Hacker Newsnew | comments | show | ask | jobs | submit login
Ask PG: Have your feelings changed about RFS3?
39 points by icey 1571 days ago | 17 comments
Given some of the recent policy changes at Twitter, do you still feel comfortable with the suggestion to build companies on top of it?



Yes, that was the first thing I thought of when I saw the news. I would be pretty cautious about building anything on Twitter after that.

-----


PG, sorry, but I am a little confused by your answer.

Is your "yes" in response to "do you still feel comfortable with the suggestion to build companies on top of it?" Or, to the fact that you would now recommend other companies and startups be "cautious about building anything on Twitter after" the new Twitter TOS and other policy changes there?

Are you removing http://ycombinator.com/rfs3.html from the Y-C suggestions?

My team and I were hacking on our site to launch http://tweeplayer.com at SXSW exactly at the moment the new TOS came out, and it freaked us out at the time. But, the gist I got from people in the know was that as long as you are helping Twitter get content they wouldn't get on their own, they will likely be happy with you.

-----


I meant yes, my feelings have changed about RFS3. I would no longer be so enthusiastic about funding companies built on Twitter.

-----


Twitter is no longer a protocol.

-----


Twitter was never a protocol. SMTP, http, XMPP are protocols - a data exchange description explained in an RFC where anyone is free to do their own implementation of it. Twitter is just a fauxpen API for a private company's massive platform. The idea that no one has created a true microblog protocol (like a realtime USENET) is a testament to the incorporation of innovation and a lock-in economy.

-----


Er. http://identi.ca ?

-----


How would you contrast this opinion to your current thoughts on RFS7 (Facebook Instant Personalization)?

Facebook has shown themselves to be developer-hostile in the past as well, and they represent all of the common risks of building on someone else's platform. Are their boundaries better drawn? Is the magnitude of the opportunity on Facebook larger to the point of justifying the risk?

-----


Sounds obvious now, but I remember advising quite a few people "not build their house on other people's land" several years ago.

It was true then, is true now, and will always be true.

-----


Clickable RFS3 link: http://ycombinator.com/rfs3.html

-----


On the other hand, it might be a good time to launch/fund a product that competes. Without its developer ecosystem, Twitter is practically an unusable product. Sure, they have their network, but so did Myspace, Digg.

Twitter can't reasonably act like Apple and get away with it. It's a very low cost-to-switch environment. If this trend continues, if they make things too hard for devs, the devs will leave, and won't be in a hurry to come back. Might be a perfect time for a lean, developer friendly startup with the right innovation to come in and claim a piece of that market.

-----


I don't agree that Twitter is a very low cost-to-switch environment. Its value only comes with scale - I think it would be difficult for non HN users of Twitter to see enough value to make a switch, at least, not without something new or quite compelling...

-----


I suppose what I meant is that, unlike Apple, there's no cost for a user to start using a competitor. But I suppose switch isn't even really the right word here; there's no reason they even have to stop using Twitter.

Basically, if something else did gain momentum, it could happen pretty fast. As opposed to an iPhone competitor, which Apple can see coming and make strategic moves to plug leaks.

-----


The fact that they're changing policies without much community discussion is certainly scary to any twitter-based company. Luckily, it appears that they're primarily opposed to clients, which is only one of the many possible applications people can build on top of Twitter.

-----


I think the general concern is that today it's clients. Tomorrow it will be something else..

-----


Actually, I feel more comfortable now with Twitter after this.

Why? They have come out and said it. We know their direction now and what they intend: don't make another client, but make a /unique way of using twitter itself/

If anything, this is exactly what we need as entrepreneurs. We now have a mandate from the source to take their platform and ecosystem to the next level. If everyone else wants to quake in their boots over this silliness, then I say go for it.

-----


My concern would tend more along the lines of "how long before they change the deal again?"

They have the right to do so, but it makes building on top of their platform unappealing, and really downright dangerous. It's the same reason building on top of iOS is tricky since you can be denied a way to make money off of your work at any moment by a whim from apple.

-----


Yeah, it may seem an odd comparison but the tone and thoughts expressed here resemble community complaints back in 2004 that serious "uber power sellers" made after eBay started really messing around with its pricing and policies too much and too often. Within a couple years, sellers realized they had to diversify away from eBay and AMZN effectively captured a lot of that value by creating the third party platform.

eBay then, now Twitter and any platform with some critical mass still has to be careful about policy changes like this; they can easily create enough animosity to alienate their most innovative users, motivating them to support other platforms and eventually flee entirely. This pattern has played out a number of times elsewhere; Twitter isn't invincible.

If a platform isn't really an open commty standard and shows it doesn't appreciate your contributions why build on them? And why not instead figure out how to organize other disgruntled users to build something better that addresses their most apparent shortcomings?

-----




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: