Isaac Castañeda avatar image
Isaac Castañeda asked

Articles behind a paywall

Before doing a deep search in the docs, I want to know if this is posible

We are a news company. Some of our digital articles are open to public, some others are behind a paywall and some others are behing a meter, this means you can read N articles before a stop. Also there are articles only to registered users

For the case of the command "Alexa, read first article from XXX section", is there a way to control what I said? Know if a user is a registered one, or a subscriptor

I was thinking on "Alexa, my token as registered user is XXXX". We provide that token to the user

Dunno if everything is from Rss or we can use some kind of webservice, save attempts, use our internal counter, and act according

Can we catch all the commands user send? Not the voice, only the text, so we can see which features are the most used or something like that

flash briefing
10 |5000

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Andy Whitworth avatar image
Andy Whitworth answered

I thought Flash Briefing skills were very limited in how you can interact with them via voice and didn't support things like "read first article from xxxx section" etc. Sounds like you need some other skill type, podcast ? Or a custom skill.

1 comment
10 |5000

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

It's ok if is not Flash Briefing, I can dig in Custom skills or the name it have. Just wondering if what I'm looking for is posible

0 Likes 0 ·
Joseph Kesselman avatar image
Joseph Kesselman answered

Normally what you'd do is ask the user to establish a connection to their account when they enable the skill, store that account information in the user's persistent data, and use the latter when your skill tries to fetch information behind the paywall. Since only you know the details of your paywall, that has to be code you write -- though it might be easier if customer identity is already being handled by a standard LDAP server.

Once they *are* signed in, I can't see any reason you couldn't do the authentication as part of a flash briefing... but you do have to do it.

(And as noted, flash briefings aren't intended to be very interactive; if you want the user to have more control, you do need another kind of skill. The two might share much of their back-end code, though.)

10 |5000

Up to 2 attachments (including images) can be used with a maximum of 512.0 KiB each and 1.0 MiB total.

Thanks for the reply, currently I'm reading how custom skill works. I'm not asking how to do it, but want to know if is posible to know a kind of device/user id

My plan is this:

- User install the skill

- User ask for 3 articles

- On the 4th, my server must tell Alexa to tell the user he can't

- User must go to our url to obtain their token

- User must says something like "Alexa, my token for XXX app is 34534"

- Alexa sends me that information so I can link the device/user

- User ask for 4th article and then our server answer according

0 Likes 0 ·

The Alexa API can tell you what Alexa user is currently signed in. Generally that's the same for all devices in a household unless the customer explicitly set it up otherwise. Your code and/or the APIs would use that to look up and store customer-specific data, including your token and the article count.

I'd really recommend that your skill be set up so the user gives it their login information on your servers and it manages the token tracking for them. Nobody wants to have to juggle multiple devices and re-enter data unless they absolutely must. Manual interaction is tolerable for two-factor authentication, but it doesn't sound like your skill needs that level of security...?

I have no idea whether or how one might get the device identity/device type.

1 Like 1 ·