised to develop for a platform only if the terms of service include some commitments which will be kept and not changed.

Over the years I have talked to a few companies (large and small) that wanted to move forward on a platform yet maintain some legacy cruft...aka sustaining engineering. I partially declined these companies at the risk of them hiring new and then the stuff you work on you maintain for it's lifetime...but that's another story.

That said, I am curious if there is a business model being missed here. Essentially for twit pic and similar, buying up/sustaining the content and trying to figure out how to get some value out of it.

Another option would be for companies like twitter to offer a mechanism where twit pic/etc content could be snarfed up by the internet archive and when a company goes away, somehow the links to content get redirected.

I'd say it's more the shortsightedness of the developers who build on others' platform. It's OK to tag on but you've got to make sure that your core value prop is not a value added feature to that platform. Twitpic could have branched to something else 2-3 years ago, but they did not, and as a result, is eliminated from the game.

Agree that it's shortsightedness.

But I think the problem goes much deeper: We keep building one-offs. Twitter - a one-off messaging service. Instagram - a one-off photo sharing app.

This is one of the reasons I started Buddycloud. We'd already built a nice location and social app but eschewed the VC cash to, dare I use the word, pivot, and build a different way of building apps.

Instead of building another one-off social-location-system like Foursquare, we decided it better to build a federated platform that others can then start building on. The federation and run-it-yourself mentality means that our users don't end up in the Twitpic scenario and that there are always other suppliers that will host your pictures in a compatible way. In a way that fosters competition between providers without needing to resort to switching friction cost to keep users.

In the Twitpic case, the buddycloud media server is designed to be a plug-in federated media hosting provider for each domain. Don't like how one provider is dealing with you data? Just switch.

It's not been easy to get this far, but we're starting to see traction from ex-app.net devs who are looking for something a bit more open that they can also run themselves (as a Docker container). And I still believe that the real solutions will be based on federated open systems that form a foundation so that developers can innovate further up the stack.

Yes, it will stop when platforms are distributed and cannot be shut down so easily. Where APIs are replaced with protocols, built on top of the open web.

Something that perhaps ought to happen is that in the UK, Australia, etc public broadcasters (those funded by the public) such as the BBC and ABC should be required to stop integrating Twitter into their programs.

For example, Q&A (on ABC1) uses Twitter as its primary channel for real-time audience engagement -- announced every week by the host and shown on-screen multiple times in each program. It's fairly clear this isn't benign as the producers at the ABC might think, but means that public money is being used to give an advantage to one commercial social networking company over any other.

This will never stop because it's profitable for SOHOs to build on the developer APIs. I remember the days of making 5K a day on Facebook API before they shut it all down.

Your best bet when you do implement to a dev API is to assume this is exactly what will happen and plan to pivot once you've established some mindshare.

Like it would hard to set up 10-20 cheap OVH servers to serve all photos accessed at least once last 6 months... Pay for them for 2 years and leave with a dignity. $15-20K spent, all happy, you are an exit hero.

But yeah, there is barely any honour in exits like that lately...

Since when does Twitter care what you tweeted last year, or even last month? Faithfully archiving your online history has never been a priority for them. The entire platform is focused on showing you what's happening right now, and they deliberately make it difficult to find older tweets without knowing the exact ID or running some sort of crawler.

We could go on a long rant about whether or not we should trust our data to cloud-based third-party services, and so on and so forth; but in the case of Twitter, they're not even trying. In fact, I wouldn't be surprised if they went ahead and announced that they would delete every tweet that is over X months or Y years old.

Here's a reminder to please donate to the Internet Archive. https://archive.org/donate/index.php They have tons of old twitpic pages! http://web.archive.org/web/20140304234132/http://twitpic.com... See that timeline at the top? You can go to any date and browse around.

And the Internet Archive is more than just the "Wayback Machine" that lets you browse old versions of websites. They have massive datasets of video, audio, and texts too - check out the examples on the homepage. And you can add your own collections now. https://archive.org/create/

They just finished ingesting 2TB of public Fotopedia photos before it went down. Do you know how much it costs to store a terabyte permanently? Not an offline backup, but accessible 24/7 forever? About $2,000! Power, cooling, redundancy, and especially dedicated people to keep it all going. https://archive.org/donate/index.php

Shoutout to Jason and #archiveteam. Great work on Fotopedia, ready to launch DO instances to go grab Twitpic!

I didn't mean to leave out credit to Archive Team :) But last I checked they don't need more help, whereas most of the data they grab ends up on the Internet Archive who really could use the money.

> whereas most of the data they grab ends up on the Internet Archive who really could use the money.

Completely agreed. The IA is a treasure, an underfunded one at that.

Two lessons:

1. Don't base your entire business on another business' product. Especially one that has had such a hostile relationship with third parties. When the other business decides to discontinue your access to its service, or otherwise change things, or basically copy what you do under their own brand, you're screwed.

2. Don't create a similar name that will give you trouble in trademark disputes. (Seriously? TwitPic? I work in technology, and even I thought they were owned by Twitter, though admittedly, I don't use Twitter much and don't use TwitPic at all).

When they launched I remember thinking they were naming themselves that way for 2 reasons: 1. cause user confusion like you suggested, 2. try to flip to twitter itself.

I believe Twitpic was a 'weekend project' that Noah put together, that grew into a successful business. AFAIK there was not this level of intentionality behind it in the early days. I suspect twitpic was just the most obvious catchy name that came to mind.

A business product that encourages data isolation, control and non-transparency is not a product anyone should ever want to use.

I don't think those lessons apply well to this TwitPic scenario, since they have been breaking them for 5 years without any problem.

Yeah, they do. It's hasn't been 5 years without any problem, it's was 2 years without any problem, and 3 years of being slowly strangled by Twitter.

Plus, if the definition of success is "last 5 years", then that's not a very lofty goal.

Inasmuch as it came to a sudden, screeching halt even after five years highlights the value of those lessons even more so, IMHO. Five years is long enough to get (falsely) comfortable and that's when ignoring those lessons really gets dangerous for a business.

Isn't that worse? 5 years of effort, all gone. I'd much rather have a project of mine shut down after a week then to go half a decade before it is shut down for reasons beyond my control.

> Don't base your entire business on another business' product.

That is a terrible generalization. That pretty much covers every ISV ever. And large companies like Microsoft, Google, and Apple all have a history of "Embrace, extend, and extinguish" when it comes to closing out 3rd parties.

However I think your point is valid for companies like Twitter, and to a slightly lesser extent Facebook. These companies don't have a vested interest in 3rd parties building on their platforms.

Indeed, the obvious counter-example to that lesson is imgur.

imgur is a good service even without Reddit, though (and I often use it just to send nice, short, copyable image links to people). TwitPic without Twitter is almost worthless, like a worse imgur. Its single defining characteristic was 'put pictures on twitter automatically!' but as soon as Twitter supported posting photos directly their entire business model was gone.

Honestly I'm surprised they've lasted this long; they had no differentiating factor other than being new, and they had no real features that anyone cared about. I haven't even seen a twitpic link in years, let alone clicked on one.

Imgur has taken clear steps to hedge itself from its dependency on reddit though.

Yet it has not succeeded. Most of users base browses the sites "Viral" gallery. Which is nothing more then a "Reddit's 24 hour greatest hits", a surprisingly low number of images make it into imgur's gallery from imgur itself.

TL;DR No reddit, no front page content on imgur.

Except there's quite a bit of content there that's not from reddit.

There isn't. User submitted is horrible. Most of the sites visitors are people uploading/sharing images normally via direct linking.

I disagree. User-sub is actually a really good time. Maybe you're just browsing user-sub/newest first? Try this: http://imgur.com/new/viral

> TL;DR No reddit, no front page content on imgur.

With the number of image macro posts that are on the reddit front page at any given point, I'd say their relationship is pretty symbiotic. Without imgur, I'd argue that reddit loses a significant number of their highest viewed posts, and vice-versa. Reddit would certainly have content in their smaller subreddits, but their largest ones (AdviceAnimals, funny, pics, etc.) would be ghost towns without imgur.

Without imgur, they'd use any of a dozen of another imagehosts, like they did before imgur got big.

Reddit can live without imgur, and has lived without imgur. Its imgur who still needs reddit.

Not since the site exploded in numbers. I'd wager someone else will take up the challenge though, just like Alan did.

I agree that they haven't succeeded in removing their dependency on reddit, but you can tell it is certainly something on Alan's mind.

I really don't understand why they have taken steps that really limit the community for no real good reason. For example, comments are limited to 140 chars, which de facto means the discussion is going to be incredibly shallow. The only reason I can think of is maybe Alan was afraid of reddit thinking he was trying to steal their users and cutting him out entirely by creating their own image hosting solution, so he purposely limited imgur's ability to develop its own community to not upset reddit stakeholders. (EDIT: It appears imgur has recently taken investment money from reddit, so now this threat is much less of a concern).

For example, why hasn't imgur adopted a subreddit style model to help organize the content on the site? It's a no brainer to help develop individual communities with actual focus instead of having one giant "random" feed. Anyone who argues that would be stealing an idea from reddit clearly hasn't been to a forum before. All forum's are divided by topic... except imgur.

|Most of users base browses the sites "Viral" gallery. Which is nothing more then a "Reddit's 24 hour greatest hits", a surprisingly low number of images make it into imgur's gallery from imgur itself.

Additionally, using a subreddit approach would allow imgur to bridge the gap from "reliance on reddit" to all content coming from within the imgur community itself. That is, if imgur was to switch off content from reddit right now, the quality of the site's content would plummet. User's would be confused (as most it seems don't actually understand where the content comes from) and there would honestly probably be a huge drop in the userbase. However, if they adopt a subreddit approach, with the first/main "subimgur" being "viral" which stays the exact same as it is now, then as the additional subimgurs that get created increase in quality they will naturally begin to take over more and more of the homepage allowing for a seamless transition away from reddit content to entirely imgur content.

Lastly, the layout of the site could use an upgrade in my opinion. I would prefer a hot ranked layout with full size images in descending order, basically reddit+RES with all images expanded, but with comments expandable under each post (or show top 3-5 comments under each post with clickable button to expand all -- or have the comments as a sidecar to each post when site is viewed on a desktop/tablet), which is a better user experience than having to click from image to image. Alan might be afraid that layout would kill their monetizable page views, since now they get a pageview per image where they can display a new ad. But the fact of the matter is, the impressions/pageview model of online advertising is dying fast. Using a feed instead of displaying one image at a time means imgur could integrate a native advertising solution, which pays much better than traditional display ads. If Alan is afraid of immediate drop in revenue that would come from transitioning to a new (though significantly better longterm) ad model, maybe it's time for imgur to actually take in some investment capital that VCs are constantly trying to throw at it.

EDIT: It appears imgur has taken $40 MM in investment this year, I wasn't aware of that.

Imgur is about instant gratification. As you cycle through the best images on the web, there's an oddly satisfying instant gratification effect after each one. This is the reason for both having 1 piece of content on each page, and having 140 char comments.

The longer the comment, the longer it takes to read. Ain't nobody got time to read walls of text on the Internet anymore. People want their Internet quick fixes, and 140 char comments allow you to skim through and continue on.

In Imgur's case, people visit for their quick fix, and end up getting lost in a rabbit hole of viral images. Fun fact: When people visit the homepage and start browsing from there, they average 44 pageviews over an average time of 22 minutes!

While Imgur don't have "subimgurs", it does have tags. Like game of thrones? Check out the best game of thrones images on the Internet: http://imgur.com/t/game_of_thrones/top/year

I don't want to split the community into subcommunities where you have to follow certain ones over others. This would lessen the affect of discovering something you didn't know you liked, and would lessen the instant gratification affect imho.

Also, I disagree that the user submitted section sucks, and for those wondering, just over 50% of front page "viral" content is from user-sub: http://imgur.com/new/viral

The tagging functionality does mirror the formal organization of a forum. It would be nice to see it formalized a bit more on imgur. Maybe a bar at the top of the screen with trending tags? Or heck, maybe even just in the sidebar.

Oh and the real exciting question... when is imgur going to branch into a "simple video sharing" service. Youtube gets more cumbersome and annoying each year (dialog boxes that pop up all over the videos? I mean come on!). The perfect opportunity is arising for imgur to step in at what it does best, keeping things simple. In addition to imgur users, I bet all OC video content on reddit would immediately switch to using an imgur video service.

50% of images on the front page of Imgur come from user-sub. The algorithm that runs the front page was designed to do that.

> Indeed, the obvious counter-example to that lesson is imgur.

Is that actually a good counter-example? Are they profitable? If so, would that continue to be the case if reddit began prohibiting linking to imgur?

Reddit is one of the bigger investors in imgur. I think that's how it's a counter- example.

Imgur was bootstrapped for 5 years and profitable the whole time.

Also, less than half of Imgur's traffic comes from reddit.

> Don't base your entire business on another product.

Possibly the most naive statement ever made about software business models.

> 2. Don't create a similar name that will give you trouble in trademark disputes

Often this is not a problem, as once you've gotten your foot in the door you can relinquish the conflicting mark and continue your business. In any case, the trademark didn't really conflict anyway, it was just a legal tactic. The API block was the end goal.

I'm not sure why anyone would dispute #1. Perhaps they interpret it too broadly.

You can have some success basing a business on another for a while maybe, but unless you have a solid agreement as to how that relationship works, the rug can be pulled out from under you at any moment. And that's usually what happens.

Twitter was very friendly to 3rd party developers. But then the big rich guys came in.

i'd go further and say don't waste a second of your precious time and energy developing for someone else's closed platform. the sharecropping metaphor has only gained in relevance since tim bray first proposed it.

"Don't base your entire business on another business' product. "

Well another angle to that could be it's ok if you plan on getting out while the going is good by selling to a greater fool.

This is one of the detriments in life to knowing to much. You tend to shy away from things (because of past knowledge and experience) that someone with less experience might do. And they might, by luck, make out.

For example if you operate in old school business the idea of going forward with an idea not knowing how you will make money simply doesn't make sense. But hey people have done that and they've made out.

1 should rather be "don't build your business on being an obvious missing feature from another product, especially such that if the product implements that feature, they make you irrelevant overnight" with 1a: "if you do, understand that your business' potential exists entirely in the short term".

It worked out fairly well for Rapportive:

http://techcrunch.com/2012/02/07/linkedin-picks-up-rapportiv...

1) What if I told you, you could base your entire business on another business' product, get shut down, and still come out ahead? Getting shut down isn't the worst thing to happen, it could even be the "exit strategy." Run the money machine until it's dry.

There's more to this story... If I had to guess I would say that it's no longer profitable (or at least, very profitable as it once was) and it's not fun to run an image host (imagine all the trouble that comes with just copyright issues, let alone criminal matters). The founder also has a new startup called Pingly which I'm sure is more rewarding to be working on.

On its face Twitpic is shutting down because they won't get their trademark, but they've never had their trademark and have had no problems operating. Twitter isn't suing them, they're just saying they are against the trademark application. They don't need the trademark and it's bizarre to shut your company down for the possibility of being forced to change your name in the future.

If Twitpic wanted to continue running they could without any changes or extra work. It's a dead end though, Twitter has its own image service and there is little reason besides legacy to use Twitpic. Here's their traffic situation which should pretty well mirror revenue as it's web based advertising paying the bills:

https://www.quantcast.com/twitpic.com

They're hemorrhaging traffic. Globally there were 85,043,992 visits in the month period a year ago compared to 13,315,016 in the most recent month. That's also just the past year, Twitpic has been on the decline for years and peaked in July 2011 with a cool 280,021,248 visits.

Expenses are likely not down nearly the same amount as they have to store more and more images every month (with less traffic there is less bandwidth, but this doesn't count images not loaded on their website).

Just in the last month they lost 46% compared to the previous period. It's falling off a cliff. This seems like a good way to get some sympathy on the way out. A rude move to their users though. Maybe Twitter can step in and offer a bottom dollar buyout so that the links don't break.

This echoes my thoughts when I read the post. Sounds like a clasdic case of simple cloud service losing profitability; answer shut down and screw your users. Afaik the service was free, so I don't have much sympathy for the users; but blaming twitter seems disingenuous.

So, there is a bunch of info we dont have... But would it not make sense to just rebrand? From a preliminary search on twitter, there is a huge amount of users still using twitpic. Seems like this is a biz that could/should make millions a year with some cost reductions and run mostly on autopilot.

Also, what happens to the millions of pictures previously hosted with these guys? Even if you can backup/export all your data, think about all those links on twitter that are going to suddenly stop working. Its basically like a URL shortener shutting down.

Shit, send the site over here and Il provide the resources to keep it going. Just seems like such a shame.

Sound to me like they are trying to see if Twitter is willing to change their mind when they threaten to shut the service down and break all the tweets. Rebranding would be an obvious option but they would still have to keep to current domain alive in order to not break the links. Could they abandon the trademark and still keep the domain or could Twitter just grab the trademark and based on that later also try to claim the domain and maybe restore all the links from a backup?

Twitter could issue them a license on their brand name, with the contract limiting its use to catching and redirecting old links. As long as there is a nominal license fee, that should satisfy the trademark lawyers.

Yeah, they really should have just looked at rebranding. A few Instagram-related services (e.g. Webstagram, Statigram) had to do so recently, IIRC, and I don't think they're significantly worse off.

Thing is, even if Twitter tacitly condoned it for years, the name "TwitPic" is pretty obviously derived from "Twitter" and is trading on their (Twitter Inc.'s) reputation. Even their logo is styled similarly to the old Twitter wordmark. The blog post didn't even say Twitter wanted to force them to stop using the name, only that they shouldn't attempt to register a confusingly-similar trademark (though granted, forcing a rebrand might well have been the next step).

Leo Laporte had "twit" first and gave Twitter permission to use it back in the beginning.

Leo Laporte likes to play up having "twit" in relation to twitter, but as far as trademarks, his claim would have been laughable. They aren't in the same business.

He should buy TwitPic and mix things up a bit.

My thoughts exactly. Why is the trademark so important to them? And as I read it, Twitter never demanded that they stop using the name, just that they do not apply for the registered trademark. This decision seems rash.

>>> My thoughts exactly. Why is the trademark so important to them?

Because it would protect them from Twitter going after them for infringement on Twitter's trademarks. I'm sure Twitter knew they had no argument for the "Likelihood of Confusion" Test and TwitPic would've been granted its trademark in due time.

From a legal perspective, it looks like Twitter felt if they got trademark protection, they wouldn't have any leverage to sue them to get them to stop treading on their business. Instead of mounting a legal defense in the courts, they simply used what leverage they did have and threatened to cut them off from their API - which is everything for TwitPic.

For you non-legalese folks:

http://marklaw.com/trademark-glossary/confuse.htm

"The terms, "confusingly similar" or "likelihood of confusion" both refer to the standard required to prove infringement of a trademark. Specifically, if the relevant consuming public will likely be confused or mistaken about the source of a product or service sold using the mark in question, then likelihood of confusion exists, and the mark has been infringed.

The likelihood of confusion test is also one of several examinations conducted by the U.S. Patent and Trademark Office in determining whether to approve an applicant’s trademark application. It is worth noting here that even if there is no likelihood of confusion, i.e. no trademark infringement, you may still be liable for using another company's trademark if you are blurring or tarnishing their mark under the state and/or federal dilution laws."

>I'm sure Twitter knew they had no argument for the "Likelihood of Confusion" Test and TwitPic would've been granted its trademark in due time.

Maybe I'm misunderstanding you, but are you saying it's not confusing that TwitPic is not owned by Twitter? (At least from a legal standpoint)?

I always assumed Twitpic was owned by Twitter, and it seems many other HN members did too. I'm amazed they lasted more than a month with that name.

At the time, Twitter gave specific guidance that ecosystem applications with the term "tweet" in it would not be allowed, but twit was OK. That was specifically covered. I know for a fact because I ran a site with "twit" in the title, and cleared it with Twitter. Dozens of modestly popular sites (around data, retweets, images, etc) were using "twit" at the time.

I just posted this comment on the blog post itself. It's awaiting moderation:

--- Could you explain more about this:

> A few weeks ago Twitter contacted our legal demanding that we abandon our trademark application or risk losing access to their API.

Why not abandon the trademark application but keep the service online?

That does seem pretty crazy that they wouldn't just buy another domain and keep things running. My guess is it actually has nothing to do with Twitter or Trademarks. They likely ran out of capital and are giving up.

>Its basically like a URL shortener shutting down

Did anything major happened after some of the early URL shorteners died? We survived just fine.

This is more akin to bit.ly shutting down imo.

Yes, we "survive," but that's a pretty lame measurement of the impact. URL shorteners leave a bevy of broken links everywhere, linking to content that is now lost somewhere on the net, twitpic's departure leaves use with a big hole of the social web empty. Thousands of tweets are now without context.

> But would it not make sense to just rebrand?

Most likely not. TwitPic's only possible exit was an acquisition by Twitter, who has since developed their own image hosting platform.

This is the danger of tying your startup to another company that has the power to shut you down. The company that you are tied to needs to either acquire you, or you are at their mercy.

On the bright side, Noah earned millions in profit with TwitPic over the years (by his own admission). If he has played it smart at all, he's not going to have to ever worry about money. On to the next.

Really? How? With advertising? I guess I am spoiled by AdBlock and didn't know they even made money.

There are already too many links to the twitpic.com domain that would break if they changed the name.

Sad news. Just the latest victim in Twitter turning on it's own developer ecosystem. :-/

They stuck it out pretty long with a good customer base, I personally kept using them even after Twitter introduced the ability to embed images.

They just never adapted, I feel like their site design/experience hasn't changed since their initial launch.

It's a shame Twitter has become so developer-hostile. When it first came out, I remember the excitement of using Twitter as a host platform for other apps. Now I couldn't imagine writing a line of code that interacts with their platform. It's a nightmare. I wish app.net had been more successful: it was clearly an attempt to ride the waves of that excitement after it'd been made clear that Twitter had jumped the shark.

> It's a shame Twitter has become so developer-hostile

Nothing about this is developer hostile. I'm rather surprised this did not happen sooner.

I suspect Twitter's demand is the straw that broke the camel's back. I'm sure Twitpic use has declined significantly since Twitter introduced their own image hosting. The team has moved on to other projects so there's not much incentive to fight it.

Pretty big straw... Not many fully functional and profitable companies can handle a "Drop your trademark or we'll sue" legal demand from a company like Twitter.

Except they're not threatening to sue, they're threatening to drop TwitPic from their API.

That seems like a much, much bigger threat in this case.

A year ago I wrote a quick Python script to export all my images. If anyone wants to use it (or improve on it) I blogged about it at http://shkspr.mobi/blog/2013/08/exporting-twitpic-images-pyt...

THANK YOU!!!! I just used a modified version of your script to download all my data. Again, THANK YOU!!!

My version: https://gist.github.com/hugs/fa7892da03ce660c212e

Changes:

  * Use the short_id as the image filename.
  * Use https instead of http when downloading the json data.
  * Save the json data to individiual "page-*.json" files.

Quick question: Is it okay to use the MIT license for this?

Go ahead :-)

I don't understand why Twitpic would be shutting down. If they are growing and doing well, then why not just change the business name? Or is it that things aren't going well (usage dropping drastically) and they've already been thinking about shutting down or moving on, and maybe this trademark issue was just the last straw?

long time coming really. follows long narrative/twitter history where twitter forced it's way into owning the photo content shared on the service....allowing only a few like twitpic to maintain the key aspect 'embeddability/previews'. What still remains interesting to me is that at one point instagram was gaining an incredible amount of speed in becoming the defacto picture sharing service on twitter -- but then they had the spat with fb/insta and twitter stopped the previews for instagram pics. Instagram has a huge community so survived fine, but it could and still could be so much more with twitter integration...but that sanction has never been lifted.

Actually it was Instagram that decided to stop including the Twitter preview of photos, because "we want images viewed on instagram.com".

http://techcrunch.com/2012/12/05/kevin-systrom-on-pulling-tw...

sorry yeah, you're right. There just seemed to be an air of 'fb v twitter' and content owning about it anyways....and really, at the time anyone will remember that instagram was becoming really popular for sharing pics on instagram + twitter...and then boom. Insta silo'd and twitter aggressively pushing pic.twitter.com native hosting

Are they just playing chicken with Twitter here?

It seems odd that, because Twitter are being trademark jerks, they're going to just take their ball and go home. This feels more like attempted blackmail.

Seems they're continuing their aggressive push towards controlling it all. Anybody have recent numbers on 3rd party client tokens remaining? And what happens to clients like TweetBot when they run out of tokens?

Here is the relevant Archive Team project for Twitpic: http://archiveteam.org/index.php?title=Twitpic

So it's time to try this backup tool https://github.com/Stantheman/Twitpic-Backup

Thanks for the shoutout! Twitpic-Backup is pretty old but will definitely get your images. Props to pcgMongo for today's PR:

https://github.com/Stantheman/Twitpic-Backup/pull/2

Thanks! It's a straight forward shell script using wget. It just needs twitter handle and working folder name, downloads everything.

They could always go back to twitter and propose dropping the lawsuit if they change their name?

Yes I know Twitpic is the brand, but if they were to simply DNS redirect Twitpic.com to ickleimages.com [1](available!) then twitter wouldn't be littered with old broken images, and the company could work on pushing the new brand. They do now support more than twitter after all.

If it's a viable business, it seems crazy to throw it away to prove a point that many probably saw as an issue from the outset (That is the use of twit* for a service that leeches off twitter's success).

I'm sure changing every single link to serve twittermademedoit.gif would be frowned upon.. but then the company isn't going to exist any more..

disclaimer. I've never used Twitpic before, nor posted on twitter so may be missing something..

[1] http://www.urbandictionary.com/define.php?term=ickle

When is Twitter shutting down its API?

Sounds like there is either more to this or this is a strategy. I'm guessing they would like to sell but can't do to possibly of being sued by Twitter over their name. Not so much that Twitpic can't continue but they can't get further funding or bought out due to risk.

I wonder how well Twitpic is doing revenue and profits wise and how much that affected the decision?

It's hard to believe only the patent issue is the cause here. Open closing down announcements are nice, but should list all the facts and circumstances, not try to spin it PR wise.

I was wondering that - maybe they are losing money and low on cash. There don't seem to be any patent issues - only that Twitter does not want them to register Twitpic as trademark which I can understand.

A 2010 interview with Mixergy, the founder Noah Everett stated: "TwitPic is generating $1.5 to $2 million in ad sales on an annual basis, with 70% profit margins. It was roughly similar revenue in 2012.

In interview last month, he said his other company, Pingly will revolutionize email. He's changing focus on something that has real revenue potential and possibly good VC?

I'd always assumed Twitter had purchased Twitpic a long time ago. Interesting plot twist.

They're going to pivot and introduce a suite of new services which use the same underlying engine to share pictures across a variety of social media sites:

  Googpic: share on Google+
  Facepic: share on Facebook
  Instapic: share on Instagram
  Tumbpic: share on Tumblr
  Pinterpic: share on Pinterest

When interviewed, the founder stated "We plan on filing trademark applications for these new service names and anticipate no further legal issues. It's regrettable that Twitter is acting so unreasonable by taking reasonable steps to protect its mark."

They weren't facing legal issues with Twitter, so that last part seems somewhat off base.

Never ever, ever build a service that is 100% dependent on another service that you don't have alternatives for and never use someone else's trademark in order to leverage your business. That's two fatal mistakes in one go, they should be rather surprised that they held on this long. But maybe the 'user outrage' will be enough to get twitter to buy them, though - assuming twitter isn't dumb - that would be a bad move because that would lead to hundreds of copycat attempts.

To be perfectly honest, I am surprised to learn that twitpic is owned by a different company than twitter itself, which is precisely what trademark law is supposed to prevent.

I must not be paying enough attention, because this doesn't make a lick of sense from Twitters side. If I had a company that was willing to host images instead of me, at no cost to me, and no restrictions on me, I would be a barking mad idiot to disrupt this relationship. Unless I was about to sell...

I don't think it mentions anything about Twitpic as a service. It's only about the trademark.

That being said it makes a lot of sense for Twitter want to control the user experience. The cost of hosting low resolution thumbnails can't be a pressing issue for Twitter. User experience and engagement on the other hand....

I'm not saying it's worth sabotaging relationships with 3rd party developers but there's a certain mindset in which this type of thinking makes a lot of sense.

If Twitter killed the API that would put Twitpic out of business - but how is that worse than shutting it down? They might as well call Twitter's bluff, if the alternative is just as bad.

https://news.ycombinator.com/item?id=4315663 I guess stocktwits is next on the chopping block.

I think the issue here is that Twitpic's business model is having people leave the main Twitter interface to take them to another domain, where Twitpic can extract advertising money from the impressions. Twitter has image previews these days, which offer a nicer user experience and keep users on the main Twitter site. With Twitter hosting images itself, it makes little business sense for them to continue letting Twitpic make ad money off a feature duplicated externally.

Hopefully somone like the Archive Team will at least manage to save most of the images. Urls used for images and links seems to be sequential numbers.

I just really hope that no-where in the shutdown does twitpic put up a restrictive robots.txt. Archive.org's retroactive robots.txt and all that.

>> "Unfortunately we do not have the resources to fend off a large company like Twitter to maintain our mark which we believe whole heartedly is rightfully ours. Therefore, we have decided to shut down Twitpic."

Is this supposed to be serious? You're shutting down your company because you're stubborn and not getting what you want??

Probably a poker move to put pressure on Twitter... If the company is doing well, they'll do what it takes to keep it running, even if it means changing its brand.

Isn't the entire reason companies aggressively pursue trademark claims against small fry precisely because lack of enforcement leads to weakening of the mark?

I don't see how Twitter has a brand confusion case against them since they've obviously known twitpic existed for many years now and haven't asserted brand confusion until now.

> "A few weeks ago Twitter contacted our legal demanding that we abandon our trademark"

Guess it's time to abandon Twitter.

Alternatives?

Well... there is app.net - But honestly I don't think that anyone beyond a certain german podcast-bubble really uses this.

German podcast bubble? I'm from Germany and I don't know what you are talking about.

I asked App.net about alternative paying methods because in Germany it's not very common to use your credit card. Nothing.

Twitter has basically lost all my loyalty because of actions against the developers that made them who they are.

Whatever the reasons may be, or even if Twitter is "on the right side of the law" I would rather see them embrace this entrepreneurship than go the route of the bigger tech companies and squash any bit of choice. Typically people use third party add-ons because they enhance the experience. What would twitter be like if they banned all the third party services that convert longer messages and post a shortened version on twitter? I can imagine a similar scenario if, for example, reddit decided all of a sudden to threaten the creator of "reddit enhancement suite" or if they decided to shutdown Imgur, which I think that symbiotic relationship was already mentioned here.

Time to look for alternatives.

This is why you want to run your own server for this sort of thing... or at least use something standardised and portable.

After having both Lightbox7 and whatever Sonys big photo-store/share platform close down and wipe out my high school/college albums I've learned to definitely not rely on any image hosting services.

Back then there wasn't twitter or any other good social media and you found things out by visiting the site or through ICQ.

Lets put this in perspective. Some twitter pictures are going to disappear. In the grand scheme of things the people who had the luxury of using this service will survive just fine. Please stop the navel gazing.

Lesson to learn: don't tie down your service and/or brand to an existing brand/service. Otherwise you're both dependent of that service and/or at risk to get sued sooner or later.

This is a good point. I've heard of many companies learning this point the hard way.

This is revealing in 2 ways.

1. Twitter doesn't care about developers. 2. Twitpic wasn't making enough money.

Why didn't Twitter acquire them instead? Did they try that and fail?

Why would they? What assets does Twitpic have that Twitter wants? They built a "better" service on their own. They don't need the brand Twitpic. What's left?

That's true. I imagined more on the lines of a talent acquisition.

thousands/millions of images form twitter users that could be converted seemlessly in to twitter's in house solution providing a good user experience for their users preventing the breakage of the site and creating a PR problem for them...

I don't get it. Why not just change the name and keep going?

i saw an article mention noah working on this http://pingly.com/

Isn't he the guy from Heello?

Twitter being dicks... As usual

Link rot

Just change the name already. Jeepers.

At what point do you forego patents for the sake of usability. Twitpic has been around for so long and they have tons of media hosted, all of those Tweets are now essentially broken. Twitter should instead foster a relationship with Twitpic.

Twitpic is DEPENDENT on Twitter to work, it's not taking users from Twitter at all (other than hosting media). Truly sad to see.

There is no mention of patents anywhere in the post. This has nothing to do with patents.

USPTO: US Patent and Trademark Office. P is not the relevant letter here, T is.

Reply on news.ycombinator.com