[twitter-dev] Re: Updates to the retweet API payload
I am a bit confused as to how we can use these new APIs. If I use the new retreet API in my app then no-one will see it if they are using any app that isnt using the new home_timeline (perhaps no- one) ... so how can it ever be used unless everyone (I mean client apps and sites) uses the new APIs all at the same time. Maybe someone can clarify this for me as I would like to add support but I cannot see how.
[twitter-dev] New Retweet API not working?
Is the new retweet posting api not working, I seem to just get Not Found error on any attempt to retweet a status: curl -u user:pass --http-request POST http://twitter.com/statuses/retweet/4492407684.xml ... when the tweet clearly exists: http://twitter.com/coderanger/statuses/4492407684 Am I going mad and not understanding this or something :-)
[twitter-dev] Searching by status id (Does it exist?)
We have a list of Twitter Status Id's and need to get the message content for all of the messages in a single xml stream. (non sequential and can be any # range) ex) 4540244431,3977530424,4544923774,4540244431,3977530424,4544923774 Options: 1. Use the Search API (or standard API) to get the messages in our list. It seems like we have looked everywhere to find an API method to let us do this. We cannot find one. Does it exist? If not, does anyone have a suggestion as to how to get around this problem? 2. Get Each Message Individually It would be very inefficient to have to call the /status/show/id.xml method for each message as their could be a hundred or more. Thank you in advance! Any help is much appreciated Ryan
[twitter-dev] streaming api documentation error (?)
Hi everyone, I'm in the process of implementing a consumer for the streaming API, but while perusing the documentation I noticed an inconsistency in http://apiwiki.twitter.com/Streaming-API-Documentation When a network error (TCP/IP level) is encountered, back off linearly. Perhaps start at 250 milliseconds, double, and cap at 16 seconds. Network layer problems are generally transitory and clear quickly. When a HTTP error ( 200) is returned, back off linearly. Perhaps start with a 10 second wait, double on each subsequent failure, and finally cap the wait at 240 seconds. both blocks tell to use a linear backoff, but both give an example of a geometric growth, not a linear one. So I guess the backof shpould actually be binary-exponential and maybe the doc should be updated to avoid confusion. Hope this helps someone else :)
[twitter-dev] Re: New Retweet API not working?
The retweet api hasn't been launched yet (hence the coming soon in the documentation) On Oct 2, 8:36 am, Coderanger d...@coderanger.com wrote: Is the new retweet posting api not working, I seem to just get Not Found error on any attempt to retweet a status: curl -u user:pass --http-request POSThttp://twitter.com/statuses/retweet/4492407684.xml ... when the tweet clearly exists:http://twitter.com/coderanger/statuses/4492407684 Am I going mad and not understanding this or something :-)
[twitter-dev] Re: Updates to the retweet API payload
All apps will eventually have to use home_timeline as the platform guys have said they will remove the old timeline in a matter of months after the retweet api launches On Oct 2, 8:08 am, Coderanger d...@coderanger.com wrote: I am a bit confused as to how we can use these new APIs. If I use the new retreet API in my app then no-one will see it if they are using any app that isnt using the new home_timeline (perhaps no- one) ... so how can it ever be used unless everyone (I mean client apps and sites) uses the new APIs all at the same time. Maybe someone can clarify this for me as I would like to add support but I cannot see how.
[twitter-dev] Re: streaming api documentation error (?)
Thanks, Gabriele. Fixed! -Chad On Fri, Oct 2, 2009 at 4:33 AM, gabriele renzi rff@gmail.com wrote: Hi everyone, I'm in the process of implementing a consumer for the streaming API, but while perusing the documentation I noticed an inconsistency in http://apiwiki.twitter.com/Streaming-API-Documentation When a network error (TCP/IP level) is encountered, back off linearly. Perhaps start at 250 milliseconds, double, and cap at 16 seconds. Network layer problems are generally transitory and clear quickly. When a HTTP error ( 200) is returned, back off linearly. Perhaps start with a 10 second wait, double on each subsequent failure, and finally cap the wait at 240 seconds. both blocks tell to use a linear backoff, but both give an example of a geometric growth, not a linear one. So I guess the backof shpould actually be binary-exponential and maybe the doc should be updated to avoid confusion. Hope this helps someone else :)
[twitter-dev] Re: Updates to the retweet API payload
So you have to wait until then before you can even support the retweet posting?
[twitter-dev] Re: New Retweet API not working?
Thought that meant publicly, but as platform developers you could actually implement, test and use it. Seems strange to even bother publishing it if you cant use it, if thats the case then does anyone know when its going to actually happen as its been coming soon since beginning of August?
[twitter-dev] Re: New Retweet API not working?
Actually home_timeline is also coming soon but that works, so your comment seems a bit incorrect.
[twitter-dev] Re: New Retweet API not working?
Except home_timeline doesn't contain the re-tweet stuff yet, it's just a copy of friends_timeline at the moment, but it just allows us to switch to use the new url ready so technically home_timeline is still coming soon. How I test is take a copy of the sample XML in the docs and use that to test against. On Oct 2, 11:49 am, Coderanger d...@coderanger.com wrote: Actually home_timeline is also coming soon but that works, so your comment seems a bit incorrect.
[twitter-dev] Re: Updates to the retweet API payload
Not at all, just that retweets won't show in any app using the new re- tweet API. However I expect many apps that care about their users will switch because retweets from the biggest source i.e. web will be using it! On Oct 2, 11:45 am, Coderanger d...@coderanger.com wrote: So you have to wait until then before you can even support the retweet posting?
[twitter-dev] Best way to implement caching of searches (with multithreading)?
Hi, just wondering about a best practice thing. Suppose I show results of specific Twitter searches on a web site. How would I go about caching the searches? The naive approach seems to be to first check in my own database, then do a twitter search with the since_id parameter to only get results I don't already have. Then store the results from twitter in the database, too, and return the merged results to the web site. The problem I see is that if multiple user run the same search on my web site, threading issues might occur (as each user starts a separate thread on my server). Not only could multiple twitter searches with the same since_id be executed (maybe forgivable), but trouble starts when said results are to be inserted in my local database. Different threads could attempt to insert the same messages into my database. One simple solution I could imagine: just use the message ids from twitter as the primary key in my local database. That way, multiple threads saving the same message would just overwrite the message with itself. I actually wonder if that is a common solution - to use the twitter ids as primary keys (also for users, direct messages...). I have kind of arrived at the opinion that this would be the way of least resistance, although I feel a bit uneasy about it. An alternative that came to my mind might be to have single threaded background jobs do the copying of the search results from twitter to my database, and only show the results from my cache to the web site. This would cause some lag in the time the search results would appear, but it would not be too bad. However, if I have a lot of different searches, it would become infeasible to update all of them periodically. It would become necessary to only trigger an update when a user does the search. At that point things might get overly complicated: presumably I would need some kind of Ajax solution to trigger the caching with the first request, show a spinner while the updating of my local db is going on, and then show the results from my local cache/db. The trickiest part being to prevent the starting of multiple update tasks for the same search. All in all the simple solution might be the better way to go? Would be interested in hearing your opinions, experiences and solutions! Thanks! Björn
[twitter-dev] Re: geolocation API update
I tried this with the search API, json version ( http://search.twitter.com/search.atom?lang=enq=devo) but did not see get any geo sub-object... is it on only for atom output or ..?!? ™hanks,ciao On Oct 1, 9:52 pm, Raffi Krikorian ra...@twitter.com wrote: as some of you may have already noticed, we've started going through the first steps to get the geolocation API out our door. there are a few more steps in the process that i want to share with all of you. if you start to pull status objects through the API, you'll notice that, for the majority of them, there is an empty geo/ tag and for the user objects there is a geo_enabled tag that is set to false. i say most, because, if you pull my user object curlhttp://twitter.com/users/show/raffi.xml you'll see that geo_enabled is true for me, and if you pull one of my statuses from yesterday curlhttp://twitter.com/statuses/show/4512367904.xml then you'll see a fully populated geo object at the end of that status. status ... geo xmlns:georss=http://www.georss.org/georss; georss:Point37.780300 -122.396900/georss:Point /geo /status for clarification: the geo_enabled will always be in a user object reflecting whether the user has opted-into the geolocation API. there will also always be a geo tag in the status object regardless of whether there is a location attached to the tweet or not. if there is no location, then the tag will be empty. if there is a location (as above), then the tag will be populated. just to lay out a timeline -- we've deployed for internal testing, and soon we'll be turning this on for the general audience. -- Raffi Krikorian Twitter Platform Team ra...@twitter.com | @raffi
[twitter-dev] Search API Not from a User
Is there a way to use the Search API to not return results from a selected user?
[twitter-dev] Re: Search API Not from a User
add either -from:user or from:-user to the query (i can't quite remember which). On Fri, Oct 2, 2009 at 06:44, Greg gregory.av...@gmail.com wrote: Is there a way to use the Search API to not return results from a selected user? -- Internets. Serious business.
[twitter-dev] Re: geolocation API update
Raffi, Could you tell me if the existing Twitter radius advanced searching will be tweaked to include those tweets with geo tags within the enclosed location? Thanks, brian On Oct 1, 3:52 pm, Raffi Krikorian ra...@twitter.com wrote: as some of you may have already noticed, we've started going through the first steps to get the geolocation API out our door. there are a few more steps in the process that i want to share with all of you. if you start to pull status objects through the API, you'll notice that, for the majority of them, there is an empty geo/ tag and for the user objects there is a geo_enabled tag that is set to false. i say most, because, if you pull my user object curlhttp://twitter.com/users/show/raffi.xml you'll see that geo_enabled is true for me, and if you pull one of my statuses from yesterday curlhttp://twitter.com/statuses/show/4512367904.xml then you'll see a fully populated geo object at the end of that status. status ... geo xmlns:georss=http://www.georss.org/georss; georss:Point37.780300 -122.396900/georss:Point /geo /status for clarification: the geo_enabled will always be in a user object reflecting whether the user has opted-into the geolocation API. there will also always be a geo tag in the status object regardless of whether there is a location attached to the tweet or not. if there is no location, then the tag will be empty. if there is a location (as above), then the tag will be populated. just to lay out a timeline -- we've deployed for internal testing, and soon we'll be turning this on for the general audience. -- Raffi Krikorian Twitter Platform Team ra...@twitter.com | @raffi
[twitter-dev] Re: Blog post on upcoming twitter geolocation api.....
I'm a developer, releasing an iPhone app (Myallo HotList, see http://myallo.com/hotlist ) that uses Twitter geolocation features. I replied in the blog. I think the main thing Twitter has to do is add some security to the location data. With lots of apps coming that will be able to stick geo data into every tweet, I think it is going to be easy for people to tell an app to add geo location to Twitter, and then forget about it. Being able to restrict who can see that info will be important. I think many people will only want their friends to see location. Maybe a few named friends, maybe just followers, maybe everyone except a few named users, something like that.
[twitter-dev] Re: geolocation API update
Hi. This will show up in search also - we're still in the process of rolling that out. Thanks! On Oct 2, 2009, at 5:44 AM, LucaPost lucap...@gmail.com wrote: I tried this with the search API, json version ( http://search.twitter.com/search.atom?lang=enq=devo) but did not see get any geo sub-object... is it on only for atom output or ..?!? ™hanks,ciao On Oct 1, 9:52 pm, Raffi Krikorian ra...@twitter.com wrote: as some of you may have already noticed, we've started going through the first steps to get the geolocation API out our door. there are a few more steps in the process that i want to share with all of you. if you start to pull status objects through the API, you'll notice that, for the majority of them, there is an empty geo/ tag and for the user objects there is a geo_enabled tag that is set to false. i say most, because, if you pull my user object curlhttp://twitter.com/users/show/raffi.xml you'll see that geo_enabled is true for me, and if you pull one of my statuses from yesterday curlhttp://twitter.com/statuses/show/4512367904.xml then you'll see a fully populated geo object at the end of that status. status ... geo xmlns:georss=http://www.georss.org/georss; georss:Point37.780300 -122.396900/georss:Point /geo /status for clarification: the geo_enabled will always be in a user object reflecting whether the user has opted-into the geolocation API. there will also always be a geo tag in the status object regardless of whether there is a location attached to the tweet or not. if there is no location, then the tag will be empty. if there is a location (as above), then the tag will be populated. just to lay out a timeline -- we've deployed for internal testing, and soon we'll be turning this on for the general audience. -- Raffi Krikorian Twitter Platform Team ra...@twitter.com | @raffi
[twitter-dev] Re: how are the ten trends born?
@secretbear did it first in the halcyon days of the PubSub Firehose... I'd ask him == Why not encrypt the mail you send me? You never know who's looking. If you use Firefox, why not use the FireGPG plugin to make it easy (http://getfiregpg.org) Get my key from: http://keyserver.pgp.com/ == On Thu, Oct 1, 2009 at 4:20 PM, Martin Dudek goosegoesgro...@gmail.com wrote: Good morning wonder if somebody knows how twitter determines the ten trends it declares every five minutes? Is this a pure word/phrase frequency algorithm or some more complexity behind. Thanks martin
[twitter-dev] Re: geolocation API update
Hi Brian. Yup - that's currently the plan. On Oct 2, 2009, at 6:41 AM, Brian b.kn...@gmail.com wrote: Raffi, Could you tell me if the existing Twitter radius advanced searching will be tweaked to include those tweets with geo tags within the enclosed location? Thanks, brian On Oct 1, 3:52 pm, Raffi Krikorian ra...@twitter.com wrote: as some of you may have already noticed, we've started going through the first steps to get the geolocation API out our door. there are a few more steps in the process that i want to share with all of you. if you start to pull status objects through the API, you'll notice that, for the majority of them, there is an empty geo/ tag and for the user objects there is a geo_enabled tag that is set to false. i say most, because, if you pull my user object curlhttp://twitter.com/users/show/raffi.xml you'll see that geo_enabled is true for me, and if you pull one of my statuses from yesterday curlhttp://twitter.com/statuses/show/4512367904.xml then you'll see a fully populated geo object at the end of that status. status ... geo xmlns:georss=http://www.georss.org/georss; georss:Point37.780300 -122.396900/georss:Point /geo /status for clarification: the geo_enabled will always be in a user object reflecting whether the user has opted-into the geolocation API. there will also always be a geo tag in the status object regardless of whether there is a location attached to the tweet or not. if there is no location, then the tag will be empty. if there is a location (as above), then the tag will be populated. just to lay out a timeline -- we've deployed for internal testing, and soon we'll be turning this on for the general audience. -- Raffi Krikorian Twitter Platform Team ra...@twitter.com | @raffi
[twitter-dev] Re: New Retweet API not working?
Actually I think it is great that Twitter published these retweet APIs even only as placeholders for the real thing. It communicates what the feature will look like well in advance of its release, which lets you plan its implementation. Here is a vote for publishing the lists APIs well in advance of them being publicly usable as well. If only some get that access to them in advance would be more than a little unfair. --ejw Eric Woodward Email: e...@nambu.com On Oct 2, 3:48 am, Coderanger d...@coderanger.com wrote: Thought that meant publicly, but as platform developers you could actually implement, test and use it. Seems strange to even bother publishing it if you cant use it, if thats the case then does anyone know when its going to actually happen as its been coming soon since beginning of August?
[twitter-dev] Re: Auditing apps actions
I have another issue, a user of mine claims is account was closed because my app was considered SPAM creatorI Have +3000 users and I'd like to prove my innocence. On Tue, Sep 29, 2009 at 8:13 PM, Paul Kinlan paul.kin...@gmail.com wrote: Funny you should say that I have raised a feature request about this earlier today. http://code.google.com/p/twitter-api/issues/detail?id=1081 2009/9/29 Cristovão Morgado cristovao.morg...@gmail.com Is it possible to know what application added a friendship, posted an update? Some miss behaved apps are hard to detect... :( thx -- Cristovao Morgado aka Saintr http://www.oMeuJogoUsado.com http://www.TweetaPorSMS.com http://twitter.com/TheSaintr
[twitter-dev] Twitter Stats
Looking for stats on which browsers and screen resolutions for visitors to Twitter. Thanks
[twitter-dev] List API Announcement.
As a concerned developer who just had two months of work pretty much relegated to null with one single announcement (same day I was going to release beta), I am very eager to know the extent that the List API will cover and what sort of functionality twitter will perform. When will we know more?
[twitter-dev] Re: geolocation API update
This is great news! On Oct 1, 3:52 pm, Raffi Krikorian ra...@twitter.com wrote: as some of you may have already noticed, we've started going through the first steps to get the geolocation API out our door. there are a few more steps in the process that i want to share with all of you. if you start to pull status objects through the API, you'll notice that, for the majority of them, there is an empty geo/ tag and for the user objects there is a geo_enabled tag that is set to false. i say most, because, if you pull my user object curlhttp://twitter.com/users/show/raffi.xml you'll see that geo_enabled is true for me, and if you pull one of my statuses from yesterday curlhttp://twitter.com/statuses/show/4512367904.xml then you'll see a fully populated geo object at the end of that status. status ... geo xmlns:georss=http://www.georss.org/georss; georss:Point37.780300 -122.396900/georss:Point /geo /status for clarification: the geo_enabled will always be in a user object reflecting whether the user has opted-into the geolocation API. there will also always be a geo tag in the status object regardless of whether there is a location attached to the tweet or not. if there is no location, then the tag will be empty. if there is a location (as above), then the tag will be populated. just to lay out a timeline -- we've deployed for internal testing, and soon we'll be turning this on for the general audience. -- Raffi Krikorian Twitter Platform Team ra...@twitter.com | @raffi
[twitter-dev] Re: List API Announcement.
Marcel gave a bit of info about the upcoming List API with the announcement: http://groups.google.com/group/twitter-development-talk/browse_thread/thread/d9e4ce113ea74668/12ef432e648d019b He told me that preliminary documentation should be hitting the wiki sometime next week. -Chad On Fri, Oct 2, 2009 at 11:39 AM, mattwdelong mattrugb...@gmail.com wrote: As a concerned developer who just had two months of work pretty much relegated to null with one single announcement (same day I was going to release beta), I am very eager to know the extent that the List API will cover and what sort of functionality twitter will perform. When will we know more?
[twitter-dev] Re: Auditing apps actions
If it could be proven it was your app, it would be your App that got banned, not his account. He's full of it IMO
[twitter-dev] Re: New Twitter feature API coming soon: Lists
I am also very glad to see this functionality added the Twitter platform, and personally dont care that we created groups already, as lists adds more functionality that needs to reside on the platform. It is sorely needed to make the data portable so users can use the best client in the browser or desktop or mobile devices, and take this data with them. However the announcement really was very vague. Avoiding the word groups seems strange, and tells me they then wont be implemented purely as subset timelines of the main timeline (suggestable/ followable or not), arranged by following IDs. If they are more complicated than this, and are more like suggested followings to help with user discovery (aka TweepML), I suspect many clients will retain their own group functionality, not really solving this problem (which maybe Twitter does not consider as much of a priority). If we are unable to easily display groups/lists, or whatever you call them, as subsets of the home timeline I fear a lot of client developers will end up supporting lists but also having their own Groups features. We likely would since we have built it already. Will we get to see the API as it is drafted and being tested for a period of time before it is released at all (not just a small group of Twitter friends) so that will know what is intended and have a chance to adjust development priorities? --ejw Eric Woodward Email: e...@nambu.com On Sep 30, 9:22 pm, Derek Gathright drg...@gmail.com wrote: Yay! Lists are a much needed feature and it's great to get some official word out of Twitter that they are coming. The real killer part of this feature is the ability to subscribe to other people's lists, you really hit the nail on the head with that one. The list subscription functionality I've toyed around with in my Tweenky client and most recently implemented at tweetgroups.net. Being able to create a list and allow others to subscribe to it is huge! Along with the saved searches API, the introduction of these features is great because it is one less thing the client devs have to worry about storing managing ourselves. And face it... interoperability was never going to happen with groups/lists between clients, so this is a great way to give choice back to the users. Boo to silos. On Wed, Sep 30, 2009 at 6:22 PM, dean.j.robinson dean.j.robin...@gmail.comwrote: Nice stuff! Glad I didn't spend a heap of time developing my own list/group solution for Hahlo, instead I can just drop in the API. sweet. On Oct 1, 9:13 am, Marcel Molina mar...@twitter.com wrote: Over on the main Twitter blog @nk has written about a new Lists feature we're getting ready to launch: http://blog.twitter.com/2009/09/soon-to-launch-lists.html. We just wanted to let API developers know that we'll be shipping an API for it on day one. You'll be able to do things like create lists, add and remove users from lists, find out who has been added to a list and read the tweet timeline for a given list. Stay tuned for documentations on the full Lists API soon. -- Marcel Molina Twitter Platform Teamhttp://twitter.com/noradio
[twitter-dev] New NEW retweet API and retweet text
Hi all, looking at this http://twitter.com/statuses/user_timeline/testiverse.xml It seems like retweeted status mesage now may contain more/different text, right, the retweet (now at the top) is: RT @luciuskwok: Ideas are cheap. Implementation is hard. #360iDev And the original is this: Ideas are cheap. Implementation is hard. #360iDev --- Sorry, but I am now really getting confused. Is this just a bad example and the text would normally be the same or does the statuses/ retweet method now let you change the text you want to retweet. If so, still 140 character limit? What is the proposed way to handle over-size messages. Cheers Sven
[twitter-dev] Re: Best way to implement caching of searches (with multithreading)?
Really a database is the way to go. Any modern database should allow you to check if a value is in there before inserting, so the same tweet won't go in there twice. Additionally, not every user search has to use the up to the minute results. They can go back just a little in time (30 seconds or so) to have batches in case there are multiple people searching for the same thing. On Oct 2, 7:36 am, Bjoern bjoer...@googlemail.com wrote: Hi, just wondering about a best practice thing. Suppose I show results of specific Twitter searches on a web site. How would I go about caching the searches? The naive approach seems to be to first check in my own database, then do a twitter search with the since_id parameter to only get results I don't already have. Then store the results from twitter in the database, too, and return the merged results to the web site. The problem I see is that if multiple user run the same search on my web site, threading issues might occur (as each user starts a separate thread on my server). Not only could multiple twitter searches with the same since_id be executed (maybe forgivable), but trouble starts when said results are to be inserted in my local database. Different threads could attempt to insert the same messages into my database. One simple solution I could imagine: just use the message ids from twitter as the primary key in my local database. That way, multiple threads saving the same message would just overwrite the message with itself. I actually wonder if that is a common solution - to use the twitter ids as primary keys (also for users, direct messages...). I have kind of arrived at the opinion that this would be the way of least resistance, although I feel a bit uneasy about it. An alternative that came to my mind might be to have single threaded background jobs do the copying of the search results from twitter to my database, and only show the results from my cache to the web site. This would cause some lag in the time the search results would appear, but it would not be too bad. However, if I have a lot of different searches, it would become infeasible to update all of them periodically. It would become necessary to only trigger an update when a user does the search. At that point things might get overly complicated: presumably I would need some kind of Ajax solution to trigger the caching with the first request, show a spinner while the updating of my local db is going on, and then show the results from my local cache/db. The trickiest part being to prevent the starting of multiple update tasks for the same search. All in all the simple solution might be the better way to go? Would be interested in hearing your opinions, experiences and solutions! Thanks! Björn
[twitter-dev] Re: how are the ten trends born?
It's pretty simple, but with a few twists. First of all, remember that everything that Twitter does is done with simplicity and efficiency in mind. For the most part its just a frequency count of words over a short time period, minus stop words, filtering out usernames (notice @foo is never a trend) and URLs. How it combines Wave OR Google Wave I'm unsure of, and then there's some basic spam filtering in there additionally. I was theorizing that some of it is based on accelleration of words above their standard volume. Apple for example is always talked about a great deal, but isn't always trending. Sometimes it has a greater volume than other trending words, but it doesn't trend. Yet some things stick around for a long time like IranElection. I need to dig into this more. I almost had it fully modeled at one point, then lost the code (damn you version control mistakes) dave On Oct 2, 10:54 am, Nigel Cannings nigelcanni...@googlemail.com wrote: @secretbear did it first in the halcyon days of the PubSub Firehose... I'd ask him == Why not encrypt the mail you send me? You never know who's looking. If you use Firefox, why not use the FireGPG plugin to make it easy (http://getfiregpg.org) Get my key from:http://keyserver.pgp.com/ == On Thu, Oct 1, 2009 at 4:20 PM, Martin Dudek goosegoesgro...@gmail.com wrote: Good morning wonder if somebody knows how twitter determines the ten trends it declares every five minutes? Is this a pure word/phrase frequency algorithm or some more complexity behind. Thanks martin
[twitter-dev] Re: Best way to implement caching of searches (with multithreading)?
Try memcached. - @NeluLazar On Oct 2, 7:36 am, Bjoern bjoer...@googlemail.com wrote: Hi, just wondering about a best practice thing. Suppose I show results of specific Twitter searches on a web site. How would I go about caching the searches? The naive approach seems to be to first check in my own database, then do a twitter search with the since_id parameter to only get results I don't already have. Then store the results from twitter in the database, too, and return the merged results to the web site. The problem I see is that if multiple user run the same search on my web site, threading issues might occur (as each user starts a separate thread on my server). Not only could multiple twitter searches with the same since_id be executed (maybe forgivable), but trouble starts when said results are to be inserted in my local database. Different threads could attempt to insert the same messages into my database. One simple solution I could imagine: just use the message ids from twitter as the primary key in my local database. That way, multiple threads saving the same message would just overwrite the message with itself. I actually wonder if that is a common solution - to use the twitter ids as primary keys (also for users, direct messages...). I have kind of arrived at the opinion that this would be the way of least resistance, although I feel a bit uneasy about it. An alternative that came to my mind might be to have single threaded background jobs do the copying of the search results from twitter to my database, and only show the results from my cache to the web site. This would cause some lag in the time the search results would appear, but it would not be too bad. However, if I have a lot of different searches, it would become infeasible to update all of them periodically. It would become necessary to only trigger an update when a user does the search. At that point things might get overly complicated: presumably I would need some kind of Ajax solution to trigger the caching with the first request, show a spinner while the updating of my local db is going on, and then show the results from my local cache/db. The trickiest part being to prevent the starting of multiple update tasks for the same search. All in all the simple solution might be the better way to go? Would be interested in hearing your opinions, experiences and solutions! Thanks! Björn
[twitter-dev] Re: How do I search using ticket number on the support site? (help.twitter.com)
You can visit it directly using the following url:
http://help.twitter.com/requests/{ticket_id}
replace {ticket_id} with your ticket number.
Abraham
On Thu, Oct 1, 2009 at 11:43, salamander j.sal...@gmail.com wrote:
My ticket no longer shows up under View your solved and closed
requests, and I was never given a response to my request. I wanted
to see if I can search using my ticket number to understand why it was
closed and why it no longer shows up.
Thanks,
J.
--
Abraham Williams | Community Evangelist | http://web608.org
Hacker | http://abrah.am | http://twitter.com/abraham
Project | http://fireeagle.labs.poseurtech.com
This email is: [ ] blogable [x] ask first [ ] private.
Sent from Madison, WI, United States
[twitter-dev] 400 status error while accessing RSS feed
I develop a website that is reading from a Twitter RSS feed for @democraticgovs (http://twitter.com/statuses/user_timeline/ 38477209.rss). Though we parse and display the results from the RSS feed on the home page of the site, a cron job only updates the RSS object itself in our system twice an hour. Since we launched two days ago, we've noticed that Tweets are not updating on our side. When I attempt to force the RSS feed to update internally, I get a 400 response. I realize there is a limit of 150 REST requests/hour, but I'm fairly certain we are not exceeding that threshold; since a cron job controls those request -- website visits have no affect on this number. Any ideas?
[twitter-dev] Re: how are the ten trends born?
It's definitely not going to be based on sheer volume, but rather delta based on some averages. You need to filter out natural language too, as that can be all over the place. Although it's a different (and also secret) algorithm, take a look at http://twitscoop.com and watch their trend cloud change, might help give some ideas for algorithms etc. On Oct 2, 9:00 pm, David Fisher tib...@gmail.com wrote: It's pretty simple, but with a few twists. First of all, remember that everything that Twitter does is done with simplicity and efficiency in mind. For the most part its just a frequency count of words over a short time period, minus stop words, filtering out usernames (notice @foo is never a trend) and URLs. How it combines Wave OR Google Wave I'm unsure of, and then there's some basic spam filtering in there additionally. I was theorizing that some of it is based on accelleration of words above their standard volume. Apple for example is always talked about a great deal, but isn't always trending. Sometimes it has a greater volume than other trending words, but it doesn't trend. Yet some things stick around for a long time like IranElection. I need to dig into this more. I almost had it fully modeled at one point, then lost the code (damn you version control mistakes) dave On Oct 2, 10:54 am, Nigel Cannings nigelcanni...@googlemail.com wrote: @secretbear did it first in the halcyon days of the PubSub Firehose... I'd ask him == Why not encrypt the mail you send me? You never know who's looking. If you use Firefox, why not use the FireGPG plugin to make it easy (http://getfiregpg.org) Get my key from:http://keyserver.pgp.com/ == On Thu, Oct 1, 2009 at 4:20 PM, Martin Dudek goosegoesgro...@gmail.com wrote: Good morning wonder if somebody knows how twitter determines the ten trends it declares every five minutes? Is this a pure word/phrase frequency algorithm or some more complexity behind. Thanks martin
[twitter-dev] Re: how are the ten trends born?
On Fri, Oct 2, 2009 at 1:00 PM, David Fisher tib...@gmail.com wrote: For the most part its just a frequency count of words over a short time period, minus stop words, filtering out usernames (notice @foo is never a trend) and URLs. How it combines Wave OR Google Wave I'm unsure of, and then there's some basic spam filtering in there additionally. I hope it isn't that naive -- do you know what they're doing, or are you speculating? For one thing, systems that count the unique individuals mentioning a term, rather than just raw term counts, are far more accurate in predictive modeling. Furthermore, Twitter has plenty of data to incorporate traffic and social network analysis to further improve this buzz analysis. FYI, I've been doing social network buzz analytics for about ten years and have some patents in that area (which don't belong to me, but to Nielsen/Buzzmetrics). Nick
