> "There should be a federated, distributed, open-source protocol that will replace Twitter and Facebook and let our data be free!"
> no average user wants that. Really, they don't. If you haven't seen Diaspora's Easy 25 Step Quick Start Guide, you should take a look. Is your mom going to do all that?
I really wish people would stop repeating this straw-man. XMPP is a federated, distributed system, and my mom can use it fine without setting up her own servers. Is it that hard to understand how the same thing would apply to status updates?
I'm surprised nobody in this thread has mentioned SMTP. It's even more widely used than XMPP, and it's a great example of a federated system (as well as the big issue one has to deal with with open, anonymous federated systems: abuse/spam).
Setting up a SMTP (or XMPP) server is not easy, but no "average user" is going to do that. They'll use their ISP, or Gmail, or their work e-mail.
Check out status.net and the most popular instance of it: identi.ca, the web ui isn't pretty but a refresh is coming with the 1.0 release and I use a CLI client to connect to it anyway.
the worst thing about it is that it maintains API compatibility with Twitter.
http://buddycloud.com/ are doing pretty much just that, with added geolocation and stuff. But basically you can use their channel server as a distributed federated Twitter replacement.
I don't know about this. I presume you mean she's using something like Google Talk, built on XMPP? While that's true, I'm not sure it has the same requirements. Twitter works because there's no fragmentation - everyone you care about is in one place. A system of loosely federated systems would lose a lot of the "authoritative source" value it seems. Of course, I could be wrong here, that's only an initial reaction, so feel free to correct me!
(I presume on Google Talk it's only commonly possible to speak to other people on Google Talk? I presume you could change the server config or set up your own XMPP servers, etc, but does anyone?)
No, you can chat with people across servers, thus the federation. Somebody using jabber.org can talk to somebody using gmail.com.
There are even bridge transports to let jabber users talk to ICQ users and Facebook users. It's quite extensible. More here: http://en.wikipedia.org/wiki/Jabber
Hmmm, thanks for the information on that, it makes sense :) It still seems like some of the authoritative aspects of Twitter which provide value would be lost, but I am grateful for the clarification on that.
"Next AOL" sounds like a pretty excellent outcome. I'm not being snarky; I'm suggesting you reexamine the stakeholders priorities. Nerd cred ranks pretty far down the list in most real businesses.
I'm feeling the same disdain about committing to Twitter. I launched http://socialgrapple.com/ a few weeks ago--a social graph analytics product built on the Twitter API, and I'm pulling a bit of revenue already (a dozen paying customers). Now I have two choices: I have some wicked ideas of how to expand further on the Twitter API and provide even more value to customers who have expressed clear demand, or do I try and diversify onto other platforms to avoid getting bulldozed.
I know I can build a better and more valuable service with less effort by continuing to build on the Twitter platform. My developer instinct says "do it" but my business instinct says "forget the opportunity and diversify somehow." Diversifying is tricky because the graph mechanics on Facebook aren't the same, it seems the value of this intelligence is lower. Only other relevant market I can think of is something to do with SEO/web search terms vs domains.
I also know Twitter is building an internal analytics product, based various leaked screenshots, to complement their business model (tracking tweet impressions) but luckily it seems there's no overlap with what I'm building... yet.
I don't mean to hijack this thread but it seems relevant and I would love some feedback.
--
On a side note, I wrote a post in another thread that questions the details about the recent shift in Twitter's spirit towards developers:
Not that you should take my advice, but I'd listen to your customers. It sounds like you have a good product and they just want more features. Being an enterprise app, you're safe from Twitter's growth for the foreseeable future; they still have a lot of consumer holes to fill.
Going the multi-platform route is overrated in my opinion. You'll just become the lowest common denominator between the two. Either become your own platform, or really nail what you're doing on one.
I thought it was interesting. Made it through to the end and pretty much agree with the points you're making. More people should consider the long term effects of building on platforms where they have no control of direction. Everyone knows the sharecropper story, not enough people consider it.
Also, many congratulations on having a wife who sounds like a saint.
I did, it was a great post. You and Paul are an amazing team, and I feel fortunate to know both you guys. Hope I can make it out to SF soon and stop by the office!
Big applause for my cofounder Chad launching his blog. I knew he was up to something.. he was asking me Jekyll questions in Campfire this weekend. Then I woke up at 7 this morning and he was still in Campfire saying he wrote a big post.
His blog is actually hosted by S3 but I'm sure he'll write a post on Jekyll stuff later.
Thanks for writing this Chad, well worth taking 15 minutes to read. As you said, the writing has been on the wall for the past year at least.. and I have to ask, why did you continue to bother?
To me "don't fill holes" says it all. Use Twitter data, but don't be a "Twitter company". If a one sentence description of your company includes the word Twitter, you're in the wrong business. There are many uses of the API that Twitter isn't going to touch. Is Twitter going to build a Zite like app (recently launched news/reader iPad app)? No. Is Twitter going to compete with StockTwits? Not a chance.
Ideas should center around a general need or want. Then see how Twitter, Instagram, other APIs can be used to leverage those ideas.
A couple of weeks later I received an email saying, "Thanks, but no thanks."
Weeks? That's pretty obnoxious.
The lack of feedback is pretty standard. Companies are scared stiff that they'll get sued for something they say, so they try to say as little as possible.
I interviewed with a company in San Jose for a job on Nov 16th last year. And guess when I got a answer back from them? Dec 23rd. Yep, that's right. And not like it was some big corporation or a shady company, its a legit publicly traded $2B up and coming firm. As an active job seeker, it has been a very frustrating past 4 months for me. People don't respond after phone interviews or after interviews. And yet if I make a single mistake I always keep thinking about if that costed me that opportunity. Sorry for the unrelated rant.
Very good writeup, Chad. I too have, apparently, flunked Silicon Valley up to this point. It's really great that you've been relentlessly pursuing it, though.
I had the same reaction about Chirp. I started to get the feeling from Twitter around the time of Chirp.
I was always wary of building apps that are completely dependent on someone else's platform to live and, indeed, my own startup was dealt a serious setback by the release of OpenVBX. So my fears were valid, after all.
My own Twitter usage has changed quite a bit lately, as well. I've given up on keeping up with what's going on. I check it less and less each day it seems. I had an idea this weekend about using small GroupMe groups to share ideas instead of Twitter because I could segment people better that way...
Anyway, It will be interesting to see what Twitter looks like in a few years...
Chad--
Marshall @KE4ZNR here. We actually met up for a few mins at the Durham Tweetup at Bronto. Just wanted to say as a satisfied user of both Notifo and Pushly that any company out west would be lucky to have your skills in their arsenal. I am glad you and Kim are doing well and you still have the support of the Triangle Twitter Community in whatever projects you tackle in the future.
Look forward to seeing what the future holds for you.
All the best from Raleigh.
Marshall @KE4ZNR
Great post Chad. I work for an SF startup and have been trying to sell my house in Texas for 6 months. It's ridiculously frustrating and stressful. Best of luck on your house selling!
Thanks Chad for sharing your story. I remember you vividly from a dinner in Raleigh with Matt Mullenweg, Wayne Sutton, etc. You're a rock star. Code strong.
> no average user wants that. Really, they don't. If you haven't seen Diaspora's Easy 25 Step Quick Start Guide, you should take a look. Is your mom going to do all that?
I really wish people would stop repeating this straw-man. XMPP is a federated, distributed system, and my mom can use it fine without setting up her own servers. Is it that hard to understand how the same thing would apply to status updates?