Thread: Cannot get a list of data centers through fed catalog

+1 on a consistent set of parameters.

I would like to see duration on the dataselect service as well.

Philip


On Tue, Aug 2, 2016 at 6:59 PM, Suzan van der Lee
<suzan<at>northwestern.edu> wrote:

Hello,

I have been wondering about the options for different web serices having to
with time series. Example: timeseries and rotation.
They both understand the options for net=, location=, starttime=, station
name=, output=, etc.
They can both convert digital counts to m/s but for one it’s via
“scale=auto” and for the other it’s “correct=scale”.
They both understand the “endtime=…” option, but only one of them can also
understand the “duration=…”, and the other doesn’t.
One of them can “demean=true”, the other cannot.
The first one shows only one channel at a time, e.g. “cha=BHZ”, while the
other can show three, “chaset=BH&components=ZNE”

I woudl love it if these options could be a bit more homogeneous. For
example
Can rotation accept the duration option?
Can the option for scaling with stage 0 sensitivity have the same syntax
between services?
Can rotation be given the demean option?

Suzan


----------------------
Web Services (http://ds.iris.edu/message-center/topic/webservices/)

Sent via IRIS Message Center (http://ds.iris.edu/message-center/)
Update subscription preferences at http://ds.iris.edu/account/profile/

Consistency is good.

You can leave the existing commands implemented for a while so people
who may have scripted them (I haven't) can change them.

On 2 Aug 2016, at 17:59, Suzan van der Lee wrote:

Hello,

I have been wondering about the options for different web serices
having to with time series. Example: timeseries and rotation.
They both understand the options for net=, location=, starttime=,
station name=, output=, etc.
They can both convert digital counts to m/s but for one it’s via
“scale=auto” and for the other it’s “correct=scale”.
They both understand the “endtime=…” option, but only one of
them can also understand the “duration=…”, and the other
doesn’t.
One of them can “demean=true”, the other cannot.
The first one shows only one channel at a time, e.g. “cha=BHZ”,
while the other can show three, “chaset=BH&components=ZNE”

I woudl love it if these options could be a bit more homogeneous. For
example
Can rotation accept the duration option?
Can the option for scaling with stage 0 sensitivity have the same
syntax between services?
Can rotation be given the demean option?

Suzan

----------------------
Web Services (http://ds.iris.edu/message-center/topic/webservices/)

Sent via IRIS Message Center (http://ds.iris.edu/message-center/)
Update subscription preferences at http://ds.iris.edu/account/profile/

================================mb===================================
Charles J. Ammon, Department of Geosciences
Penn State University / 440 Deike Bldg / University Park, PA 16802
VOICE: (814) 865 2310 / FAXES: (814) 863 7823 or (814) 863 8724
http://eqseis.geosc.psu.edu/~cammon/
http://www.personal.psu.edu/faculty/c/j/cja12/

Hi Suzan (and others that responded),

You brought up a number of different issues of varying complexity, let me try to address them one at a time.

1) A demean option for rotation. We can probably add this without too much difficulty. If there are other, specific options that you would like to see in rotation please let us know.

2) Consistent parameter for applying Stage 0 (total sensitivity) scaling to data. Most of our parameters are pretty darn consistent, this processing action is admittedly an exception. In two of the three services where we do Stage 0 scaling the option also does other and different things, so we struggled a bit with names. Here is the current situation:

irisws-timeseries http://service.iris.edu/irisws/timeseries/1/ service:
scale parameter, accepts a number to scale the sample values or auto to apply stage 0 sensitivity
correct parameter, a boolean for doing full instrument correction

irisws-rotation http://service.iris.edu/irisws/rotation/1/ service:
correct parameter, accepts one of none, scale, or response (where scale does stage 0 sensitivity)

irisws-timeseriesplot http://service.iris.edu/irisws/timeseriesplot/1/ service:
earthunits parameter, a boolean to apply stage 0 sensitivity.

Ideally, we would add a parameter to one or two of the services to provide a consistent way to request the application of the scaling in Stage 0. For example, we could add earthunits (a boolean) to irisws-timeseries and irises-rotation. Alternatively we could add scale (that accepts auto) across them all. We can't use correct, it already has conflicting values. I'd like to avoid a completely new parameter if possible. I am leaning towards adding earthunits, but it's not a perfect description. Feedback encouraged.

For those worried about us removing or changing options, don't, we will avoid that whenever possible. Options might become deprecated, even undocumented, but the current ones will work for some time barring a major conflict.

3) The duration parameter. This is a legacy parameter that we stopped using in order to provide clarity and to reduce the number of alternative ways of doing this to reduce bugs and complexity. After all, this is just an alternate way to specify an end time. The irisws-timeseries service is one of the few remaining original services still running and so still has it. If there is enough interest in this bit of syntactic convenience I would prefer to add it the same way we did in the irisws-timeseriesplot service where, if the end time is specified as a number it is interpreted as a duration in seconds instead of a date-time. As opposed to propagating a whole new 'duration' parameter.

If we really want to be consistent we would do it in every service that has 'endtime'. For the FDSN services it would be an extension of the standard and certainly not supported at any other data centers anytime soon. So if you are writing software that uses FDSN services you'd have to calculate the end time anyway since that's the only way it would work beyond IRIS (or until the specification changes and everyone updates). Worth it?

Again, feedback welcome.

4) irisws-timeseries can plot only a single seismogram while irisws-rotation can plot 3 seismograms. Making irisws-timeseries handle more than a single time series is unlikely any time soon; this is basically asking for a "bulk processed time series" service, which we simply do not have the resources to support. The alternative is to simply call irisws-timeseries multiple times as many already do.


A bit lengthy, but I hope that clarifies a few things.

Thanks for the feedback, we'll start working on a #1 and continue to discuss the rest.

Chad

On Aug 2, 2016, at 3:59 PM, Suzan van der Lee <suzan<at>northwestern.edu> wrote:

Hello,

I have been wondering about the options for different web serices having to with time series. Example: timeseries and rotation.
They both understand the options for net=, location=, starttime=, station name=, output=, etc.
They can both convert digital counts to m/s but for one it’s via “scale=auto” and for the other it’s “correct=scale”.
They both understand the “endtime=…” option, but only one of them can also understand the “duration=…”, and the other doesn’t.
One of them can “demean=true”, the other cannot.
The first one shows only one channel at a time, e.g. “cha=BHZ”, while the other can show three, “chaset=BH&components=ZNE”

I woudl love it if these options could be a bit more homogeneous. For example
Can rotation accept the duration option?
Can the option for scaling with stage 0 sensitivity have the same syntax between services?
Can rotation be given the demean option?

Suzan

----------------------
Web Services (http://ds.iris.edu/message-center/topic/webservices/)

Sent via IRIS Message Center (http://ds.iris.edu/message-center/)
Update subscription preferences at http://ds.iris.edu/account/profile/