First off, I LOVE the idea. I think serial entrepreneurs like myself NEED this and it will grow into a thriving community of benevolent contributors (and hopefully less flaming)

while I think it would be great to see it built from scratch putting it all in the hands of one person seems to be asking too much, then comes the question of who does the upkeep, who fixes the outages etc.

I think Open Source is probably the best way see this project through

Why not just start a git repository on github (private) and start a group on a site like fridge to discuss what's going on/have the beginnings of the community while we all work on it together.

Anyone who forks the project and submits accepted code gets some sort of badge in their profile along with credit for helping and being a "founding member" so to speak.

Open a WePay group for the SG so donations can be accepted for server costs, hardware etc.

I think ultimately this can be a passionate community of people who do lots of good, who knows what can grow out of an idea like this. Let's take the time to do it right and make sure all who come can receive sound advice from seasoned professionals in a manner which is better for all parties

Zach

I think that if you're building a start up community, your goal is to engage talent in computer science across the country/planet. I don't think there is a better way to do that then to start from the very beginning as having it that people build together. Zach has probably the best idea so far.

I think adding advertisements to help reduce the costs along with donations would also be a good start.

I really like Zach's approach. Seems more practical. Use stuff that is freely available to stitch something together. All hosted in a private github repo and build custom stuff - if needed and nothing available exists.

I think this is one of the better suggestions in the thread. Find some sort of free software, throw it on github, accept pull requests of good things, and those who are to be trusted give them access.

You are building a community and being as transparent as possible in this manner will win over a lot of people instantly.

Zach's idea is on point. If you are looking for a committed community, the best way is to get them committed from the start. The best will survive the test, and this will result to a richer community.

Quality precedes quantity

I really think we should build the community first using one of the many tools out there before we start investing time into building something custom. Perhaps we can even use something like basecamp/campfire or talkerapp (I'm willing to pay for the initial costs personally).

The community is the important part but I think something as simple as a Google Group would be enough to prove that it's a community that will be worth being part of. Just my 2c.

Agree with grace 100%.

Also signing NDAs might turn people away. I would try to be as casual and organic with this as possible to start with.

Also agree with grace. A simpler way to create and engage community (Google Groups, Posterous Groups, ...) may be enough for the early alpha stage.

About the software, I'd usually go with the "use existing software" approach. However, with the specs outlined (game mechanics and so on) there's also a big risk on ending creating a big patch "hacking" the existing software.

A third option in between would be to use some kind of framework as Django (or Ruby) with as much already done apps for the functionality as possible. For example, a comparison page for forum appz solutions on Django: http://djangopackages.com/grids/g/forums/

I think the alpha stage option is a good point. Get things moving asap with Google/Posterous Groups, and then turn some attention to a custom site including the desired features.

Use existing tools and as (if) the idea gains traction then consider building tools meeting it's specific needs. There's momentum here, capitalize on it by getting people engaged as opposed to spending a month or two building out a site. You may not have the momentum a month from now!

How about stack exchange? There exists one for onStartups that's pretty good. I'm not suggesting using their community but there's probably a ton we could learn by looking at their questions and answers.

I would recommend Wordpress/Buddypress to start. It's simple to setup, has a ton of plugins, and isn't hard to extend. If not BuddyPress, use something else existing. Building your own software from scratch would be a terrible use of your time given the goals you have for the Startup Guild.

I say focus on the structure of the "program", then take specific votes on the details. Do you want levels of members to roughly correspond with Y-Combinator's classes/timeline? The statuses could be: Applied (need social validation to get it), Level 1 (approximately 1 month into Y-Com), Level 2 (mo 2), Level 3, Alumni.

I definitely say use off the shelf software wherever possible. I also think there needs to be some kind of "voting off the island" type situation. Maybe of a given group that starts, only x-percentage can finish. Part of Y-combinator's appeal is that it selects positively for good companies. If it is completely open, you will loose some of this.

I agree that you will need some kind of identification walls in order to make people comfortable working within it, and to make it special. My company, signnow, has a product notarynow that I would gladly donate our service to solidly identify people, if you like.

No brainer. Open source CMS. They are extensible and rich plugin markets. Current leaders are wordpress, joomla, Dotnet nuke, Umbraco and drupal, but there are many others. Word press has around 13,000 plugins for everything from ecommerce to social networks but all most all them have that.

One consideration is that dotnetnuke and umbraco target the enterprise so can be somewhat heavy weight. More modern CMSs are lighter weight. Orchard.net, for example is a .net mvc3 application whose core is very small and even the basic features are design as external modules.

Existing (Open Source) tools at first for sure. Capture the audience, then extend and provide customized solutions / tools. I like zbruhnke idea as an initial project / mastermind group any one can join to get their feet wet if they dont already know a project or have one of their own they want to work on. But initially to get things started I think it would be best to use existing (open source) tools.

Once the choice of existing or new is made then the choice of technology / tools needs to be made. There are alot more existing solutions that meet the working spec needs in alot of differing technologies. So once the new/existing choice is made you really need to focus on tech (.Net, Python, Ruby, Java, etc) as this choice will help inform the number of potential contributers.

<ThoughsOnConcept> I see this working best kind of how Google did in the early days and how I hear facebook currently works. Developers post project pitches to try and gather other people (Devs, PMs Marketing, etc) to their idea, then working together to make it happen. All members of SG would be under a blanked NDA for pitch ideas and each project once started could have their own. </ThoughsOnConcept>

The mere fact that you ask this question is usually an indicator that you should go with existing solutions. If you had a compelling reason in mind not to use off-the-shelf solutions, you'd not bother asking, and if you don't have a compelling reason in mind, why bother? This is obviously not always true, but is rather a general rule, but it's a good one to keep in mind.

Overall: My vote goes to using pre-existing tools like Google Groups, HN Userbase to get started

If a component of the existing tools requires a paying element, perhaps a very low joining fee could be in order. Taking money always changes how people react of course.

e.g. USD$1 to join --> $400 kitty to spend on a paid-hosted plan

.

.

If you need to build something out, consider:

http://pinaxproject.com/ A python and django OOtB framework

.

or Tiki... which has an awful lot of features that don't require writing code:

http://info.tiki.org/Tiki+Wiki+CMS+Groupware

.

They say:

    * Wikis (like Wikipedia)
    * Forums (like phpBB)
    * Blogs (like WordPress)
    * Articles (like Yahoo News)
    * Image Gallery (like Flickr)
    * Map Server (like Google Maps)
    * Link Directory (like DMOZ)
    * Multilingual (like Babel Fish)
    * Bug Tracker (like Bugzilla)
    * RSS Feeds (like Digg)
    * Free Open Source software (LGPL) 

And much more...

    * all integrated (like no one else)

Don't start from scratch! That lesson is reiterated over and over.

Start as soon as possible with a minimum viable solution, get community growing, and the software solution will create itself based on real needs.

I think off the shelf is probably fine for the most part. Mailing lists, IRC, blogging are all available. Live with the limits they impose (embrace constraints!).

As for contracts and online signing, my current gig is with a startup here in Vancouver (BC) that's doing just that for small companies that might be useful if some people/groups did want to do NDA's or contracting/employment agreements, etc. (Not linking as I don't want to be too promotional ;)

I agree with the current majority. Use off the shelf software to get started.

My company uses an install of Wordpress with the GTD theme (P2 Theme is also an option) for communication - it is basically a twitter/yammer clone. We have 20 people and it has been an awesome tool. Of course you can easily add forums (bbpress) and a ton of other functionality with 10k+ plugins available.

This could be built/ready in 1-2 days.

http://p2theme.com/

At this point, I'd agree with the 'off the shelf' solution that most are supporting here. Either starting with something like Reddit's code base, as suggested, or even the Google groups option could be a good start initially.

If the need/desire to roll your own becomes an issue, Kickstarter could be a good way to look at funding this, seeing as there are already 400+ people wanting to see this happen, who I'm sure would all happily pitch in.

I don't know of any existing systems that offer everything you're looking for. Maybe we can start with something that meets most needs, and build on it? e.g., start with Reddit's code base.

Building from scratch would be less desirable, and risky -- in terms of potentially killing the project before it even launches -- but it would also get you exactly the package you want.

I don't want in here to sound like an echo chamber, but building from scratch is not a good idea when there are so many open-source/free tools available that fill all the current needs. As the community feels itself out the need for specific features will arise and then it's time to look at extending and developing the existing deployment.

As a general principle, I think the software platform should be off-the-shelf if possible - to contain costs and management resources. That said, the platform needs to do everything the group requires without too much tweaking.Maybe we could take the open-source remnants of Google Wave and build a customization from there?

If anybody managed to build something actually usable on top of the carcass of Wave, I would love them forever. That was easily one of my favourite "great, but not right now" ideas of recent times.

As a big fan of Google Wave, I like this idea a lot. Although if I remember correctly it had huge performance issues with waves that had many concurrent users. That might take a little too much tweaking to fix for it to be worth it.

This is great Justin, thanks for getting it together.

I vote for Open Source. I think everyone is champing at the bit to join and share with each other, so getting it off the ground fast is best.

Also, as others have already said, this seems to be a predominantly altruistic group, so funding a better solution later won't be an issue.

The main reason to purpose build is to have total control and address scaling issues that pop up with many out of the box solutions. But you may not need total control or have issues with scaling if you don't build a strong community.

I agree open source is probably the best solution at this stage.

I'll add my 2 cents to the idea of using a pre-existent solution for the main community site.

And my vote goes for using buddy press and extend from it.

As others have said, it would be great to start this community opening up the space to code, design and ideas contributors from the very beginning.

I completely agree with graceyang that the community is the important part. While it would be nice to have a custom tool, that adds months to the startup process full of debating how the tool should be made and what it should do.

Start with BuddyPress or something else off the shelf you can customize later. You can leverage experience from the group on whatever solution you pick and the excitement you've started. I think it is imperative to move quickly.

Def start with existing software.

Even a simple forum (Google group) could work. All the way up to stack exchange site (Q&A). Possibly http://braintrust.io

It is most important to begin developing the community. Many communities have been formed using existing tools. Use them until the need for something new/different becomes apparent.

I'd be happy to evolve http://ongoingfeedback.com, it currently already does private mastermind groups.

I reckon make use of existing software first. Manual curation should be ok initially, if necessary. Automate and build as needed in the future.

For a start up it would be good to build up from existing software like everyone before me has said and tweak around it to your liking.

UPDATE:

The Startup Guild now exists and is Live!

http://bit.ly/e2vAee

Convore?

Convore may be the best solution for right now...it's super easy to organize a public/private group of people, and the daily email digests are perfect.

When There's a clearer idea of the form the SG is going to take, BuddyPress is probably the best off the shelf solution since it's open source (free & constantly being improved) easy to set up (under 15 minutes now that it's a plugin) and easily customized with themes and plugins.

I've been using it for the past few months on client projects and it's been really awesome to work with, as well as easy to use for admins and users.

This was the first idea that popped into my head, but I have not been using it yet. My only worry is that it is still not proven reliable or stood the test of time. Its seems nice and pretty, which is great, but I wouldn't want its growing pains interfering.

A more proven alternative may be simply to use google talk (with history tracking if you want that), google groups, and (gasp) email.