Scienceblogs.com Reader Community

They want to turn ScienceBlogs into a Facebook clone? Not interested. I can see most of the bloggers using most of those features, but the benefits to the readership are not that clear.

In any event, scrutinize the hell out of the privacy policy. Somebody will want to make money off of the profiles and member information. I think that's crap, and it could undercut readership. Even if that's not the plan, unless they handle the policies effectively, they risk losing a lot.

I'm not excited by the prospect of a user registration system, but what I've wanted from ScienceBlogs for some time is a better search facility. The one powered by Google isn't really good enough, because it can't sort results by the date of publication of the post.

Sometimes I read an article on a science news website (such as PhysOrg) and want to know whether any ScienceBloggers have commented on the research. Finding out would be a whole lot easier if I could restrict search results to recently published articles only, and reverse chronological sorting would be the simplest way to achieve this.

The biggest plus for me would be comment-tracking. Privacy issues don't worry me because I'm not planning on putting anything sensitive in my profile. I can't imagine using many of the other features.

I'm just going to repost what I wrote on a different scienceblog:

Registration Profiles
User Pics

As has been shown by the past disasters with typekey and its relatives, scienceblogs has a long history of getting everything wrong when creating a user login system.
I already have about 10 user profiles floating around on the internet, which is way more than I have time or energy to maintain.
It's just more garbage to fill out, which takes away time from reading and commenting.

Now here's my list of recommended features:
(a) There are a few scienceblogs on which preview does not work. Get that fixed.
(b) If you use html character entity references (those thingies allow you to put funny foreign symbols in your post) , and you preview your post, the preview will display your html character entity references as if they are working, but the text it puts in edit window will *not* be what you typed, and thus, will not be what you previewed. It will be dysfunctional garbage. So you see that the preview looks good, and you click "post", and then, the preview code posts something you didn't write, and when you go to look at you post, the html character entity references you spent so much time doing correctly, are all fucked up. This bug has been in scienceblogs forever, and it's high time somebody fixed it.
(c) Although most scienceblogs typically have short threads, the most popular blogs have often very long threads, and these long threads take a ridiculously long time to load. Please, Please, Please make the loading of long threads more efficient.
After writing a comment, and clicking "post", I find that it's often much faster to hit Ctrl-L, and find the url for the original article in my history bar, than it is for scienceblogs to load the page in the normal fashion. This too is a long standing bug which needs to be fixed.
(d) I've run out of patience, but there is a host of other bugs in scienceblogs, most of which have been present since day 1, that should be fixed first, before adding new features.

This will be done by SixApart, not the scienceblogs.com techies. Thus it should work. And registration + use of any of these tools will be optional.