The Fourth Facebook Goldrush Just Started | Hacker News Reader

The Fourth Facebook Goldrush Just Started

medium.com

58 pointsjeremyliew46 comments

I think he is more on the mark about streaming video than conversational commerce / bots. If you are a regular Facebook user, you can get an idea of where they are headed with the notifications that get greater priority and what is more likely to appear in your feed.

I have noticed greater visibility of Events and Live Video streaming. When a friend starts a live stream, I get a notification that shows up on my lock screen.

I'd put my money on Facebook going into video in a big way. Maybe they will be the one streaming sports games around the world.

Maybe - though interesting that Twitter just beat them to the NFL rights[0], suggests they have a way to go yet, and a clear fight on their hands.

[0] http://www.businessinsider.com/twitter-beats-facebook-amazon...

I don't know, just by the articles that have been bubbling up on HN lately, the dam seems about to burst on chatbots.

I'm seeing more an more articles about them. Amazon (via Echo/Alexa), Microsoft (via bot framework), and Facebook all seem to think it's about to become big, and I saw an article about China's mobile experience that suggested it's already pretty big over there.

Developers seem to be less interested in it, but businesses seem to be pretty hot on it.

Just the current in vogue snake oil to sell to khaki wearers. Devs aren't interested because they know they don't work.

Yes, you too can be the next startup to place your complete destiny in the hands of an organization that owns the walled garden you are sharecropping, that can watch your every monetizing success, copy it without penalty at any time of their choosing and then cut you out going forward.

You better have completed your "liquidity event" before that happens.

A better attitude is to get in while it's good, and get out in time. This can be big profit.

I agree with what you're saying, but if you don't fool yourself into thinking you're safe, and you have a proper game plan (or simply don't invest too much!) you'll be fine. If you make money, sweet! If not, eh.

Pretty much every startup is forced to work under those conditions. Microsoft and Apple grew up in the shadow of the IBM monopoly, and Apple was almost killed by it. PayPal was a feature of EBay's, with 90+% of their users being EBay sellers. Google and Facebook grew up under the IE monopoly - Microsoft had total control over the channel to their users. Twitter's product was copied by Facebook a couple months after it came out.

Big companies aren't in the habit of leaving virgin territory where new startups can take root, and every startup needs users.

The startups that succeed are the ones who make their userbase passionate enough to resist the initial competition from the big company who they're sharecropping on, long enough to build footholds on other competing platforms that'll give them leverage.

The difference is that in this case Facebook controls the complete platform including access to the customers that your startup requires to survive and prosper.

Paypal had independent access to markets. It just grew fastest via eBay because it solved a problem that eBay initially could not.

Clone PC makers had independent BIOS and MS-DOS so IBM could not snuff them. It finally lost when it tried to move PC users to the PS/2 platform and customers would not follow.

Microsoft negotiated and blocked exclusive licensing and control by IBM of MS-DOS before the first IBM PC shipped.

So yes, you can succeed if you have a way out of the inherent trap of being just another sharecropper. And to do that you better have negotiated a much better deal where you have options that cannot be easily blocked that are not being offered to everyone else. But you cannot expect to win much if you accept the publicly offered terms.

Facebook's terms of use for their platform allow them to cut you off from "your passionate users" at Facebook's whim.

Unless I misunderstand, conversational commerce bots seem to want to solve the problem of the interface between users' natural language and software's logic (i.e., it's an application of NLP).

But I doubt AI will be good enough to do that, and I expect that I'll still have to think about fitting my words to the software's understanding. Why would these bots work better than Siri?

In my mind it's not really solving any 'problem' at all... We have a browser, and we have Google... It's generally pretty easy to buy stuff online, so what's the problem??

I know several people who lack the skills to buy things online. For others it's a chore. Finally, what if by lowering the barriers for customers, you could increase sales 5%? 10%?

Sure, it *might increase conversions - I'd hope that FB have tested this assumption... Will be interesting to see how this turns out

They will work far better than Siri because they have a constrained world to work in. Every call center has scripts for their predefined scenarios. AI chat bots navigate a world that fits in a book, but Siri has to navigate the whole world.

"Conversational commerce" No thank you.

Why? I am looking forward to it. For example now, if I order a hotel, I might book it via phone or email, and then I get a notification email about the order. And... let's say one year later I want to book again - I think it is more convenient if I can do it via chat application, where I would just chat with the hotel, and whats most important, the previous communication is still there in history and I can refer to it - while with email, there is no continuity of my relationship with that hotel.

It's called a browser... Why stay in FB to do this?

I think more than conversational commerce the true power of BOTs will be conversational planning assistance within chat. Imagine a personal assistant always available in your chats with friends that can help you plan dinners by finding restaurants everyone would like, making reservations, booking movie tickets, putting the event on everyone's calendars, sending reminders, giving directions to locations when you leave for the event, etc. now that's powerful and useful.

I have always wanted this, so badly.

Facebook buys MAGIC?

Here's how I envision that use case with a bot called JEEVES.

ME: Hey have you guys watched Zootopia yet?

FRIEND 1: Nah I want to but didn't have the time. I'm free tonight if you want to go.

FRIEND 2: Yeah count me in too

ME: JEEVES find us a movie theater close to us that's playing Zootopia tonight.

JEEVES: There are 3 theaters close to all of you that's playing Zootopia tonight. Theater 1 at (map)Location 1 is playing it at 9PM, Theater 2 at (map)Location 2 is playing it at 8:30PM, Theater 3 at (map)Location 3 is playing it at 10:PM

ME: you guys wanna get dinner before the movie?

FRIEND 1: Sure

FRIEND 2: Nah having dinner with GF's parents today.

ME: JEEVES find us some chinese restaurants near Theater 3.

so on and so on.

JEEVES can then book tickets, make a reservation, put the event on calendars, send restaurant location to just the 2 people who agreed to have dinner, send reminder in the group chat, etc.

How does the company behind Jeeves make money?

How do OpenTable and Fandango make money?

Exactly! Monetization through partnerships with service providers or affliate programs or service promotion (e.g. a new chinese restaurant opened if you would like to try it)

I believe this violates the single responsibility principle...

But seriously, do you really want all the Jeeves text showing up in your chat? Isn't that kind of annoying - like showing your friends your working when trying to give them the answer to a math equation?

As it should violate because single responsibility is not applicable to this use case.

JEEVES text showing up in our chat is useful when I ask JEEVES to do something for us. Remember JEEVES does not show up till we ask for "it". Also remember this conversation is about planning to meet up for a movie not discussing some topic or sharing some content or having witty back and forth.

My point isn't whether it could be useful, but that it might be more annoying to have that in a chat conversation...

I was half joking about the single responsibility thing - I do kind of think though that chat should be for chat, and not for searching for movie timetables.

> now that's powerful and useful

Maybe, but how could a company apart from FB monetise that?

This seems like a very risky opening for all but a few strong brands ("Get me an Uber to blah blah blah", etc) since they'll have no control over the platform and I imagine be quite difficult to remind users they exist. At least apps have icons to remind users they installed them.

Chatbots like JEEVES will have the same problem that apps currently do: discoverability. But once discovered and installed JEEVES could potentially chime in in a conversation with offer to help e.g. if JEEVES sees friends planning to meet up JEEVES can start by offering to put it on their calendars.

if u have an hour, watch https://www.youtube.com/watch?v=tABT6GdygnI and listen for the distribution strategy part. This is a wide open one.

Can someone provide a narrower indication of where that part occurs?

it's the "Scalable Distribution Model" part of the talk https://scontent-sjc2-1.xx.fbcdn.net/hphotos-xpa1/t31.0-8/12... 15:56 but then there's more at end

Thank you

But if you want to cash in on this you need to create compelling content more than be a hacker. The top Snapchat Stories are not there because of hacking.

I remember back in the old days when bots were enabled by open protocols …

We've had IRC and eggdrop (among many other bots) for decades but we've dropped the ball - it's just never been a priority to build nice interfaces on top of these things.

Heck, when I was working with a distributed startup we were using Skype group chat which was actually surprisingly good. Almost Slack good at times. The Skype API was powerful enough to build a bot, but clunky and nowhere as good as the APIs available today. I wrote eggdrop code decades ago and it was fragile and picky.

Welcome to Walled Garden As A Service, the game where the rules are made up and openness doesn't matter!

That's right, open architectures are like corporate tax laws!

There's a lot of cool bots I can come up with but I none that I think you'd be able to monetize. For example, if I'm in a group chatting away, it'd be cool to have a bot collect certain posts that I can reference later instead of having to scroll/search. Or, I can ask the bot to live blog a certain event (or if there's video show a certain event) and we can live discuss.

"Conversational commerce" has a critical problem: in whose interests is the bot operating?

Is it an unbiased-as-possible purchase agent working on your behalf, or it is a salesman on commission? Will it route you to the best option or the one that makes the most money for its owners? How do you know?

Currently the nearest thing I do to conversational commerce is Skyscanner and Digikey-style selective refinement.

I'm pretty sure that Facebooks in the first place.

I'm not a Facebook user. I like to think I'm still well-informed as a professional in the industry - i.e., I understand this major platform - because I know what social networking is about and how Facebook works on a technical level. Am I wrong? What don't I understsand by not being a user?

I say all of this who has been out of the Facebook scene for about 4 years, so things might have changed:

At any given point in time Facebook favors certain integrations more than others. By not being a user you are more likely to waste time on integrations that are less useful than other ones. Before you design Facebook integrations you should become a user and observe these things.

Beyond being a user, unless you're also an active developer in the ecosystem I doubt you actually understand Facebook. There's lots of behind the scenes changes that you won't notice as a user but as someone who relies heavily on Facebook traffic, you will notice. So beyond becoming a user, you should also look at other successful companies and see how they handle integrations to get an idea of where the playing field currently stands.

Doubtful, because NLP is not yet at the point where it can help users faster than a normal website. If that was the case, google would be there already.

It was posted couple days ago, that people are sharing less and less private things from their lives on Facebook.

So now they'll start live streaming it instead ?

Yes, ever heard about LifeLogger?

All that shines is not gold

Nah

Reply on news.ycombinator.com