API only shows messages from last 7 days

[Flashing Moose]

Hello, having some trouble with the API because only the messages from
the last 7 days show up:


example:
feed://search.twitter.com/search.atom?q=from%3AstimulusHome

Yes, there must be older posts in there... but how do i get to them?

I read about the Operator Limits:
filter:links operator:
results are limited to 7 days
source: operator:
results are limited to 7 days
queries must contain at least one keyword parameter with the source:
operator
lang= operator:
results are limited to 7 days
location operator:
results are limited to 7 days

but i'm not using filter, source, language or location do i?

regards, Moose

[Chad Etzel]

Hi Moose,

The documentation may be a bit out-dated. Right now the limit for all
searches is pretty much 7 days b/c of performance/storage reasons.
They are working on extending that window, but there is not a specific
date for having that accomplished.

-Chad

[Dewald Pretorius]

I believe the tweet retention in Twitter Search has always been 7
days.

[Abraham Williams]

It was at least a month when Twitter acquired Summize.

Abraham

2009/7/25 Dewald Pretorius <dpr...@gmail.com>

--
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.

[Flashing Moose]

Hmm, then i can't use the API for this project, thx for replies guys.

[David Fisher]

I am a bit concerned. I remember at one point it being between 30-45
days. Now it seems to be getting smaller by about 1-day per month.
Last month it was closer to 10 days.

Is it basically going to keep getting smaller and smaller until we get
V2 of the API, or will we be forced to all use only streaming services
and then locally cache everything that we'd want to search for any
time period?

I know there are a LOT of problems inherent in the massive scaling out
of Twitter, and this is just a symptom of them- but at the same time I
can only imagine how unusable Google would be if you only had a 7-day
window to Search in, and couldn't get any content made prior to that.
Very worried about this soon being a 2-3 day window.

dave

[owkaye]

I agree with you Dave. I have several thought about new
services based on searching Twitter's historical data.
Unfortunately my ideas appear to be getting less and less
practical.

Twitter claims to have all its data stored in disk-based
databases from what I understand ... yet without access to
this data it is worthless.

It seems to me they could allow searches of this historical
data via a new "History API" then let us cache the results
on our own servers. Most of the services I've conceived
would do this infrequently -- never in real time -- and
would not impact their existing "cached" server data because
this historical data would exist on separate data storage
servers ... theoretically anyways.

Owkaye

[David Fisher]

I would do anything (including paying good amounts of money) to be
able to purchase access to older datasets that I could transfer to my
database through non-rest-api methods. I'm envisioning being able to
download a CSV or SQL file that I could merge with my database easily,
but only have to make a single request to the server to get a month of
data. I'd sign agreements and pay money for such.

dave

[owkaye]

I'm sure others feel the same way Dave, but it looks and
feels like Twitter is moving in the opposite direction.

The load on a server to extract a big dataset once a month
would be minimal, and both you and I can see the value in
this approach. But I'm not sure the folks at Twitter do, or
if they do maybe they just don't have the people who can
(and will) get things like this implemented. Is a shortage
of competent staff the cause of this type of problem?

Even though I have the capabilities I do not have the
'resume' to get a job there and help them deal with some of
this stuff, nor do I have the contacts within the Twitter
organization to put a good word in for me and help me get
hired so I could do good things for them.

I'm 52 years old too, and my age seems to be a negative to
most of the Web 2.x companies hiring these days. This is
kind of a shame considering that people like me frequently
have broader-based experience and insights that are
sometimes lacking in younger people, and because of this we
can add a lot more value in the areas of planning and
structural development than people half our age. Our coding
skills are honed after so many years of experience too, not
to mention the thousands of code snippets we have collected
over the years to contribute to making us even faster.

But since jobs like this are basically not open to me and
many other folks my age, my alternative is to remain self-
employed and try to build something on top of their existing
available source data and API's ... and then deal with the
issues and frustrations created when building a service on
top of a 'moving target' that sometimes seems to be moving
in funny directions.

I hear about Twitter having lots of money to work with, and
I'm probably wrong here but it almost seems like there's too
little of this money being dedicated to paying new talent
with long term views of some of these issues, and who will
implement wise policies to help support and encourage rapid
growth in the areas that are lacking. But once again this
might just be due to a shortage of the right staff.

Obviously we cannot do anything from the outside except
point out these issues and ask questions, or beg and plead
for changes, but it sure would be great if a few of us could
actually get in there as employees and implement a couple of
the new features we really need -- such as a new Historical
Search API for example. Then developers like you and I
could proceed with some of our plans now, instead of months
or years from now ... or possibly never. I would love to
lead a team on a project like this, or even be one of its
members, but until it happens I'll focus on building my own
little space in the Twitter universe and continue to hope
for the best.

:)

Owkaye

[David Fisher]

I don't think that adding more people to the staff at Twitter is the
solution. In one startup I saw a thing posted on the refrigerator that
had the adage, "Adding more people to a project already behind
schedule will only slow it down more". Surely for support and customer
service issues having more people on the team to deal with growth is
good, but I doubt throwing more programmers at it will help fix most
issues. It just never seems to work that way.

While many startups do tend toward younger employees (I personally
think because being younger normally means that you can work a lot
with minimal life impact), I'm sure that someone with a strong
background would be able to get a job at Twitter if they were local to
the company (or willing to move).

A lot of this surely comes down to priorities inside the company.
While Doug and Team want to support us developers as much as possible,
much of our initial 'value' that we've offered in helping push twitter
to the masses has already happened. We aren't the core business
strategy, and with a fixed amount of resources and focus they aren't
working to push mainly for developer access, but for standard user
access. This 100% makes sense. Users are what is going to make twitter
happen, not 3rd party developers. They want to provide a stable
experience on both fronts, but users come first.

In my private discussions with some team members, I've gotten the
sense that they have good stuff in the pipeline for us and that they
are working hard to make it happen. However we're only a small part of
the overall strategy of a quickly growing company that is still
dealing with massive growing pains which is no fault of theirs and
something they are dealing with as best they can.

david

[owkaye]

You're probably correct when you say that throwing more
programmers at the problem is not the solution. That's not
what I was suggesting ...

My thought is that there may be no one at Twitter actually
planning or developing a plan for historical data access,
and if this is true then hiring someone with the skills and
the desire to implement this in a practical manner would go
a very long way towards providing people like us with a
workable solution now.

Having said this, I agree that in the absence of enough
people in the company who can be trusted to make wise
decisions and accomplish a wide variety of projects all at
the same time, it ends up becoming a priority issue. When
there are too few people available to actually take charge
and make progress on projects like the one we've been
discussing in this thread, it all comes down to priorities
-- and when those priorities focus on things we do not need,
the things we really want are set aside and ignored, with no
progress being made.

In other companies money is a significant limiting factor,
but I tend to question this at Twitter given all the reports
of their financial condition, so I really think it's a
priority issue in Twitter's case.

Now, if only someone at Twitter could see how important
historical data access can be to real businesses, and how
these businesses might be willing to pay for this data, then
all it would take is to hire the right person to implement
it. Twitter simply needs the money, the current ability to
recognize the future value of such a project, and the
commitment to "make it happen" ... and then they hire a
leader who gets it done.

Easier said than done of course, but there are excellent
people available who can accomplish such goals when given
the chance -- and the support they need from within the
company of course.

Then again, if these people are already working on it (as
you may have suggested) then it's going to happen one of
these days anyways ... :)

Owkaye