[twitter-dev] Retweet count for status?

2010-05-03 Thread Tim Bull
Hi,

I noticed that http://search.twitter.com/search?q=Lakers+OR+%23Lakers
is returning a recent retweets count which is neat.  I can see this in
the search API, but I can't see where this would be in the normal API.

I'd like to call a specific status and see the retweet count.  Is
there a way to do this (short of executing a search for the specific
tweet)?

Cheers

Tim



[twitter-dev] Re: Who's using up my rate limit?

2010-09-12 Thread Tim Bull
I'm sure I read somewhere that the API limit applies to the account
and not the application?  Perhaps someone here can confirm.

In that case, if you have multiple applications registered under the
same account, it might be these other apps that are using the limit.

I once had a similar issue, turns out that it was my local client
(which I was using for monitoring the account and replying to users)
which was using the other portion of the limit.

Try disabling any other apps on the account and also shutting down any
local monitoring client etc. and see if that helps.

Tim



On Sep 13, 5:31 am, Tom van der Woerdt i...@tvdw.eu wrote:
 To answer your question: I have no idea who's using up your rate limit.

 However, I would recommend using OAuth to sign each request. This will
 get your application 350 (I think) API calls per hour, and that's not an
 IP-based limit.

 Tom

 On 9/12/10 10:37 AM, crystalchris wrote:



  I constantly get the exceeding limit message when calling twitter api
  even though I only make 2 calls in an hour. So I use rate_limit_status
  to find out what happened. I found out that right after the reset
  time, I would have about 68 hits remains. Fifteen minutes after that
  all 150 hits will be used up. That leaves me only 15 minutes in an
  hour to call the api before reaching the limit.

  I thought that's probably I am on a shared hosting plan. If I get my
  own dedicated IP, the problem will go away. So I change my ip from 7X.
  126.92.192 to 7X.126.92.194. The problem remains!! Does the limit
  apply to the whole network? Someone help me!

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk?hl=en


[twitter-dev] Re: search results - how far does it go back

2010-10-04 Thread Tim Bull
The time window depends on how busy Twitter is as a whole - the search
is not a fixed timeframe.

On Oct 5, 7:16 am, Quy quyten...@gmail.com wrote:
 When try to search on results from a user like from:mashable, I only
 see results going as far back as 24 hours? I thought the archive went
 back further for a search.

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] How long is it taking for support tickets to be reviewed and is it possible to expedite them?

2010-10-04 Thread Tim Bull
Our application (twitter.com/distlr) has had it's account suspended
and I need this reviewed.  I (think I) understand why it was
suspended, which was stupidity on my part rather than anything
malicious or bad on the part of the application itself.

We follow a very lean start-up approach:

* We had the idea for Distlr and immediately interviewed people and
asked them if they'd use it.
* When they said Yes, we prototyped it with @anywhere and put it in
front of them, then surveyed them and asked if they'd use it.
* When they said yes, we developed a very alpha version with full
OAuth integration, incorporating their feedback and put it front of a
bigger audience.  Around 60 people registered and 400 used it
anonymously.

Unsurprisingly if you follow the thread here, I then went to survey
these 60 registered users - I created a Google Form and then (being
very lean and not trying over engineering things) started sending the
messages from the Distlr account to each of the registered users.

@usera @userb @etc Thanks for using Distlr. Would love your feedback
via this short survey: http://bit.ly/distlrsurvey;

After several of these, the Distlr account was suspended.

Immediately when this happened, I've slapped my forehead with the
biggest Doh! you've heard.  As I've explained in the ticket, this is
more user stupidity on my part rather than anything malicious the
application itself is doing.

It's holding us up because while the ticket is Open, I'm loath to move
the app to a new account as it may look like we are trying to be
underhanded, which we absolutely aren't.  Unless Twitter knows
differently, I really don't think the app itself is misbehaving, I
think it's just these survey links which have caused the concern.

How long are these support tickets taking to be reviewed (it's been
open over 17 hours now) and is it possible that someone from the
Twitter dev support team can help out with getting it reviewed sooner
please?  If we have to wait, we have to wait, but we'd love to
progress forwards with the feedback we did get from the few people
that filled out the survey before the account got suspended.

The ticket is http://support.twitter.com/tickets/1256917

Thanks!

Tim Bull

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: How long is it taking for support tickets to be reviewed and is it possible to expedite them?

2010-10-04 Thread Tim Bull
Thanks Taylor,

I appreciate you taking a look.

It's interesting that the account is suspended, not the
application.  One side effect of this appears to be that while users
who have authenticated continue to remain authenticated to our
application, the OAuth API appears to be refusing to issue tokens for
new users - presumably because our account is suspended?

Anyway, on to how we would avoid this in the future.

Simply put, in this proving stage for a new application we try to
quickly approach a Minimum Viable Product (MVP) and we were quite
ruthless in throwing out features / enhancements if there was another
manual way of doing something in the very short-term.  In terms of
gaining feedback from our users, the approach we took of sending
@messages to the authenticated users clearly upsets the Spam Gods and
therefore isn't a suitable mechanism - we should of invested in a
different approach.

Here's four things we could of done differently to achieve a similar
outcome that wouldn't upset Twitter:

1. Collect the users email address before, or after they authenticate
with OAuth and make it part of the profile we store.  We could make
this a mandatory or optional step for using our application.  Either
way we'd have at least some email addresses and permission to email
users directly so we could send the survey to them.  We would make it
optional given we allow anonymous users anyway!
2. Provide a link or callout on the homepage with the survey for
returning users.
3. After some period of use, popup a call out in the application
asking users to provide feedback.  Annoying yes, but valuable to us in
this early stage and we need only do it once.  This has a big
advantage in that it would capture both Authenticated and Anonymous
users.
4. On close of the application, popup a survey link.

Of these options a combination of 2 and 3 would probably be the most
suitable for us, especially if we made sure that once a user had seen
a message they didn't see it again (or perhaps if they saw message 3
and cancelled, they'd see message 2 on return and never again if they
cancelled after that).

Cheers,

Tim



On Oct 5, 8:14 am, Taylor Singletary taylorsinglet...@twitter.com
wrote:
 Hi Tim,

 We looked into your request but unfortunately cannot expedite resolving it
 right now. In this case, the account used to post the tweets was suspended
 -- not your application. While there's obviously a good deal of overlap
 between API policy/enforcement and account policy/enforcement, this kind of
 suspension falls squarely in the account policy camp. The support team's
 response time is actually quite good right now, and I imagine you'll be
 hearing back from them soon.

 Thanks for the keen understanding on what went wrong here -- since there are
 obviously many developers who might find themselves facing the same scenario
 at some point, can you share a bit on what actions you'll take to avoid this
 in the future?

 Thanks!
 Taylor



 On Mon, Oct 4, 2010 at 1:48 PM, Tim Bull tim.b...@binaryplex.com wrote:
  Our application (twitter.com/distlr) has had it's account suspended
  and I need this reviewed.  I (think I) understand why it was
  suspended, which was stupidity on my part rather than anything
  malicious or bad on the part of the application itself.

  We follow a very lean start-up approach:

  * We had the idea for Distlr and immediately interviewed people and
  asked them if they'd use it.
  * When they said Yes, we prototyped it with @anywhere and put it in
  front of them, then surveyed them and asked if they'd use it.
  * When they said yes, we developed a very alpha version with full
  OAuth integration, incorporating their feedback and put it front of a
  bigger audience.  Around 60 people registered and 400 used it
  anonymously.

  Unsurprisingly if you follow the thread here, I then went to survey
  these 60 registered users - I created a Google Form and then (being
  very lean and not trying over engineering things) started sending the
  messages from the Distlr account to each of the registered users.

  @usera @userb @etc Thanks for using Distlr. Would love your feedback
  via this short survey:http://bit.ly/distlrsurvey;

  After several of these, the Distlr account was suspended.

  Immediately when this happened, I've slapped my forehead with the
  biggest Doh! you've heard.  As I've explained in the ticket, this is
  more user stupidity on my part rather than anything malicious the
  application itself is doing.

  It's holding us up because while the ticket is Open, I'm loath to move
  the app to a new account as it may look like we are trying to be
  underhanded, which we absolutely aren't.  Unless Twitter knows
  differently, I really don't think the app itself is misbehaving, I
  think it's just these survey links which have caused the concern.

  How long are these support tickets taking to be reviewed (it's been
  open over 17 hours now) and is it possible that someone from the
  Twitter dev

[twitter-dev] User Streaming API and use of OAuth from web browser

2010-10-06 Thread Tim Bull
Hi,

We are building an application client that is browser based.  We're
very comfortable with using OAuth from our server side code and are
using it fine with the REST API (users sign in, authenticate with
Twitter, we store their access tokens and reuse as requested - at the
moment we mimic the required Twitter API on our server and when a user
does something like a POST, we call our stub, use their token to then
make the call via OAuth to Twitter).

So far so good, but we'd like to implement User Streaming directly
into the client side application.

I've been browsing the Twitter Development documentation and there's a
couple of points I'd like clarification on:

* http://dev.twitter.com/pages/auth_overview says Streaming supports
Basic and OAuth.

* http://dev.twitter.com/pages/user_streams says that the user streams
supports OAuth only HTTPS, OAuth and JSON only.  No problems here, I
just raise it to point out the auth_overview doco is slightly out of
date.

* http://dev.twitter.com/pages/oauth_libraries talks about a JS
library but says Javascript really shouldn't be used for OAuth 1.0A
with respect to websites in web browsers. Ideally, you'll only use
Javascript to perform OAuth operations when using server-side.

The points I'd like some clarification on:

1. Given user_streams API is the intended way for clients to access
Twitter going forwards, I presume it's intended not just for desktop,
but also web clients too?
2. If 1 is correct, then is it OK to use JavaScript for the OAuth?  If
it's not, what is the recommended approach for a client side web
application to connect and authenticate to the user_stream?

Thanks,

Tim

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: Error / lack of clarity in the OAuth documentation regarding the oauth_callback parameter

2010-11-12 Thread Tim Bull
FYI I see the correct flow is documented here:

http://groups.google.com/group/twitter-api-announce/browse_frm/thread/472500cfe9e7cdb9?hl=en

It's just not in the actual formal doco.

Cheers,

Tim

On Nov 13, 4:29 pm, Tim Bull tim.b...@binaryplex.com wrote:
 Hi,

 I'm building a site that integrates with a single Twitter application
 from a series of sub domains under the same main domain.

 In the past I've only ever used a single domain, so never bothered
 with oauth_callback, but I saw it mentioned when I registered the
 application and figured it was exactly what I need, however I
 struggled to implement it for a while.

 After much trial and error, I did further research and seems the
 oauth_callback parameter is to be sent with the REQUEST, not with
 AUTHORISATION.

 If you refer to this documentation on getting a Twitter request token,
 oauth_callback is NOT listed as an allowable parameter.

 http://dev.twitter.com/doc/post/oauth/request_token

 Confusingly oauth_callback IS mentioned here on the authorize step,
 despite it's seeming irrelevance here?

 http://dev.twitter.com/doc/get/oauth/authorize

 This was where I was misled, although it's not clear at all, I
 interpreted this as meaning the oauth_callback should be sent with the
 authorisation step.

 I've now successfully implemented what I want by passing the
 oauth_callback with the request step - seems like an easy fix to
 clarify this in the documentation, it appears to be supported and
 mentioned in several places, just not clearly documented.

 Thanks!

 Tim

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Any progress on Favorites API?

2010-11-22 Thread Tim Bull
I see posts from several months ago, so I thought I ask again
http://groups.google.com/group/twitter-development-talk/browse_thread/thread/134d3bf90a717f8d/806fa7325dd1c6e7?lnk=gstq=twitter+favorites#806fa7325dd1c6e7

I need to regularly extract and process a users favorites and as noted
in that previous link they appear to be stored by date of Tweet ID,
which means to be fairly sure you get all the tweets you need to page
through all of them ( a user could although admittedly unlikely,
favourite a tweet that was older than last time you checked them).

I note that the count parameter e.g. count=200 works but is
undocumented which is helpful (that it works) - in one use case we
have a user with over 1,500 favourites.

It would be great to have this under-utilised feature in Twitter get a
bit of API love and be modernised some more.  I can achieve what I
want by caching on my end, but it means I have to hit Twitter to fetch
data I already have which isn't ideal - you don't want more API calls
I don't want to take the time fetching data I almost certainly already
have for the one in 100 chance it's changed.

The documentation here is also a bit sketchy - for example, is there a
page limit like with other calls? http://dev.twitter.com/doc/get/favorites
If so, it would be great to have it documented here.

Thanks - I guess I'm just highlighting that there's at least one
developer who's still interested in seeing the favourites updated, and
I'm optimistically wondering if Twitter can provide any comment on
where it's at in the planning.



-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Minor bug in Application Settings / Configuration... (http://dev.twitter.com/apps)

2010-12-07 Thread Tim Bull
I've successfully implemented a few OAuth implementations with Twitter
now and was setting up a new application.  Got to the callback URL
when registering the app and thought nah, not sure what it will be,
will leave it blank and either enter it later or just pass
oauth_callback anyway.  So I saved the application with a blank
callback URL.

Turns out this was A bad idea TM.

I'm using the Ruby OAuth gem which is new to me, but I was getting the
following problems:

On my request token step - if I didn't pass an oauth_callback, it did
an OOB / PIN authentication for me.

If I did pass an oauth_callback it returned a 401.

Eventually I remembered about the callback URL I didn't create and
tried to change it.  Having not entered a callback URL in the first
place, it now seemed like I can't edit it at all.

So, eventually I tried to delete the application (fail whale), so I
just created a new one, and this time entered a Callback URL.

Changed the consumer keys and the exact same code now works perfectly
and the oauth_callback is doing exactly what I expected (which is
returning to a URL that's not the one registered with the App).

Looks like that step of registering the Callback URL is critical and
shouldn't be allowed to be left blank, although also I think if I pass
an oauth_callback even if it IS blank it should acknowledge it.

Cheers,

Tim

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: oauth_callback

2010-12-07 Thread Tim Bull
There is a required OAuth parameter step which is unclearly documented
by Twitter. When Twitter returns from your /oauth/authorize It returns
an oauth_verifier token. Make sure that you pass this oauth_verifier
token (along with the other parameters) along to
your /oauth/access_token call.


Make sure you are passing this oauth_verifier in and see how you go.
I've found that if you DON'T set a callback, it doesn't enforce the
verifier, but if you do, then the verifier is essential (just be aware
Twitter are planning to change to always require this in the future, so
it's more compliant with the spec; worth making this change regardless,
a lot of Twitter libraries don't implement it).


Hope this helps...


Tim

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: register twitter app which is on intranet

2010-12-07 Thread Tim Bull
Use the OOB process - so pass oauth_callback=oob and you should get a
PIN from Twitter which you then use in fetching your access_token.


http://dev.twitter.com/pages/auth_overview#oob

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: register twitter app which is on intranet

2010-12-07 Thread Tim Bull
Oh, and while I think of it - if you just need the access token to make
calls as your app (i.e. it's some kind of bot) then you don't even need
to do that - just go to http://dev.twitter.com/apps, view your app and
select my access token on the right. This will give you the access
keys you need without doing the 3 step OAuth dance.


Just use these to sign your requests and you'll be sweet.

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: oauth_callback

2010-12-08 Thread Tim Bull
Dave-tweinds,

It is mentioned in passing and buried in some documents which discuss
the full flow, but if you're relying on Twitter's own API
documentation (http://dev.twitter.com/doc/post/oauth/access_token)
you'll not see a single mention of the oauth_verifier here.  Nor will
you see oauth_callback mentioned here 
http://dev.twitter.com/doc/post/oauth/request_token

I think it would be more useful to the community if Twitter did a more
complete job of describing the flow and parameters supported and
required on each of their own end points which is where most people
look first, even if they are implicitly supporting these parameters
because they comply with oauth and therefore documented elsewhere.

T

On Dec 8, 8:07 pm, Dave-twiends i...@davesumter.com wrote:
 Hi Tim, I'm pretty sure the oauth_verifier is documented in their
 oAuth articles.. I'm speeking from memory here, but I'm sure I saw
 last week when we were investigating our own oAuth issues..

 But, nonetheless, you are correct, oauth_verifier should be passed
 back every time.

 Dave
 Twiends

 On Dec 8, 2:27 am, Tim Bull tim.b...@binaryplex.com wrote:







  There is a required OAuth parameter step which is unclearly documented
  by Twitter. When Twitter returns from your /oauth/authorize It returns
  an oauth_verifier token. Make sure that you pass this oauth_verifier
  token (along with the other parameters) along to
  your /oauth/access_token call.

  Make sure you are passing this oauth_verifier in and see how you go.
  I've found that if you DON'T set a callback, it doesn't enforce the
  verifier, but if you do, then the verifier is essential (just be aware
  Twitter are planning to change to always require this in the future, so
  it's more compliant with the spec; worth making this change regardless,
  a lot of Twitter libraries don't implement it).

  Hope this helps...

  Tim

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Upgrading from Read to Read / Write access for OAuth API Key

2011-01-29 Thread Tim Bull
We must be about the only developers in the universe that requested
users grant only read access when we first got people to connect
http://trunk.ly to Twitter (I think of the 40 or so apps authorized on
my account, Trunk.ly is the only one that asks for Read only).  Never
ask for more access than you need is my philosophy.

Doh!

Of course now, we want to add some Tweet out functions which require
users grant us Write access.

A couple of questions for the Twitter people.

1. If we change the access in the application from read to read/write
does this reset the API key, or will it stay the same (hoping it stays
the same).
2. How can I work out if existing users have authorised us for read/
write?  I looked at 
http://developer.twitter.com/doc/get/account/verify_credentials
but it doesn't show me what access they have.  Do I have to write,
fail, force them to step through OAuth then post? Or is there a way of
knowing before hand it will fail and asking them to upgrade?

Thanks,

Tim

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: Upgrading from Read to Read / Write access for OAuth API Key

2011-01-30 Thread Tim Bull
OK, that's more or less what I expected.

Just one last confirmation - the API key won't change though right?
So if I add read / write the read users won't suddenly be de-
authenticated?

Cheers,

Tim

On Jan 31, 6:19 am, Taylor Singletary taylorsinglet...@twitter.com
wrote:
 You'll have to re-ask your users for permission for write mode and you won't
 have any way via the API to track who is ready to read/write yet -- you'll
 want to manage the conversion process yourself and track whether you've
 converted your users yet or not.

 The thinking behind this is that when your users authorized your app, they
 only authorized it for read-access. Wanting write access requires a new
 agreement with the user.

 The oauth/authorize step should now upgrade to read/write from read-only
 tokens when the user is re-challenged.

 Taylor







 On Sun, Jan 30, 2011 at 8:32 AM, Adam Green 140...@gmail.com wrote:
  So if a user authorizes an app for read access, the app can switch to
  read/write at any time without asking the users permission? Is this
  true? Anyone from Twitter have any input on this?

  On Sun, Jan 30, 2011 at 11:04 AM, Patrick Kennedy kenned...@gmail.com
  wrote:
   Tim -

   1.  Changing from read to read/write won't change you API consumer
   keys or tokens.

   2.  Your application's users don't authorized for read or read/write;
   they merely use your application, which you offer as read or
   read/write to the world.  That is to say, if it's read, your
   application can only read its tweets, and if read/write, it can both
   read its own tweet and post to the world.

   I'd say go ahead and switch to read/write, given the fact that you now
   want that functionality.

   ~Patrick

   On Sat, Jan 29, 2011 at 10:24 PM, Tim Bull tim.b...@binaryplex.com
  wrote:
   We must be about the only developers in the universe that requested
   users grant only read access when we first got people to connect
  http://trunk.lyto Twitter (I think of the 40 or so apps authorized on
   my account, Trunk.ly is the only one that asks for Read only).  Never
   ask for more access than you need is my philosophy.

   Doh!

   Of course now, we want to add some Tweet out functions which require
   users grant us Write access.

   A couple of questions for the Twitter people.

   1. If we change the access in the application from read to read/write
   does this reset the API key, or will it stay the same (hoping it stays
   the same).
   2. How can I work out if existing users have authorised us for read/
   write?  I looked at
 http://developer.twitter.com/doc/get/account/verify_credentials
   but it doesn't show me what access they have.  Do I have to write,
   fail, force them to step through OAuth then post? Or is there a way of
   knowing before hand it will fail and asking them to upgrade?

   Thanks,

   Tim

   --
   Twitter developer documentation and resources:
 http://dev.twitter.com/doc
   API updates via Twitter:http://twitter.com/twitterapi
   Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
   Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

   --
   Twitter developer documentation and resources:
 http://dev.twitter.com/doc
   API updates via Twitter:http://twitter.com/twitterapi
   Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
   Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

  --
  Adam Green
  Twitter API Consultant and Trainer
 http://140dev.com
  @140dev

  --
  Twitter developer documentation and resources:http://dev.twitter.com/doc
  API updates via Twitter:http://twitter.com/twitterapi
  Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
  Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk


[twitter-dev] Re: Upgrading from Read to Read / Write access for OAuth API Key

2011-01-31 Thread Tim Bull
While this makes me happy (from a developers point of view), surely
this is a bug and therefore not to be relied on?

As a user, I agree with the logic that if I authorised Read only, the
application shouldn't be able to turn this into Read/Write without
some subsequent approval.

Tim

On Jan 31, 1:46 pm, Abraham Williams 4bra...@gmail.com wrote:
 Taylor,

 Confirmed. I just upgraded read only tokens and was able
 to successfully send a DM.

 Thank you for finally allowing read only access tokens to be upgraded to
 read and write access tokens. This issue has been plaguing developers for
 almost a year now. Both forcing applications to ask for permission they
 didn't need if there was even a remote possibility they might want write
 permissions in the future and biting devs in the ass if they unknowingly
 built up a customer base of read only tokens.

 I hope we will continue to see fixes coming down the pipe to keep Twitter
 API a viable platform for further development.

 Thank you again,
 Abraham
 -
 Abraham Williams | Hacker Advocate | abrah.am
 @abraham https://twitter.com/abraham | github.com/abraham | blog.abrah.am
 This email is: [ ] shareable [x] ask first [ ] private.

 On Sun, Jan 30, 2011 at 11:19, Taylor Singletary 







 taylorsinglet...@twitter.com wrote:
  You'll have to re-ask your users for permission for write mode and you
  won't have any way via the API to track who is ready to read/write yet --
  you'll want to manage the conversion process yourself and track whether
  you've converted your users yet or not.

  The thinking behind this is that when your users authorized your app, they
  only authorized it for read-access. Wanting write access requires a new
  agreement with the user.

  The oauth/authorize step should now upgrade to read/write from read-only
  tokens when the user is re-challenged.

  Taylor

  On Sun, Jan 30, 2011 at 8:32 AM, Adam Green 140...@gmail.com wrote:

  So if a user authorizes an app for read access, the app can switch to
  read/write at any time without asking the users permission? Is this
  true? Anyone from Twitter have any input on this?

  On Sun, Jan 30, 2011 at 11:04 AM, Patrick Kennedy kenned...@gmail.com
  wrote:
   Tim -

   1.  Changing from read to read/write won't change you API consumer
   keys or tokens.

   2.  Your application's users don't authorized for read or read/write;
   they merely use your application, which you offer as read or
   read/write to the world.  That is to say, if it's read, your
   application can only read its tweets, and if read/write, it can both
   read its own tweet and post to the world.

   I'd say go ahead and switch to read/write, given the fact that you now
   want that functionality.

   ~Patrick

   On Sat, Jan 29, 2011 at 10:24 PM, Tim Bull tim.b...@binaryplex.com
  wrote:
   We must be about the only developers in the universe that requested
   users grant only read access when we first got people to connect
  http://trunk.lyto Twitter (I think of the 40 or so apps authorized on
   my account, Trunk.ly is the only one that asks for Read only).  Never
   ask for more access than you need is my philosophy.

   Doh!

   Of course now, we want to add some Tweet out functions which require
   users grant us Write access.

   A couple of questions for the Twitter people.

   1. If we change the access in the application from read to read/write
   does this reset the API key, or will it stay the same (hoping it stays
   the same).
   2. How can I work out if existing users have authorised us for read/
   write?  I looked at
 http://developer.twitter.com/doc/get/account/verify_credentials
   but it doesn't show me what access they have.  Do I have to write,
   fail, force them to step through OAuth then post? Or is there a way of
   knowing before hand it will fail and asking them to upgrade?

   Thanks,

   Tim

   --
   Twitter developer documentation and resources:
 http://dev.twitter.com/doc
   API updates via Twitter:http://twitter.com/twitterapi
   Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
   Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

   --
   Twitter developer documentation and resources:
 http://dev.twitter.com/doc
   API updates via Twitter:http://twitter.com/twitterapi
   Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
   Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

  --
  Adam Green
  Twitter API Consultant and Trainer
 http://140dev.com
  @140dev

  --
  Twitter developer documentation and resources:http://dev.twitter.com/doc
  API updates via Twitter:http://twitter.com/twitterapi
  Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
  Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

   --
  Twitter developer documentation and resources:http

[twitter-dev] Re: Upgrading from Read to Read / Write access for OAuth API Key

2011-01-31 Thread Tim Bull
The way I read Abraham's note, he's saying that by simply upgrading
the token from read to read/write he was able to write?  I didn't take
it to mean he had also sent the user to reauthorise?

T

On Feb 1, 8:46 am, Tom van der Woerdt i...@tvdw.eu wrote:
 Actually, since the user needs to re-authorize the application, I do not
 think that this is a bug.

 Tom

 On 1/31/11 10:45 PM, Tim Bull wrote:







  While this makes me happy (from a developers point of view), surely
  this is a bug and therefore not to be relied on?

  As a user, I agree with the logic that if I authorised Read only, the
  application shouldn't be able to turn this into Read/Write without
  some subsequent approval.

  Tim

  On Jan 31, 1:46 pm, Abraham Williams4bra...@gmail.com  wrote:
  Taylor,

  Confirmed. I just upgraded read only tokens and was able
  to successfully send a DM.

  Thank you for finally allowing read only access tokens to be upgraded to
  read and write access tokens. This issue has been plaguing developers for
  almost a year now. Both forcing applications to ask for permission they
  didn't need if there was even a remote possibility they might want write
  permissions in the future and biting devs in the ass if they unknowingly
  built up a customer base of read only tokens.

  I hope we will continue to see fixes coming down the pipe to keep Twitter
  API a viable platform for further development.

  Thank you again,
  Abraham
  -
  Abraham Williams | Hacker Advocate | abrah.am
  @abrahamhttps://twitter.com/abraham  | github.com/abraham | blog.abrah.am
  This email is: [ ] shareable [x] ask first [ ] private.

  On Sun, Jan 30, 2011 at 11:19, Taylor Singletary

  taylorsinglet...@twitter.com  wrote:
  You'll have to re-ask your users for permission for write mode and you
  won't have any way via the API to track who is ready to read/write yet --
  you'll want to manage the conversion process yourself and track whether
  you've converted your users yet or not.

  The thinking behind this is that when your users authorized your app, they
  only authorized it for read-access. Wanting write access requires a new
  agreement with the user.

  The oauth/authorize step should now upgrade to read/write from read-only
  tokens when the user is re-challenged.

  Taylor

  On Sun, Jan 30, 2011 at 8:32 AM, Adam Green140...@gmail.com  wrote:

  So if a user authorizes an app for read access, the app can switch to
  read/write at any time without asking the users permission? Is this
  true? Anyone from Twitter have any input on this?

  On Sun, Jan 30, 2011 at 11:04 AM, Patrick Kennedykenned...@gmail.com
  wrote:
  Tim -

  1.  Changing from read to read/write won't change you API consumer
  keys or tokens.

  2.  Your application's users don't authorized for read or read/write;
  they merely use your application, which you offer as read or
  read/write to the world.  That is to say, if it's read, your
  application can only read its tweets, and if read/write, it can both
  read its own tweet and post to the world.

  I'd say go ahead and switch to read/write, given the fact that you now
  want that functionality.

  ~Patrick

  On Sat, Jan 29, 2011 at 10:24 PM, Tim Bulltim.b...@binaryplex.com
  wrote:
  We must be about the only developers in the universe that requested
  users grant only read access when we first got people to connect
 http://trunk.lytoTwitter (I think of the 40 or so apps authorized on
  my account, Trunk.ly is the only one that asks for Read only).  Never
  ask for more access than you need is my philosophy.

  Doh!

  Of course now, we want to add some Tweet out functions which require
  users grant us Write access.

  A couple of questions for the Twitter people.

  1. If we change the access in the application from read to read/write
  does this reset the API key, or will it stay the same (hoping it stays
  the same).
  2. How can I work out if existing users have authorised us for read/
  write?  I looked at
 http://developer.twitter.com/doc/get/account/verify_credentials
  but it doesn't show me what access they have.  Do I have to write,
  fail, force them to step through OAuth then post? Or is there a way of
  knowing before hand it will fail and asking them to upgrade?

  Thanks,

  Tim

  --
  Twitter developer documentation and resources:
 http://dev.twitter.com/doc
  API updates via Twitter:http://twitter.com/twitterapi
  Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
  Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

  --
  Twitter developer documentation and resources:
 http://dev.twitter.com/doc
  API updates via Twitter:http://twitter.com/twitterapi
  Issues/Enhancements Tracker:
 http://code.google.com/p/twitter-api/issues/list
  Change your membership to this group:
 http://groups.google.com/group/twitter-development-talk

  --
  Adam Green
  Twitter API Consultant and Trainer
 http

[twitter-dev] Re: Are embedded videos available through the API?

2011-03-19 Thread Tim Bull
Have you looked at embed.ly?

You can use the entities to extract the URLs really easily too
http://developer.twitter.com/pages/tweet_entities

Tim

On Mar 20, 10:44 am, Scott Wilcox sc...@dor.ky wrote:
 Hi Adam,

 I've not seen anything API side for it (for public use), I think mostly its 
 built into the NewTwitter UI. Probably rendered inline.

 It'll be interested to see Ryan or Taylor respond to this, but I doubt there 
 is anything for us to use.

 Scott.

 On 19 Mar 2011, at 23:38, Adam Green wrote:







  I have a client who wants to extract videos that are embedded in
  tweets and displayed in the new Twitter UI. I realized that I have
  never seen anything here about this issue. A check of the docs shows
  nothing on this, and using the relevant API calls for statuses doesn't
  return any fields related to embedded media. Is this available through
  the API?

  The other way I can see doing this is looking for entity URLs from
  YouTube and other video sites, but I was hoping there was something
  more direct.

-- 
Twitter developer documentation and resources: http://dev.twitter.com/doc
API updates via Twitter: http://twitter.com/twitterapi
Issues/Enhancements Tracker: http://code.google.com/p/twitter-api/issues/list
Change your membership to this group: 
http://groups.google.com/group/twitter-development-talk