Warning: this post consists of discussion of management of queues that matter only to myself, and is therefore extremely unlikely to be interesting to anybody else. Despite which I insist on writing it, because of my excessive fascination with the effects of and management of queues.
Two months ago, when discussing my lists of saved items in Google Reader, I claimed that:
my bet is that I’ll be down to 50 items in another couple of months, and will be down to 10 items in half a year.
Which, two months later, turns out to be false. I’m down to 89 items, and the only reason why the number is under 100 is because I took a sick day today. (My brain is not up to programming, but blogging and blog reading are not so demanding.) The tags that I’m currently using are:
- blog: 4 items. (Would be longer were it not for my last post.)
- commented: 5 items. (A bit longer than normal.)
- flash-game: 8 items.
- long: 6 items.
- podcast: 11 items.
- read: 0 items. (That’s the imperative form of the verb, not the present or past tense.)
- recommendation: 33 items.
- think: 10 items.
- video: 12 items.
The numbers add up; as is normally the case, I don’t have anything double-tagged. (Other than numerical tags used for indexing purposes, which I’m not worrying about here.)
These lists have a different feel from each other. The simplest is “read”: that’s a temporary overflow area, nothing should stay in there for more than a day or so, it’s usually empty. The categories “blog” and “commented” are short-term queues for a specific purpose. (Well, they should be: all the current posts sitting in “blog” are from September. I should do something about that. Hmm, I’ll go and delete one of them right now!) Some of “podcast” fits into the specific, short-term category, too.
There are also queues of stuff that I want to get around to going through, but which is too long to fit into my normal rhythm: “flash-game”, “long”, “video”. (Some of “podcast” fits in here, too.) I’ve been making my way through that: these days, items that take 15 minutes or so to process don’t last in there for very long, but longer stuff can stay there for a while. Even so, the numbers in these categories are dwindling, but it wouldn’t surprise me if it takes a year to go through it all. (Hmm, does that add up? A net removing of one every two weeks? Seems plausible, but hopefully I can do a bit better than that.)
Looking at those three queues, it’s not clear to me that they’re the most useful choice anymore: am I getting anything out of segregating them based on content type? Maybe I should merge them all into “long”? Or maybe have “long” and “medium” categories, where “medium” means “about 15 minutes”? Or maybe I should put items of 15 minutes or less into “read”, and accept that occasionally items will stay in there for up to a week? Hard to say; the flip side is that I do like to have categories that don’t get much larger than about 10 items, so if I have reason to feel that way, then I shouldn’t merge them just yet. So I’ll leave them as-is for now, but I probably will merge them in a few months.
And then there’s “recommendation” and “think”. These, especially the former, are the big sinks. I use “recommendation” for blog posts that mention a book that I might want to read, a CD I might want to buy, etc., while “think” is for things that I’m considering doing: upgrading my WordPress installation, taking Alexander Technique lessons, learning a new way to tie a tie, various random stuff.
Those two queues don’t move very fast at all. My rate of reading books is dreadfully slow these days, for example, and I have other sources of book recommendations than blogs, so it’s a rare month when I knock off more than two or maybe three books of the recommendation list, while I can just as easily add two or three new recommendations back on. Similarly, I buy a new CD every week or two these days, but a lot of those are from artists that I’ve discovered elsewhere or artists that I already know I like and am working through the back catalog of. (Fortunately, or rather “fortunately”, I don’t subscribe to many blogs that give me good music recommendations.) To make things worse, some of the posts contain lists of recommendations where I’m interested in more than one item on the list.
Honestly, I’m not sure if the “recommendation” queue is growing or shrinking these days. And its size rather stands out. I should attack both of those problems by breaking it up into multiple categories; that makes since, after all, since I drain the different sorts of recommendations at different rates. Let me go and do that right now…
Okay, I’m back. It turns out that the oldest 8 saved items all fit into that category, which is further evidence that it was a bit out of control. (It would have been the oldest 10 if I hadn’t just shed a few.) The former “recommendation” category has turned into:
- book: 18 items.
- music: 9 items.
- recommendation: 7 items.
So now I’m left with three decent-sized categories instead of one large one; that’s better. And one of the remaining ones is still the largest category on the list! I should clearly spend more time reading books.
I guess I should make another prediction, if for no other reason than that it will let me continue saying false things. Over the last two months, the list has more than shrunk in half, but I don’t think I’m going to be able to keep that up. I’ll predict that, over the next two months, the list will shrink by 20 items, that I’ll eventually reach a steady state of 20 items in total, and that it will take me a year to reach that steady state.
Post Revisions:
There are no revisions for this post.
Posts
I’m not sure I buy the logic here. Suppose, hypothetically, that the rate at which books you might want to read–heck, it’s a hypothetical, so make it books you are certain you want to read. Suppose these books enter your universe at a rate that exceeds your ability to read them even if you could read continuously without sleep or other breaks.
Clearly, in that hypothetical case, “spend more time reading” is not going to achieve the goal of keeping the queue size under control. But the hypothetical isn’t so far-fetched. I think it’s safe to say that books you would want to read are being produced faster than you can read them, it’s just that technology has conspired to make you aware of them (ignorance was bliss?).
So maybe a better strategy is to find a way to reduce the queue size without reading everything in the queue.
11/5/2007 @ 6:04 pm
Yes, you’re quite right. The advantage of breaking up the queue is that it will give me information as to whether or not I’m draining my book recommendations faster than I’m adding new recommendations. Once I have that information, I can take a few courses of action. If it turns out that my reading isn’t keeping up, I can look at my priorities and either drop recommendations unread or read more. And I’ll probably do some of the former, if for no other reason than, in times when I was reading significantly more books than I am now, those books in turn generated more to-read items, so I’m not confident that reading more books would actually help a lot. (Curse you, bibliographies!)
If it turns out that my reading rate is actually keeping ahead of the entry rate, on the other hand, then I can wait until the queue is down to a smaller level (5 items?) and then lower my mental threshold for adding items to the list. Which I’m not sure is completely impossible: I’ve modified my “visit the library” algorithm to include a pre-step of taking the oldest book or two on this queue, searching the library catalog for them, and putting in a hold or interlibrary loan request before necessary. (And then checking out one or two fewer books when I go to the library in person.) This means that, for better or for worse, my reading is now biased towards books that other people blog about. (So if you think I’d be interested in a book, don’t tell me, just blog about it!) Actually, then, maybe when the queue gets short, I’ll change my library visit algorithm to eliminate that bias? Or set up a special rss feed (using delicious or tada lists or something) for other book recommendations, that I could then subscribe to? So many possibilities, which is why I like thinking/talking about this sort of thing: queue management isn’t a static process at all.
11/6/2007 @ 11:19 am
[…] About two and a half months ago, I had 89 saved items in my feed reader. I noted that I wasn’t shrinking that number as quickly as I expected, and predicted that, two months later, I’d have shrunk the list by a further 20 items. It’s a more than two later; how am I doing? […]
1/27/2008 @ 10:48 am