back
46 comments
wrote the same thing when they first released social reader (it has been here on HN):

http://github.com/nikcub/frictionless

the problem is that they change the HTML elements every few days and it is a pain to keep up with it. I am in the middle of rewriting frictionless to use an external ruleset which will be updated, not requiring the extension to be updated.

you can see what we go through to get this to work consistently across loggedin/loggedout users, all the different apps (what you do wont work for all apps) and all the different use cases:

https://github.com/nikcub/frictionless/blob/master/webkit/fr...

(we also anonymize referral and click data)

Edit: to add, i'd be more than happy to merge the efforts and we are definitely seeking contributors. would rather one extension that works all the time than two that don't. we have already put in a lot of time keeping this extension up-to-date since launch with all the changes and updates made by the social readers.

Thanks Nik. I agree that it makes much more sense to join forces and merge efforts.

1) This extension looks at all wpsocialreader links, so unless they change the URL paths, it should still function.

2) Is your extension still in the webstore? I tried the link in the github repo, but it's 404'ing.

we took it down from the webstore temporarily so that we wouldn't be adding new users while the extension is in a state of flux. I will email you to work out how we can merge efforts - we have a plan for what we are going to implement in the new version. It would be good to have more eyes on the project and on the social reader keeping it updated.

Edit: just realized that I can't find your contact info. can you ping me on nikcub at gmail? I am also nikcub on GitHub - are you on there?

Cool, I like this better since I can read the source! Forked it and added support to another app, already filed a pull request. Thanks!
thanks for that! will be checking it out in a moment
Has no one noticed that you can set permissions to the app posts. You can set it to visible to only yourself at the time of installation.
Yes. My friends who are sharing their gossipy and sex-related readings have not noticed it. They're all shocked by what ends up on Facebook.

As for just using the social reader as a user — well, I shouldn't have to agree to an app and share my social graph just to read news stories. Friction Sharing is great and working already.

I've noticed it, but I'm still wary of that data entering Facebook at all.
Why?
Because they've consistently shown they can't be responsible with it. That's why.
I'm interested in specific examples or theoretical concerns, rather than "I said so." This is HN, not Reddit.
Facebook will disclose to Alice a list of all people who have Alice in their contact list on their phones (and use the Facebook phone app). I was told by Facebook engineers that this is working correctly.

HN discussion: http://hackerne.ws/item?id=3145857

There have been other things, but most of them were fixed rather than being called features.

As an FYI: the canonical URL is http://news.ycombinator.com or in this case: http://news.ycombinator.com/item?id=3145857
I understand philosophers like Sarte and Kant would agree that the moral thing to do here is not use a plug-in that lets us nerds cheerily use links generated by an evil feature without harming ourselves (by auto-sharing what we read), but rather choose not to follow such links and to warn our often-unwitting friends that they are auto-sharing content merely by reading it.

Because the question is, what is the effect if others were to act like ourselves?

One action still rewards these evil social reader / frictionless sharing apps.

The alternative helps impede them.

I know which world I want to help create.

It actually has no effect on the market, it's the same as if I searched for the name of the article in Google. Companies can't tell that I'm using the link for the app or not. I do agree though on discouraging your friends to use these things and I think it's important to point out to people that they are autosharing without really understanding what they are doing.
If an article looks interesting, I usually comment saying, "Sounds interesting; I'm not going to use that social reader app, but I will look up the article later." That lets people know that if they really want their friends to read an article they'll get more bites if they post it as a link. Fewer people using the app does affect the market, and the plugin removes the main disincentive to using it, which is that your friends will ignore the links you post. It's better to raise awareness of that disincentive than to bypass it.
Often I don’t even see an option to comment on these stories! — Especially when they appear as little lists of articles people have been reading.
All the plugin does is launch a Google search. It's the same manual bypass that I would have done anyway.
Pet peeve: conjecture stated with the adverb "actually."

It’s a guess, man. And I am not convinced. Consider that the traffic for that page is still higher now – a data point in favor of "frictionless sharing" if, say the company is testing randomly enabling it for certain articles. Separately, the fact is now you contributed a page view and perhaps a tiny bit of ad revenue in response to seeing an auto-shared article.

It probably does still matter.

I feel like the quality of the 'read' articles is even lower then of 'shared' ones.
Am I wrong in thinking that you can just click cancel and it takes you to the article anyway?
From personal observation, its inconsistent. A lot of the guardian links take you to the article on pressing cancel, but the yahoo feed doesnt.
...and that's why they're failing. I never read anything from Yahoo anyway.
Well lots of people still do, or this wouldn't be an issue in the first place. The Washington Post is the same way.
That does not work for me.
The weird thing is this is so automated that it doesn't even convey any social proof. It's not "articles my friends actually thought were interesting" so much as "linkbait my friends fell for".
maybe im just paranoid, but a bit wary of installing chrome-ext without a quick perusal of the source. Is this github'ed?
Whether it is or not, chrome extensions are just ZIP files with a .CRX extension - with a quick rename, it should be trivial to unpack and peruse it yourself.
You could look at the source, but it's obfuscated (sorry). Also, it's not github'd (sorry).

If it makes you feel any better, a different extension of mine has 7000+ users and no complaints of evil things being done:

https://chrome.google.com/webstore/detail/bgonpegbhnjepleakg...

Posting what articles one reads is a very bizarre feature- waiting for when this bites a political dissident someplace
Mediocre but (afair) working alternative solution. Create a friendlist with no-one (but yourself) And give those apps only stream_publish permission to that friendlist.
No need to create a friend list for that - you can just set the visibility to "Only Me" when you first authorize the app.
Cool stuff. What I usually do is click the link and then set the privacy level to "only me" so that no one knows what I'm reading.
If I see a headline that interests me, I tend to copy-and-paste it to my browser's search box, and find the article that way.

Actually, more often than not, I've forgotten about "social reader", so click on the headline -- then, when they try to get me to give them access to my account, re-think whether I can be bothered searching for the article. About half the time, that stupid popup costs them at least one pair of eyes.

I agree with the giant text on the link: "Worst. Feature. Ever." It's just such a blatant, obnoxious attempt to manipulate your users.

I just highlight the title, right click and select "Search Google" usually... takes me to the same place without giving FB any info. Interesting app, but I'm weary of using Chrome because of the data it gives to Google. (Can't win these days - I just want to browse in peace! D: )
Isn't Chromium = Chrome minus Google?
No. Google Chrome is Chromium + closed source components like the PDF reader. Wikipedia has a list of differences here:

http://en.wikipedia.org/wiki/Chromium_(web_browser)#Differen...

Both Google Chrome and Chromium let you choose a search engine other than Google when you first start it.

The privacy policy is a good read if you want to know exactly what information Google collects:

https://www.google.com/chrome/intl/en/privacy.html

My read of this indicates that Google is not tracking what sites you go to, only sites that don't resolve or attempt to feed you a bad version of Google's SSL certificates. And, if you believe that Google is actually tracking the Safe Browsing queries when they say they aren't, you can ask either browser to not check for malware:

https://support.google.com/chrome/bin/answer.py?hl=en&an...

This is great. I wrote an article[1] on this exact topic after I felt that it was not getting the attention it deserved.

[1] http://www.saewitz.com/recently-read-articles-on-facebook

I wrote on it, too: http://alanhogan.com/facebook-pseudo-links

And aggregated a few (more noteworthy) bloggers’ reactions: http://alanhogan.com/facebook-frictionless-sharing-pushback

This doesn't work on Yahoo links, only on Guardian and WaPo links from what I can tell.
FYI - Updated the extension (due to a suggestion from a HN reader) to just do an "I'm feeling lucky" search, so now it's a one-click affair.
Working great!
Thanks for this, this whole frictionless sharing/social reader stuff is the worst!
Agreed. Every time I see it I get very angry. It's like seeing software betray your friends.

I've had at least five friends who have unknowingly shared their readings on Facebook. A few of them were guys who were reading about sexy gossip and celebrity bikini shots and whatnot. The result was embarrassing and annoying for them.

Seems to die if I try to open up chat, but otherwise its been solid for me.
Social might become a bit of a negative word online in a few years.