Learning to love RSS Bandit
Reading thousands of RSS feeds is not easy, as Robert Scoble, the avid reader of sub-1300 feeds, points out in his Weblog. Most of the people are a bit saner and prefer not to deal with hundreds of feeds daily, but some of us are just that crazy. I generally used SharpReader to do most of my reading, but yesterday I have finally re-discovered RSS Bandit.
Issues with SharpReader
What’s wrong with the SharpReader? - you might ask. Well, here are some of the issues, some mysterious, some quite obvious, that I had to fight while trying to read more than 800 feeds in this RSS aggregator.
- Sometimes, for no apparent reason SharpReader just stops updating the feeds, of which you’re not informed, but have to sort of figure out on your own by saying “Wait a minute, it tells me everything is updated, but these News.com are two days old”, which means that the feed has not been updated. Okay, down the feed list, choose News.com, right-click, Update this feed. Nothing happens, except the little updating icon appears by the feed name, without the feed getting ever updated. It’s a mysterious crash somewhere inside SharpReader, which behaves like a zombie, having no knowledge that some threads have died.
- However, if at any point the feed is missing or contains an incompatible character, which invalidates the RSS, SharpReader is sure to inform you about it with big frigging red title in the pane where you expect to read titles, i.e. useful information. Which is not that bad if the breakdowns are occasional and if you have like 5 feeds. However, here’s the case that occurred more than once. I am sitting at work, and my employer’s network connection is down for a minute, or I am at home, and I need to use the phone line, so I disconnect. God forbid that happens while the SharpReader is updating, because instead of 800 useful feed updates you’re getting 800 red messages “Feed could not be retrieved” or something like that, making the window impossible to read and putting you at risk of getting carpal tunnel by pressing the Del button repeatedly.
- Memory consumption is unbelievable.
Good things about SharpReader
- It supports folders, so you can read headlines from the entire folder, instead of browsing feed by feed. I stopped reading individual feeds long time ago, I just read folders now, especially since a bunch of my feeds come from Yahoo! News and Google News requests (there is a cool PHP hack for Google News to RSS conversion that you can download and host on your server).
- There’s internal search (for feeds you’ve downloaded) as well as external (Feedster).
- You can specify the global refresh speed (for me it’s 1 hour) as well as per-feed refresh rate (MSDN blogs and ASP.NET blog feeds are updated every 15 minutes on my desktop, since they’re refreshed more often due to multiple authors involved).
Part of the issue that I suspect is specific to my box, is that while the links from SharpReader used to open in my FireBird 0.7, they somehow do not open anymore in FireFox 0.8 (yeah, they changed the browser name in between the versions, in case you didn’t hear the story). I’ve spent some time trying to fix that problem, but then got fed up and decided to search around, keeping in mind a cool project Dare Obasanjo once started that now evolved to this.
Good things about RSS Bandit
- Folder browsing is there (although the default sorting is descending alphabetic, what gives?)
- Local (via View - Find or Search button) and global (via Search pane) search is supported
- Global as well as local refresh rates are supported.
- The processes do not die mysteriously.
- All the errors are saved into a separate Feed Errors folder, which is definitely nicer than SharpReader’s approach.
- It’s open-source.
- It stores your feeds remotely, as long as you give it the FTP server username/password.
- Post Reply button for those supporting the CommentAPI.
- Blog this using w.bloggar, if you happen to use this Windows GUI client for access to a dozen of blogging back-end engines (like Movable Type and LiveJournal). I am using w.blogger as you could probably learn from my previous posts, haven’t used this feature yet, but it’s rather convenient.
February 22nd, 2004 at 1:49 am #AeiGiio
Naoa Iineae?e a naiai aiaeiycu?iii
Naoa Iineae?e a naiai aiaeiycu?iii aeiaa iienuaaao ie?nu e ieionu Sharpreader, iainiiauaay naie ia?aoia ia RSS Bandit. ? oi?a ?aoee…
February 22nd, 2004 at 12:44 pm #Luke Hutteman
Hi Alex,
Regarding your issues with SharpReader:
* I haven’t had the feed downloads lock up on me since I fixed a bug a couple of versions ago that did sometimes have this unfortunate side-effect. I’m not tracking anywhere near your amount of feeds though (how do you find the time???) so it’s possible there’s still some other bug that only comes out with very large subscription-lists. If you still have SharpReader installed, could you email me your opml or the sharpreader.log file when you’re experiencing this behaviour? this may give me some leads on how to fix this. Thanks!
* No need for carpel tunnel syndrome deleting the red error-items. SharpReader automatically removes them the next time the feed in question is succesfully retrieved.
* I’m aware of the Memory consumption issue and have some ideas on how to potentially fix this. This fix will not be in the next release (due out this week hopefully) though so for now keeping your total number of items limited is your best option to keep memory consumption within bounds. I can see that with 800 feeds this may be a difficult task though…
February 22nd, 2004 at 1:25 pm #Alex Moskalyuk
Hi, Luke, wow, I feel honored:-)
The OPML is here http://www.moskalyuk.com/moskalyuk.opml, the thing is that while I subscribe to 800 feeds, SharpReader allowed me to do folder browsing, so regularly I’d just go through the list, paying attention to headlines that interest me.
Ok, problem solved with red error message. It would still be nicer if they disappeared maybe after 10 minutes or so. Perhaps this behavior should be among the user-selectable settings.
As for the memory consumption, hard for me to give any advice here. It bogged down my office machine, my home machine with 1 GB of RAM seemed to behave fine. But don’t feel completely bad - RSS Bandit is currently taking up 133 MB on my home box as I am typing this.
February 22nd, 2004 at 1:26 pm #Alex Moskalyuk
The URL for OPML feed does not contain comma, it’s
http://www.moskalyuk.com/moskalyuk.opml
February 22nd, 2004 at 2:05 pm #Lockergnome's RSS Resource
Learning to love RSS Bandit
From: Alex Moskalyuk Weblog: “Reading thousands of RSS feeds is not easy, as Robert Scoble, the avid reader of sub-1300 feeds, points out in his Weblog. Most of the people are a bit saner and prefer not to deal with hundreds of feeds daily, but some of…
February 22nd, 2004 at 2:58 pm #David Nash
Hi guys, Like Alex I use Sharpreader with a rather large feed list and have siilar problems regarding lockup so it would appear that there is still a bug somewhere. One feature in Sharpreader that Alex didn’t mention, and that is really excellent, is the grouping of feeds with the same content so that you only need to read them once rather than having the same data over and over again…this only works when the source of the data is properly identified in the feed so it isn’t perfect but it’s still a boon. I also use another free aggregator, Vox Lite. This one has an interesting user interface but is buggy and still in fairly early development. If you shut down Windows without manually closing Vox Lite it tends to corrupt the programme and the stored feed data is wrecked so it isn’t really up to the level that is needed for casual users. The reader does have some interesting ideas as regards using the web wheel to scroll through the feeds and has no delay before a feed is marked as read (annoying in Sharpreader as one can speed read the feed data and then have to wait while the timeout for it to be marked as read expires). My ideal reader (which doesn’t, yet, exist would offer the Vox Lite interface as an alternative and integrate a system based on Spam Bayes whereby one could mark each item one reads out of 10 (0-9 so as to make it one key press) and thus build up a statistical database which could be used to assess new feeds so as to present them filtered according to the users probable interest in the data…basically use the Bayesian filter technique to sort the feeds for the user based on previous user input. Cudos to Luke for offering such good software and fast responce to problems (he’s helped me out in the past too). There’s my wish list and maybe Luke will have a read of it and see if any can be added to Sharpreader. Thanks, David
February 23rd, 2004 at 7:38 pm #Luke Hutteman
“annoying in Sharpreader as one can speed read the feed data and then have to wait while the timeout for it to be marked as read expires”
Undocumented SharpReader feature: If you use the spacebar to jump to the next unread item, it will mark the current item as read even if the timeout has not yet expired.
March 3rd, 2004 at 4:55 pm #Datagrid Girl
re: RSS Reader Review: IntraVnews
July 12th, 2004 at 5:28 am #Ebbe Kristensen
I’ve changed to RSS Bandit from SharpReader too. While I do like the UI better, I do not like RSS Bandit’s inability to retrieve RSS from BBC or Wired, something that SharpReader has no problem doing. It is however a known problem so I hope that it will be fixed i due course.