James Governor's Monkchips

Bloglines: on the verge of losing its third redmonk analyst

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

Cote and Stephen have both adopted Google Feedreader. I tried it and wasn’t all that impressed. No really Stephen, I had an open mind, I just didn’t like the UI and I was surprised that the service was as slow as bloglines.

That’s my main issue with bloglines: Too. Darned. Slow.

I realise I am quite unusual in the number of feeds I read, but that’s exactly why I need a high performance solution. I would probably pay for it if you offered a premium service. Most confusing to me is there is an occasional bug, which when it occurs gives bloglines Google search class response times… says database isn’t available but runs as fast as a big male spider in the Autumn looking for a mate.
I think I am going to try some rich clients. Its a shame really, not least because if Bloglines was smarter it would have people writing rich clients to its APIs. Bloglines could have been a contender. Losing three RedMonk people may not seem such a loss, but its indicative of issues for the Ask subsidiary, and we’re certainly not aloneSpeed is a feature.

5 comments

  1. interesting. Google Reader’s given me no performance issues whatsoever, and i’m at 515 feeds (according to Google Reader trends).

  2. James,

    Could you email me your Bloglines User Name, I am wondering how many feeds you have setup?

    If you could also give me specific examples areas you feel are too slow, it would be very helpful.

    For general reading, one of the major areas we take a hit is when you load an entire folder, in which hundreds of items download all at once, and many browsers don’t render it until everything is downloaded. Google Reader definitely does this part better with the progressive downloading of large number of items.

    Thanks,

    Paul Querna
    Bloglines Engineer

  3. about 500. bloglines user name is jgovernor.

    its just time to update feeds, even if only a few need to be read. but no i dont make extensive use of folders.

  4. Hmm. Looking into it quickly, the feed list is taking about 12 seconds to generate. It looks like something in your feed list is causing a bug in our cache layer, and the request isn’t being serviced as quickly as it should be.

  5. James,

    We found a bug in the EPoll system call in Linux used by our APR/apr_memcache layer. We have switched back to using Poll, and you should notice a significant performance increase when our next release is pushed to production. (Currently looks like Thursday of this week, but I can’t promise anything).

    Thanks,

    -Paul

Leave a Reply

Your email address will not be published. Required fields are marked *