|
|
Thursday, January 10th, 2008
| |
3:53 pm
|
tithonium
|
I haven't had much reaction from the post on my journal, so I thought I'd toss it out here too:
I'm thinking of shutting down sixmore. It's not seeing a lot of traffic, I've got other projects I need to spend time on, nobody seems to be interested in contributing to it.. In short, it seems to be a waste of database and processor time. And while I'm not exactly running short of either, I feel bad leaving it sitting out there to rot.
Thoughts, anyone?
|
|
(7 comments | comment on this)
|
| Thursday, October 18th, 2007
| |
3:33 pm - Power Outage
|
|
| Saturday, October 13th, 2007
| |
9:55 pm
|
tithonium
|
Servers are moved, everything's back to normal. And, as soon as I find some appropriate rails and install a new operating system (and possibly a new hard drive), I've got a dedicated server to move sixmore to. Whee.
|
|
(comment on this)
|
| |
5:52 pm
|
|
| |
12:34 pm - Down time today
|
|
| Wednesday, October 10th, 2007
| |
2:49 pm - Development Priorities
|
tithonium
|
So, here are the current development priorities, in current-priority order:
- Write a privacy policy
The privacy policy currently in place is something I tossed up there around the time the site first went public. We really need something better, though there isn't a whole lot more to say. If y'all have specific things you'd like to see or not see in the policy, feel free to comment. I'm not promising I'll put them in (or leave them out), but your input is - mostly - welcome. If anybody would like to volunteer to actually /write/ something, that would be great too.
- CSS updates & Better browsing/navigation
The site needs a redesign, and bad. The 'user' and 'home' pages still don't share enough code despite having nearly identical layout. The colors and fonts are /terrible/ on Windows (the site was developed from my 17" Powerbook). Lots of links aren't obvious enough, or their purpose is unclear. If you have /specific/ examples of things that need fixing, and particularly if you have specific suggestions of how to make them better, please comment.
- Write Terms of Service
This might be able to be rolled into the privacy policy, but basically we need something that says "this is what we expect you to do here", "this is what we expect you not to do here", and "this is what we'll do if you ignore either of the first two".
- "Leave me alone" flag
It would be good to be able to allow people to say "I never want to sign up, please stop spamming me". And, for that matter, "I've signed up, and I'll check regularly, please don't bother emailing me". A fairly simple flag for users, probably equally simple for non-users. Just need to create a handler and add some columns in the database.
- User-editing for Admins
A good example is the problem of registering no-longer-valid email addresses. There's no way to verify them, other than the admins seeing the verification email bounce. Since so many different mail servers send different messages for a bounce, I really don't want to try to write something to differentiate, say, "mailbox full" bounces from "mailbox doesn't exist" bounces. So it's probably going to remain a manual process. But going into the database to mark something as validated is /too/ manual. So, adding some bits and bobs here and there to make life easier for the not-Marty admins would be nice.
- vCard export
Partly this ties to the social-network-portability thing, but it's a pretty easy thing to do, and might be useful. Esp if we add data types for more of the sort of thing vCards tend to contain (like, say, phone numbers).
- OpenID authentication
You should be able to tie your account to an OpenID login. Simple enough. When signing up, you'll still have to pick a username, I think, and you'll probably have a local password, even if you don't use it. Mainly, it's just an alternative method of logging in; hopefully without the problem of forgetting passwords.
- More AJAX
There are a lot of places now where you have to /go/ to a form, fill it out, and go back. Like adding link types to someone you're already linked to. That could be sped up a bit by doing it inline. Also, adding mass-edit would be good. If you switch jobs, link-decay helps with the 'Works With' -> 'Worked With', but you still have to do it for /every single one/. Would be easier to be able to select a bunch of people and apply the same edit to all of them. Not that this necessarily implies AJAX, but that's a good way to make it more usable.
- vCard Import & Social Network Portability
I'm not really interested in the walled-garden model of website. It's just easier to write. But there's no reason we can't add APIs to make it easy to get your data in and out and share it across other sites. Your linked-in connections should show up in sixmore. Your sixmore connections should show up in facebook. Ok, neither of those is likely to happen any time soon, but we should still lay the groundwork for it.
- Advertising
Definitely not a high priority thing for me. We get enough traffic for right now, and I don't even want to think about making an active attempt to increase it until I move the site onto a dedicated server. That said, it's something to keep in mind. The definition of success on this project is pretty vague, but if nobody uses it, I don't think it qualifies.
- Financial Sustainability
Sixmore is run by just the two of us right now, on my personal hardware, on my personal T1 line. All of this is paid for out of our pockets. Even if I pushed harder for donations, that's not likely to cut it. Now, I'm happy to keep paying for it as long as I can afford to, and there's no sign of that being a problem any time soon. But, if the site starts to draw a lot of traffic, it's likely to start taking up a non-trivial portion of our bandwidth. At some point, I might need to add another T1, or move it to a colo. Not cheap. Again, we're doing fine right now, but it's still something to be kept in mind. At some point, I might need to find a way to make the site pay for itself. I'm not looking for this to make me rich - I wouldn't be /against/ that, mind you - but at least breaking even would be good. Cover a proportional chunk of the T1, cover the electricity, maybe even cover some of my time, or that of future admins.
Unfortunately, I don't see a lot of ways to do that that I'm happy with. I really HATE advertising on websites. Hate hate hate hate. With a fiery burning passion. I hate the idea of nagging and whining for donations. And I really can't see any possibility of getting any traction on any sort of pay-for-features or other service-level scheme. So, if you've got any bright ideas, let me know.
- OpenID provider
Well, if we're going to take it as auth, there's no reason we couldn't provide it as well. Not that there aren't /quite/ enough providers out there already. But, whatever. It's on the list, if somebody gets bored enough to implement it.
So, that's it for the moment. I'm sure there's all sorts of other stuff missing, and some of these can definitely be broken down into smaller pieces. But, this is where my thinking, though not much of my time, is at the moment.
In tangential news: we're in the very slow process of moving the server farm out of the dining room and into the basement. I'm running into a few problems with moving power from point A to point B, particularly with the big 240V servers, but things are getting sorted out. I might actually /have/ a dedicated server to move sixmore onto within a month or so, and a new - faster, better - database server as well. We'll see how it goes.
|
|
(1 comment | comment on this)
|
| Wednesday, October 3rd, 2007
| |
11:09 pm
|
tithonium
|
Shouldn't "Used to trust" have a more-negative score than "Distrusts"? I can distrust someone I never met, but no longer trusting someone I used to means they /lost/ trust. Which seems, at the moment, to be worse. Thoughts?
|
|
(4 comments | comment on this)
|
| |
9:46 pm - New relationship type: Would like to know better
|
loree
|
Description: You know them now, but you want more. For whatever value of "more" that is appropriate. Or inappropriate.
|
|
(1 comment | comment on this)
|
| |
3:48 pm - To Do List
|
|
| Tuesday, October 2nd, 2007
| |
9:11 pm - New relationship type: My Favorite!
|
loree
|
Description: They're the bestest, most awesomest, nifty-keenest of people. In short, they're your favorite. (You can have more than one.)
|
|
(11 comments | comment on this)
|
| |
10:10 am - Moving forward
|
tithonium
|
So.. the site is back. It wasn't exactly cutting edge when I wrote it. Write a social networking site over a weekend, it's not going to be the slickest thing ever, you know? And then it was dead for a year. It's feeling a little dated.
It IS gratifying to see so many people coming back and signing up. But there were a lot of things I wanted to do to make it better, and so many things have happened in the outside world that I'd really like to apply to it.
At the same time, I've got a full time job that, while relatively low stress and fun most of the time, is still sucking my brain dry. I come home and I /don't/ want to code. I've got a dozen projects, both hardware and software, that are languishing because I just don't want to work on them by the time I get home.
I'd love to take sixmore to the next level or two. Improve the UI. Let people change the color scheme (even /I/ am sick of those particular shades of purple). Make it easier to navigate people and link types and relationships and networks. Add more visualization tools so you can SEE your network. Add in APIs and web services so you can get your data in and out more efficiently, and share that data with other networks. Add openID so you don't have to remember Yet Another Damned Password.
But I can't do it all in anything like a reasonable amount of time. So, what I'm saying here is, I need help.
I really don't want to make sixmore into a commercial enterprise. I don't want to charge money for it, and I'm really averse to putting ads on it. I hate ads. I'm sure you do to. The site doesn't pay for itself. Not remotely. It's on a network shared with other sites; it's on a server shared with other sites. But it uses more bandwidth and processor and storage than all but one of those others; it's not free to run. Donations help offset that a bit, tho we never got very many of those. But that's ok. I'm not doing this for the money, and I'm happy to pay for it out of my own pocket as long as I can afford to.
But what this means is that I can't pay people to work on it. I thought of putting up one of those pages, "here's a list of features, and the reward we'll pay for it", but.. I can't pay a reward. I have, off the top of my head, nothing to give people who do work for the site. Well, ok, that might not be true. I have a system to flag people as 'developers', and we could work up some kind of badge for it that shows up on your user page. But that's about it.
Long story short, I'm asking for two things:
a) Ideas for rewards that I /could/ give people, that would actually be sufficient to encourage them to do development work for the site.
b) People willing to do work for the site, for free or whatever reward we can come up with. I'm happy to come up with a list of things we need, but I'm just as happy to have people work on the parts they care about. I'm also happy to explain just WTF I was thinking in various places in the code.
|
|
(8 comments | comment on this)
|
| Thursday, September 27th, 2007
| |
2:49 pm - And We're Back.
|
tithonium
|
Good news: I'm re-opening the site to the public. We haven't recovered the data that was lost, but the database structure is back to where it needs to be.
There's still a lot of work to be done to get all the features we'd like. All of the code is available via anonymous SVN at svn://svn.midgard.org/sixmoredegrees. Please feel free to check it out and play with it. Feel free to email me patches.
"What happened to the site?"
A series of increasingly problematic events. First, a power outage. After the power was restored, the database server wouldn't boot. In the process of installing a new operating system, the database files were lost. It was then discovered that the backups hadn't actually succeeded in nearly 9 months.
So, we lost not only 9 months of users and data, but also all of the database structural changes that had been made in that time. So, I shut the site down until I could at least recreate the database structure. Well, I finally got around to it, and we're back.
edit: Also: if you happen to come across any problems, please let us know.
|
|
(12 comments | comment on this)
|
| Wednesday, September 12th, 2007
| |
9:34 am - Reopening test in progress
|
tithonium
|
I have reconstructed the database /structure/, and reopened sixmore to a small group of testers. When we think we've got all the problems fixed, I'll reopen it to the world. Please stay tuned.
|
|
(3 comments | comment on this)
|
| Monday, January 22nd, 2007
| |
10:40 am - oh yeah
|
|
| Saturday, January 6th, 2007
| |
5:33 pm - So then...
|
tithonium
|
For those of you who haven't been following the drama in my personal journal, we suffered some Problems here overnight, the final upshot of which is that the SixMore database has been rolled back to 9 May 2006. Which means that most of the users are gone, as well as lots of other things, like link types added since then, multi-axis scoring, etc.
So I ask you again: Does anybody actually /care/? At this point, my first impulse is to just shut it down. Anybody have a reasonable argument against that?
|
|
(14 comments | comment on this)
|
| Wednesday, January 3rd, 2007
| |
1:08 pm - How about...
|
tithonium
|
a) Real Name. Required, must be legal name. Private, for internal and/or legal purposes only. (internal purpose might, frex, be customer service related or disambiguation problems)
b) Friends Name. Optional, the name to show to people you have a positive trust score to. If absent, use C.
c) Public Name. Required, the name to show everybody else.
|
|
(6 comments | comment on this)
|
| Saturday, December 16th, 2006
| |
10:53 pm - Nerdvana 2: Electric Boogaloo
|
|
| Friday, December 15th, 2006
| |
12:06 pm - Nerdvana is off the air
|
loree
|
[crossposted to ![[info]](http://p-stat.livejournal.com/img/userinfo.gif) loree and ![[info]](http://p-stat.livejournal.com/img/community.gif) nerdvana]
For those of you who aren't local to us, Seattle has been hit by a massive wind storm which has resulted in power outages across half the city. About a million people are without power in the Puget Sound region, 170,000 in Seattle proper, including us.
As of about 01:30 Pacific time this morning, any services or hardware we're hosting are off the air. We have no ETA on when power will be restored, and chances are we will be more focused on getting heat and hot water back in operation before we look at the server farm.
This outage affects all of our domains and services: Midgard.org, Epiglyph.com, Invitotron.com, sixmoredegrees.com, the Invitotron, MidgardMOO, DownMOO, Cyborg Folly, and probably several dozen other things I'm forgetting.
|
|
(comment on this)
|
| Sunday, October 15th, 2006
| |
12:37 pm - So, um...
|
tithonium
|
Does anybody actually care anymore? Other than switching over to the official domain name, I haven't really touched the code since that whole I'm-quitting-my-job thing back in June. The todo list still has a bunch of stuff on it, but if nobody is actually using the site anymore, I'm not sure I should invest the time. Any thoughts?
|
|
(7 comments | comment on this)
|
| Thursday, July 13th, 2006
| |
9:39 am - SVN access
|
tithonium
|
I've now got an SVN repository, anon readonly, no web interface, set up to hold the sixmore code. If you want to take a look, lemme know and I'll tell you where it is. I'll announce the location publicly later, when I'm sure it's working properly.
|
|
(1 comment | comment on this)
|
|
|
|
|
![[info]](http://stat.livejournal.com/img/community.gif){kind=small}