Skip to content

Instantly share code, notes, and snippets.



Created Oct 23, 2016
What would you like to do?
A few months ago, Google Reader added a friends list with a sharing feature.
I immediately became interested in using it. It was so much better than delicious.
I would read a cool article in my reader, then I would click one button, and my friends
would have that article added to their reading list. Its very simple, and very powerful.
I'd say that these days, most of the really interesting articles I read are because
someone I know shared it. However, the feature was missing a "big piece" of functionality:
The ability to add and remove friends. Google was using some kind of magic formula to pick
people with reader that you communicate and chat with often. This was a blessing and a curse.
First off, I never had to think about who I wanted to share with. They just kind of appeared.
It massively lowered the barrier to entry on the feature. One day, it just started working.
I didn't have to do anything to get setup with it. Its kind of similar to the way that
facebook apps let you just kind of steal from all of the hard work users have already done
with signup/friends/etc.
The down side is... I wanted to remove some people, and add others... and I couldn't. There
was no management feature baked in. Honestly, I can't imagine selling this kind of thing
to a client. "Yeah, we'll have that... and the user will be powerless to change it." I can't
even imagine many other companies that would have let that through the door. People are obsessed with
being feature-complete. I hear all the time.. "Yeah, but what if you want to delete _____?" Actually,
a much better example of this is pagination. Sure, rails+plugins makes it kind of easy to do, but
I hate doing pagination.... WHEN WE HAVE NO USERS OR THINGS TO PAGINATE! Why solve a problem
that doesn't exist yet?
The idea isn't that you should spend less time on the quality of your app, its that you should
take time to think about what is most important. Obviously, in this case, Google decided that both
launching the feature and making its core functionality were more important than administration
of the feature. I give a huge "BRAVO!" to the folks at Google in charge of this process.
After the launch of the basic feature, BEFORE friend management, they added "notes" to your shared items.
Why? Because sharing is the core-business of the feature... not friend management.
Today, I just launched up my reader, and lo-and-behold there is friend management. It has help messages,
feature notices, and is very well thought out. Obviously better than what would have been there before
launch. Small details of it give away that the eventual management feature is based off usage. How do *I*
want to use this app? What problem are our users reporting?
Point is, don't prematurely solve problems. The only way to solve a problem right is to really understand it
and you can only understand it by launching it.
Google has grown, but apparently they haven't lost their way.

This comment has been minimized.

Copy link

@wmorland wmorland commented Oct 23, 2016

In google reader you could hit a share button on articles in your feeds to add them to your list of shared articles. A folder in Google Reader would then have the articles your friends had shared that you could then read. That way, even if you didn't follow a particular feed, the articles could be surfaced to you by friends.

It appears that when Google introduced this feature the friends list was not manually editable and so who you shared with or followed was algorithmically generated by google based on some unknown set of factors (gmail, google talk, who knows).

This might help illustrate

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment