Goread - Open Source Google Reader clone in Go | Hacker News Reader

Goread - Open Source Google Reader clone in Go

goread.io

196 pointsobilgic69 comments

For me the end of Google Reader meant that I could finally stop using my Google account altogether since Reader was their only service I was using. It's a shame that so many of the Google Reader alternatives want me to sign in with my Google account again. In that case I simply won't use them (I'm looking at you, Feedly).

Well, if you find it worthwhile, feel free to submit a merge request that adds the authentication method of your choice, right?

This attitude is what's wrong with the open source software. It is important to him, but not that important to spend time getting to know code base in a language he might not be familiar with in order to implement it himself. Or he just might not be a programmer.

If you want actual users for your project, not developers, than either implement what they ask for or say you don't want to. If enough people complain, it is a good sign that you are missing important features that prevent adoption. Build it yourself is not an answer to those users, and is just plain annoying.

If you want to be able to answer with "send me a pull request" then just stop creating end users software, and build some library instead that is used by developers.

I'm in no way sure who you replied to was the owner, so your entitlement rage may have been aimed incorrectly.

That said, maybe the software isn't targeted at you? One of the beautiful things about open source is that contributors get to contribute on their own terms. Unlike many commercial products where targeting anyone and everyone willing to purchase it is the goal, having a completely free product allows the author to do away with all of that if they desire.

This isn't the market you are used to. You, as an end user may have very little power. I understand that the complete change of being catered to by commercial products and then ignored in an open source product can be frustrating. Just keep in mind, that frustration is built by your own misunderstanding of your place in the open source ecosystem. The sooner you accept your place, or change it, the better.

As a "potential" end user you have a complete right to complain about the missing feature, as much annoying it can be for the developer of the open source thing. The developer has the same right to just answer "send me the patch". In the end of the day, patch won't be sent, and feature will still be missing. Other users might also miss that feature, and not become users. If that's fine to the project authors, it's also fine for me. Just wanted to say, maybe not that clearly, that "send me the patch" is perfectly appropriate answer to developer, and not a good answer to the end user. And you can't just assume that everyone complaining about the feature is the developer.

That's a good attitude to have. I just wanted to point out that open source projects don't have to follow the normal market economics of regular, for sale software, so the developer/user relationship can be very, very different. There seems to be friction when the users don't notice this, and think they are still being courted as possible sales.

Entitlement? That's abusing the term. Users are within their right to ask for features, and within their rights to throw their hands up in the air and say "WTF" when a dev turns around and snaps "Well make it yourself"

After all, go read never said "The is MY rss reader"

it's tagline is "this is an rss reader." rss readers normally don't require single sign on.

The GP was extending a specific instance to open source projects as a whole, and making a statement about common problems with them. My comment was in response to that.

Narrowing my argument to a single specific case is unconstructive. I have no interest in speaking for the author of this software. They should be capable of doing that themselves.

Users are well within their rights to do just about anything, including insult the developers mother. The issue isn't what their rights are, the issue is with etiquette.

Politely requesting features from developers you are not paying is not a breach of etiquette. Saying "No." in response to those requests is not a breach of etiquette. Making demands of developers that you are not paying is a breach of etiquette.

>Just keep in mind, that frustration is built by your own misunderstanding of your place in the open source ecosystem

And what place is that? A place from which asking for features is inappropriate?

You know why I hate the "feel free to submit a merge request" kind of response? It's because most of the time it doesn't mean what it says. What it usually really means is: "Either write the feature yourself or shut up!". And knowing that most people do not have the time or the skills to actually add the feature, it boils down to "Shut up!"

And that is not an adequate (first) response, because it is disrespectful and hurts the open source movement.

This seems to be what's wrong with the consumer culture when it comes to open source software. It is a sense of entitlement. If you want a feature, request it. The correct reply from the developer should be "thanks for the feedback, it's been noted, in the mean time you could add it yourself and send me a pull request". Done and done. It's not up to the developer to implement everyone's request. That has nothing to do with "end user software" if so far you are the only one requesting it. If there's no community demand for a feature, feel free to do it yourself. Don't blame the developer for not implementing your feature if 1) there's no community demand outside of yourself and 2) you have no plans to make it happen yourself.

What would be interesting is if issue trackers allowed users to pay actual money for features. Many times I find myself saying "I care strongly enough about ____ that I'm willing to pay ____ but it's not worth my time to read the codebase of ____ and fix it myself"

It's been done a few times. Bounty Source is probably one of the more successful ones that comes to mind.

https://www.gittip.com This is sort of their M.O.

How hard can it be to add Persona?

That depends, do you want a site which can be used without Javascript?

In this case, you really couldn't use the site/application without JS anyway.

I have the exact opposite reaction: a site that allows me to use my google account and not create another one will get me to sign up more readily. Adding persona support is being considered.

I have to agree, so long as the permissions are limited... I don't like what a lot of sites/apps ask as default permissions... they tend to seriously overreach... as long as it's essentially for authentication, and maybe my verified email address, I'm okay with that.

What do you use as email?

Not the OP, but I self-host. I do have a Google account, I don't like using it for SSO. I don't use many Google properties nowadays -- Maps seems to be the biggie.

Website is powered by Go and AngularJS. Source code is available on Github [0]

[0] : https://github.com/mjibson/goread

Thanks. I was wondering why it being powered by Go would be relevant when there appeared to be no way to self-host or contribute...

This is the most approaching clone of Google Reader I've seen so far.

It has one big problem, however:

A feed can have only one tag (or, if you see a tag as a folder, a feed can be only in one folder).

This is a big problem because Google Reader's UI encouraged to add multiple tags on a feed (you didn't moved a feed from one folder to an other : you selected a list of tags the feed appear in). And the OPML format allows this too.

So, if you import an OPML export, half your feeds may appear to be missing.

---

An other great Google Reader alternative is http://yoleoreader.com/ . Apart from its UI sometimes freezing for a second (maybe due to synchronous I/O or too long script execution), it's an other great Google-Reader-like RSS reader.

I personally use https://www.inoreader.com/ and found it to be a simple and brilliant Reader-like replacement.

I've been enjoying https://digg.com/reader as a "close-enough to Google Reader" substitute, as well. They launched a mobile-browser-friendly update a few days ago, and the Android app is supposedly "very soon" to release.

Same here. And most important to me, it implements the Google Reader API, so my filtering scripts continue to work.

Big is relative, of course. I don't think I ever used that feature of Google Reader, and it's unclear how many others depended on it as well.

That being said, of course, it's a great feature to have.

I'm rather happy with http://bazqux.com/

Granted, it's a one man shop, but he's charging $9-$29/year, so hopefully enough to keep it running. It's written in Haskell and Ur/Web, is quite fast, took my Google Reader takeout with no issues (400 feeds scattered across 35 folders) Allows login via Twitter, Facebook, Google or OpenID.

Very cool to see my co-worker's pet project front and center on HN when I wake up :) Matt's also working on an Android app for Goread at the moment, which is also open source: https://github.com/mjibson/goread-android/

Do you have to self host? If not, how are they going to keep it around once more people start to use it? Looks nice and clean. Thank you!

It's hosted on Google App Engine. He has ads on the free version and an ad-free subscription option. You can self host if you want and it would probably stay under the free tier on AppEngine if it's just you (depending on the number of feeds and update frequency)

You can either pay or self host (which is quite simple).

I switched to Goread.io as my full-time Reader replacement since I first heard about it after the announcement of Reader's demise. It's been a decent alternative, but it lacks the polish that Reader had, especially when it comes to read/unread stories.

After weeks of use, I haven't bothered to figure out the consistency of how to mark a story as read, so sometimes it will take frequent clicks on "Mark all as read" to register that I've read something. When you click a story, it awkwardly jumps your scroll position around.

These are minor complaints, and it's still very usable otherwise, but I'm thinking about upgrading my subscription to encourage development.

The jumping around got fixed last week. No one has reported the read/unread issues you've had. Could you report them to the github issues page? https://github.com/mjibson/goread/issues

Well... since everyone are sharing their readers, I'll share mine: http://theoldreader.com/

The one feature that most others don't have, sharing and commenting with friends.

On the other hand this one is closing its registrations, and deleting users registered after some date X.

They were talking about letting unpaid customers go, but they found backing, to let them keep up the good work: http://blog.theoldreader.com/post/57274499607/the-new-old-re...

It's also really slow, can't seem to keep track of my read items, and has been down an awful lot the past month. I had high hopes, but it's close to the point where I'm starting to look for something else. Again.

Looks like Matt Jibson is on a front page roll.

Yeah. I was surprised to see these since they are both old news.

I tried to import my feed.ly feeds to this but the page stuck on "OPML import is happening. It can take a minute. Don't reorganize your feeds until it's completed importing. Refresh to see its progress." and I got no errors in the javascript console. It looks promising though. I find feed.ly to be too gimmicky and noisy. I like the minimal style of go read.

One thing that's missing from every google reader replacement I've tried is the ability to infinite scroll through a feed history. Somehow google knew how to request RSS feed pagination parameters.

I made a feed reader called Bulletin that has infinite scroll that works just fine. Can't see why others couldn't do it as well. We also went for minimal on the UI.

bulletin.io

I really wanted to like this but it was missing a couple key features for me. I described by problems in the github issues page and wondered back the theoldreader. When theoldreader got slow and was down for over a day I came back and found he has continued to make fixes and addressed most by my issues. Very happy now, and the website is consistently responsive which I care about much more than fancy layouts.

The only thing I don't understand about this otherwise awesome project is that it being tied to the Google app engine.

That's why it's so awesome. What don't you understand? I get infinite scaling and the datastore. When goread hit the HN front page and the gizmodo front page a few weeks ago, it wasn't any slower: everything scaled perfectly. Consider some of the other readers that have hit the HN front page - they became unusable for a day or two until the traffic died down. App engine is an incredible platform for developing scalable websites.

Yes, that's true when you want to scale. What I'm looking for is something that can run efficiently in my cheap VPS, me being the only user. For that purpose, I'd guess, App engine would be an overkill.

Does someone have a more in-depth description of what exactly happens when I'd log in with a google account there, besides "Google will share your email address with Go Read" and "Go Read may use your email address to personalize your experience on their website"? Or is that really all there is to it?

Yes, this permission only shares your email.

Sharing an email address is not the same as sharing email, is it? At least your reply just confused me more. :)

You're only sharing your email address. Anyway, this is well documented and this level of paranoia is rather ridiculous anyway.

Here is what happens on Go Read's end: https://github.com/mjibson/goread/blob/master/goapp/user.go#...

Here is the information that Go Read stores per user: https://github.com/mjibson/goread/blob/master/goapp/types.go...

Great job. I have been working on a similar Google Reader replacement using Ruby and Ember.

Reader clones are the new Todos.

And written in Go? HN catnip.

I tried fever http://feedafever.com/. Its a nice selfhosting feed reader. But after a while using online feed readers i still use Operas build in Reader.

Matt, this is really kind of fantastic... nicely done! Would love to see a breakdown blog post of how you architected/implemented this... extremely responsive, pleasing UI, no surprises. Just works.

This is really awesome. I've been using this for a few days now and really like it. I've been programming in Go alot these days, so the fact that it's in Go is a big plus for me.

I have been using this as google reader replacement. Very happy with it.

It looks nice and it's FAST! Plus it has a clean mobile UI. I think I'll have a Go and leave Feedly on the shelf, at least for a few days.

Wow goread.io is much more responsive then when I first tried it around a month ago .. definitely will start contributing on Github.

HTTPS URL is https://go-read.appspot.com/

Are the unread RSS feed entries stored in db then called to be read later?

Sort of. goread tracks a global unread date. Anything newer than that is unread. Any item you have read since then is added to a big list that is used to filter returned items. When you've read everything, it moves up the unread date and clears the list.

So far very impressive. Definitely worth $30 a year

Can I set this up on my own server from github?

Yes. https://github.com/mjibson/goread has details.

It's a nice alternative.

Very good ! i also checked http://yoleoreader.com/ which is amazing too.

i'm building my own with nodeJS. As usual my requirement are that it should work without javascript on , so i implemented the static site first ,the rest will come later :

http://feedpress.herokuapp.com/

click on about for the source code.

Reply on news.ycombinator.com