Why not use Twiml?

[dave]

I just found out about Plivo, it looks awesome and am looking at using
it as an alternative to Twilio for my app. Just a question though, why
use Plivo RESTXML instead of Twiml? Using Twiml would make porting
applications that are already on Twilio so much easier. I understand
that there are now probably a number of applications written for Plivo
RESTXML so you would have to still support it but what would the
community think about the development of a gem which allows you to
write your apps in Twiml? E.g. similar to if you use the haml gem for
a rails app. If this would be a welcome addition I'll go ahead and
build a gem for Twiml support.

[Venky]

Hi Dave,

We have not followed Twiml as a standard for a reason. Our APIs and XML allow for much more functionality which Twiml does not.

So following some standard which does not cover the complete functionality as a benchmark would only let us do so much.

However, one could build a twilm to RESTXML converter for those who want to make porting easier. Would love to see something like that.

Regards

Venky

[David Wilkie]

Hey Venky,

Thanks for the explanation. I'm still deciding whether to switch to Plivo or to Verboice (https://bitbucket.org/instedd/verboice/wiki/Home) which uses Twiml and Asterisk/FreeSwitch. Both seem like cool projects

Dave

[mphill]

I've done extensive programming with Twilio.

Venky points out not following their API allows Plivo more
flexibility. I agree with this 100%

For instance, when you are doing an IVR with Twilio you must program
all of your error handling for inputs. I.e.

"Press 1 for sales, 2 for technical support" Twilio has nothing that
would allow you consider 3 as a key press to be invalid. You'd have to
code that yourself, not a big deal but it would make sense to expose
this via the API

There are subltle differences like that make programming cleaner,
easier and less prone to regressions with Plivo potentially.

Verboice looks interesting, I know nothing about it but it looks like
SaaS so you would not have the developer functionality per se. Twilio
is a great product, and has allowed me to incredible things in
rediculously low amounts of time. Of course the Twilio premium is well
worth it to many people. Their pricing is competitive and they give
you their APIs. Our company is starting to spend almost 500 a day
though on Twilio so we are looking into Plivo as a possible
alternative to save costs and also because it would allows us to
connect sip devices directly.

[Clark S]

In my eyes, using Plivo over Twilio is a no brainer. Why - much cheaper. I'm not a programmer, but if you're spending $500 a day, then do your company a favor and switch. You'll be a hero with the savings you create. My programmer prefers Plivo over Twilio and so does my pocket book…and no, I'm not getting paid to say this :)

[mphill]

Clark S,

If you have heard of 2600hz's project Bluebox I am going to test to
see if A.) Plivo and Bluebox will play nice and if so B.) write a
module. I just joined this group today and I hope to become an active
member.

[David Wilkie]

@Clark S,

I wasn't comparing Twilio with Plivo here, I just wish it was easier to port from one to another and I thought that supporting Twiml would make it easier.

FYI Verboice is an open source project not a SaaS here is the source code:

https://bitbucket.org/instedd/verboice/wiki/Home