MySpace setting a dangerous precedent by not complying with OpenSocial

[Alex Epshteyn]

The MySpace platform, unveiled today, has too many limitations and
differences from the OS spec to allow developers to build cross-
(social)-platform apps. This is setting a very dangerous precedent
for the future of OpenSocial and Google needs to step in right now and
force MySpace (and the other networks) to comply with the spec. Allow
me to explain:

As it stands now, MySpace is the least common denominator among the
OpenSocial implementations. This means that in order to write cross-
platform apps, developers have to limit
themselves to only using features that MySpace chooses to support. If
you let every site pick and choose their own pieces of OpenSocial, you
will soon end up with a pretty small common denominator.

Here are the current limitations of the MySpace platform (as far as I
can tell after spending the entire afternoon reading their docs and
forums):

1). No way to host your own gadget XML file (for now, it only lets you
paste the HTML/JS content into a web form). You also have to enter
code for each surface separately.

2). Can't import external JS files from your servers. MySpace
provides their own version of jQuery but if you want to use GWT or any
other JS library (or, more importantly, your own libraries), you're
out of luck. I guess you have to mash everything into one giant
file. Furthermore, Myspace is crippling your ability to write
Javascript by running your code through Caja even though it's already
in an iframe.

3). The Persistence and Activity APIs are not implemented. If
developers have to build and host their own database just for MySpace
you might as well get rid of Persistence API support from other
networks. No one wants to work with 2 or more different data models.

This is Google's chance to step in and prevent OpenSocial
implementations from mirroring what happened with the web browsers
(and worse).

(Note: this is a cross-posting from http://groups.google.com/group/opensocial-api/t/025e87bd6869696f
I wanted to also share my thoughts here as there isn't much activity
on the OpenSocial forum)

[jai...@gmail.com]

I agree, what is the use of them having open social if you can "use"
open social.

On Feb 5, 10:32 pm, Alex Epshteyn <alexander.epsht...@gmail.com>
wrote:
> TheMySpaceplatform, unveiled today, has too many limitations and

> differences from the OS spec to allow developers to build cross-
> (social)-platform apps. This is setting a very dangerous precedent
> for the future of OpenSocial and Google needs to step in right now and

> forceMySpace(and the other networks) to comply with the spec. Allow
> me to explain:
>
> As it stands now,MySpaceis the least common denominator among the

> OpenSocial implementations. This means that in order to write cross-
> platform apps, developers have to limit

> themselves to only using features thatMySpacechooses to support. If

> you let every site pick and choose their own pieces of OpenSocial, you
> will soon end up with a pretty small common denominator.
>

> Here are the current limitations of theMySpaceplatform (as far as I

> can tell after spending the entire afternoon reading their docs and
> forums):
>
> 1). No way to host your own gadget XML file (for now, it only lets you
> paste the HTML/JS content into a web form). You also have to enter
> code for each surface separately.
>
> 2). Can't import external JS files from your servers. MySpace
> provides their own version of jQuery but if you want to use GWT or any
> other JS library (or, more importantly, your own libraries), you're
> out of luck. I guess you have to mash everything into one giant

> file. Furthermore,Myspaceis crippling your ability to write

> Javascript by running your code through Caja even though it's already
> in an iframe.
>
> 3). The Persistence and Activity APIs are not implemented. If
> developers have to build and host their own database just forMySpace
> you might as well get rid of Persistence API support from other
> networks. No one wants to work with 2 or more different data models.
>
> This is Google's chance to step in and prevent OpenSocial
> implementations from mirroring what happened with the web browsers
> (and worse).
>

> (Note: this is a cross-posting fromhttp://groups.google.com/group/opensocial-api/t/025e87bd6869696f

[Daedeloth]

It looks to me this whole OpenSocial thing is entirely failing.
I really hope google will prevent this project from failing.