ljreview. Luckily for me, I think the whole system of lj-review is, bluntly, a load of crap. I shall apologise in advance if I offend those people on my friends list who've had journals reviewed - you're on my friends list because I like to read you, after all.
I could start by criticising the ljreview journal itself, since it commits the biggest sin against web-design since the blink-tag, customizing the mouse cursor. Or I could go through the list of people who believe themselves qualified to review the journals of others, and point out the background images that make the text unreadable, the eye-gouging overuse of the colour pink, or the framed styles that scream "My stock anime-girl picture is more important than the words I'm writing!" But I won't, because that's just shooting fish in a barrel.
What annoys me about ljreview is its "review-by-numbers" system. In the fashion of Livejournal's annoying obsession with filling in surveys, each review is a checklist of questions, and points values for each answer. This, of course, means that each review is written to a standard set of criteria. It also means the reviewers don't have to stress themselves writing too much detailed prose.
It also means that 4% of your total result is going to be based on how many userpics you have. 4% on whether you want people randomly ICQ'ing you.
The score-sheet presupposes that certain things make a journal "good". Like having lots of userpics and allowing random strangers to contact you. Assumptions are hidden in the questions. You get 12% for "frequency of posting", so obviously, posting more often is considered to make a journal better than posting less often. Other assumptions are that answering "quizzes and polls" are a requirement for a good journal (although not in excess), that using "customized fonts, colors, links" in your bio is a requirement for a good journal, and that allowing everyone to comment is a requirement for a good journal.
lonita and I were wondering what would happen if reviewers in The Real World used checklists such as this. You could imagine the review of Nirvana's Nevermind. "Can the lead singer carry a tune? (0 out of 6 points) Quality of guitar solos? (2 out of 6 points). Oh and -2 points for the naked baby on the cover. Eww.".
So, say the teeming masses. What would I do better?
- Replace the checklist with a Reviewer's Guide that describes what aspects of a journal the reviewer should look at, but that avoids making any claim that one thing is good, and another is bad.
- Each review should be a couple of paragraphs long, maybe around 500 words, written in prose form, not bullet points.
- The review should close with four marks out of ten. One for userinfo, one for the written content of the journal, one for the visual style of the journal, and one overall mark.
- The overall mark should not be based in any pre-determined way on the scores in the previous three, but should be based on the reviewer's "gut feeling" of how good the journal was.
I give ljreview 3/10.