Community Platforms Built vs Bought Does it Matter?
TL;DR; No.
Building Pros: customization
Cons: takes more work today and later
White label pros: customization
Cons: easy today takes more work later
The end.
As 2020 wraps up, it's pretty clear; the next generation of online social communities will be smaller and more focused than Facebook or even Reddit. The great unraveling of Facebook has begun and I'm all for it.
The platform options are becoming better everyday. It's even easier than launching an old phpBB forum thanks to cloud services. Even vBulletin wasn't so hard to get started with and it's old af.
Today you can make a website and publish it in the time it took to boot a computer up, and wait for a dialup modem to download one image.
Starting a community platform has always been relatively easy or we wouldn't have an internet to use. Nobody would've stuck around. It just took more effort back then.
Before the internet there were only bulletin board systems unless you had access to a government, corporate or university intranet.
It was all distributed and private.
You used a modem to call a computer in a room somewhere operated by the person who set it up, connecting directly was just how it was done.
In the early days it was just people making stuff or posting stuff that most people never gave a shit about.
Some things change and others remain the same 😂
Starting a community platform, chat room or forum from scratch has never really been required. We're all building on someone else's work in one way or another.
Why build a custom platform when so many exist? You could ask that question about everything and never do anything.
I can only tell you why I went this route. I'm not saying it's better, it takes tons of ongoing effort.
For one, it's probably not the brightest idea to start with building something completely custom. So I'm doing this out of foolishness... but I'm cool with that.
I haven't even launched it yet so I haven't actually done much of anything.
I decided to go with a custom built community platform mainly because I want to include unique features for the niche I'm building it for.
It's currently a Java Android app and a web app with a Mongo database and a Node backend. Pretty simple. I use Fresco to load images and Volley for REST bs.
Kotlin? Nah. Go? Nope. Roblox? Maybe next time.
I don't doubt niche specific features are probably easy to develop on some white label platforms. For no-code platforms its probably built in.
But at what cost?
If a solution is easier today that means it might have a trade off later.
Transferring data between platforms five years from now might not be an option.
Adding completely custom features, or just not knowing exactly what happens with community data are potential downfalls of an easy to setup platform.
And the grass is always greener...
A custom built platform has issues too is the understatement of the decade.
For starters the work involved pretty much never ends. The newest platforms have all the features. Things like live video and audio conferences take tons of effort to develop. So the competition is always ahead.
But modern SDKs make it easier and easier. I'm pretty sure Twilio is on a mission to make communication as easy as possible at a cost.
Custom built means OS updates and data security, the same problems facing every app. But then, this is what I do so it's not like I don't learn something along the way.
No-code fans all agree:
"Nobody cares about the platform unless it sucks" so I can rest knowing the same thing applies to me.
Sure WE might care about websites because we're into this stuff. Nobody else really does. It's similar to cars, if you care what the badges mean (because you know what they mean) doesn't make everyone else care 🙄
And that's ok.
What are my goals?
My goal is to start a Mustang and classic Ford car community platform. I don't think saying "starting a community" is the right way to say it because there's no I in community.
Well other than the I.
There's no ME in community 🤦♂️ how cheesy can I get? It's true though.
We already exist. I'm not starting a community, it's already out there. And as a community we've seen it all.
On a regular basis the car community helps each other out, it's cool. I'd like to make a place to hangout and encourage people to help each other off of the major social media networks.
The groups that exist make it clear, people might show up to chat or hangout but we stay for technical help. The niche specific features I'm planning are part of that.
Well that's why I'm going custom. I don't think it's the next Facebook, phpBB or the next vBulletin. It might fail. But it's a similar risk as any other social media group, forum or electronic bulletin board system 😂
cool! i'm buidling one too! we need more!
Right on! I think I've seen it on twitter. When you gonna open it up? The day people quit creating new social media platforms with various names is the day the internet begins to die haha.
we need more. we have folks using it in the
alpha... these are my#yenOGfam — the original folks who validated the concept back in early feb/march. you can read more about that here: commsaas.orgLooks cool man I look forward to signing up. I'm taking just about the opposite approach here starting with coding and work my way towards early users at an mvp stage.
but, you talk to users / pre-community members, right?
Nope! well not talking to them at scale about it, I'm not building an early signup list for it or anything. I'm just winging it for the moment. I told some friends about it locally, nothing organized. It sounds counter intuitive but it's a side project. I do for motorift.
cool. no worries. just trying to understand! good luck!
thanks!
Not strictly on topic here, but considering that there are so many efforts and choices, I'd like to throw a little wrench in those gears as a disruptor and suggest that perhaps it may be the underlying infra and stack that communities are built upon, which is considered one of the more important considerations moving forward, in this era of stolen privacy concerns.
Having said that, I've launched and am currently testing an alpha Cerulean instance.
I'm quite excited. Not really so much a bonafied community platform per sé, yet very reddit-esque in one view type as a microblogging platform. Another view type is more like Twitter.
Strong community like aspects are possible in those environments.
But that's not really my point at all. My interest is two fold...
First - it's FOSS (@8bit saw that coming).
Second, and most importantly, it's Matrix protocol. And that spells all sorts off possibilities - e2e encryption, privacy, devoid of tracking, decentralization, federation, interoperability with other Matrix derived platforms, and even non Matrix platforms, as is currently possible using bridges in current implementations, say, with IRC or Slack or Gitter.
love that it's FOSS.
... and i love that you're into crypto /
defi!Customization is not a pro of white label solutions when compared to ground up purpose built solutions. Customizing within the confines of a cookie cutter hardly ever really yields the intended result or objective.
The last few generations of online social communities, by and large, were and continue to be smaller and more focused than Faceplant, as is the current trend; yet for the next? Well, Faceplant is merely the new MySpace. Something will supplant it too.
Faceplant was really intended for those who think that Faceplant is the Internet, and there's always been surprisingly huge amounts of people who know otherwise - Kinda like what AOL tried to present itself as back in the dialup days you so fondly mention.
I dunno about your choice to go with MongoDB, especially if what you're bootstrapping exceeds your wildest expectations (as most success stories usually do), but I certainly would advise against second guessing yourself when it comes to the value of your current efforts.
Something fresh is usually a good thing.
I hope that helps :)
⛵
.
I appreciate the encouragement!! There are tons of options out there. I don't mind the wrench in the gears. I'm more afraid of the day everyone is so afraid of being offensive people quit expressing ideas.
I wouldn't have been able to afford an IDE and SDK every year to learn to develop software if it weren't for FOSS, so I'm all for it 👍
I hadn't heard of Matrix, it looks cool - a modern IRC? I would think having multiple client apps working with the same servers would leave more potential for privacy issues vs having one trusted client app.
I lean towards closed systems being more secure and secure from tracking. The fewer hands it passes through the better.
One of my biggest goals for what I'm doing was being able to tag a post with a type of content ie is it a "post or question" and with a specific car make model and engine to describe the content being discussed. A generic style communication platform doesn't typically allow for that. Tbh this is really a side project, developing it is the fun part.
I'm not really sure about mongo either lol but I'm using their atlas cloud service which helps with some early issues it had.
Yes that can be a problem for some instances where you want the privacy, so you're absolutely right about information leaking out depending on what platforms you decide to bridge a particular Matrix room with.
Matrix offers the choice during room creation (and later, but this is a one time, one way choice), of e2e encryption, and that's on top of your TLS (SSL) connection to the server.
For example, if you bridge with a chan on Freenode, your not going to be able to convert that room into an encrypted room unless you disconnect from the Freenode chan, even if that channel is invite only and SSL only, because IRC doesn't know how to speak Matrix Protocol.
And that's a potential leak.
Some platforms do however, so you can bridge them to an encrypted Matrix room. Matrix is highly secure out of the box, but for the sake of interpretability it does allow you to ratchet down that level of security.
So you are are given the choice to weigh your options against those considerations :)
Just remember though, you can't unencrypt a room once you have implemented the e2e encryption go for that particular room.
⛵
.
nice! I'll have to check it out further.