Firefox 51: SocialAPI deprecation
Mozilla introduced the SocialAPI in Firefox 17 as a way to make interactions with social network services like Facebook easier and faster.
The API offered access to various browser components, a chat window, sidebar, or share buttons, and allowed social providers to create add-on like extensions that users could install in Firefox.
One of the first services to take advantage of Firefox's new SocialAPI was Facebook. Messenger for Firefox added support for Facebook chat directly in Firefox's interface. This allowed Facebook users to use chat functionality without being on Facebook at all.
SocialAPI deprecation
Mozilla plans to deprecate a large part of the application programming interface in Firefox 51. The following SocialAPI parts will be removed:
- Chat Window support.
- Socialmarks and socialstatus.
- Social Sidebar.
This leaves only the share functionality which will remain for the time being.
SocialAPI is not the first controversial part of Firefox that Mozilla plans to remove in the next couple of release cycles.
The organization announced last week that it plans to remove Firefox Hello in Firefox 49. The removal of parts of SocialAPI is related to that, as Hello uses SocialAPI's chat window.
The main reason for the removal is lack of use. The chat window is only used by Hello for instance according to Mozilla. Socialmarks and socialstatus are not used at all, and the few services that make use of social sidebar functionality have an "insignificant userbase" according to the bug report on Bugzilla. This leaves share functionality which "has a good set of users and providers" according to Mozilla.
All providers that provide Social Sidebar functionality have been informed about the deprecation according to Mozilla.
The organization plans to begin with the removal of SocialAPI components after the removal of Firefox Hello. This makes sense considering that Hello relies on SocialAPI to function.
The components will be removed in Firefox Nightly first. The change is then pushed to other channels during updates.
The removal of the SocialAPI in Firefox 51 has no effect on traditional browser extensions for Firefox. These may continue to use the sidebar just like before.
Closing Words
I never used SocialAPI providers apart from running tests to write articles about the functionality. I have to admit that I'm not overlay active on social networking sites and thus not really part of the target audience.
It appears however that the functionality that the SocialAPI provided, apart from sharing, was not overly popular.
Considering that Mozilla has been criticized in the past for integrating the API in Firefox in first place, I think it is a good idea to remove it after all again. (via Sören)
Now You: What's your take on the removal?
Good riddance… why Mozilla thought it was a good idea to bake this useless addon into the browser is beyond me…
What about an Email API ? Or a Video API ? Do you see the use better now ?
But yes, the experiment failed, so it is wise to move on.
(See Opera Neon for an example of what I mean with video API. Even if in their case it’s not implemented as an API to be used by websites, that’s just implementation details. I think Firefox has a similar add-on too. Dunno about other browsers.
For an example of Email API, I guess Web Notifications are a small example of it, it’s a web standard that allows a website to notify users of stuff even when they’re not interacting with the website, after permission has been granted of course. e.g. “You just received an email from Handsome Jack. Click to read it.”)
Good news
Glad to see this social stuff removed.
Why did Mozilla invest all the time and money into this API when they could have been — gasp — actually fixing bugs? It appears that they perform inadequate market research before investing large sums in R&D? Is anyone held accountable at Mozilla for this waste of resources?
held accountable? waste of resources?
Repeating (different wording) what I opined in my earlier comment:
Seems like a pattern — opensource funded efforts are expending in developing, or half-developing, the pet projects of various mozilla employees. Mozilla drops/halts the project, the proponent moz employee(s) hire on with privateers interested in polishing and using the project’s code. Like, it’s a “perk” available to moz employees.
http://www.mozillalabs.com/en-US/projects/
(search in page for)
“Synbiota: a BioLab in Every Browser”
https://developer.mozilla.org/en-US/docs/Archive/List_of_Mozilla-Based_Applications
Zimbra Desktop
The CODE from abandoned projects remains opensource, yet it’s useless without expert consultation and hella heavy-duty backend server infrastructure. Follow the Money. It’s a win-win-win-lose scenario.
Mozilla wins. Perked mozilla employees (or volunteer interns) win. Privateers and partners win. Who loses (and how/what do they lose)? Users? Donators? Volunteer coders?
If you don’t see the strategic advantage of spreading relationships across the industry, or the advantage there is in creating and abandoning a lot of projects, there’s nothing I can do for you.
>the pet projects
First I read this as “the pajeets”
‘kktkkr’ wrote: I guess this is going to be the Firefox way now:
Dream up a “nice†feature with very little relation to web browsing…
That’s MOZILLA CORP. Give more to the gullible userbase, while lacking the essential feature of an open source, de-botnetted browsing. My ol’Firefox was a browser. I used to browse the pages on the net… anyone remember the good times?
Finally they are removing UNNEEDED BLOATWARE in the browser. Hope this direction continues towards making it a fast browser.
That the heck is going on at Mozilla(!?)
I read that one of the goals which Mozilla had/has is to provide Firefox users a one stop platform that would allow them to substitute many of the functions that can be made available on the desktop using third party clients, with the removal of many useful functions from the browser, it would appear to me as though they are backing-down on that goal.
If the functions provided by the API are not being used then there is not much point of it being there, right.
People are just not willing to substitute the mediums that they are currently using in favour of the medium that is being offered in Firefox, why use Hello when you got clients like Whatsapp and Skype? I would use Hello if I knew people who were actually willing to use it over the more popular aforementioned clients.
I agree. I would have used Hello if it was more widely used; though one question was privacy, which I never bothered to check for Hello because the feature wasn’t used by anyone anyway.
But the keyword is experiment. You see them as full fledged features, but they were attempts to see if users would adopt them. They did not, even though there was toolbar buttons to access them in one click and I think even UI tours to quickly explain what they were for and how to use them.
The features got removed because the experiments did not take off, that’s the way innovation go: You try out a bunch of things, most of them fail, when they do you throw them out.
As for the goal of all browsers including Firefox to keep the web public, as opposed to proprietary where each site has its own app, it is mostly accomplished through web standards nowadays. As for non-standard experiments, Mozilla has recently started Test Pilot, and beyond that they have updated their development process last year in a way that I think allows for quicker and non-intrusive experimentation-iteration-failure process (= innovation process). One downside of doing it the Hello/Pocket/Social API way was that it was a little intrusive to Release users and so their amount had to be limited.
I guess this is going to be the Firefox way now:
1. Dream up a “nice” feature with very little relation to web browsing.
2. Spend months developing it into an API. Release and then ignore it.
3. Let it gather negative publicity for being a mostly useless core component.
4. Remove it for being underused.
5. ???
6. Nonprofit!
Yeah, kktkkr, your analysis, as disputable as it is, is above all 5 years late.
“the way now”?
A year ago (July 2015), Martin blogged about mozilla’s “Great or Dead” initiative
and I believe that’s what we’re finally seeing now/lately
https://www.ghacks.net/2015/07/16/these-firefox-features-could-be-part-of-mozillas-great-or-dead-initiative/
Your number six item, “Nonprofit!”, might just have been intended as sarcasm… but spun slightly differently, I think there’s something to be said. Some of the built-and-abandoned mozilla projects wind up being co-opted by for-profit companies… and the mozilla devs who worked on those projects become $$$ consultants/employees of those privateers. A prime example of this is the “FirefoxOS” project.
Why, all of a sudden, am I seeing the “Mobile” version of the site?
Mike, scroll all the way down and click on the desktop button to switch. Sorry for that.
Martin, I did that at the time – click on “desktop” – but it didn’t do anything. Just to check, I opened ghacks in Chrome and Vivaldi (I’m normally running Firefox) and got the same Mobile version. Also closed Firefox and cleared the cache but that didn’t help.
But the normal view came back on its own (or you fixed it!) shortly thereafter, so no big deal. :)
Oh, one other minor issue: previously my name and email were pre-entered in the comment section, now I have to type them in. Maybe that’s Firefox?
“Insignificant userbase” …isn’t that how Mozilla has referred to Firefox users in general? You know, like everyone who installed an add-on? (There just weren’t that many of us after all–according to Mozilla.) Removing an API, though… well, I guess it’ll make the download size of Firefox a little smaller. Meh.
According to Mozilla, 40% of Firefox users installed add-ons. Now show me where you found them (or anyone else) claim that 40% of anything is insignificant ?
Also, removing useless stuff, as in use-less, is good for the entire program. It has a bunch of advantages that one can’t understand without getting technical.
“The main reason for the removal is lack of use.”
I don’t even know FIrefox has that feature.. maybe because I rarely user social media
This is an overreaction. They were only enabled for users who wanted to use them. As such, they were not privacy issues until that point, nor were they “forced” on anyone. There are many features in browsers that people have no problem with which fall into the same category.
The Social API was not different from say Opera Neon, it was an experiment, except it was added to the release version of Firefox. I don’t think that was bad so long as there was no privacy issue, but I disliked the fact that Facebook and social networks received special treatment that other sites did not. It had a smell of non-neutrality which is contrary to what the web is and what Mozilla itself stands for. On the other hand it’s a fact of life that social networking is very popular and people use it all the time, so experimenting with ideas that make using them more comfortable is good. Creating an Email API would be a similar idea, and it wouldn’t appear as controversial. It was also very bad for the web that websites started using native apps instead of web browsers, so there was a need to provide user experience that was as flexible as native.
But experience has shown that distributing such experiments as new features in Firefox release was not always good for PR and as a result possibly skewed feature adoption chances. Mozilla apparently decided to do it through Test Pilot now, which is opt-in.
This “overreaction” is probably just a collateral damage due to the term “idea”, a bit exaggerated imo.
While I agree that there was an overreaction, I still believe to this day that this did not add enough value to Firefox to justify native integration in the browser.
I get it that Mozilla wanted to try it out, and that is all fine and good. Did not work out this time though.
Social, hello, pocket.. all these gone from Firefox are good news. Make them available as addons for those who want to use them, but don’t force them on everyone. Especially when you’re aiming at privacy.
Exactly. All these features are good – if you need them. Therefore add-ons are the best solution.
I’m still using Pocket as add-on (and thus with disabled built-in feature). Add-on in this case is actually even better: it has a drop down menu and minimalistic button to add/remove pages in address bar.
Bravo ! One less entry point available to data mining sniffers.