Hacker News new | past | comments | ask | show | jobs | submit login

I used to use Google Reader, but I found that I missed seeing the original site and that I couldn't easily filter out stories I didn't want to read.

So I built NewsBlur -- http://www.newsblur.com. It's open-source and I'd love if more folks followed my development: http://github.com/samuelclay.

It's been a weekend project but it's been a wild ride so far. One of my main goals is to have much of the functionality of Google Reader, most of which has slowly made its way into NewsBlur. I'm currently working on a free iOS app for NewsBlur. I also put out an API so that others could also make their own feed reader.

But really, I missed seeing the original site more than anything. It looks like a security hole, but NewsBlur actually just proxies the site and displays it back to the user. And the intelligence slider keeps me sane, since I slide it towards green and my unread count comes down to a manageable level.




That's pretty cool. GReader does leave a lot to be desired... I imported my feeds and here are a few initial thoughts (I realize this is a weekend project but it is a huge want of mine):

+ I prefer to see the RSS text before I see the original site. Easier reading (no nav or ads) and I'll click through if the content catches my interest. So it would be nice if I clicked the feed tab that tab would stay selected as I browsed my other feeds (rather than switching back to the default original tab)

+ I use folders in GReader to group my feeds. I have a ton of feeds and a lot of them are not updated often. One of my favorite features of GReader is the ability to click on a folder and see that folder's feed. Saves a ton of individual feed reading and the folder mark as read is a time saver (I read the headlines and don't want to read the individual topics)

+ I need better marked as read (a button). The context menu is nice but requires two clicks (give me a hot key).

+ The config icons in the bottom sidebar get lost behind chrome's stupid appearing status bar making it frustrating to use these when the feed is loading.

+ Minor - I'd like to see the feed list above the site or rss txt body (more like a email client) Is there a way to configure the layout of the panes?

In all this is pretty exciting!


Yup, everything's already there, just a bit disguised:

* Feed before original site? No problem, set Feed as the default in Preferences. Also, if you like one view or another, it saves per-site.

* To read all stories in a folder, just click the River of News icon next to the folder. It's exactly the same. You can also right-click the folder (or click the little arrow) and mark the folder as read.

* Plenty of Keyboard Shortcuts. Type `?`

* I'm working on more customization of the interface as we speak. You'll be able to move panes around, change sort order, etc.


okay this just got a lot better.

I don't see a mark folder as read shortcut (did I miss it). I can context click on the river of news icon, but I'd love a button (and shortcut) next to the "Next Unread" in the dividing bar. I wouldn't text an icon button would do to save space (maybe a preference shut off text in toolbars)

real nice.


That's funny because I was planning to set up on documentcloud this week. This is beyond awesome, and if you do not become insanely rich from it then there is no justice. I had grown to dread the Reader page, and you've made it enjoyable again.


I just tried this, it is awesome! Gratz. and to think I was just about to launch into my very own weekend project for the same reasons :) Now I don't have to, keep up the good work.

One thing, I can't see an obvious way to become a premium member. I think you might get more conversions if it was way more obvious on the front page.


NewsBlur is awesome. I think it's the visual refresh that google reader needs. It also has a public documented API which is awesome. Google readers API users have to cobble together documentation of a private API. After a public API was promised years ago.

But one gripe, when is newsblur going to get google reader sync. As many have mentioned on this thread they like using google reader as the backend sync between many clients. There is already a issue in github.

https://github.com/samuelclay/NewsBlur/issues/19


About two-way sync: it wouldn't make sense. Since NewsBlur hides stories below your intelligence threshold, many stories would be unread in Reader but simply hidden in NewsBlur. That, and the whole story sync problem would be a huge bear. It comes down to NewsBlur being its own reader. The import from Google Reader is the only time NewsBlur talks to Reader.

I'm thinking about further integration with Google using Google Contacts (for a Email Story feature), but that's about it.


Thank you for taking the time to explain that. I can see now why it wouldn't be simple.

This poses a problem for me, and possibly others though. I like to attempt to read almost all of what comes down my funnel. I have built a priority system already inside of google reader. I have folders basicaly 1 - 10, that ranks the importance of things. For example, I can usually read my folder 1 everyday, 100% of it. As I have time I will read through lower folders.

So, with this system I have done part of what newsblur does which is to help me seek the good stuff. This even works across clients. So I can use the google reader web UI, and Reeder.

I would love to be able to use the newsblur web UI in place of the google reader API, but without the sync it wouldn't fit into my system.

I know that my system may not be something you want to cater to, but hopefully you can understand how for some people google reader sync is very important.


Wow! This is a _weekend_ project? It looks very slick! How many weekends if I may ask?


Been working on it for 2 years now. But only as a side-project. I also worked on it on the A train in NYC for a year, about 30 minutes each direction. And I now get to work on it for an hour and a half a day when on Caltrain (SF -> Mountain View).

Check the git history on GitHub. This has been by no means an easy feat, but little by little, you can build a 6-server service (and learn about every single component in the meantime).


You, sir, are an inspiration! /me takes hat off.


Wow... One of the few websites for which I immediately bought a premium account.


Great tool !

but something about seeing the original site. I totally agree you're missing something in a RSS reader. But well, when the connection is slow (my residence is in Bolivia ...), what matters most is the content and not the container, right? I'd rather receive the news quickly than waiting the website logo.


When I first saw this was open source, I downloaded it immediately. Was pretty sure you were gonna take it to become a billionaire, so I wanted to have it as a reference for my own projects. Glad to see I can still get the updates. :)


Hey, not my thing but the way you detect the current post and synchronize the list and content is really slick.

Also, no offense, but you should ask some designer friend to polish the UI because this looks very good already and it can be even better.


I'm a hacker, what can I say? The design I think makes a lot of intuitive sense. The visual layout leaves a little something to be desired. But then again, that's how you know NewsBlur is legit and not some fly-by-night.


Yes, I was talking mostly about visuals. But again, very good job.


Very Impressive! I've been working on a number of 'content extraction' tools - cQuery.com and webXtract (experimental), but maybe you would find them a useful way to extract news and headlines from hundreds of online sources.

Here is an experiment that uses 'profiles' as a basis for content extraction:

http://webxtract.com/content-extractor


If you are interested, feel free to get in touch - anything to help NewsBlur!


If nothing else, Google's recent rework of the way they do apps accounts makes this an instant win for me. It's no longer been possible for me to have my reader open (logged into personal account) and my work email open in the same browser. I'm a new customer for that alone.. and it looks like there are plenty of other things to be excited about here. Thanks!


Unless the administrator of your work gmail has disallowed it, you should be able to stay logged in to both work email and google reader.

http://googlesystem.blogspot.com/2010/08/google-multiple-sig...


Some people at work have had luck with using Chrome private browsing for their home account. Since the cookies are not shared with the 'main' session, you shouldn't run into any conflicts.


I'm a premium user on Newsblur and higly recommend it. I've tried about a dozen RSS readers and it's the one I like best so far.

I'm not a big fan of the original-site-view myself but I suppose it can come in handy sometimes. The addition of the API is quite welcome, opens up a lot of new possibilities.

Anyway, give Samuel some money so he can keep working on Newsblur, it's awesome.


Wow. I am blown away just in the first 10 minutes of using it. It is perfectly designed. You have a good collection of reads when someone comes in, that makes people want to stay on and explore a bit more.

The tiny things w.r.t to the UX are subtle and yet there. And there the option to view the content in the original site design.

You just inspired me again!


Wow, this is pretty excellent. Is there an easy way to subscribe to things similar to how Reader works?


You mean like a bookmarklet? Manage > Goodies has a bunch of browser-specific hacks to make the RSS button go to NewsBlur. There's also a bookmarklet that lets you add a site to a folder (and create a folder if you want) from the site itself.


Hah, now all you're missing is a feature to erase 6 years of muscle memory and habit to always immediately check Reader.


Excellent, thanks, I think that would probably do it. I was thinking specifically of that screen you'll see that lets you choose whether to add something to Reader or your home page.


Very nice! NewsBlur looks way better then Google Reader and I will use it from now on.


I love the concept, but I find it's slow to use in practice. Any plans to preload pages to improve the speed? For example, you could preload upcoming posts in hidden iframes so they're cached locally by the time you visit them.


It depends on the site. It can be slow on ridiculously huge sites (ahem, TechCrunch, Gothamist, Macrumors, etc), but in that case you can just switch them to the Feed view instead of the Original view. The Feed view is instantaneous and it's how you get the speed back on those sites.

It's not ideal, but that means you can still use Original view for the 90% of sites that you still want to see. Plus you can set the Feed view as the default and just switch to the Original view on a per-site basis.


Do you know if there is a way to export all my data from google reader (including stars and likes) and import'em in newsblur? I know google would let me export the list of my subscriptions but what about the rest?


The API does offer an import OPML option, so I assume the website does too. Unfortunately, I've gone and forgotten my newsblur login/password combo, and so can't verify that for you.


Good work there! Just wanted to point out that if you cycle through the feeds quickly on Chrome, the memory footprint keeps increasing and drags the system. Worked fine on IE though.


Didn't cause me any problems Chrome 13.0.782.99 beta, Mac OS X


I am on Chrome 12.0.742.122


I'll give it a go. I'm in the process of setting up a large number of feeds to help filter content for our product websites.


Very nice, I luv it! The one thing I can say I don't like is the blurred logo... gives me a headache :)


Frickin' awesome site. It takes a lot to move folks like me from Google Reader. But you cracked it


Wow....this is EXACTLY what I've been looking for. Awesome work. Followed!


Very, very clean looking site. Progress bars and stuff. Nice job.


this is super AWESOME. Keep up the great work Sam.


Woo, it looks awesome


awesome~ The style is nice.


Fab!




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: