As mentioned previously, the Twitter operations team will artificially increase the maximum status ID to 4294967296 this coming Friday, September 11th. This action is part of routine database upgrades and maintenance.
If your Twitter API application stores status IDs, please be sure that your datastore is configured to handle integers of that size. Thanks.
> As mentioned previously, the Twitter operations team willartificially
> increase the maximum status ID to 4294967296 this coming Friday,
> September 11th. This action is part of routine database upgrades and
> maintenance.
> If your Twitter API application stores status IDs, please be sure that
> your datastore is configured to handle integers of that size. Thanks.
Sorry, an error in phrasing. It was previously mentioned that this change was pending. We had not previously announced a date for the change.
Normally, we prefer to provide more advance notice where possible, but I'm letting you all know immediately after our operations team informed me that it was necessary to make this change on Friday.
On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com> wrote:
> May I know when and where was it mentioned that it will be > artificially increased this coming Friday?
> -- > Hwee-Boon
> On Sep 10, 2:49 am, Alex Payne <a...@twitter.com> wrote: >> As mentioned previously, the Twitter operations team willartificially >> increase the maximum status ID to 4294967296 this coming Friday, >> September 11th. This action is part of routine database upgrades and >> maintenance.
>> If your Twitter API application stores status IDs, please be sure that >> your datastore is configured to handle integers of that size. Thanks.
> As mentioned previously, the Twitter operations team will artificially
> increase the maximum status ID to 4294967296 this coming Friday,
> September 11th. This action is part of routine database upgrades and
> maintenance.
> If your Twitter API application stores status IDs, please be sure that
> your datastore is configured to handle integers of that size. Thanks.
And nobody thought about the significance of accelerating anything called a *pocolypse to be on the anniversary of a date that thousands died in a terrorist attack.... Tactful Twitter... Real Tactful
On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com> wrote:
> Sorry, an error in phrasing. It was previously mentioned that this > change was pending. We had not previously announced a date for the > change.
> Normally, we prefer to provide more advance notice where possible, but > I'm letting you all know immediately after our operations team > informed me that it was necessary to make this change on Friday.
> On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com> wrote:
> > May I know when and where was it mentioned that it will be > > artificially increased this coming Friday?
> > -- > > Hwee-Boon
> > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com> wrote: > >> As mentioned previously, the Twitter operations team willartificially > >> increase the maximum status ID to 4294967296 this coming Friday, > >> September 11th. This action is part of routine database upgrades and > >> maintenance.
> >> If your Twitter API application stores status IDs, please be sure that > >> your datastore is configured to handle integers of that size. Thanks.
> And nobody thought about the significance of accelerating anything > called a *pocolypse to be on the anniversary of a date that thousands > died in a terrorist attack.... Tactful Twitter... Real Tactful
Now, now - this is the Twitpocalypse, not the Twittocaust.
-- Dossy Shiobara | do...@panoptic.com | http://dossy.org/ Panoptic Computer Network | http://panoptic.com/ "He realized the fastest way to change is to laugh at your own folly -- then you can let go and quickly move on." (p. 70)
Twitter is in league with Al Qaida! You heard it first here, folks!
Ok, seriously, this message I wrote wasn't worth the electrons it took to transmit it... let's see if I can increase the s2n ratio:
4294967296, that an unsigned 32-bit int? ok, fair enough. i know some of my apps use signed 64bit ints, but i'm not sure about the db... will need to check... might be signed32...
Nicholas Moline wrote: > And nobody thought about the significance of accelerating anything > called a *pocolypse to be on the anniversary of a date that thousands > died in a terrorist attack.... Tactful Twitter... Real Tactful
> On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com > <mailto:a...@twitter.com>> wrote:
> Sorry, an error in phrasing. It was previously mentioned that this > change was pending. We had not previously announced a date for the > change.
> Normally, we prefer to provide more advance notice where possible, but > I'm letting you all know immediately after our operations team > informed me that it was necessary to make this change on Friday.
> On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com > <mailto:hweeb...@gmail.com>> wrote:
> > May I know when and where was it mentioned that it will be > > artificially increased this coming Friday?
> > -- > > Hwee-Boon
> > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com > <mailto:a...@twitter.com>> wrote: > >> As mentioned previously, the Twitter operations team > willartificially > >> increase the maximum status ID to 4294967296 this coming Friday, > >> September 11th. This action is part of routine database > upgrades and > >> maintenance.
> >> If your Twitter API application stores status IDs, please be > sure that > >> your datastore is configured to handle integers of that size. > Thanks.
if you were on signed32 you'd have had a problem a long time ago. not quite sure why people haven't just taken to treating/storing as strings -- sure there's a bit more overhead mem/storage-wise, but you don't have to change your code every few months.
On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
> Twitter is in league with Al Qaida! You heard it first here, folks!
> Ok, seriously, this message I wrote wasn't worth the electrons it took > to transmit it... let's see if I can increase the s2n ratio:
> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some > of my apps use signed 64bit ints, but i'm not sure about the db... will > need to check... might be signed32...
> Nicholas Moline wrote: > > And nobody thought about the significance of accelerating anything > > called a *pocolypse to be on the anniversary of a date that thousands > > died in a terrorist attack.... Tactful Twitter... Real Tactful
> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com > > <mailto:a...@twitter.com>> wrote:
> > Sorry, an error in phrasing. It was previously mentioned that this > > change was pending. We had not previously announced a date for the > > change.
> > Normally, we prefer to provide more advance notice where possible, > but > > I'm letting you all know immediately after our operations team > > informed me that it was necessary to make this change on Friday.
> > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com > > <mailto:hweeb...@gmail.com>> wrote:
> > > May I know when and where was it mentioned that it will be > > > artificially increased this coming Friday?
> > > -- > > > Hwee-Boon
> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com > > <mailto:a...@twitter.com>> wrote: > > >> As mentioned previously, the Twitter operations team > > willartificially > > >> increase the maximum status ID to 4294967296 this coming Friday, > > >> September 11th. This action is part of routine database > > upgrades and > > >> maintenance.
> > >> If your Twitter API application stores status IDs, please be > > sure that > > >> your datastore is configured to handle integers of that size. > > Thanks.
I've always just stored as 64bit integers, I'd assumed that 32bit wouldn't be enough.
Now, if it goes above 64bit then I'm screwed, because neither my language or database have built in support for that! :P
From: JDG Sent: Thursday, September 10, 2009 4:21 AM To: twitter-development-talk@googlegroups.com Subject: [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday, September 11th - make sure you can handle large status IDs!
if you were on signed32 you'd have had a problem a long time ago. not quite sure why people haven't just taken to treating/storing as strings -- sure there's a bit more overhead mem/storage-wise, but you don't have to change your code every few months.
On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
Twitter is in league with Al Qaida! You heard it first here, folks!
Ok, seriously, this message I wrote wasn't worth the electrons it took to transmit it... let's see if I can increase the s2n ratio:
4294967296, that an unsigned 32-bit int? ok, fair enough. i know some of my apps use signed 64bit ints, but i'm not sure about the db... will need to check... might be signed32...
Nicholas Moline wrote: > And nobody thought about the significance of accelerating anything > called a *pocolypse to be on the anniversary of a date that thousands > died in a terrorist attack.... Tactful Twitter... Real Tactful > > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com
> <mailto:a...@twitter.com>> wrote: > > > Sorry, an error in phrasing. It was previously mentioned that this > change was pending. We had not previously announced a date for the > change. > > Normally, we prefer to provide more advance notice where possible, but > I'm letting you all know immediately after our operations team > informed me that it was necessary to make this change on Friday. > > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com
> <mailto:hweeb...@gmail.com>> wrote: > > > > May I know when and where was it mentioned that it will be > > artificially increased this coming Friday? > > > > -- > > Hwee-Boon > > > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com
> <mailto:a...@twitter.com>> wrote: > >> As mentioned previously, the Twitter operations team > willartificially > >> increase the maximum status ID to 4294967296 this coming Friday, > >> September 11th. This action is part of routine database > upgrades and > >> maintenance. > >> > >> If your Twitter API application stores status IDs, please be > sure that > >> your datastore is configured to handle integers of that size. > Thanks. > >> > >> -- > >> Alex Payne - Platform Lead, Twitter, Inc.http://twitter.com/al3x > > > > > > -- > Alex Payne - Platform Lead, Twitter, Inc. > http://twitter.com/al3x > >
> -----Original Message----- > From: twitter-development-talk@googlegroups.com [mailto:twitter- > development-talk@googlegroups.com] On Behalf Of Alex Payne > Sent: Wednesday, September 09, 2009 1:01 PM > To: twitter-development-talk@googlegroups.com > Subject: [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday, > September 11th - make sure you can handle large status IDs!
> Sorry, an error in phrasing. It was previously mentioned that this > change was pending. We had not previously announced a date for the > change.
> Normally, we prefer to provide more advance notice where possible, but > I'm letting you all know immediately after our operations team > informed me that it was necessary to make this change on Friday.
> On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com> wrote:
> > May I know when and where was it mentioned that it will be > > artificially increased this coming Friday?
> > -- > > Hwee-Boon
> > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com> wrote: > >> As mentioned previously, the Twitter operations team > willartificially > >> increase the maximum status ID to 4294967296 this coming Friday, > >> September 11th. This action is part of routine database upgrades and > >> maintenance.
> >> If your Twitter API application stores status IDs, please be sure > that > >> your datastore is configured to handle integers of that size. > Thanks.
> > -----Original Message-----
> > From: twitter-development-talk@googlegroups.com [mailto:twitter-
> > development-talk@googlegroups.com] On Behalf Of Alex Payne
> > Sent: Wednesday, September 09, 2009 1:01 PM
> > To: twitter-development-talk@googlegroups.com
> > Subject: [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday,
> > September 11th - make sure you can handle large status IDs!
> > Sorry, an error in phrasing. It was previously mentioned that this
> > change was pending. We had not previously announced a date for the
> > change.
> > Normally, we prefer to provide more advance notice where possible, but
> > I'm letting you all know immediately after our operations team
> > informed me that it was necessary to make this change on Friday.
> > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com> wrote:
> > > May I know when and where was it mentioned that it will be
> > > artificially increased this coming Friday?
> > > --
> > > Hwee-Boon
> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com> wrote:
> > >> As mentioned previously, the Twitter operations team
> > willartificially
> > >> increase the maximum status ID to 4294967296 this coming Friday,
> > >> September 11th. This action is part of routine database upgrades and
> > >> maintenance.
> > >> If your Twitter API application stores status IDs, please be sure
> > that
> > >> your datastore is configured to handle integers of that size.
> > Thanks.
On Thu, Sep 10, 2009 at 00:12, Rob Ashton <robash...@codeofrob.com> wrote: > I've always just stored as 64bit integers, I'd assumed that 32bit > wouldn't be enough.
> Now, if it goes above 64bit then I'm screwed, because neither my language > or database have built in support for that! :P
> *From:* JDG <ghil...@gmail.com> > *Sent:* Thursday, September 10, 2009 4:21 AM > *To:* twitter-development-talk@googlegroups.com > *Subject:* [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday, > September 11th - make sure you can handle large status IDs!
> if you were on signed32 you'd have had a problem a long time ago. not quite > sure why people haven't just taken to treating/storing as strings -- sure > there's a bit more overhead mem/storage-wise, but you don't have to change > your code every few months.
> On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
>> Twitter is in league with Al Qaida! You heard it first here, folks!
>> Ok, seriously, this message I wrote wasn't worth the electrons it took >> to transmit it... let's see if I can increase the s2n ratio:
>> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some >> of my apps use signed 64bit ints, but i'm not sure about the db... will >> need to check... might be signed32...
>> Nicholas Moline wrote: >> > And nobody thought about the significance of accelerating anything >> > called a *pocolypse to be on the anniversary of a date that thousands >> > died in a terrorist attack.... Tactful Twitter... Real Tactful
>> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com >> > <mailto:a...@twitter.com>> wrote:
>> > Sorry, an error in phrasing. It was previously mentioned that this >> > change was pending. We had not previously announced a date for the >> > change.
>> > Normally, we prefer to provide more advance notice where possible, >> but >> > I'm letting you all know immediately after our operations team >> > informed me that it was necessary to make this change on Friday.
>> > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com >> > <mailto:hweeb...@gmail.com>> wrote:
>> > > May I know when and where was it mentioned that it will be >> > > artificially increased this coming Friday?
>> > > -- >> > > Hwee-Boon
>> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com >> > <mailto:a...@twitter.com>> wrote: >> > >> As mentioned previously, the Twitter operations team >> > willartificially >> > >> increase the maximum status ID to 4294967296 this coming Friday, >> > >> September 11th. This action is part of routine database >> > upgrades and >> > >> maintenance.
>> > >> If your Twitter API application stores status IDs, please be >> > sure that >> > >> your datastore is configured to handle integers of that size. >> > Thanks.
Call me crazy, but I store any data from a 3rd party in strings.
Typically, I used a text blob to store some serialized object (like
json or a python pickle) which maximizes flexibility. For the tweet
id, I think I used 64 chars.
In about 10 years, after I've cleared all the other higher priority
and more impactful optimizations, I might think about dealing with
this again.
> and if they are, just store the twos complement of the ID in the DB and do
> the math when you retrieve if it's negative. :)
> On Thu, Sep 10, 2009 at 00:12, Rob Ashton <robash...@codeofrob.com> wrote:
> > I've always just stored as 64bit integers, I'd assumed that 32bit
> > wouldn't be enough.
> > Now, if it goes above 64bit then I'm screwed, because neither my language
> > or database have built in support for that! :P
> > *From:* JDG <ghil...@gmail.com>
> > *Sent:* Thursday, September 10, 2009 4:21 AM
> > *To:* twitter-development-talk@googlegroups.com
> > *Subject:* [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday,
> > September 11th - make sure you can handle large status IDs!
> > if you were on signed32 you'd have had a problem a long time ago. not quite
> > sure why people haven't just taken to treating/storing as strings -- sure
> > there's a bit more overhead mem/storage-wise, but you don't have to change
> > your code every few months.
> > On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
> >> Twitter is in league with Al Qaida! You heard it first here, folks!
> >> Ok, seriously, this message I wrote wasn't worth the electrons it took
> >> to transmit it... let's see if I can increase the s2n ratio:
> >> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some
> >> of my apps use signed 64bit ints, but i'm not sure about the db... will
> >> need to check... might be signed32...
> >> Nicholas Moline wrote:
> >> > And nobody thought about the significance of accelerating anything
> >> > called a *pocolypse to be on the anniversary of a date that thousands
> >> > died in a terrorist attack.... Tactful Twitter... Real Tactful
> >> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com
> >> > <mailto:a...@twitter.com>> wrote:
> >> > Sorry, an error in phrasing. It was previously mentioned that this
> >> > change was pending. We had not previously announced a date for the
> >> > change.
> >> > Normally, we prefer to provide more advance notice where possible,
> >> but
> >> > I'm letting you all know immediately after our operations team
> >> > informed me that it was necessary to make this change on Friday.
> >> > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com
> >> > <mailto:hweeb...@gmail.com>> wrote:
> >> > > May I know when and where was it mentioned that it will be
> >> > > artificially increased this coming Friday?
> >> > > --
> >> > > Hwee-Boon
> >> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com
> >> > <mailto:a...@twitter.com>> wrote:
> >> > >> As mentioned previously, the Twitter operations team
> >> > willartificially
> >> > >> increase the maximum status ID to 4294967296 this coming Friday,
> >> > >> September 11th. This action is part of routine database
> >> > upgrades and
> >> > >> maintenance.
> >> > >> If your Twitter API application stores status IDs, please be
> >> > sure that
> >> > >> your datastore is configured to handle integers of that size.
> >> > Thanks.
We have been able to work with our operations team to delay the forced update until around September 21st or 22nd (over a week away). Since this twitpocalypse is based on the tweet count, it is impossible to predict exactly when it will happen and therefore we can only make projections based on current usage and possible spikes. With that being said, it *could* happen as early as Sept 16th (Wednesday), so please start updating your applications now to handle the change. We will be able to give you better estimates as the event moves closer and we will be sure to update the list when we know the exact time of the update.
Let us know if you have any questions and be sure to stock up on water and non-perishable goods :)
On Thu, Sep 10, 2009 at 9:10 AM, Ivan Kirigin<ivan.kiri...@gmail.com> wrote:
> Call me crazy, but I store any data from a 3rd party in strings. > Typically, I used a text blob to store some serialized object (like > json or a python pickle) which maximizes flexibility. For the tweet > id, I think I used 64 chars.
> In about 10 years, after I've cleared all the other higher priority > and more impactful optimizations, I might think about dealing with > this again.
> On Sep 10, 5:48 am, JDG <ghil...@gmail.com> wrote: >> and if they are, just store the twos complement of the ID in the DB and do >> the math when you retrieve if it's negative. :)
>> On Thu, Sep 10, 2009 at 00:12, Rob Ashton <robash...@codeofrob.com> wrote: >> > I've always just stored as 64bit integers, I'd assumed that 32bit >> > wouldn't be enough.
>> > Now, if it goes above 64bit then I'm screwed, because neither my language >> > or database have built in support for that! :P
>> > *From:* JDG <ghil...@gmail.com> >> > *Sent:* Thursday, September 10, 2009 4:21 AM >> > *To:* twitter-development-talk@googlegroups.com >> > *Subject:* [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday, >> > September 11th - make sure you can handle large status IDs!
>> > if you were on signed32 you'd have had a problem a long time ago. not quite >> > sure why people haven't just taken to treating/storing as strings -- sure >> > there's a bit more overhead mem/storage-wise, but you don't have to change >> > your code every few months.
>> > On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
>> >> Twitter is in league with Al Qaida! You heard it first here, folks!
>> >> Ok, seriously, this message I wrote wasn't worth the electrons it took >> >> to transmit it... let's see if I can increase the s2n ratio:
>> >> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some >> >> of my apps use signed 64bit ints, but i'm not sure about the db... will >> >> need to check... might be signed32...
>> >> Nicholas Moline wrote: >> >> > And nobody thought about the significance of accelerating anything >> >> > called a *pocolypse to be on the anniversary of a date that thousands >> >> > died in a terrorist attack.... Tactful Twitter... Real Tactful
>> >> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com >> >> > <mailto:a...@twitter.com>> wrote:
>> >> > Sorry, an error in phrasing. It was previously mentioned that this >> >> > change was pending. We had not previously announced a date for the >> >> > change.
>> >> > Normally, we prefer to provide more advance notice where possible, >> >> but >> >> > I'm letting you all know immediately after our operations team >> >> > informed me that it was necessary to make this change on Friday.
>> >> > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com >> >> > <mailto:hweeb...@gmail.com>> wrote:
>> >> > > May I know when and where was it mentioned that it will be >> >> > > artificially increased this coming Friday?
>> >> > > -- >> >> > > Hwee-Boon
>> >> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com >> >> > <mailto:a...@twitter.com>> wrote: >> >> > >> As mentioned previously, the Twitter operations team >> >> > willartificially >> >> > >> increase the maximum status ID to 4294967296 this coming Friday, >> >> > >> September 11th. This action is part of routine database >> >> > upgrades and >> >> > >> maintenance.
>> >> > >> If your Twitter API application stores status IDs, please be >> >> > sure that >> >> > >> your datastore is configured to handle integers of that size. >> >> > Thanks.
One suggestion: similar to API changes, it seems more appropriate that
if you want to force it, to do it earlier in the week, starting
Monday, rather than Friday. That leaves enough resources and hands to
stock up water and non-perishable goods rather than on a Friday.
--
Hwee-Boon
On Sep 12, 12:27 am, Ryan Sarver <rsar...@twitter.com> wrote:
> We have been able to work with our operations team to delay the forced
> update until around September 21st or 22nd (over a week away). Since
> this twitpocalypse is based on the tweet count, it is impossible to
> predict exactly when it will happen and therefore we can only make
> projections based on current usage and possible spikes. With that
> being said, it *could* happen as early as Sept 16th (Wednesday), so
> please start updating your applications now to handle the change. We
> will be able to give you better estimates as the event moves closer
> and we will be sure to update the list when we know the exact time of
> the update.
> Let us know if you have any questions and be sure to stock up on water
> and non-perishable goods :)
> Ryan
> On Thu, Sep 10, 2009 at 9:10 AM, Ivan Kirigin<ivan.kiri...@gmail.com> wrote:
> > Call me crazy, but I store any data from a 3rd party in strings.
> > Typically, I used a text blob to store some serialized object (like
> > json or a python pickle) which maximizes flexibility. For the tweet
> > id, I think I used 64 chars.
> > In about 10 years, after I've cleared all the other higher priority
> > and more impactful optimizations, I might think about dealing with
> > this again.
> > On Sep 10, 5:48 am, JDG <ghil...@gmail.com> wrote:
> >> and if they are, just store the twos complement of the ID in the DB and do
> >> the math when you retrieve if it's negative. :)
> >> On Thu, Sep 10, 2009 at 00:12, Rob Ashton <robash...@codeofrob.com> wrote:
> >> > I've always just stored as 64bit integers, I'd assumed that 32bit
> >> > wouldn't be enough.
> >> > Now, if it goes above 64bit then I'm screwed, because neither my language
> >> > or database have built in support for that! :P
> >> > *From:* JDG <ghil...@gmail.com>
> >> > *Sent:* Thursday, September 10, 2009 4:21 AM
> >> > *To:* twitter-development-talk@googlegroups.com
> >> > *Subject:* [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday,
> >> > September 11th - make sure you can handle large status IDs!
> >> > if you were on signed32 you'd have had a problem a long time ago. not quite
> >> > sure why people haven't just taken to treating/storing as strings -- sure
> >> > there's a bit more overhead mem/storage-wise, but you don't have to change
> >> > your code every few months.
> >> > On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
> >> >> Twitter is in league with Al Qaida! You heard it first here, folks!
> >> >> Ok, seriously, this message I wrote wasn't worth the electrons it took
> >> >> to transmit it... let's see if I can increase the s2n ratio:
> >> >> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some
> >> >> of my apps use signed 64bit ints, but i'm not sure about the db... will
> >> >> need to check... might be signed32...
> >> >> Nicholas Moline wrote:
> >> >> > And nobody thought about the significance of accelerating anything
> >> >> > called a *pocolypse to be on the anniversary of a date that thousands
> >> >> > died in a terrorist attack.... Tactful Twitter... Real Tactful
> >> >> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com
> >> >> > <mailto:a...@twitter.com>> wrote:
> >> >> > Sorry, an error in phrasing. It was previously mentioned that this
> >> >> > change was pending. We had not previously announced a date for the
> >> >> > change.
> >> >> > Normally, we prefer to provide more advance notice where possible,
> >> >> but
> >> >> > I'm letting you all know immediately after our operations team
> >> >> > informed me that it was necessary to make this change on Friday.
> >> >> > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com
> >> >> > <mailto:hweeb...@gmail.com>> wrote:
> >> >> > > May I know when and where was it mentioned that it will be
> >> >> > > artificially increased this coming Friday?
> >> >> > > --
> >> >> > > Hwee-Boon
> >> >> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com
> >> >> > <mailto:a...@twitter.com>> wrote:
> >> >> > >> As mentioned previously, the Twitter operations team
> >> >> > willartificially
> >> >> > >> increase the maximum status ID to 4294967296 this coming Friday,
> >> >> > >> September 11th. This action is part of routine database
> >> >> > upgrades and
> >> >> > >> maintenance.
> >> >> > >> If your Twitter API application stores status IDs, please be
> >> >> > sure that
> >> >> > >> your datastore is configured to handle integers of that size.
> >> >> > Thanks.
That is definitely part of the plan and hence why we are aiming for that Monday / Tuesday. We know what a strain it can be to push stuff out at the end of the week.
On Fri, Sep 11, 2009 at 10:27 AM, Hwee-Boon Yar <hweeb...@gmail.com> wrote:
> One suggestion: similar to API changes, it seems more appropriate that > if you want to force it, to do it earlier in the week, starting > Monday, rather than Friday. That leaves enough resources and hands to > stock up water and non-perishable goods rather than on a Friday.
> -- > Hwee-Boon
> On Sep 12, 12:27 am, Ryan Sarver <rsar...@twitter.com> wrote: >> To give everyone an update --
>> We have been able to work with our operations team to delay the forced >> update until around September 21st or 22nd (over a week away). Since >> this twitpocalypse is based on the tweet count, it is impossible to >> predict exactly when it will happen and therefore we can only make >> projections based on current usage and possible spikes. With that >> being said, it *could* happen as early as Sept 16th (Wednesday), so >> please start updating your applications now to handle the change. We >> will be able to give you better estimates as the event moves closer >> and we will be sure to update the list when we know the exact time of >> the update.
>> Let us know if you have any questions and be sure to stock up on water >> and non-perishable goods :)
>> Ryan
>> On Thu, Sep 10, 2009 at 9:10 AM, Ivan Kirigin<ivan.kiri...@gmail.com> wrote:
>> > Call me crazy, but I store any data from a 3rd party in strings. >> > Typically, I used a text blob to store some serialized object (like >> > json or a python pickle) which maximizes flexibility. For the tweet >> > id, I think I used 64 chars.
>> > In about 10 years, after I've cleared all the other higher priority >> > and more impactful optimizations, I might think about dealing with >> > this again.
>> > On Sep 10, 5:48 am, JDG <ghil...@gmail.com> wrote: >> >> and if they are, just store the twos complement of the ID in the DB and do >> >> the math when you retrieve if it's negative. :)
>> >> On Thu, Sep 10, 2009 at 00:12, Rob Ashton <robash...@codeofrob.com> wrote: >> >> > I've always just stored as 64bit integers, I'd assumed that 32bit >> >> > wouldn't be enough.
>> >> > Now, if it goes above 64bit then I'm screwed, because neither my language >> >> > or database have built in support for that! :P
>> >> > *From:* JDG <ghil...@gmail.com> >> >> > *Sent:* Thursday, September 10, 2009 4:21 AM >> >> > *To:* twitter-development-talk@googlegroups.com >> >> > *Subject:* [twitter-dev] Re: Alert: "Twitpocalypse II" coming Friday, >> >> > September 11th - make sure you can handle large status IDs!
>> >> > if you were on signed32 you'd have had a problem a long time ago. not quite >> >> > sure why people haven't just taken to treating/storing as strings -- sure >> >> > there's a bit more overhead mem/storage-wise, but you don't have to change >> >> > your code every few months.
>> >> > On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
>> >> >> Twitter is in league with Al Qaida! You heard it first here, folks!
>> >> >> Ok, seriously, this message I wrote wasn't worth the electrons it took >> >> >> to transmit it... let's see if I can increase the s2n ratio:
>> >> >> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some >> >> >> of my apps use signed 64bit ints, but i'm not sure about the db... will >> >> >> need to check... might be signed32...
>> >> >> Nicholas Moline wrote: >> >> >> > And nobody thought about the significance of accelerating anything >> >> >> > called a *pocolypse to be on the anniversary of a date that thousands >> >> >> > died in a terrorist attack.... Tactful Twitter... Real Tactful
>> >> >> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com >> >> >> > <mailto:a...@twitter.com>> wrote:
>> >> >> > Sorry, an error in phrasing. It was previously mentioned that this >> >> >> > change was pending. We had not previously announced a date for the >> >> >> > change.
>> >> >> > Normally, we prefer to provide more advance notice where possible, >> >> >> but >> >> >> > I'm letting you all know immediately after our operations team >> >> >> > informed me that it was necessary to make this change on Friday.
>> >> >> > On Wed, Sep 9, 2009 at 12:13, Hwee-Boon Yar<hweeb...@gmail.com >> >> >> > <mailto:hweeb...@gmail.com>> wrote:
>> >> >> > > May I know when and where was it mentioned that it will be >> >> >> > > artificially increased this coming Friday?
>> >> >> > > -- >> >> >> > > Hwee-Boon
>> >> >> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com >> >> >> > <mailto:a...@twitter.com>> wrote: >> >> >> > >> As mentioned previously, the Twitter operations team >> >> >> > willartificially >> >> >> > >> increase the maximum status ID to 4294967296 this coming Friday, >> >> >> > >> September 11th. This action is part of routine database >> >> >> > upgrades and >> >> >> > >> maintenance.
>> >> >> > >> If your Twitter API application stores status IDs, please be >> >> >> > sure that >> >> >> > >> your datastore is configured to handle integers of that size. >> >> >> > Thanks.
> That is definitely part of the plan and hence why we are aiming for
> that Monday / Tuesday. We know what a strain it can be to push stuff
> out at the end of the week.
> Best, Ryan
> On Fri, Sep 11, 2009 at 10:27 AM, Hwee-Boon Yar <hweeb...@gmail.com> wrote:
> > One suggestion: similar to API changes, it seems more appropriate that
> > if you want to force it, to do it earlier in the week, starting
> > Monday, rather than Friday. That leaves enough resources and hands to
> > stock up water and non-perishable goods rather than on a Friday.
> > --
> > Hwee-Boon
> > On Sep 12, 12:27 am, Ryan Sarver <rsar...@twitter.com> wrote:
> >> To give everyone an update --
> >> We have been able to work with our operations team to delay the forced
> >> update until around September 21st or 22nd (over a week away). Since
> >> thistwitpocalypseis based on the tweet count, it is impossible to
> >> predict exactly when it will happen and therefore we can only make
> >> projections based on current usage and possible spikes. With that
> >> being said, it *could* happen as early as Sept 16th (Wednesday), so
> >> please start updating your applications now to handle the change. We
> >> will be able to give you better estimates as the event moves closer
> >> and we will be sure to update the list when we know the exact time of
> >> the update.
> >> Let us know if you have any questions and be sure to stock up on water
> >> and non-perishable goods :)
> >> Ryan
> >> On Thu, Sep 10, 2009 at 9:10 AM, Ivan Kirigin<ivan.kiri...@gmail.com> wrote:
> >> > Call me crazy, but I store any data from a 3rd party in strings.
> >> > Typically, I used a text blob to store some serialized object (like
> >> > json or a python pickle) which maximizes flexibility. For the tweet
> >> > id, I think I used 64 chars.
> >> > In about 10 years, after I've cleared all the other higher priority
> >> > and more impactful optimizations, I might think about dealing with
> >> > this again.
> >> > On Sep 10, 5:48 am, JDG <ghil...@gmail.com> wrote:
> >> >> and if they are, just store the twos complement of the ID in the DB and do
> >> >> the math when you retrieve if it's negative. :)
> >> >> On Thu, Sep 10, 2009 at 00:12, Rob Ashton <robash...@codeofrob.com> wrote:
> >> >> > I've always just stored as 64bit integers, I'd assumed that 32bit
> >> >> > wouldn't be enough.
> >> >> > Now, if it goes above 64bit then I'm screwed, because neither my language
> >> >> > or database have built in support for that! :P
> >> >> > *From:* JDG <ghil...@gmail.com>
> >> >> > *Sent:* Thursday, September 10, 2009 4:21 AM
> >> >> > *To:* twitter-development-talk@googlegroups.com
> >> >> > *Subject:* [twitter-dev] Re: Alert: "TwitpocalypseII" coming Friday,
> >> >> > September 11th - make sure you can handle large status IDs!
> >> >> > if you were on signed32 you'd have had a problem a long time ago. not quite
> >> >> > sure why people haven't just taken to treating/storing as strings -- sure
> >> >> > there's a bit more overhead mem/storage-wise, but you don't have to change
> >> >> > your code every few months.
> >> >> > On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
> >> >> >> Twitter is in league with Al Qaida! You heard it first here, folks!
> >> >> >> Ok, seriously, this message I wrote wasn't worth the electrons it took
> >> >> >> to transmit it... let's see if I can increase the s2n ratio:
> >> >> >> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some
> >> >> >> of my apps use signed 64bit ints, but i'm not sure about the db... will
> >> >> >> need to check... might be signed32...
> >> >> >> Nicholas Moline wrote:
> >> >> >> > And nobody thought about the significance of accelerating anything
> >> >> >> > called a *pocolypse to be on the anniversary of a date that thousands
> >> >> >> > died in a terrorist attack.... Tactful Twitter... Real Tactful
> >> >> >> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com
> >> >> >> > <mailto:a...@twitter.com>> wrote:
> >> >> >> > Sorry, an error in phrasing. It was previously mentioned that this
> >> >> >> > change was pending. We had not previously announced a date for the
> >> >> >> > change.
> >> >> >> > Normally, we prefer to provide more advance notice where possible,
> >> >> >> but
> >> >> >> > I'm letting you all know immediately after our operations team
> >> >> >> > informed me that it was necessary to make this change on Friday.
> >> >> >> > > May I know when and where was it mentioned that it will be
> >> >> >> > > artificially increased this coming Friday?
> >> >> >> > > --
> >> >> >> > > Hwee-Boon
> >> >> >> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com
> >> >> >> > <mailto:a...@twitter.com>> wrote:
> >> >> >> > >> As mentioned previously, the Twitter operations team
> >> >> >> > willartificially
> >> >> >> > >> increase the maximum status ID to 4294967296 this coming Friday,
> >> >> >> > >> September 11th. This action is part of routine database
> >> >> >> > upgrades and
> >> >> >> > >> maintenance.
> >> >> >> > >> If your Twitter API application stores status IDs, please be
> >> >> >> > sure that
> >> >> >> > >> your datastore is configured to handle integers of that size.
> >> >> >> > Thanks.
> On Sep 11, 3:00 pm, Ryan Sarver <rsar...@twitter.com> wrote:
> > Hwee-Boon,
> > That is definitely part of the plan and hence why we are aiming for
> > that Monday / Tuesday. We know what a strain it can be to push stuff
> > out at the end of the week.
> > Best, Ryan
> > On Fri, Sep 11, 2009 at 10:27 AM, Hwee-Boon Yar <hweeb...@gmail.com> wrote:
> > > One suggestion: similar to API changes, it seems more appropriate that
> > > if you want to force it, to do it earlier in the week, starting
> > > Monday, rather than Friday. That leaves enough resources and hands to
> > > stock up water and non-perishable goods rather than on a Friday.
> > > --
> > > Hwee-Boon
> > > On Sep 12, 12:27 am, Ryan Sarver <rsar...@twitter.com> wrote:
> > >> To give everyone an update --
> > >> We have been able to work with our operations team to delay the forced
> > >> update until around September 21st or 22nd (over a week away). Since
> > >> thistwitpocalypseis based on the tweet count, it is impossible to
> > >> predict exactly when it will happen and therefore we can only make
> > >> projections based on current usage and possible spikes. With that
> > >> being said, it *could* happen as early as Sept 16th (Wednesday), so
> > >> please start updating your applications now to handle the change. We
> > >> will be able to give you better estimates as the event moves closer
> > >> and we will be sure to update the list when we know the exact time of
> > >> the update.
> > >> Let us know if you have any questions and be sure to stock up on water
> > >> and non-perishable goods :)
> > >> Ryan
> > >> On Thu, Sep 10, 2009 at 9:10 AM, Ivan Kirigin<ivan.kiri...@gmail.com> wrote:
> > >> > Call me crazy, but I store any data from a 3rd party in strings.
> > >> > Typically, I used a text blob to store some serialized object (like
> > >> > json or a python pickle) which maximizes flexibility. For the tweet
> > >> > id, I think I used 64 chars.
> > >> > In about 10 years, after I've cleared all the other higher priority
> > >> > and more impactful optimizations, I might think about dealing with
> > >> > this again.
> > >> > On Sep 10, 5:48 am, JDG <ghil...@gmail.com> wrote:
> > >> >> and if they are, just store the twos complement of the ID in the DB and do
> > >> >> the math when you retrieve if it's negative. :)
> > >> >> On Thu, Sep 10, 2009 at 00:12, Rob Ashton <robash...@codeofrob.com> wrote:
> > >> >> > I've always just stored as 64bit integers, I'd assumed that 32bit
> > >> >> > wouldn't be enough.
> > >> >> > Now, if it goes above 64bit then I'm screwed, because neither my language
> > >> >> > or database have built in support for that! :P
> > >> >> > *From:* JDG <ghil...@gmail.com>
> > >> >> > *Sent:* Thursday, September 10, 2009 4:21 AM
> > >> >> > *To:* twitter-development-talk@googlegroups.com
> > >> >> > *Subject:* [twitter-dev] Re: Alert: "TwitpocalypseII" coming Friday,
> > >> >> > September 11th - make sure you can handle large status IDs!
> > >> >> > if you were on signed32 you'd have had a problem a long time ago. not quite
> > >> >> > sure why people haven't just taken to treating/storing as strings -- sure
> > >> >> > there's a bit more overhead mem/storage-wise, but you don't have to change
> > >> >> > your code every few months.
> > >> >> > On Wed, Sep 9, 2009 at 16:45, Joseph Cheek <jos...@cheek.com> wrote:
> > >> >> >> Twitter is in league with Al Qaida! You heard it first here, folks!
> > >> >> >> Ok, seriously, this message I wrote wasn't worth the electrons it took
> > >> >> >> to transmit it... let's see if I can increase the s2n ratio:
> > >> >> >> 4294967296, that an unsigned 32-bit int? ok, fair enough. i know some
> > >> >> >> of my apps use signed 64bit ints, but i'm not sure about the db... will
> > >> >> >> need to check... might be signed32...
> > >> >> >> Nicholas Moline wrote:
> > >> >> >> > And nobody thought about the significance of accelerating anything
> > >> >> >> > called a *pocolypse to be on the anniversary of a date that thousands
> > >> >> >> > died in a terrorist attack.... Tactful Twitter... Real Tactful
> > >> >> >> > On Wed, Sep 9, 2009 at 1:00 PM, Alex Payne <a...@twitter.com
> > >> >> >> > <mailto:a...@twitter.com>> wrote:
> > >> >> >> > Sorry, an error in phrasing. It was previously mentioned that this
> > >> >> >> > change was pending. We had not previously announced a date for the
> > >> >> >> > change.
> > >> >> >> > Normally, we prefer to provide more advance notice where possible,
> > >> >> >> but
> > >> >> >> > I'm letting you all know immediately after our operations team
> > >> >> >> > informed me that it was necessary to make this change on Friday.
> > >> >> >> > > May I know when and where was it mentioned that it will be
> > >> >> >> > > artificially increased this coming Friday?
> > >> >> >> > > --
> > >> >> >> > > Hwee-Boon
> > >> >> >> > > On Sep 10, 2:49 am, Alex Payne <a...@twitter.com
> > >> >> >> > <mailto:a...@twitter.com>> wrote:
> > >> >> >> > >> As mentioned previously, the Twitter operations team
> > >> >> >> > willartificially
> > >> >> >> > >> increase the maximum status ID to 4294967296 this coming Friday,
> > >> >> >> > >> September 11th. This action is part of routine database
> > >> >> >> > upgrades and
> > >> >> >> > >> maintenance.
> > >> >> >> > >> If your Twitter API application stores status IDs, please be
> > >> >> >> > sure that
> > >> >> >> > >> your datastore is configured to handle integers of that size.
> > >> >> >> > Thanks.
