I have a question about the brouhaha surrounding the availability of more user icons. I've read through some of the comment threads in this post, (and the previous one, in which Brad made the announcement of the availability of 10 extra icons for ten dollars a year for paid users) and much of the discussion is about how the increase in the number of user icons is going to effect LJ's bandwidth. Clearly, there is something I don't understand.
This is what I see: when I load my friends page, it has 25 posts on it, and each post has a user icon. Whether my friends are each choosing from among three icons, or ten icons, or fifty icons, each post will still have but one icon on it. One icon=x amount of bandwidth. So how does an increase in the number of icons available have any effect on bandwidth at all, given the fact that LJ must fetch only one icon per post? Granted, some icons may be larger files than other icons, but even with fifty icons per user, that would average out. LJ must still fetch only the maximum of 40K of icon file for each post made. As far as I can see, this means that the option of more icons per user will have no effect on bandwidth. Essentially, this new option allows us to spend ten dollars a year for a maximum of 1600K of additional icon storage space.
I can see how more icons could effect bandwidth usage, but only if LJ currently fetches a given icon only once for multiple posts using the same icon on a given page. With more choices, a user might use more icons in, say, a long thread on a comments page, thus using up more bandwidth. But even that doesn't seem like it would have any huge impact. How many times would that rare long page be accessed? Even now, a paid user could use all ten of their icons in a thread. How many threads go beyond ten comments by each participant? Very few that I've seen. All the angst this change has provoked among some users strikes me as terribly overwrought. Even if the a greater variety of icons that will be employed by some users does increase bandwidth usage, I doubt that there would be more than a modest increase in overall bandwidth usage on the site.
If I'm right, of course, then this new option is going to be quite profitable for LJ. Storage is fairly cheap, and getting cheaper all the time. I'm not begrudging Brad the additional income, even though ten bucks a year for 1600K of storage is very profitable. I know he's unlikely to spend it on hookers and beer -- well, not all of it, anyway -- and will put at least some of it back into improving the site. Still, since a paid account with extra icons is now very close to the price Blogger charges for an account with multiple paid journals ($36.00 a year, last time I checked), I think Brad will have to start rolling out more features to stay competitive. And I'm pretty sure we haven't heard the last of the complaints about the price of extra icons, or of fretting over the strain they place on LJ resources. Ah, LiveJournal: home of drama for over 1/25th of a century!