Wrong feed after GReader import

I used the Google Reader import and it seems NewsBlur picked the _comments feed_ for a few blogs instead of the article feed that I subscribe to in Reader. I just thought, I’d mention it…


This happens when the blog has the Comments feed above the Posts feed. I suppose I should be smarter about it, since I see both feeds and choose the earlier one (when a specific rss url is not given, I go to the site itself and try to detect which rss feed to use). I’m thinking of iterating through them and ignoring one with ‘Comments’ in the title if there are other options. I’ll get this in soon.

Haha, this was the first explanation that I thought of but I instantly refuted it as naive. :slight_smile:

(Doesn’t Reader let you fetch the subscribed feed URL directly, instead of the main website from which you need to figure out the feed?)

Anyway, thanks for answering and no hurry…

Hey, can you do me a favor a write out a few of the URLs that had trouble? I want to see if I can reproduce it, then fix it.



Here ́s more:


Hope you can fix that easily


I’m working on a more general solution that will allow everybody to fix these feeds with no problem. The solution will allow you to:

  1. Change feed link and address and have it lock so it doesn’t revert on you.
  2. See why a feed/page won’t fetch and what there is you can do about it.

This kind of exists, but not really. To tell the truth, I’m working on this before I hit Social features because it’s so important to get this right.

Hmm. This is still happening. I just upgraded to premium yesterday and have been trying to switch over from google reader. I instinctually went to reader this afternoon for my news fix and saw some stories I hadn’t seen in newsblur yet.

I tracked down at least one specific example:

I am subscribed in google reader to:
MacRumors : Mac News and Rumors http://www.macrumors.com/macrumors.xml

Newsblur got:
MacRumors : Mac News and Rumors - iOs Blog

Gah, kind of frustrating. Doesn’t google reader give you access to the actual feed url in the api or opml? Do I have to check all 180 sites by hand?

Reader gives me a URL that may or may not correspond to what the site itself reports. So often NewsBlur has to go to the site and ask what the feed is, and then it reports the iOS blog before the Main Feed url. It’s frustrating, because Reader is giving me poor urls that don’t exactly match what the site is reporting.

What’s your username? I’ll see if I can take a look later today.

It’s CDMoyer.

Well. I’ve got subscriptions accumulated over the past… decade? I suspect that many sites have probably changed their feed systems and whatnot, but I’d rather have whatever feeds is being pulled from, even if that’s not what they are put in the their meta tags.

On the other hand, it’s been cool to see some sites start having a feed again, because their old feed did change.

Obviously, it’s a much bigger task, but it seems like the ideal would be to import it all, then show the user a screen where they can see imported sites that are using a different feed URL than the “get url from site” option would select.

Another Example:

In google reader:
Evernote Blogcast

In Newsblur:
Evernote Blogcast:

Yeah, this is an ideal feature, but it’s a bear to build and it will have a somewhat minimal impact. What would be nicer is if you can right-click on a site and “disambiguate” it and change it to one of the other rss feeds that points to the same url. That’s pretty easy to build. But i’d want to put it in Site Settings, and I’m not sure it would be used there.

Think about this: when a user is new, a few of their sites may be incorrect, but I don’t know which ones. So how do I prompt them, on a per-site basis, to correct the wrong ones?

I guess, my question is, why not just import the actual feed from google reader? Grabbing my opml file, they are right:

It seems like the importer is grabbing the htmlURL and trying to guess what feed I want, rather than just using the xmlUrl.

I mean, am I crazy in expecting that importing from google reader will result in all my feeds being imported? I don’t mean that insultingly. I’m just wondering what the use case is for not doing that. Do people generally have google reader settings filled with broken feeds?

(I guess, as someone trying to switch, I want to be able to be confident that I’m not losing data… but now I know I am, having had to manually fix several URLs after continuing to use google reader in parallel)

If feeds are imported form google reader, and are bad, won’t the user get the exclamation point to know and try and fix them with “fetch url from site” ?

(Sorry to be so annoying… I’m psyched about newsblur, but frustrated by trying to switch… and i have nearly a decade of baggage I’m bring over. ;))

Oh believe me, I try to use the feed that Google Reader provides. It’s much faster to do it that way. But if it doesn’t match up perfectly with what I have, then I have to do investigative work on the feed that often results in finding one of several feeds that could match and deciding on one. I already have it set to discourage Comment feeds when both are available.

This affected me on http://projects.metafilter.com/ which became a subscription to “best of metafilter” instead; and on a few blogs where I got comments feeds instead. I hope I have all of these cleaned up, but it’s hard to tell if the substituted feed is also next-to-no-traffic…