The Changelog: Software Development, Open Source - Feedbin and RSS resurgence (Interview)
Episode Date: February 21, 2017Ben Ubois, the creator of Feedbin (a simple, good-looking online RSS reader) joined the show to talk about the indie web and developers, how RSS usage has changed over the years – particularly since... Google Reader shutdown. We also talked about RSS vs the social web that we're in now and the idea of an RSS resurgence and taking back control over the content we choose to subscribe to.
Transcript
Discussion (0)
Bandwidth for Changelog is provided by Fastly. Learn more at Fastly.com.
I'm Ben Newboys and you're listening to The Changelog. Welcome back, everyone.
This is The Change Log, and I'm your host, Adam Stachowiak.
This is episode 240, and today we're talking to Ben Uboys, the creator of Feedbin, a simple, good-looking
online RSS reader.
Feedbin is also open source.
We talked to Ben about the IndieWeb and app developers, how RSS usage has changed over
the years, particularly since Google Reader shut down.
We also talked about RSS versus the social web that we're in now
and the idea of an RSS resurgence and taking back control over the content we choose to subscribe to.
We have two sponsors today, Linode and Hired.
On Valentine's Day, our friends at Linode made some big announcements.
They now offer a one gig plan for $5 and high memory plans starting at 16 gigs of RAM for $60 a month.
The super popular two gig plan they have, which is $10, now has upgraded storage from 24 gigs to 30 gigs.
This is a huge announcement for Linode and puts them in areas they've never been.
At this point, their RAM offerings across all their plans are double what most competitors offer at the same price.
What does this mean?
Simply put, you get more for less.
Of course, Linode is our cloud server of choice,
and everything we do is hosted on Linode servers.
Head to linode.com slash changelog.
Use the code changelog2017 to get $20 credit.
This code has unlimited uses.
Share it with your friends blog about it tweet it whatever
once again head to lyndo.com slash changelog and now on to the show
all right we're back jerry got uh ben you boys here and this is a an interesting show because
of how far back it goes like it goes back to 2014 2014. Wow. Oh my goodness. So yeah,
we have to give a shout out to an old friend of ours,
Alex Kessinger.
You might know him as Void Files on the interwebs.
Alex suggested this in our Trello board to have a show up on Feedbin.
And Ben,
you may know that as the Feedbin.me days,
not even Feedbin.com.
He said we should have Ben,
you boys on,
talk about Feedbin.
This was,
yeah, January of 2014.
And it sat in our queue for however long that is.
Long time.
And not that we were not interested.
It was just, you know, one of those things.
That's awesome.
I had no idea.
I worked for Alex for a time.
Oh, really?
And I like him a lot.
You worked at app.net?
No, it was after the app.net thing
kind of collapsed. And we both joined like a media startup kind of company. And he was the
head of engineering and I was just working for him on whatever he needed. And we've just kept
in touch since then. Very cool.
What were you, was that around the January 2014 time range?
Hmm.
I'm bad with like dates and remembering things.
All right.
But I'm going to say.
Where were you at yesterday?
I don't know.
I think I was.
Yeah, I was at home.
Likely, sorry.
Yeah.
So we had, I had known him since 2013.
And I think, yeah, 2014 sounds about right for when we work together.
Interesting.
Well, we're here to talk about Feedbin, which is your open source RSS reader.
So here to talk about RSS, the business of it, the change in time, what we're all using,
what Feedbin is, all the technology and stuff.
And I just say as a bit of a disclaimer that I've been a Feedbin subscriber slash user
probably ever since I think it was after the Google Reader shutdown.
So a longtime user, Adam has not, as I introduced it to him, what was it, Tuesday, Adam, or
Wednesday?
Tuesday or Wednesday, yeah.
Sounds about right.
Tuesday or Wednesday.
Not the concept of RSS.
You have one long time user
and then you have one person
who's not a user at all.
Right.
So the gamut of potential users here.
Very gamut.
Yeah.
But Adam, you said you do use RSS though, right?
I do.
Yeah.
And I've been using RSS for forever.
I think when Google Reader did its thing, I started using it less.
And especially when Twitter became more of the thing to share news and to follow people and news.
And now Twitter's like this whole new thing where it's like just everything.
It's mostly angry people and politics these days.
I don't even know what the heck to do these days.
It's harder to find the tech stuff there it's totally fine but whenever you're trying to use twitter as a rss
reader ish kind of thing or at least a news collecting thing you have a hard time defining
what news you're in taking you know it sort of defines what you take in because you have no
control over your feed whereas with feedbin or the concept of RSS,
obviously you're taking the URL or pointing to a URL and saying subscribe,
and you have much more control over what you take in.
I agree with that.
Yeah, I think RSS totally lets you
take control of your newsfeed.
Maybe I'd say at least compared to Facebook,
who's going to algorithmically insert a lot of stories.
But you still get to choose who you follow.
That's true.
Right.
But the content is still a little more wide ranging than someone's blog, I would think.
Right.
I mean, you can make lists and things like that, but we give the control to social networks at first
because they you know they they give us the i don't know they they care they seem to care more
about the end user at first like i can remember a day when twitter actually cared about developers
maybe they do more so these days but they've had a bad reputation with their uh you know early
adopters so to speak where they've kind of like turned their backs on like the early adopters of Twitter,
you know, where you don't have,
you know, a lot of people using the API
like they were originally,
where Twitter wants to do a lot more of that stuff themselves.
But the point I'm trying to make is that
we, as a user of that platform,
give a lot of control to them,
and we consume the timeline in real time, chronological,
and then they go and change the game
and they throw an algorithm behind it, and they say, this is actually what you care about more this is what you should
care about more and instead of saying like well i actually subscribe with that person so i want to
get all their tweets not just the ones you think i should be checking out right it's interesting to
hear you say twitter used to care about developers or currently does not care about developers. I look at it as like
corporations don't really have feelings or really care about anything except for profit and
surviving that sort of thing. Some of them are socially conscious around the environment and
that sort of thing. But more or less, I don't think corporations have feelings, which is why I'm a big fan of indie software, because indie software is made by individuals and individuals care about things and have feelings.
Yeah, that's true.
It's one of these continuums, right?
Because the smaller the corporation, the more the individuals have to say, and especially once you get to a point,
Twitter size, Facebook size, IPO,
now you have shareholders and everything else.
But it's one of the things,
I mean, Twitter started off very small.
It was a group of a small subset of, what was the Evan Williams podcast company?
Odeo.
Odeo, yeah.
I mean, it was three people inside of Odeo building it.
And when it first came out and all the other adopters, you could talk to the individual
people who were not just working at Twitter, but were actually making the decisions.
And so over time it grows and turns into something I very much agree with you that has no feelings
or, you know, it doesn't care in terms of individuals care, but it's one of these things that as tech users and people
on the forefront we see these companies incubate and start really small and then we see them turn
into something that they weren't which is inevitable if they're going to be successful
yeah and with twitter that that can be painful because people remember when it was small and
you could name everybody who
worked there right if you're following it closely and they were trying to
attract people that you identify with instead of celebrities and that sort of
thing so right that that transition can be painful but I think at a certain
point it's like we just have to recognize it's not for you anymore
specifically right which is the nice thing so let's get back to RSS here it's not for you anymore specifically. Right.
Which is the nice thing.
So let's get back to RSS here because comparing and contrasting
and I'm with you, Ben,
in terms of liking indie developers,
you know, indie everything really.
And I think the open source mindset is very indie.
It's like, let's just get together
and build a thing together.
And we've seen even that grow up
over the last five, 10 years
into something that it wasn't before. We've seen a lot of seen even that grow up over the last five, 10 years into something that it
wasn't before. We've seen a lot of fruit from that growing up and with that comes problems as well.
But RSS has been all about indie. It's been all about control. It's been all about a really pure
technological solution that's simple. And right there in in the acronym tell us about your perspective on rss
and why you began feedbin especially in light of the google reader shutdown around 2013 was it
before or after that give us some historical context with you and rss feeds yeah so i think
like i personally go back uh with rss to around 2004 i was starting to get into web development at that time. And I was visiting
a lot of blogs like that was the pretty peak time around web developers talking about web
development on their blogs. And I found myself visiting a lot of these sites and just kind of
reading archives. But then I would come back to the sites. And then I feel like I was
aware that like in the sidebar, people would link to their RSS feed. And I was like, what is this?
This makes no sense. I click on it and it looks like machine barf because it's all XML.
Machine barf is a good word.
But then I came across NetNewsWire, which is a great Mac app developed by Brent Simmons at the time.
And that let you put the machine barf into a very nice, consistent, well-designed format.
And it let you be a completionist for any website you're interested in. So if they post a new
article, then you see that article in a very timely manner and you can be sure that you're
not missing anything. That's Jared's love language right there. Completionist. I am a completionist.
Yeah. It also becomes like a single destination then because you're no longer visiting dozens of websites individually.
You just go to one place, which is your RSS reader, to read all the content that interests you.
So right away, I was hooked and I used that for years.
When the iPhone came out, I think I started using Reader, which is a wonderful iOS app, probably one of my favorite iOS apps to this day.
And then I eventually switched to using Google Reader, which would sync with both NetNewspire
and Reader.
And I was kind of down on Google Reader.
Like I thought it was an interesting product and I liked the sync service, but I was thinking
about it maybe in like late 2011, early 2012.
I was thinking like Google isn't really like dating this all that much.
In fact, they had like a big update that moved a lot of features just to kind of like prop up Google Plus or something at the time.
I don't really know.
But I was like, I kind of want something on the web that's as good as NetNewsWire is on the desktop and Reader is on iOS and Google
Reader is not that. And that combined with, you know, as the time I was working at a web
development agency, but I was building mostly static marketing sites and I loved doing the
backend development and system administration and all that stuff. I just felt like I wasn't
getting enough of it. So I thought like, you know, my number one passion and hobby is reading these
RSS feeds. Let me see if I can build something for that. So then I started building it in Rails,
which is also fairly new to me in late 2012. And then along the way, I was thinking like,
this thing needs to have great uptime because,
you know, if I'm going to be visiting it frequently, it needs to always be up.
So that's going to cost me money, basically, just to keep the servers going.
So I thought, well, I'll build in like a billing component so that maybe I can convince like
some friends and family to sign up just so I can keep it running basically so I can
use it still. So then Feedman launched, I think in like early March of 2013. And about two days
after launch, Google announced that they were shutting down Google Reader, which was insane.
Like the timing of that was just like really crazy for me because crazy good, right?
Yeah. In a good way. I mean, it was, it was like overwhelming in a way also, because right away,
there is a ton of attention in that space and on Feedbin. And I didn't know what I was doing
around running a web service and, uh, had a lot of catching up to do.
And I still had a day job at the time.
And I was just working really hard to keep everything running and keep it up.
Yeah.
And try to, you know, spread the word that this is an alternative out there.
You know, like two days before is great timing. But if I had launched like six months before that would have been even better
yeah because you would have been a little further along like your experience of running the service
not yeah just launched and have to take in all this massive traffic yes i'm assuming you had a
hockey stick right uh yeah so the google reader shutting down was huge. But the next amazing thing that happened was pretty much out of the blue,
the developer of Reader, the great iOS and Mac app, contacted me and said that he wanted to
support Feedbin in his next version. So he has like a huge built-in following and Feedbin was going to be an exclusive service to Reader, at least for the time.
So that really put Feedbin on the map as a viable alternative because everybody was already using Reader
and they would need something for it to synchronize with.
And Feedbin had an API and he to uh to get sync working with that
and uh that kind of established feedman as a viable business i think a very similar story
to yours minus the open source business that you began as a rss reader fan i never actually
used net newswire i think i started on google Reader, but used Reader, the app, on both iOS.
And I hated Google Reader's website so much that I actually used the Reader app for Mac as well.
And so when I found Feedbin, it was very much because I was using Google Reader as my sync service, effectively, right?
My dumb pipe, which worked very well for that purpose. And just for some historical context, this was like the summer of 2013
is when the whole Google Reader announcement
and shutdown on July 1st, I looked it up, happened.
And so around that time, I mean, there were thousands of us,
maybe tens of thousands that were scrambling
because as a daily user of Google Reader Sync Service,
like that was just going away.
And, you know, many of us, I think, never actually found new homes.
I think there's a lot of people that just quit using RSS at that point.
I think so, too.
I see that.
I see that a lot.
And I kind of feel like those people just need to Google it.
I feel like there should be some sort of resurgence.
I mean, I almost brought that up a little earlier, but I didn't want to go that direction
too early because I feel like for the reasons you know that we just
talked about whether it's instagram whether it's facebook whether it's twitter however you consume
your community right and it's kind of weird to say to say you consume your community but
the point is is the news the information coming from the people you follow if instagram facebook
or twitter those places then they've all begun to dictate to you what information matters more to you rather than you having the control
which is what we talked about earlier right and so i feel like there needs to be some sort of
resurgence because if it weren't for the uh what was the term you used ben you were you said the
machine barf because that's the that's the hindrance right the machine barf is what stops
people they hit that url it pukes onto their screen they're like i have no idea what to do
with this let me get out of here as fast as possible back back back and they never come
back to rss or they've been there and they've used google reader and then it bailed and it
stopped doing what it did and right they're like well i'm done too i'll just use twitter
well there was a time where the the browsers were doing more to make i mean there was this whole movement where
like consumer let's just call consumer facing rss right was like a thing that was gonna happen
yeah i also was gonna blame the the browser vendors because if you look at like something
like safari they used to have a nice like xslt style sheet that they would
put on top of rss feed so when you click on one it looks like formatted links and text instead of
raw xml yeah however i you know i i recognize that there's like a strategy i don't know if
that's the right word strategy tax though i want to say there because they all have like apple
specifically now has apple news which launched is basically an RSS reader.
Yes, it's definitely that kind of style.
They have a custom format, but it crawls XML feed.
Right.
So with those kind of conflicting interests, it never gained a great foothold at like a browser level or OS level.
Do you feel like page views is a concern here though?
Yeah.
I mean, cause that, that to me is a reason.
You have a choice whether you want to publish an RSS feed or not.
Right.
And for the sites that do choose to publish an RSS feed, I think it makes sense for the
browsers to do something with that.
Cause the browsers aren't making anybody do anything they don't want to do.
And the website operators have full control over whether they want to have an RSS feed,
whether they want to have full content, RSS feeds, all that kind of thing.
However, I think that the kind of business that would not want an RSS feed also is the kind of business that would not have very
interesting RSS feed. The interesting content out there, in my opinion, is all the personal blogs
of people who are not really in the business of producing content. Right. They do it because they're interested in what they're writing about
or recognize that there's value in what they have to say.
And that's exactly how I use RSS, which is why I still love it to this day.
I was never, because I am a completionist, Adam,
and I absolutely have to get to inbox zero, RSS zero, Twitter zero, all the zeros.
Because of that, it limited how I used RSS. I couldn't subscribe to the people that, Ben, you said wouldn't have, well, you can't
subscribe if they don't have a feed. But there's a lot of sites that publish feeds, but they're
not good RSS feeds, like Mashables, if you subscribe to a hacker news feed where you're just getting aggregated content
or the quick hit news in huge batches throughout the day, it's not a good way to go about things.
But like you said, Ben, if you find the indie bloggers, the guy over here who likes to write
about this thing and the gal over here who loves web animations and she publishes maybe... I mean,
you don't have to keep up.
Like she publishes once a month and you're going to see that.
You're going to see that.
There's so many goodies.
Yes.
That come out of RSS if you do it that way.
Low volume is a key to enjoying yourself with RSS.
Yeah.
Well, especially if you plan to get to the end of the feed. I mean, in the case of what you had said, I mean, even if you're not a completionist right right even if you just plan to keep up let alone complete right you know trying
to subscribe to say hacker news some sort of like front page feed consider yourself just consumed
with that i mean like it's going to be a fire hose right it consumes you and yeah exactly and
i think the point of at least at least what i like most about RSS is that it puts you in more control over the things because you can easily just delete something from your feed and now you don't subscribe anymore because, well, hey, the frequency got too high or it kind of moved you away from this personal relationship with the indie artists out there in whichever niche you're hanging out in and you're actually able to
get through that news versus feeling overwhelmed i think which is a whole different topic but i
think in general people are overwhelmed today i tend to agree let's let's loop back into the
business side of it here ben but back to the google reader thing you had the you had we had
a bit of a cambrian explosion you had great timing there's other people that were scrambling to get together their alternative services.
There were some little small players like yourself.
There were big ones giving it away.
Here you are charging.
But like Feedly, for instance, was one that was really popular around the closing because it was free.
Maybe BC backed.
I don't recall.
I haven't used it since.
But you have all this competition.
Then you had the reader thing where we were exclusive, or at least right up there up front.
And so you had these two big moments early on where you probably got a lot of customers,
but you're very much in a dwindling, right?
We even said a lot of people just moved on.
We are the few, the proud, the RSS readers, right?
Like the users are small and so
here you are trying to build a thing i thought you had a separate full-time job but probably
because your website might be outdated um where you said you looked worked at flicker box is that
no longer the case you're full-time on feedbin yeah so i switched to doing feedbin full-time actually in early 2014.
Okay.
Because, yeah, it had gotten to the point
where I felt comfortable financially to do that.
Awesome.
Yeah, and that was awesome.
That was a good feeling.
And it certainly was not my intention.
It wasn't anything that I was shooting for.
It just kind of fell on my lap,
which I hate to say, because I know that
like a lot of people try very hard to make that happen and I and it doesn't work out for them.
So I recognize that I was very lucky and I try to be thankful for that. Yeah. Especially,
you know, when when things aren't going well. Yeah, sure. But yeah, so RSS, I would say, is dwindling.
And there are a few people who are into it.
But the people that are into it are passionate.
So like Feedbin has probably more customers
that signed up before Google Reader shut down
than like after.
So there's like more people who have been with Feedbin since the beginning than there
are that have like signed up since then.
So they stick with it.
They come back like every day or, you know, at least several times a week.
I think this is probably like a industry leading number, but like the average time on the site is like over 70 minutes for Feedbin, which I think Facebook and anything else out there.
That's a lot.
So these people are into it.
Yeah.
And that's all I need for Feedbin to be sustainable.
Very cool.
Well, I want to dig into Feedbin, the product, talk about what it
offers, talk about the open source and the technical side of how it's all wired together,
because that's very interesting to us, too. Yeah. So we're going to take our very first break and
we'll take that up on the other side of this. Our friends at Hired want to help you get your
next job that much easier. Searching for a new job is stressful, scary and time consuming.
Recruiters try to sell you on all the roles you don't want, and job boards feel like a black hole.
Sometimes you get through the entire process only to find out at the very end that salary, company culture, or other details don't match what you're looking for.
Hired is different.
They're an intelligent talent-matching platform for full-time and contract jobs in engineering, development,
design, product management, and data science. Instead of endlessly applying to companies
hoping for the best, Hired puts you in full control of when and how you connect with
interesting opportunities. After completing one simple application, top employers apply to Hired
you over a four-week time frame. You'll receive personalized interview requests and upfront salary information so you can make informed decisions about the opportunities you'd
like to pursue. And if you're concerned about privacy, don't worry. Hired hides your profile
from your current and your past employers. The best part, Hired is free. It doesn't cost you
anything. In fact, they pay you to get hired. Our listeners get double the
normal $1,000 hiring bonus to find your next chapter on Hired. Head to Hired.com slash changelog.
Don't Google it. This URL is the only way to double the hiring bonus to $2,000. Once again,
Hired.com slash changelog. And now back to the show. All right, we are back with Ben Uboys
talking about RSS,
talking about Feedbin,
talking about open source and software.
Ben, give us the lay of the land
of what Feedbin is.
You gave us kind of the start of it
and a little bit of the business angle
about some of your fortuitous timing.
Tell us about the product,
what it is today,
and maybe even tell us, you know,
where you'd like to take it. Yeah, so FeedBend is primarily a web-based application. It's a
full-fledged replacement for something like Google Reader. You could also use it instead of
any native app. It works great on the desktop and will work on mobile as well. A lot of people also do like to combine it with a native app like Reader is probably one of the most popular ones out there because they just like the native app experience.
So Feedbin will synchronize with Reader and then everything will be reflected when you log back into the website and you can
pick up right where you left off. The way that I like to think about Feedbin is just as a nice
place to read on the web. I think the web has gotten increasingly hostile with ads and popovers
and just de-emphasizing the content in an unhealthy way. So Feedbin is all about the content.
There's nothing there except for your content that you choose to subscribe to and tools
to help you enjoy that more and manage it better.
Very cool.
How about in addition to it being a nice place to read on the web, does it have any unique
features beyond what you'd get with a reader perhaps?
Or using it as a...
The one thing about Google Reader
is the way at least I used it was,
it was just my sync service.
Does Reader offer things above and beyond
like a sync for your OPML
if you wanted to use it with an app in front of it,
like the web interface?
I guess I'm asking this really poorly,
but are there other features that set it apart, perhaps from other RSS things? So if you are primarily
or exclusively using RSS through a native app, I would say the experience of using any RSS
service is largely undifferentiated. And that comes back to what the app authors want to
implement. Like I think an the app authors want to implement.
Like, I think an app author would want to provide a consistent experience across any service that they support.
So, well, Feedman does offer some extra API endpoints that would enable app developers to support some of the extra features that Feedbin has. Often they don't always choose to do that just because it is then inconsistent
with other RSS backends.
So that is a real challenge.
Like the question then becomes
why use Feedbin over any other service
if you are just using a native app?
Right.
And I think about that all the time.
And I work towards that goal of differentiating it.
I don't think I've been like super successful in that regard
because Feedbin does cost money and there are free alternatives.
However, like, right.
You know, with like a larger company like Feedly, I kind of get nervous around like what their intention is.
A freemium product.
And I just feel like with that, they have to make compromises around features to get you to upgrade and how they think about and treat your data.
So like with FeedBend, there's this really,
I think of it as a premium product
and you get a lot out of it
because I don't have to hold back anything.
Like every idea I have for what would make it better
is available to every customer.
And I treat it as well as I can
around like uptime and performance.
And I've put a lot of money into the infrastructure
and servers and all that stuff.
So I think it's just like a very well crafted experience that is not matched by any of the
other services.
Feedly's got some interesting language on their homepage.
Their selling point, at least, which I'm assuming is that's what this is, because it's just
right underneath the welcome to Feedly. It says the content you need to accelerate your research, marketing, and sales.
It says it twice at the top and at the very bottom.
There's no other links to like a blog or about or any details to kind of research what they are.
But I can't imagine that's the best selling point to say here's a way to consume RSS feed. Yeah, I kind of. I'm not saying this is bad. It's just it's point to say, here's a way to consume RSS feed.
Yeah, I kind of.
I'm not saying this is bad.
It's just it's confusing to me.
You know, I don't I don't get it.
It's an approach that I recognize, which is how do you explain what RSS is without using the word RSS?
Yeah, right.
So that language to me sounds like, oh, you're like a business professional.
Whereas I think Feedbin is more targeted at like web professionals, savvy users and just passionate people in general with diverse interests.
Right. It also doesn't shy away from being techie.
Like there is stuff in Feedbin that you wouldn't understand unless you have like a
pretty firm understanding of how the web works. However, there's a great interface on top of all
that that still makes it look nice and be enjoyable to use. So what's interesting to me is you have
multiple pressures on Feedbin as a business. And just to put it out there, Feedbin is $3
per month subscription fee.
So that's the way Ben supports himself with it.
You have free alternatives like Feedly and others,
where, like you said, if you're using it as a sync service
and you have a native apps on all your devices,
you can't actually do better than a free service in many ways.
Then you have the dwindling use of RSS, right, which we've discussed,
and the increasing use of social networks. And then you have the inability to explain it to
people who may love RSS. They could just get over that technical hurdle. And so you have all these
different struggles as an independent developer trying to support himself with a business.
And in light of all that, you also open source sources so like the very last thing you could think of is you know reasons why i may not subscribe to
feedbin is now you get the person who's super technical and they love rss and they think feedbin
is a superior experience but they're a hacker and so now they can say oh i can just run it on my own
server it's here open source that seems like your last holdout.
And so tell us about the decision to open source it and why you did that and how that's going.
I didn't, you know, I didn't put too much thought into it.
You know, I thought about how it would impact the business and that sort of thing.
And I kind of decided that I don't feel like it's a threat that someone can run it
themselves i think that that should give any customer confidence that you know they can look
at the code and see that nothing is happening with their data that they don't they don't know about
so it's more for transparency yeah they can add features and they can, you know, run it themselves if like I get hit by a bus or whatever. And also
I was interested in like trying to see if it could be a successful open source project. I was,
I guess I was doubtful about that. Like when I, when I look at open source in general,
I kind of feel like the successful projects are, I'd say like frameworks and utilities.
I can't really think of any like full-fledged applications
that are also successful open source projects.
Like if you...
I'm trying to think of a counterpoint, but I can't do it.
Yeah.
Like what's a...
Maybe you can think of one as like,
what's a very successful open source
app or product one that i was thinking of is um that i use for my business is airbit
which is an open source version of airbrake which used to be called hop toad i think it's
an error catching service i've never heard of any of those things okay so when feedbin throws a you
know 503 or whatever you know where do your errors go and so
maybe to your logs and now you're tailing logs and all this kind of stuff well you have
error catching services which do a nice job of parsing and displaying and you know notifying
and all these fancy things for other applications and so they are application like it's a web app
that i run on heroku and i use it to catch errors and i log
into it and all these things and there's lots of people that use that one in fact i think that's
probably also in our backlog adam of people to talk to sure is uh is the airbit folks so there's
one they don't sell i mean there's a there's a for sale you know alternative to it and so that's
an example but i tend to agree with you you know it changed log.com is an open
source it's not a product but it's an open source application you know we haven't seen a cambrian
explosion has the second time i said cambrian explosion we haven't seen a huge influx of prs
we've seen plenty and we're thankful for those enough to make it worth it to oh definitely worth
it but it's not like the whole community is building this thing right and i'm assuming that's that's how feedbin is it's not like well you have 24 contributors
just looking at your github page and so it's not like the feedbin user base is building feedbin
with you yeah there's actually an upcoming sponsor we have that has a similar story in terms of open
source and a parallel product that's hosted and sold as a hosted product but you could take this
thing and run it on your own it's called century it's for error tracking since we're talking about
error tracking i was going to mention century yeah that's in the same vein as what you're just
talking about very similar in terms of yeah product development open source product you can
totally take century and host it yourself and run it yourself. It's the same exact thing that you would buy or try for free on their site.
So it seems to me like where you have this open source core or open core, which is becoming more and more of a trend where you have this open core thing and you build on top of it as a hosted thing. But, you know, the open source version is, hey, if you want to use that or you want to use it differently
and it adheres to your licenses
or, you know, for whatever reason,
you know, it's open to you to use
however you like.
You know, but you get the service
that parallels it.
Right.
It sustains it, frankly, you know?
Yeah.
So like another hero of mine,
I believe someone who has been
on the changelog in the past
is the author of Sidekick, Mike Perham.
Oh, yeah.
His business model was inspiring to me because Sidekick is an open source background job
processing utility that integrates great with any Ruby environment.
And he also offers a commercial license for it.
But again, I think of that more as a utility rather than a full-fledged application, although it does have a very nice web interface.
And I think he's been very successful with that as a business.
And I think it's an interesting example of sustainable open source because he has built-in motivation to keep making it the best background job processing framework out there.
Yeah.
It seems like you're trying to downplay the thing you've done, though, to me.
Does it seem like that to you, Jared?
In terms of open sourcing it or building it?
Well, just like that it has less ability to be successful.
I almost, Ben, that you seem surprised to some degree that you're successful
or that RSS could be or a
feed reader could be that important to people that it being open source and a paid service could be
more successful than it currently is it almost feels to me like you're just not sure of it yeah
i would i would say that's true i'm i am not sure of it i don't, it's hard to do any kind of attribution tracking around like
how big of a bonus is that for customers that it's open source. And, you know, just looking at
the pull request history, it's like, yeah, there's been a few awesome contributions, but there's far
from like a thriving community there. Right. So, so I would say it is somewhat unproven,
whether it's like a positive or a negative thing.
I like it.
I think it is interesting.
However, I don't know how much that's worth.
Well, you have 66, six open PR, 66 closed.
I'm not going to get a count on how many of those were merged,
but these aren't terrible numbers.
And now, of course, you have your own subjective opinion on what's been good or bad or it hasn't been a support burden.
You do have 53 open issues.
So there's people that are reporting bugs.
And I agree, especially as I hit your read me and I see that you have, you know, a gazillion environment variables to define and all sorts of other things to get it going.
Wow.
That is a gazillion. Well, he has all these of other things to get it going. Wow, that is a gazillion.
Well, he has all these services
you can share with and stuff.
Yeah.
I would probably look at this and say,
yeah, I'm going to pitch him three bucks a month
because this isn't worth my support, my own,
even though I could definitely run the server
if I wanted to, right?
Yeah, it's still cheaper
than the cheapest digital ocean instance.
So I think it's a pretty good value to just pay for it.
I mean, back to the indie thing, though, I think that, you know, as anybody who's probably in that realm of RSS, they probably have some kindred spirit to the indie software developers, indie web world, indie blogger, where they are going to want to support you.
And it's less about value and it's more about it's value plus, you know, it's like, right.
I get value from not running my own digital ocean server or, you know, name your host here
place to put this at. But at the same time, they want to support you as a developer to keep making
it happen because they value the end all product they're using, not just the fact that like they get some money or they save some money.
Yeah. And I think that's huge. I think there's a lot of, like I said, you know,
RSS people are passionate people and they do care about who they're giving money to
and that design matters and everything, everything else like that. So I almost think about Feedbin
as like IndieMac software, even though it's very much web-based well it's a good time to take a break
actually so we'll take this break when we come back from it we're going to get that a little
deeper into how this thing actually works so when you do use it on site or if you run the
open source version on your own server you you know how it works. We'll be right back.
I've got good news for you.
This Friday, February 24th, we're launching a new show called JS Party.
It's a live celebration of JavaScript and the web
every Friday at 3 p.m. Eastern.
The show is hosted by Michael Rogers,
Alex Sexton, and Rachel White
at thechangelaw.com slash jsparty to subscribe subscribe and here's a quick teaser of what's to come take a listen
progressive enhancement we've been talking about progress behind some for
more than 10 years right it's just been this this general good thing that
everybody should be doing that everybody talks about at conferences and then
people go away and some of them do it and some of them don't progressive
enhancement now is is the same, but fundamentally different than what people were talking about.
Or maybe not fundamentally, but in practice, very different than what people were talking about 10 years ago.
Yeah, I think the thing that people have been getting mad about in the past month is more accessibility focused than uh any other kind of progressive enhancement stuff
from what i've seen yeah it's not even about speed right now it's just like can everyone
who is using the internet use your site well i guess that would be speed if depending on where
your internet is perhaps we're on slightly different twitters then uh yeah my twitter is is if you
have uh nolan lawson and alex russell in your feed it's it's entirely uh speed related like
javascript first applications that like require javascript to run before you can see things
versus server rendered with very fast interactivity sam sam, Sam Saccone is another person really on there.
The stuff that I've been seeing lately is, um, making tools, uh, for the, um, you know,
like the, you might not need JavaScript stuff, but then not making it accessible for everyone to view,
but using JavaScript to display the page somehow. I don't know.
Let's rewind just a little bit and unpack this.
So that if anybody's not on our exact Twitter feeds, they can figure out what we're talking
about.
This goes to show how much people confirm their own biases.
All right. That was Michael Rogers, Alex sexton and rachel white in our upcoming show
js party february 24th is the first live show hit the changelog.com slash js party and now back to
the show all right we're back with ben you boys we're talking about this awesome thing called
feed bin and i've never used it. I'm going to sign up.
I'm going to check it out.
But one of the things
you want to know about
is like behind the scenes.
You mentioned it's a Rails app.
What else?
What zillion environment of variables?
What else we got going on here?
Yeah, so I was,
I think I mentioned earlier,
I was pretty new to Rails
when I was starting out with Feedbin.
I'd actually used it way back
in like 2006 for a college thing.
And then a very long diversion,
working in PHP for many years, and then started getting back into Ruby and thought I'd build Feedbin and Rails. So I tried to make it as vanilla as possible, just because I didn't have
a better opinion than what was already there. I like the philosophy of Rails and all that.
So that's really the main app. And the other things that are in use are Sidekick, which really
like even had scaling problems early on. Like I didn't I didn't know what I was doing. And there
was just way more stuff that needed to be done than I had anticipated
because of the growth.
So Sidekiq kind of solved scaling for me.
The biggest issue was how do I check hundreds of thousands or millions of feed URLs efficiently
and in a timely way. Because, you know, if you're on a single thread in a Ruby process, every HTTP request is going
to take like on average, like a couple of seconds.
So that just does not does not work out if you're limited by number of processes.
So because of Sidekick's threading model, I was able to really parallelize that and actually push
that out onto a handful of cheaper DigitalOcean servers.
So that runs on eight different servers right now with a few Sidekick processes per piece.
And then it can have maybe, I don't know, 1, 1500 feed requests per second that way,
which is the number I'm proud of.
I mean, I know that like there's probably
Go and Rust people who are like, whatever,
but that works well enough for me.
And it does it in a way that like,
I feel like it's never far behind anything.
So I've been really happy with that.
The only other like separate service
that's similar to that is this
image processing system. So Feedbin, when possible, will try to like pull out a relevant image for the
article and like put that next to the article summary. And when I was building that, it was
really important to me that the image be high quality because you know i've i've worked
on designs where it was like spec that like okay every like piece of news here has to have an image
next to it and i've seen other readers designed in that way also and the result is that you get
like some really like low quality like pixelated logo of some company.
And that just like doesn't look good.
What I wanted to do is like make sure it was like
frequently photography and that there was like
a minimum size formatted as a JPEG.
I think that's a good indication
that it's like gonna be better quality than like a PNG
or more indicative that it's going to be a photograph.
And then so it goes through all these filters and criteria and then it gets resized with image magic and then pushed to S3.
And then FeedBank can pull that back in and display it alongside the content. It also one thing I'm really happy with is I got this idea from Twitter ethic, the iOS Twitter client where they blogged about Apple has a bunch of cool image APIs and one of them does face detection.
So it'll tell you in like an X, Y coordinate where in the image faces appear. And what they did with that where they were like, okay,
we're going to crop all these images.
So they fit into our template,
but we don't want like cutoff heads in, uh, in the crops.
So using the facial detection in this case, you know,
Feedman does not use Apple stuff, so it can't use those APIs,
but it uses another, um,
project that does the same thing
and it'll figure out where faces are.
What project's that?
Do you know?
The name is escaping me at the moment.
Let me get back to you on that.
Noku Gumbo?
No, that's more of a general purpose
HTML5 parser.
I'm just looking at your gem file trying to decide which
one of these gems is what he's using for the for that part and that was my guess yeah open cv
that's what it says open cv that's correct there you go yeah so open cv has that feature
as well with the face detection so feedman can make a intelligent crop on any image that has a face in it and keep that centered.
And the results of that are amazing because you just get like it looks like it was done
by hand rather than by a machine.
Yeah.
Yeah, I can really say that I've been as a user, I've been very happy with those results.
I remember reading your blog post and thinking, I don't know if this is really going to work,
but it almost looks like these images are art directed because it does, it finds the
interesting part. Um, maybe it's just with, especially with the faces and, um, turns out
remarkably well. So I just say job well done on that. Tell us about the process. It's like,
does that image detection and uploading, you know, cropping and associating into S3. Is that part of the crawling
process or does that happen when you hit the Feedbin homepage? Where is that in the flow of
the application? So both the feed refreshing service and the image processing service,
they're Ruby apps, but they're just based on Sideq. So the only process that's running there
is the Sidekiq process.
Okay.
So those get from the Rails app,
those get queued up as Sidekiq jobs,
but it's set up in a way so only these specific servers
on DigitalOcean will grab them,
and then that service will then take over from there.
So the part that does the work is just a sidekick job.
Gotcha.
Keep it simple.
Yeah.
So you have those three major sections.
You have the web app, you have the refresher,
you have the image service.
It's like you also have some sort of HTTPS proxy,
probably because you're pulling in images
that are mixed content or something.
Yeah.
I almost feel like it's less of a big deal now like
browsers have de-emphasized uh the mixed content thing but that was a cool little project i found
from github called camo which will rewrite image source attributes to point to their proxy server
so then your proxy server terminates ssl and then makes the request to the image and
then returns that so then you don't get the mixed content issues cool i guess on the technical side
i would just have one more one other line of questioning which is around the web interface
itself it's almost it's hard to say like it seems like it's maybe like a single page app once you're
logged in or it's at least a kind-page app once you're logged in,
or it's at least kind of a rich JavaScript environment.
There's a lot of keyboard shortcuts.
There's a lot of navigating
without actually changing the URL.
Can you talk about maybe
the front-end architecture a little bit?
Yeah, so the thing that I'm the most proud of here
is the design,
and that's the work of a very talented designer
I work with named Todd Collins,
who works at the company I used to work at, Flickr Box.
And the way it all works technically is for the most part through like Rails remote JavaScript execution.
So like when you click on something that'll render, you know, a.js.erb file, which will come back over HTTP and then execute,
you know, in the full JavaScript environment.
However, that hasn't like always offered the performance that I wanted.
So then there's also like extra JavaScript on top of that to like, when I click this
thing, I need something to happen now, not when the HTTP request finishes. It's a little bit of a weird hybrid between the remote JavaScript and like JavaScript
on top of it.
I don't love it.
Like it's been flexible enough that I can get everything done that I need to get done.
But I'm also like keeping an eye out for like maybe like something like Ember might
be a better fit eventually because it is very much like a single page app, at least, you know.
Right.
The primary app, not so much like the setting screens and all that, but the main interfaces.
Exactly.
Yeah, I would tend to agree.
And that's why I was curious how it was put together, because it seems like it's very much your poster child for a framework like ember or some sort of react based thing real quick tell us about actions because
this is a we talk about differentiation it seems like a you know a place where you're really
embracing your nerdiness which is really why i even asked you specifically about this feature
it's almost like this uh if this then that type of a thing that you've built into it can you can you tell us about that yeah yeah so people people love actions and uh it was a the often requested feature around like how can i
how can i filter this stuff it worked out really nicely because i was looking into adding full
text search to feedbin um and i settled on elasticasticsearch, which I'm a huge fan of. And Elasticsearch
has so many features that like I couldn't even begin to like understand at all. But one of them
that jumped out at me was something called the percolator. And the way percolator works is it's like a almost like a saved search.
You say like, here's like the terms that I'm interested in.
So now when I when I send you content, also go ahead and tell me if it like matches any of the registered searches that you know about. So that was like a perfect fit for what actions does because it's acting on
content as it's published.
So if you say like,
like a super common one is people don't want to read a RSS article that has
the word sponsor in it.
Sorry.
Sneaky.
Yeah.
I mean, I thought about that.
That doesn't make sense.
So you're essentially performing a grip on on the article content.
And if it contains something, do something with it.
Yeah.
So what are the what are all the actions you could do with it?
A more positive one.
Like, I, I don't love that because I think people should support the content. So like a more positive one and one that I use or have used in the past is like,
let's say you follow a website that has multiple authors,
but I like to follow like individuals
rather than like websites.
So like if there's only one author that I'm interested in,
but their content is mixed in
with like a bunch of other authors,
then using actions,
I can use a negated search like author is not John Doe and then mark that as read.
And then that feed will only show me articles that are written by the author that I'm interested
in.
Wow.
That's actually really cool.
I didn't even know you could do that.
I have a specific use case for that that I'm going to go do.
Awesome.
Yeah, so you can then scope that to a specific feed
because you only want it working on that one feed
that you're interested in.
And then you can set up what you want the action to actually do.
So if it matches, then go ahead and mark that as red or alternatively you
could like star it and if you're using safari on the mac you can actually have it send you a native
push notification for that article so i'm looking at that same interface right now
on chrome so i'm assuming that it would say mark it is red, start or notify if I was in Safari.
Yeah, that's right. That's interesting. So essentially you're creating an action,
doing some sort of grip on it, which is, you know, it could be a negative or a positive
towards a search in there. You can include certain feeds that you've got obviously in your feeds
and do those three things with it. Is that the only three reactions you could do with it?
Yeah.
That's all that matters really to a feed reader, right?
Yeah.
So I kind of have a, there's an idea I've been kicking around and, you know, I don't want to promise anything
and I generally never like pre-announce features.
But one idea that I've always wondered about is like
something that is more along the lines of like,
if this than that, which is that Feedman also has like a lot of along the lines of like if this than that which is that
even also has like a lot of built-in sharing services so like it'll hook up
with your pin board account and then you can easily save articles to pin board
that makes sense so the idea that I had is like combining that with actions to
like automatically send stuff that matches that to
pinboard. I kind of the downside of that for me is like it's not very like purposeful that way.
Like if you're saving something to a service like pinboard, I kind of want that to be like
very intentional and not like in an automated way. That's just like spewing stuff over to that account.
Yeah.
Like I know like I'm down on GIF in that way.
Like I know a lot of people love it,
but I also like feel like you should really be really thoughtful about what
you're archiving.
Cause there's a tendency towards like just wanting to save everything,
which is almost like a hoarding tendency in my mind.
Also the same thing on the sharing side too.
I mean, if you think about looking it up to a Twitter
or to a Facebook share,
which you could say, well, that'd be really nice.
But then is it though?
Like, are you just adding to the noise
of like automatically sharing?
Yeah.
So I would have agreed with you. Like Feedbin So I would have agreed with you.
Like Feedbin,
I would have agreed with you in the past
because Feedbin launched
without any sharing features.
I was like, I don't like sharing stuff.
I'm an introvert.
I think that like social is kind of lame
and that sort of thing.
But what changed my mind about that
is I was thinking about like
you were saying earlier
that people are moving
away from RSS and to sites like Twitter and Facebook. But if you think about like where
content on the web originates, it's not from Facebook and Twitter. It's from websites that
usually have RSS feeds. So if you're going to be the one that originates content,
then you need tools to help you do that. And Feedbin can do that for you using those integrations.
I see what you're saying. As a publisher, you're basically syndicating yourself via
your RSS feed through Feedbin if you wanted to.
Well, if you're like, I don't have a non-awful term to describe this but if
you're like a thought leader type or just someone who likes to share interesting stuff then i think
a very good way to surface that stuff is through rss yeah because how else are you gonna like stay
on top of all that content like if you're just on Twitter and looking at what other people are linking
to,
then you're,
you're already a step behind.
But if you're on the RSS feed,
then you're,
you're right on the source.
That is certainly true.
Cause I mean,
especially if you've been doing a great job of curating various groups or
sites,
you know,
that you can put into folders.
So if you've been doing a great job of like curating great sources over the
years,
whether it's for yourself or for your team,
or if you're like us,
where Jared and I,
like we keep our finger on the pulse on a lot of stuff and,
you know,
he uses RSS far more than I do.
I think though,
Jared,
it's probably more personal for you,
but then that does trickle obviously into the work we do here at changelog.
Oh yeah.
I never really considered that you know using rss more heavily gives me more control over
hearing about the things i might typically hear about i guess more at the source perspectives
versus hearsay or retweets or whatever on twitter and already filtered from somebody else. Yeah. Yeah. So like one, one integration that Feedbin has is a way to post links to Tumblr.
So if you are like authoring a link blog or something like the change blog
used to be,
you could see an article that's interesting in,
in Feedbin and then pull down the Tumblr share sheet and then post that directly to your link
blog all without leaving Feedbin. So you've been doing this for a little while now, right? Like
this is your full-time thing, right? Yeah. You do nothing else. Nobody else is vying for your time.
It's simply Feedbin is what you spend your art on your your creativity is all poured into feed bin is that
right yeah so it's pretty important for you to have a great outlook on where rss is going wouldn't
you say yeah or at least you should yep okay are you leading the witness if that's the case then
i'm kind of curious what your thoughts are on where things are going like earlier i think
may have been in the break i'm not sure sure. I think we, I said we,
we need to have a resurgence because clearly the three of us see some great
utility in this.
Clearly you and Jared are far more than RSS users than I am.
Although I still use it,
just not quite as often.
You know,
I'm kind of curious where RSS is going.
How can we bring this?
How can we create research?
And how can we bring this back to a more personal relationship
with a lot of the indie creators out there? What do you think is happening here? Where are we going?
Yeah, so I do think, unsurprisingly, that RSS is more important than ever, and especially around
taking control of your own newsfeed, because there are people who are out there who are intentionally
misleading you. And the only way to prevent that is to read content from people you trust.
And I think RSS is the best way to do that. That's certainly the truth. Happens a lot in
the news these days. It's a big topic at least politically so i think a personal
relationship with your content is pretty awesome i mean there's often i'm talking to jared he's
like you know i'm and he's a completionist so he's sharing often these stories with me
of how he's going through his rss reader and i'm just envious of that and i just
have been waiting to like get back into that groove and i and i do it i do it kind of in
fits and starts right where i yeah get back into that groove and then I kind of, you know, it's like reading or
writing, you know, like I want to read more often.
I want to write more often, but then it kind of wears off after a bit and I kind of fall
off the horse.
And then I talk to Jared again, Jared's like, oh, it's so awesome.
This horse is so great.
I'm riding this horse with RSS feeders.
My metaphor is dying on me.
I never say that to you.
I never say that to you.
I think the first step with rss is
to is to be passionate about something and once you have that the rest just kind of happens
naturally like really think about like what you're passionate about and what you would like to know
more about and then go go find that stuff and subscribe to it so real quick before we let to
let you go where can people find you at?
What's the URL for Feedbin?
Is it feedbin.com?
That's right.
You were feedbin.me.
Is that right, Jared?
Back in the day?
Feedbin.me back in the day.
There's kind of a funny technical reason
behind that switch, actually,
which is so feedbin.com was not available.
It had already been purchased
and was offered for sale
by some kind of domain
reselling service, you know, like one of the leech type of services. But I was pretty happy with me.
But for some reason, not all top level domains are treated equally. So I started getting bug
reports that people connected on like Wi-Fi services offered through
AT&T at like Starbucks could not connect to Feedbin.
And the issue actually turned out to be that like AT&T could not connect to any like.ME
top level domain.
It was like a really weird thing because I was like, I assume top level domains are one
of those things that always just worked, but that that's not really the case. So I was like, well, I want the
dot com because I'm like old enough that dot com means something to me. And, you know, I have the
money now. So let me just get that and also solve this other technical issue. So, yeah, really long
way to answer your question of, yeah, Feedbin uh feedbin.com there you go did you have
to pay a lot of money for it since they were selling it yeah i think it was like two or three
thousand wow okay i mean that's a lot for an indie i mean it's a lot of money yeah straight
off the bottom line it was meaningful to me it was meaningful enough to do it so yeah it's worth
it i think well ben it's been a pleasure
to have you on the show, man.
Thank you for sharing the story
of Feedbin with us
and potentially the future of RSS.
Maybe the revolution
will be charged by you.
We'll see.
But thanks for coming on, man.
Yeah, thank you guys.
This was a lot of fun.
All right, that wraps up
this episode of The Change Log.
Thanks to our sponsors,
Linode and Hired. Also, thanks to Fastly, The Change Log. Thanks to our sponsors, Linode and Hired.
Also, thanks to Fastly, our bandwidth partner.
Thanks to Jonathan Youngblood for his editing skills on the show.
Great master cylinder for the awesome beats.
And if you're excited about our upcoming show, JS Party,
it goes live on Friday, February 24th
at thechangelog.com slash jsparty.
Check that out.
And thanks for listening.