How do I start making my own social bookmarking site, is it worthwhile?

November 16, 2005 7:18 PM

I want to create my own social bookmarking site. I don't know much about coding, but I've got some good ideas. Is this possible? How do I begin? Am I insane/idiotic for even contemplating this?

I've investigated the dozens of sites offering social bookmarking+ (del.icio.us, furl, shadows, myweb 2.0 etc. etc.). They are all very similar, simple variations on a theme.

I have lots of ideas about how this concept can be extended and improved, and if I knew how to do it, I'd be working on it right now, for the greater good of all the net. Unfortunately I know very little about coding.

So where do I go from here? Reveal my ideas and hope someone wants to help? Pay some indian programmers to sort it out for me? Or just forget the whole thing and wait until the big boys (google, yahoo, delicious) get their act together?

Any help/ideas would be greatly appreciated, as I need to know if driving myself nuts thinking about this is a big fat waste of time or not.

posted by MetaMonkey to computers & internet (25 comments total)

Have you considered starting from one of the open source tools that does this like unalog [I use it, I did not create it] and then building out your desired features?
posted by jessamyn at 7:22 PM on November 16, 2005

This sounds like something you could get added to an existing similar project or branch from a framework like XOOPS. There's going to be a tradeoff between letting your ideas out in the wild where more technically adept people can implement them and taking on the burden of learning the tech stuff yourself. You don't have to hand it off completely to whatever your racist stereotype of a softwear developer is and you don't have to become an expert in PHP and SQL, but at some point there has to be an intersection of concept and implementation or it will remain a big fat waste of time.
posted by moift at 7:29 PM on November 16, 2005

I think that the power of social network software is to leverage the social. Not just would you have to roll your own software, or buy some, or extend it, but then you'd need thousands (maybe tens or hundreds of thousands) of users to make it useful. And I think you're too late to this party. All of the early adopters have already gotten themselves wedded to some social bookmark system or other.

So I'd submit a lazy web request. Reveal your idea on your blog. And if it's a good idea, lots of bloggers will find it and pick it up and blog about it. And then hordes of firefox extension writers, Joshua Schacters, or whoever will come out of the woodwork to build your idea. And you'll be famous. Or at least respected. Being famous / respected is not being *paid* of course. But hey, you'll make the world a better place, and that's something to snuggle up with at night (instead of snuggling up to large trashbags filled with benjies!).
posted by zpousman at 7:34 PM on November 16, 2005

Thanks jessamyn

I like the idea of building on the open-source unalog, but I don't know enough about programming. Also, I feel that what I am imagining really needs to be build from the ground up. I could spent the next 6 months learning to code, but I'd rather concentrate on the underlying idea/theory.

For the record I've had a look at ning, and though cool I don't think its right for this project.

I guess my real issue is finding people to collaborate/give me money to pay people to collaborate.
posted by MetaMonkey at 7:37 PM on November 16, 2005

I don't know how much money you're interested in spending, but linkroll is for sale. Buying pre-made would be an easy way to get around your lack of coding ability.
posted by Espy Gillespie at 7:42 PM on November 16, 2005

Thanks for the link to xoops, moift. Looks well worth further investigation. I'm not too worried about giving the idea away, as it still has a long way to go before its robust enough to be worth stealing. Hell, I'm having difficulty even explaining it right now.

Also, please be a little more careful when accusing people of racism. I mentioned indian developers as I am seriously considered hiring one. For the record, the only other person working on this project is a good friend of mine who happens to be indian.
posted by MetaMonkey at 7:43 PM on November 16, 2005

I develop database driven websites and applications in my spare time.. perhaps if the idea piques my interest enough and doesn't sound like it will consume my entire life to work on, I could help you code it...

Drop me an email if you like... steve AT openingbands DOT com...
posted by twiggy at 7:59 PM on November 16, 2005

oh and just to add.. developing websites used to be my full time job.. I moved into business, but still love making web gadgets, soo... yea.
posted by twiggy at 8:00 PM on November 16, 2005

Sorry if I misconstrued, the tone just seemed condescending ("some indian"). Also FYI saying that some of your best friends are X as counter-evidence of racism is pretty much famously trite and meaningless.

On topic, XOOPS has some very strong implementations, has a large community, and it designed to be modular. If your idea is a good one you should have no problem getting it made. Although we're off to a bad start, I'm writing my own modular social framework and if you'd care to comment more on what exactly your idea entails I can probably give you a good idea of its feasibility and development difficulty so you'll know what to expect.
posted by moift at 8:00 PM on November 16, 2005

Just remember to read some Paul Graham. Ideas are all but valueless. You ever see an eBay auction for an idea? Mm-hmm. Execution is critical. Make sure you have the drive to make your idea a success whatever it is, and make sure you assemble people who share that drive (or whom you can similarly motivate).

Regarding outsourcing, it's probably most unsuited to a project like this, where you want a collaboration around an idea and response to customer needs. Offshore coding is best done when you can break a project down to discrete elements and shop out those parts that are best handled by journeyman programmers. Developing an interactive site is a much more flexible, ill-defined process.

You'll wnat to look into becoming some sort of programmer. Frameworks like Ruby on Rails are obscenely easy to implement simple ideas in, although polishing them is probably still as difficult as ever (think fractals here). At the very least you'll be able to have better conversations with your developer team ...
posted by dhartung at 8:04 PM on November 16, 2005

Great response, zpousman. Please allow me address some of your points directly;

“…you'd need thousands (maybe tens or hundreds of thousands) of users to make it useful”

This doesn’t really worry me as this is the foundation of my idea. In fact, I am hoping for many more users than that.

“I think you're too late to this party. All of the early adopters have already gotten themselves wedded to some social bookmark system or other.”

I did worry some about this, but I suspect the same thing could have been said about altavista, or excite, or yahoo, back in the day. I believe there is always room for a better tool.

“Reveal your idea on your blog. And if it's a good idea, lots of bloggers will find it and pick it up and blog about it. And then hordes of firefox extension writers, Joshua Schacters, or whoever will come out of the woodwork to build your idea.”

I’ve had similar thoughts. My problems with this idea are:

1. I’m having great difficulties explaining my idea. I’d hate for the whole net to simply misunderstand and dismiss the concept. That would be pretty demoralizing.

2. It would seem (to me) even worse if it did get picked up, but then implemented badly.

3. My dream is to turn this idea into an income. Though I suspect it could grow real big, I would be very happy just getting a salary for doing something useful, meaningful and worthwhile.

All that said, this is probably the way forward.
posted by MetaMonkey at 8:09 PM on November 16, 2005

Thanks for the interest, twiggy. I've been thinking and I reckon I should knuckle down and write out some sort of overview of the concept for the net. It make take a few days, but I will be sure to contact you when ready.
posted by MetaMonkey at 8:12 PM on November 16, 2005

Thanks also for the heads-up on linkroll, Espy Gillespie, but I really want something a lot more than a del.icio.us clone.
posted by MetaMonkey at 8:14 PM on November 16, 2005

Please forgive the mix-up, moift, I can see how my initial comment was a tad tactless and could easily be misconstrued. I really do abhor racism, and was a little taken aback. Now that we’re friends, I will be sure to give you some more details, as soon as I’ve worked out how to make them intelligible. I would be very grateful for any assistance you could offer.

Oh and regarding your comment,

"If your idea is a good one you should have no problem getting it made"

Could you please elabourate a little on what this means. Will xoops users really work on my idea for free if they like it? How does this typically occur? This would be a dream come true for me.
posted by MetaMonkey at 8:22 PM on November 16, 2005

Good comments, dhartung. I suspect you are right in that outsourcing is not the best solution. I would really like, as you suggest, to build a community/team around this idea. It had seemed a little far-fetched at first (that is with no capital).
But now I realise, as befits a social networking idea, that social/distributed development may well be not only possible but the best/only way to make this happen. Oh, and I’ve read a good deal of Paul Graham. In fact his writing gave me the spur to believe I could actually do this.

You are probably right also that I should at least get myself up to speed on coding so I can talk from some knowledge with devs.
posted by MetaMonkey at 8:30 PM on November 16, 2005

How does this typically occur?

It depends which ground you're breaking.

XOOPS is primarily for forums, but with its modules it does a ton of stuff. Its flexibility comes from the fact that social sites tend to look very similar on the data end... Ie., there's not a lot of difference between storing a forum post or a blog entry or a del.icio.us style link: they all have descriptions, titles, full text, tags, references to parent and child messages, and a few numeric fields for things like ratings, views, etc. The difference comes in the way the data is interpreted and presented to the user.

If your idea means a fundamental change in the basic units of data storage, it will be harder to adapt an existing system than to build from scratch. If the innovation is on the interpretation end (and hence can be built on top of the data management systems we have), and it is good, it shouldn't take more than a forum post on a (or some) community boards associated with a social framework to find some takers with programming skill. You've already got a few here without even giving up any details :/

You can probably tell I'm finding it pretty hard to comment on this intelligently without knowing anything about it, I'm just trying to get across that if this isn't something like a whole new paradigm, much of your work is already done.

And will they really work for free? Well... they'll do it because their community will gain status with innovative modules, they're tired of updating their own stagnant projects that are probably already better implemented elsewhere, they'll get in on the ground floor of a potentially profitable enterprise, and they're looking to wrest credit from you and ride the wave to web 2.0 superstardom as soon as it's practical. Doesn't mean you have to let them or anything, but it probably will be a motivating factor, especially when you start to mention things like making a modest living off it. No one is going to work for "free", but you also don't have to pay. Just think about where the motivation is coming from before you accept seemingly generous gifts from random strangers.
posted by moift at 9:00 PM on November 16, 2005

I asked a slightly similar question a while back. You may want to download and install a few different options to compare features before you roll your own.
posted by lewistate at 9:04 PM on November 16, 2005

Thanks for the detailed response moift. You’ve told me pretty much what I was hoping to discover. FYI, without trying to explain the whole concept, it really isn’t too much of a new paradigm, at least the initial implementation anyway. You’ve certainly given me confidence that the development is well within the bounds of possibility.

I do think the idea is something a lot of people would be interested in using/developing, if I were able to communicate it effectively and develop the concept further.
posted by MetaMonkey at 9:39 PM on November 16, 2005

So how do I get the ball rolling without

a) Someone else stealing the idea and doing it badly
b) The usual troll types savaging it before it even gets going
c) Having the project hijacked and ruined

Any takers? Also, I’m curious if there is any existing precedent/model for a start-up enterprise created across the net, with distributed development.
posted by MetaMonkey at 9:39 PM on November 16, 2005

You are quite right, lewistate, that I need to really research this thing properly before even thinking of starting development. I’m starting to wonder if I shouldn’t have waited until I’d worked a lot of it out on paper before even asking MeFi.
posted by MetaMonkey at 9:44 PM on November 16, 2005

Again, without knowing anything about it, I really doubt distributed development is really required. As soon as you loose it on a community your stake in it is really going to be diluted. Unless this is something that really takes a village (I can't imagine how it could be this difficult, but who knows), I'd recommend finding a contract developer like twiggy or myself (*flutters eyelashes*) and formalizing the terms and stakes before line one is written.

In my suggestion re: XOOPS and other community driven enterprises I was under the assumption that your primary concern was to see this done, and that is probably the best way to just get it done. But, if maintaining control and dibs on the fruits of its future performance is a major concern, I don't think you'll get the allegiance to your authority you're looking for via the distributed development method. It might be possible; I don't anything about how open source enterprises handle finances except that the Apache Group apparently takes junk cars as a valid form of payment. If you're confident in the strength of the idea, consider speaking with a lawyer about how you can keep it from getting away from you.
posted by moift at 12:05 AM on November 17, 2005

*don't know anything
posted by moift at 12:06 AM on November 17, 2005

No offense, but it sounds like you aren't able to articulate your idea very clearly (at least to the point where someone else says, "Brilliant! I'd pay you to work on that!" Doesn't mean it's a bad idea, but you might be better off learning enough about programming to create the "proof of concept" yourself. As dhartung says, ideas are cheap. Proven ideas, though, are gold.

Pick up a copy of "Agile Web Development with Rails" and work through the tutorial. You will be floored (and inspired) by how much you can do quickly and easily. Of course then you'll have to backtrack and get a grounding in Ruby and MySQL before you can implement your own but I think you'll be much more motivated with the end in sight.

An alternate suggestion: try to hook up with students at your local business school. They've probably got a business plan competition. You'll find dozens of bright, talented, motivated people eager to get involved in a startup but absolutely starved for vision.
posted by zanni at 5:07 AM on November 17, 2005

Thanks once more for the info moift. I'm going to give myself a few days to think it all through. I may post some sort of follow up question also, if thats not against some unwritten AskMeFi etiquette.

Thanks for the suggestions, zanni. Proof of concept via Ruby on Rails sounds like a good idea to me. Not sure about the biz kids though, although maybe some comp sci grads would come in handy.

One final question if anyone is still reading this thread:
if this does go anywhere, is it considered bad etiquette to post a link on MeFi proper to any site I set up to explain/develop the concept?
posted by MetaMonkey at 5:42 AM on November 18, 2005

Oops please ignore the last question, I just found projects.metafilter.com
posted by MetaMonkey at 5:52 AM on November 18, 2005