Channel Mob Debugging

Well the Channel Mob seems to have picked up 1 posting, but not the others yet. We are looking into why this is the case. I have contacted the people at that make Mystack, which is the agent we are using for the Channel Mob — as you can see by the comments on this post, they are now working on the issue and hopefully we will have this up and running shortly.

Minding the Planet Channel Mob

Social tagging:

4 Responses to Channel Mob Debugging

  1. Bob Wyman says:

    Sorry, but I didn’t get the email you sent “last week.” Anyway, we’re connected now and I’m trying to figure out what might be the problem. Note: We completely replaced our hardware (i.e. every server, cabinet, etc.) on Sunday and that meant some downtime over the weekend. But, I think we should be running fine now. I’ll take this to email for further discussion. Thanks for letting us know there might be a problem.
    bob wyman

  2. Bob Wyman says:

    I think I’ve fixed the problem… MyStack is a distinct application from It is basically a sample application that demonstrates what one can build on top of the kind of “prospective” matching engine that we provide at Unfortunately, there were changes to some interfaces that I failed to propogate to the MyStack code. The result was that your Channel Mob subscription wasn’t matching all the items that it should have matched. I’ve got MyStack caught up now and things should work much better in the future. Please accept my sincere apologies for this problem.
    bob wyman

  3. Nova Spivack says:

    To Bob Wyman: Great, thanks Bob! Looks better. It still isn’t catching everything though — for example a few postings in my own blog have the link in them and they aren’t showing up yet. How long is the time delay for indexing?

  4. Bob Wyman says:

    Nova, you asked: “a few postings in my own blogs have a link in them and they aren’t showing up yet.”…
    Sorry, we won’t/can’t “catch up” on links that were written in the past. implements a “prospective search” using a matching engine. Thus, we only see events that happen in the future. We only hold any particular entry for the few milliseconds that it takes to match against existing subscriptions. This is unlike a “retrospective search engine” that keeps a history of entries found and can re-inspect that history on demand. Since we’ve already read all the past entries on your site, and the entries on many of the sites of those who may have tried to get into the Channel Mob in the past, we can’t match those entries again. However, we *will* catch any *new* links that are created in the future. Sorry about that…
    You wrote: How long is the time delay for indexing?
    Well, it depends on where you post your entry. For instance, if you post an item on LiveJournal, we’ll typically find and match it within about 25 seconds. That’s because we get an aggregate feed of blog updates from LiveJournal which is updated every 15 seconds. On the other hand, if you create an entry in a blog that pings us at or via, ping-o-matic, etc., then the delay is likely to be on the order of a minute or two. Finally, if you write your entry in a blog that doesn’t ping either us or one of the sites that forwards pings to us, then it can take a few hours to get an entry into our matching engine — assuming that we know your blog exists. (Please visit if you create a new blog that doesn’t ping us.) Once an entry is in our system, it only takes a few milliseconds at most to match it against all the existing subscriptions. (for more on the speed at which we match see:
    bob wyman