Well, first of all, I set out a bunch of milestones for Hypothetical, so that I could measure my progress a bit more. Today I knocked M4 down and got two thirds of the way to M5.
M4 was basically the groups section. That is functionality to browse groups, join them, leave them, create new groups, kick people out of groups and all kinds of funky stuff. That's all running now. I'm pretty happy with the way it works as well.
M5 is:
Visiting
People
Files
The people section isn't done yet and I haven't really thrashed the whole visitor thing that much. That needs some real testing ideally as it's potential security risk-o-rama. But I'll get my lovely punters to do that. I've used a declarative system, where the view class for a page decides if it's visitor-friendly. By default all pages are not visitable, so I can tell easily where visitors are able to go and as I add new pages, they're blocked off automatically. I decided to let visitors see the full set of navigation sections, but get a message if they click any they can't visit (e.g. post). The page that results in wandering too far reminds me of playing shareware games as a kid. It's cool.
At points I've felt a bit uninspired by some of the interface design I'm doing. Looking at other social sites and taking cues, Hype is a bit flat sometimes. In places where Yahoo would have 15 million links, I have two. But my main target is to get a working version of the software live ASAP so that I can get people on board and take comments from them. In that respect it often just makes the most sense to copy the old Hypothetical and add a few bells and whistles. After all, the old Hype has stood the test of time and still works.
Then there are some other bits where I've had no mould. The groups section is an example. I took cues from Yahoo, Tribes and the work I did for TfL for that and it came out looking pretty good. Good enough that I then spread some of the layout style used there to other parts of the site to liven them up a bit. Some areas of the site are pretty utilitarian in design, until I think of something better. But as it's taking me a moderate amount of time to build, I'm thinking of better stuff as I progress. Here's to incremental improvement!
I've just made the user name at the top of a message into a link to that user's info page as well. It's the little things...
3:49, Thursday April 15th, 2004 • feeling perplexed • no comments
I think social software ain't all that or a bag of potato chips. Allegedly the toast of ETCON and, like, the biggest thing ever, I'm not totally convinced.
Social software is software, mostly websites, that allow people to hang out and stuff. It's the first round that hasn't convinced me: tribes.net, Friendster and so on. In reality, the term is ridiculously broad, covering loads of different potential application areas. It's not even something new either. I guess it's just a camel's back accident, suddenly it's moved out of the core and gone overground within the startups. I consider myself just outside of that sphere, so the fact that I'm talking about it now says something. That I developed a stoopid simple piece of social software three years ago is a random connection really. That I'm rewriting it now with some urgency says more.
I found an interesting blog post via Microsoft Research of all places, which discusses some of the problems. What use are systems which provide tools to communicate with everybody you knew ever? It's like the Christmas mail-merge newsletter, or the village notice board that nobody reads. I don't believe people can really have 300 friends. Rael Dornfest has 115 people in his AIM buddy list, how often does he communicate with each of them?
Hanging out on Hype for so long nips any potential desire to join a friend site like tribes in the bud immediately. I have have joined tribes, just to see what's up, and I'm not really impressed to be honest. Not that I've really given it much time. In my limited bit of exploration, I failed to see anything that really puts it head and shoulders above your run -of-the-mill forum site. Particularly, it seems that the main axis of group activity centers around subjects again. I haven't built my own tribe, maybe I should just to see. Everything else up there is on my desktop - email, calendar and so on. I already have ways of controlling and sharing those content blobs and they're things that not really all that social softwarey IMHO.
My thing with Hype at the moment is that the axis of group activity should not be a specific subject, but just group background noise. There's enough of that and that's what makes me use Hype. I get a regular feed of hacking, news, links, games, events, blog-like experience entries and many other types of content. I don't actually visit many other blog like sites on a regular basis, Slashdot and recently boingboing, but that's it. I've no need to stray outside my group.
So I think everybody will be up for that. I'm hoping to attract people who aren't necessarily so net literate. I think a lot of heavy net users have shoehorned their community fix into comments on blogs, forums and other similar but slightly wrong-headed systems. People who are new will hopefully find Hypothetical easy to use and basically addictive in the way that I do. The test is my 17-year-old sister. If I can get her and her friends to use it, I'm in business.
If you drag a file to an app in OS X, the system will check that that application supports opening that file type. If not, you won't be able to open it. That's great, except that sometimes you need to force it, particularly with raw text files of various kinds. I just found out that by holding down alt you can force it.
Also, I just bound my mouse wheel click to Exposé and the various combinations to shift/command middle-click. Enabling Exposé on the mouse makes sense. Maybe I'll find it fits into my M.O. more naturally now.
<edit date="15.04.2004">Scratch that, doesn't seem to work at all. Wonder how it worked that once...</edit>
5:48, Monday April 12th, 2004 • feeling reflective • no comments
Well, it's not going quite as quickly as I hoped. I was looking forward to having most of it done by now, but the more I do, the more I realise there is to be done.
On the plus side, the core groups functionality, posting features (replying, editing, removing) and personalisation stuff is all there and pretty solid now as well. I was hoping to have the archive up by now, but this has eluded me. That's tomorrow's job. Then after that there's visitors, file uploads, the news system, a set of pages for browsing groups and tools to create and manage them at times other than first login. I'd like to do a swishy theme manager, but that can probably wait until a later date. Autolinks-like functionality needs to go in as well, even if it's just adding target="_blank" to links for now. Ooh, that can be an option.
So, when I've done most of the stuff in the list above, I want to go live. Which means shoe-horning the 35,000 messages from the current database into the new schema first. All in, I reckon I have five or six days work left to do at least.
