Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

It works for me.

6 (31.6%)

It does not work for me.

1 (5.3%)

I'd be willing to help test it.

6 (31.6%)

I'm not interested in testing it.

1 (5.3%)

I might be willing to help test it.

5 (26.3%)

Total Members Voted: 14

[Angle]

Sounds good.

[Anvilfolk]

The database planning work is in progress. Hopefully we'll have something soonish, and then we can create a bogus database somewhere and start playing with it. Diana should describe it in the doc at some point, hopefully.

In the meantime I am wondering whether it wouldn't be nice to get more people involved. I've mentioned to my research group that this is going ahead as an open-source project, and they seemed interested. I'm waiting for the right time to let them know officially this is happening.

I'm also thinking of giving a small talk/seminar about the entire project here at uni, from SAA (the academic perspective) to Agora (the application perspective), and inviting participation. I feel there might be some really smart people getting interested about the entire idea and hopping on board. You could probably skype in if you're available at the time.

What do y'all think?

[Eagleon]

I'm for additional expertise. I tried to get my sysadmin friend in on the project, but he hates his job too much to have any motivation to contribute, haha. I've got skype - it could be really productive just to get people even peripherally interested in the project, with the page bookmarked and ready to pass along.

The last thing I want is for this project to be bogged down, and I'm starting to see it's somewhat of a problem when I'm proposing things to you guys that aren't necessarily even needed or fitting with established CS. Or worse, trying to reinvent the wheel, and succeeding somewhat but with ass-pulled terminology. This list is quite humbling, and makes me wish I had gotten into school again for the winter, because at the same time it's exciting to see so many that I've worked out in the past through experimentation on my own.

I think working with a dummy database for a while will solidify what knowledge I do have of the process a bit better, but I'm not about to try to work out the model for the protocol and database on my own. I'll still be able to help debug and implement what you guys come up with. Consider me your pet code-monkey intern, because I believe this project could be fantastic if finished.

[Anvilfolk]

I'm for additional expertise. I tried to get my sysadmin friend in on the project, but he hates his job too much to have any motivation to contribute, haha. I've got skype - it could be really productive just to get people even peripherally interested in the project, with the page bookmarked and ready to pass along.

Hahahah, poor sysadmin!

We should definitely try using google hangouts at some point to discuss things more easily, especially if we're working together on a proposal. Bouncing ideas off each other is amazing.

The last thing I want is for this project to be bogged down, and I'm starting to see it's somewhat of a problem when I'm proposing things to you guys that aren't necessarily even needed or fitting with established CS. Or worse, trying to reinvent the wheel, and succeeding somewhat but with ass-pulled terminology. This list is quite humbling, and makes me wish I had gotten into school again for the winter, because at the same time it's exciting to see so many that I've worked out in the past through experimentation on my own.

I don't know most of those patterns myself! Doing so comes with time, from writing code and from pursuing self-improvement while you do so. Heck, if you pick up a book on the subject and read a bit every night, you'll probably know a lot more than I do in a month's time.

Here's my personal take on "established CS", or any other field for that matter. The more you know of it, the less creative you become. You try to pigeon-hole what you want to do into one of the ways you already know of doing stuff, even though it might not fit perfectly.

What you proposed so far over on google docs *is* being used, including the site-server idea which had absolutely not occurred to me. I really think part of the open-source collaboration experience is learning from everyone else, so you should really propose any ideas you have for the comms protocol, or whatever else.

It's just important to realise that every proposal is meant to be very critically analysed and discussed by everyone. They're not personal attacks, nor personal shortcomings, nor "bogging down". It's just how you should do science, research or any piece of reliable and quality software.

Consider me your pet code-monkey intern, because I believe this project could be fantastic if finished.

I absolutely will not, nor should you.

I'm going to stop here, but I'm going to send you a more personal PM that I'd like you to read!

[Eagleon]

Believe me, the fact that the terminology for CS is overdeveloped and more extensive than reasonable is not lost on me. They seem to particularly love their bland alphabet-soup style documentation. Engineering lingo is intrinsically and unavoidably ridiculous, I think 90% for job security and 10% because English is a terrible language for science.

It's one of the reasons I'd rather major in informatics, though I'm not sure yet that it's much better. I think that if you put all the systems and techniques and algorithms developed in software engineering into a graph, and started building connections between them based on papers that utilize those elements, and then start drawing new elements out of those papers, you'll see a mind-boggling amount of redundancy that's been missed from lack of complete understanding by its participants. It wouldn't be perfect, but at the very least it would be visible, literally, to anyone that cared to look. This project would allow very rapid collaboration on something like that.

The same goes for, say, chemistry, which has had an explosion of papers as the cost to work on it goes down, but not much work done to unify their relationships. Manual navigation to references that may be completely unavailable. Review of papers requiring the same navigational effort. All kinds of redundant and repeated efforts that bog down the research process to a standstill, until someone, as you said, steps outside this comfort zone and gets a little creative or (re)forges a path on their own.

It's not that this would eliminate the ability or need to be innovative/review, either - it would just break the papers up so that you could see all of the work and data that's been checked by complete accident in distantly related work that researchers are insulated from through this barrier.
I suppose instead of giving up entirely, I could try harder to avoid using words I don't have the background for. It comes from needing an internal set of terminology to work quickly and self-document. At the same time I do know I've been a little more egotistical with flashy nonsense words than is helpful. Trying to stopper that and just communicate is not my strong point yet, heh.

[Anvilfolk]

I think what usually happens is that as you specialise in academia, you get to know the names and acronyms of people and concepts in your area, but little else. I only know the names of a few researchers for the areas I've dabbled in for a longer time. Don't know anything about other areas :\

It's even worse in the IT industry though. They have acronyms for EVERYTHING...

Also, research is very distributed in nature, and terminology ends up evolving more organically. It's not an organised process at all, so a lot of it is definitely idiosyncratic. But then again, it's impossible to keep up to date on everything being published, so it's hard to merge names.


We should make an effort to link all the concepts we talk about from now on to wikipedia articles. That could help us slowly develop names for everything we mean!

Speaking of which, the names I used for the architecture are completely random and totally not standard, as far as I know. We should totally come up with names that make sense! I'll try to add some suggestions to sane names to the google doc later.

[Anvilfolk]

Trying to set up the webserver at home. Will let you know how it's going. I've got a database created, apache, etc, but I'm having some trouble with port forwarding.

---

Update: stuff is working after all. Apparently the router won't let me access local network machines by their outside name. FINE, ROUTER, I CAN USE ITS FREAKIN' IP.

Currently upgrading the server from squeeze to wheezy. Hopefully it won't break all the things.

Here's the base setup we should get from this:
- Apache
- php
- mysql
- phpmyadmin
- SSH access

I will distribute passwords to those interested in participating in the project.


---

Server upgrade successful! Getting Oracle JDK 7 there as well now, though it's a bit of a manual process. We might have a basic database up today at some point.

[Anvilfolk]

Database has been created! Hopefully we can get together a little documentation on it some other day - it's late!

Also, created a new repository for a Java implementation of the outcome computation, and pushed some initial code in. Might be able to finish it up in the next few days.

If anyone feels like they understand the DB enough, we could really use sample debates to start testing. Perhaps something from the SAA papers?

By the way - more collaborators = more better, so if you're interested, it's a great time to start participating before there's a large codebase!

[Solarspot]

So, if Agora sites are supposed to be able to connect any arbitrary Agora-structured statements to any forum thread, should the servers be referred to using some sort of peer-to-peer protocol?  Or have a client-server connection with one server, then have servers communicate with each other with a common format / protocol?  Representing NYT articles, blogs and everything else not on the host server would be a little tedious if each one had to be added onto a monolithic list of servers / named sites to cooperate with.

That's not exactly what I was thinking of here though. What we should make possible, eventually, is for an argument here on Bay12 to reference or be referenced by something on some other website unrelated to Bay12.

So you want something like the pods from Diaspora? Is that even possible?
How would searches work? Every search by a user makes a request to all pods?

let's go ahead and create the repositories for each component and start a'hackin'

We still need to decide on one language/framework to use.

While I'm engaging in architecture-astronautism, could a network like this be searched from a client-side index?  Something like http://www.yacy.net/en/ that records everything any users sees in every client's own index, then users search their local indexes.  This avoids querying every server ever a googol times a second in the name of searching the network.

PS:  Reading this thread while listening to ambiance music was trippyyyy

[Anvilfolk]

Good points!

I guess yeah, you could definitely see the Agora servers as a P2P network!

Here's the cool thing though. Every argument is annotated by its source, meaning its website. So whenever anyone - client, server, computation node, etc - needs to go fetch information about that argument, it always knows exactly where to go. You don't need a directory to search in!

We will need a smart way to update argument acceptabilities across servers though, and that's pretty much an open question still!

Get involved if you're interested - there's... well... everything left to do

[acetech09]

Watching with interest. Saw this from the beginning, am impressed with how it's come along. I just personally can't see a point to it, but I never really enter in actual online debates so it's out of my usefulness range. If I ever find time in my day I'd be interested in helping out though.

P.S. I really just have to point out that when I see this thread the first two things that pop into mind is the Agera, a swedish hypercar, and Amora, a personal game project I've been working on. I do get the whole root thing though - I'm mainly suggesting that five-letter words capped by A's with a consonant-vowel-consonant center are tried and tested good names.

[Angle]

It's for more than just debates. Think of most bay12 threads - You'll frequently have the conversation going in several different directions at once, but because of the linear structure, it's only really possible to discuss one thing at a time. I'm putting together a graphic to show what I mean. I'll post it here when I'm done.

Edit: Ta Da!


Note, in an actual implementation, there would probably be several main indexes, one for each subject, like we have here on bay12. But for the purpose of this graph I only needed the one. As you can see, there is some debating going on here, but there is also a lot of other communication as well. There are of course a great many things that will have to be determined. For example, what are the best practices for using this forum? Should posts be broken up into many smaller ones, or should they be consolidated? What content is valuable, and what is waste? etc, etc.

[Anvilfolk]

Watching with interest. Saw this from the beginning, am impressed with how it's come along. I just personally can't see a point to it, but I never really enter in actual online debates so it's out of my usefulness range. If I ever find time in my day I'd be interested in helping out though.

Well, project's open-source, so you can just come and play around with it as well. We can definitely set it up so that there's lots of small tasks, and you can tackle one whenever you have time. I'm thinking the GitHub Issue tracker can be used for posting all sorts of small tasks and to-dos Your call! Either way, we'll be here!


The project can serve two purposes. First of all, it helps organise people's thoughts on ongoing discussions. Like Angle posted, the graph really tells people who is talking about what, what each post is responding to, and generally encourages people to be more clear in/ the way they communicate.

Another purpose, which I personally find more interesting, is that it helps you in any decision-making progress. It doesn't need to be a debate, it can be anything where you want to reach a consensus with a group of people. If you use this attack relation to propose things or counter things other people have said, the system can give you how much each proposed thing should be accepted, based on votes and the discussion structure.

I can absolutely imagine this being used in debate teams in high schools, local government, to make company decisions, and just generally all over the interweb, where people care about what they are talking about.

[Anvilfolk]

Development is progressing slowly, but progressing.

We've gotten together a basic infrastructure for the server, which should be run as a daemon, and are now at the part where we are starting to implement communication between client and server. We have a database up to test things, and all looks good.

It'd be great to get some more interested people to participate. We are trying to keep the GitHub repos filled with issues representing features that still need to be added, so hopefully there's always something you can help do.

We tend to meet and organise in the Google Doc chat, linked in the OP. If there's someone around you can join the chat thing and talk to everyone. If you are interesting, please do drop by and say hello!

[Antsan]

I am really very sorry for dropping out. I will continue watching this, of course.