Wikipedia:Peer review/Computational phylogenetics/archive1
This article is a recently failed GA. Although the review was not terribly...detailed, it makes the good point that the article is very dense and technical. There's also the problem that I'm essentially the only author (before my first edit it was a one-sentence stub), and that's never good for either comprehensiveness or accessibility. Any thoughts from knowledgeable non-experts (or of course anyone familiar with the subject) would be great. (I'm guessing the tree rearrangements are a particularly sticky spot?) Opabinia regalis 03:59, 16 October 2006 (UTC)
- I just made some comments at the article thinking I was still reviewing it for GA. Feel free to move those here if this is more apropriate. --Aranae 05:56, 16 October 2006 (UTC)
- Please see automated peer review suggestions here. Ruhrfisch 03:48, 21 October 2006 (UTC)
- This article races into jargon a bit too quickly. I think the lead of an article should be a description that one could read to a room in a nursing home (or 8th graders) and most of the folks would be able to follow what is going on. You might say what the purpose of CP is and what methods there are to approach the problem. You can then introduce the idea of the different methods (all in one sentence) and then expound on each of them with the detail you have in the lead. Foreshadow what the article is going to say and add a little bit of color. In the second lead paragraph, you mention progressive sequence alignment as a new idea without any context. If there is a common usage of this (for paternity testing?), it might be good to mention this, or even who does this kind of work. --Chrispounds 00:38, 29 October 2006 (UTC)
- Thanks. I added a couple of examples to the lead to clarify what types of problems these techniques are used for, though I don't think the subject will ever be interesting to 8th graders and grandmothers. Although these techniques can be applied to pretty much any data that has a natural tree structure and discrete characteristics, there are relatively few problems that have that form, so there's not much practical application. Paternity testing doesn't have that formulation; you just need a kid and a few putative fathers - I suppose you could try plugging in sequences from a bunch of family members and attempt to reconstruct how they're related, but there probably isn't enough useful variation in the sequences to bother. Opabinia regalis 05:19, 2 November 2006 (UTC)