I upgraded to WordPress 2.7.1, but I had to do it manually for some reason. This is one of the reasons that people prefer things like LJ or social networking sites, because they don’t have to mess around with this kind of thing, and go poking at it when it spits out some error. Of course, unlike WP, LJ doesn’t really seem to update or add new features very frequently.

Was talking to a friend about Dreamwidth and new potential features which would make Dreamwidth more useful to people. She mentioned two.

The first was metadata. Say someone is publishing an online literary journal. On LJ, they would have to tag the post as something like “author: charmian” a it to include the name of the author and the illustrator. But then this would make the list of tags unwieldy, if you have a lot of authors. So what would be better, she suggests if there were a field, like the music or mood field, where you could enter in the name of the author. This would be separate from tags.

The second was a thorough revamping of the tags feature to merge it with the memories feature and allow for post discovery. The memories system as it currently exists, sucks. The tags system isn’t much better. With the memories system, you can categorize posts and save them in your memories.

The potential new system would be like delicious: You can memory posts with multiple tags. What is more, if you should choose to publicize them, others may see them. Well, now they can see your memories, but what I mean is, if they look at the post, they can click a link or something, and see who else has memoried it, and using which tags. Then, they can click on that tag in order to see either which posts the user has also memoried using that tag, or which posts all users have memoried using that tag. (I’m not sure which the default would be).

So, if I have a post about pumpkin pie, other users could label it as “pie” or “pumpkin pie” or “thanksgiving food.” They could use this to search for all tags across the site which are tagged as “pie” or “pumpkin pie.”

The only problem I can foresee is that users may be offended by others labeling posts like “this sucks” or “grade: D.”

Anyway, the people doing the project are really, really busy, so I don’t think this is the best time to suggest the thing. I may write it up later in a post to the ML, or if it is appropriate, post it to the wiki.

Am on the Dreamwidth ML, and the topic of the day seems to be the idea of introducing a setting where posts can be locked to only DW users.

I was against this idea because a) it creates a false sense of security (people are continually using LJ features to solve problems they were not intended to solve, people are definitely not going to think things through and use it as a security measure), and b) DW is going to be an invite-code system, which IMHO means they should err on the side of friendliness to non-DW users, and c) if OpenID users can see the posts anyway, what is the point anyhow?

Since potentially people who want to join DW may NOT be able to join, because of invite code availability issues, this isn’t fair to them.

Also, the problem that some people describe ‘I want my relatives to be able to see my posts on innocuous topics but keep other topics secret’ is one which IMHO, it is extremely risky to solve in this manner. All mom and dad need to do is be all like ‘oooh, if we register we can leave comments easily’ (or your cousin is all like ‘hey, auntie, you can comment more easily if you join the service, let me help you with that’…), and then whoops, big embarrassing situation. I can’t help but think that this is a form of privacy/security, and I agree with what someone said that if something is sold as a ‘privacy/security’ measure, then it really needs to be private.

However, as a robust method of preventing indexing by unscrupulous bots, it does seem to be effective.

I suppose some of the related problems of wanting to restrict access to a post, yet not wanting to, or being unable to grant the other person access could be solved by implementing a ‘password this post’ feature, similar to WordPress’s.

I blogged last time about why people might not use WP vs. LJ on technical grounds, namely that it doesn’t have some of the most important privacy features of LJ. (There is also the icon commenting versatility thing, but so far I haven’t seen anyone really talk that up as a problem, and I’m not sure how important it is to the average LJ user. Before anyone comments, yes, I know you can have as many icons you want when posting on your WP blog thanks to various plugins, but Gravatars and other similar services only allow you one icon when commenting.)

Thing is, the “push” factor (LJ’s various missteps) has obviously not been strong enough. Similarly, there is a lack of a “pull factor,” something to convince people to leave LJ. In other words, it’s the “all my friends are on LJ and so there is no reason for me to leave.” Basically then, until LJ falls down, or someone invents something that is an improvement, technically, based on the existing needs and wants of the LJ users then LJ users have a compelling reason to switch. WP is much more powerful as a blogging software, but much less powerful as a social networking software.

I think that Dreamwidth may be one such alternative, because it is a fork of LJ, which is attempting to preserve the benefits of LJ, and build on them, while fixing LJ’s defects. I’m on the discussion ML for the project, and they seem to be full of ideas of how to better improve the features and interface, both in terms of blogging and social networking.