Means of contact ? Who is taking decisions etc ?

[Alexis Metaireau]

Hi,

Thanks for your project, it's useful and something I've been looking for since some time, so I really appreciate it.

However, I'm seeing that the issues that are opened on github doesn't get any feedback so far. I'm wondering if it's because you are not looking at them or for some other reason? Maybe that's not the right media to discuss things?

I would like to go ahead and discuss about the format of parameters in the URI [0], so feel free to indicate me what's the best way to get this discussed :)

[0] https://github.com/SPORE/specifications/issues/5

Thanks for your attention,

Cheers,
Alexis

[M Edward Borasky]

The project seems to have been abandoned.

[Marc Chantreux]

On Fri, Sep 21, 2012 at 10:26:11PM -0700, M Edward Borasky wrote:
> The project seems to have been abandoned.

hello,

the project isn't abandoned at all:

Spore is daily used in companies like linkfluence or weborama, the
university of strasbourg plan a massive usage of it.

A python implementation on its way on github (spyre) and was presented
at pycon.fr 2012.

i'm aware of new descriptions and invited the authors to make a pull
request on the description repo (i guess they did not).

I think the users are happy with the existing codebase but what we miss
is an active communication.

regards
marc

[M. Edward (Ed) Borasky]

I wanted to use it to build a Lua interface to the Twitter API,
specifically updating to the new 1.1 version of the API and adding a
connection to the Streaming API. But if I'm the only one who wants
this, there isn't much point in doing it.

--
Twitter: http://twitter.com/znmeb; Computational Journalism Publishers
Workbench: http://j.mp/QCsXOr

How the Hell can the lion sleep with all those people singing "A weem
oh way!" at the top of their lungs?

[JT Smith]

The only way projects like this get off the ground is if people contribute. You aren't the only one on this list. If you have a need for SPORE, then use it. That will encourage others to do the same and so on. I plan on writing out a massive API in SPORE. I just need to get the round-tuits to do so.

[M. Edward (Ed) Borasky]

I don't have a "need for SPORE". I have a need for a software package
that connects to the Twitter REST API, version 1.1 and the Twitter
Streaming API in Perl, Ruby or Lua. Right now there is nothing in the
open source world that does anything with the 1.1 API and there's no
Lua code for Streaming.

I was hoping the modularity of SPORE and ability to talk to multiple
languages would be a win, at least for the REST piece. But again, if
there isn't an *active* community - at least multiple on-topic
messages per day on the Google group - I'm better off reaching out to
the Perl or Ruby Twitter API developers or writing my own from scratch
in Lua.

[JT Smith]

Clearly you've made up your mind, but "one or two on-topic messages" per day, rules out 99% of all open source projects. Even big projects like Dist::Zilla which is used by thousands of developers don't have that.

[Alexander Gladysh]

Hi,

On Sat, Sep 22, 2012 at 9:57 PM, M. Edward (Ed) Borasky <zn...@znmeb.net> wrote:
> I don't have a "need for SPORE". I have a need for a software package
> that connects to the Twitter REST API, version 1.1 and the Twitter
> Streaming API in Perl, Ruby or Lua. Right now there is nothing in the
> open source world that does anything with the 1.1 API and there's no
> Lua code for Streaming.
>
> I was hoping the modularity of SPORE and ability to talk to multiple
> languages would be a win, at least for the REST piece. But again, if
> there isn't an *active* community - at least multiple on-topic
> messages per day on the Google group - I'm better off reaching out to
> the Perl or Ruby Twitter API developers or writing my own from scratch
> in Lua.

IMHO, "Multiple on-topic messages per day" is too high a bar for a
limited scope project.

Lua-Spore is alive and well (if not very lively developed). We needed
a couple of API bindings, we contributed them to lua-Spore, François
helped us to get them in shape and merged them, we're using lua-Spore
with these modules happily.

Anyway, there are some lua-Spore specifications for Twitter API, are
they outdated?

https://github.com/SPORE/api-description/blob/master/services/twitter.json

My 2c,
Alexander.

[M. Edward (Ed) Borasky]

On Sat, Sep 22, 2012 at 11:05 AM, Alexander Gladysh <agla...@gmail.com> wrote:

> IMHO, "Multiple on-topic messages per day" is too high a bar for a
> limited scope project.
>
> Lua-Spore is alive and well (if not very lively developed). We needed
> a couple of API bindings, we contributed them to lua-Spore, François
> helped us to get them in shape and merged them, we're using lua-Spore
> with these modules happily.
>
> Anyway, there are some lua-Spore specifications for Twitter API, are
> they outdated?
>
> https://github.com/SPORE/api-description/blob/master/services/twitter.json

https://dev.twitter.com/blog/current-status-api-v1.1

In less than six months, all your current Twitter code will be
inoperative. As will the Perl Net::Twitter module, which is what I'm
using now, unless the developer or one of his users rewrites his code.

My current plan is to write my own Twitter REST API 1.1 code from
scratch in a language yet to be chosen and not released as open
source. Obviously if I fork someone's project I'll need to release as
open source, though. There's no immediate need for me to move away
from the Perl Twitter Streaming code I'm using; that API is not
changing. But if I switch languages on REST I'll need to switch
languages on Streaming as well to make maintenance easier.

By the way, the only reason Lua is in the picture at all is that Redis
uses it for scripting as of 2.6. My preference is for languages that
lots of people know and use on a daily basis.

[Nils Grunwald]

Hello,

This is indeed the right channel to discuss issues with SPORE as a specification. It is indeed quite low traffic, but this is, as you said, a specification, not a software package, and I take that as a sign that the spec is enough for people to implement it without too much external help. SPORE is a deliberately short and simple specification. As for seeing the spec evolve, we have been discussing possible extensions with François Perrad at the last French Perl Workshop, but seeing breaking or major changes is very unlikely, we want our existing implementations of the client and API descriptions to remain valid.

> In less than six months, all your current Twitter code will be

> inoperative.

Not really, as the separation of concerns between using APIs over HTTP and authentication or caching was the whole point of SPORE. Now that Twitter mandates using OAuth, all you have to do is add to your code initializing the SPORE client with the OAuth middleware. As Twitter as, as far as i know, has not modified its calling parameters, the SPORE description should still be valid (if not, patches welcome!) 

As for the original question about URI-Templates, I have now answered it on Github, see you over there!

[M. Edward (Ed) Borasky]

On Sun, Sep 23, 2012 at 6:12 AM, Nils Grunwald <nils.g...@gmail.com> wrote:

>> In less than six months, all your current Twitter code will be
>> inoperative.
>
> Not really, as the separation of concerns between using APIs over HTTP and
> authentication or caching was the whole point of SPORE. Now that Twitter
> mandates using OAuth, all you have to do is add to your code initializing
> the SPORE client with the OAuth middleware. As Twitter as, as far as i know,
> has not modified its calling parameters, the SPORE description should still
> be valid (if not, patches welcome!)

1. The endpoints have changed. In less than six months, code that
talks to the old Search endpoint and the old REST endpoint will stop
working.
2. Search is now included in the REST API, rather than being a
separate endpoint.
3. All calls to REST / Search need to be authenticated via oAuth.
4. Only JSON is returned.
5. The rate limits have changed.
6. A few calls do have different parameters.

A detailed test-driven recoding to the API as documented by Twitter is
a *minimum* level of effort required for "patches". I haven't scoped
it out because I don't want to do it; it's easier to pick a language
with an existing open source library and *test suite* and tweak that.
;-) Sadly, I've talked to a couple of developers of said libraries and
they are busy getting paid to do other stuff.

There's also the issue of Twitter developer relations. Most of the
developers I know are either in business for themselves or are
employed by a business, and Twitter pretty much insists on win-win
negotiations up front. I'm going to propose on the Twitter developer
forum that Twitter get involved in the SPORE effort, since it's open
source and they love open source. There's a chance they'll help out if
I can convince them that it's a win for Twitter to support SPORE.