Gravitational Wave Open Science Center “shares” data for earth and solar system gravitational imaging research
The Auxillary data might contain data useful for earth and solar system gravitational imaging. Not some distant event (GW170814). So all the useful reference signals and events are at different times. Can we get data for times that mean something on earth? – earthquakes, tsunamis, volcanic eruptions, focus on ocean currents, correlate with superconducting gravimeter and MEMS gravimeter arrays? Link to “big G” experiments that are also sensitive enough to pick up strong or well characterized signals?
You just dumped the 68 GB full data without an index. What is in the dataset, how many tables and records, what are the units. One stupid lump is so discourteous to anyone I cannot imagine how you cannot know better.
“A large number of sensors” – can you be more vague?
“data recorded in around 500 channels at each LIGO site” – still not a single useful data structure, counts of field values, sizes, units.
“three hours of time centered on” — have you not human empathy at all? You are presumably trying to attract new people and groups to work with you. You HAVE to use common language. Not jargon that only you know. What is the exact date and time period? That event was important to your PIs and people who happened to be working then. But if it is across the universe somewhere it is not going to be useful for earth and solar system gravitational imaging, development and characterization of reference signals, helping save lives, calibrating climate models, imaging magma flows. Those are here on the earth. You have to think differently and you need to use much better practices on the Internet.
If you have a choice, give plain English terms and common references. Remember that English is going to be the second language for many. And they are NOT going to have already memorized all your jargon. Be liberal with the use of hoverboxes (hover ANYTHING that has depth and more related things and show that in a hoverbox that can be moved, resized, minimized, dragged and grouped. or hovered itself. Try it. You can stay one one screen and hover ten levels deep if needed, save that sequence for reference, and then stay in context on the piece you are trying to build or use.
Your download is only a tiny 68 GB and should download to my computer in a couple of hours. But you only allocate a tiny bandwidth, and it will take half a day. Without the data structure, field value counts, summary statistics, units and dimensions and known mathematical relations, it is not much better than a lump of coal or a dirt clod.
Give me ten minutes worth of summary data and a complete view of the kinds of data. ACTUAL DATA, not words talking about data. Let people see the data. Let them get used to it. Learn its ranges and frequencies. Learn what connects, where the data comes from — all without ever leaving one screen (or several if you have more than one physical display). Or a group, if you are running a large group exercise.
Your NDS is problematic. I have no idea what tools are there. It jumps into “install” and “use” and i have no idea if it will even be useful. You are probably heavily biased toward strain data, and have never really tried to correlate with anything on earth or the sun or solar system. I dread even trying. What is the average time it takes anyone to set up, learn and then actually apply the tools you are offering. Is it minutes or hours? Or are you expecting people to get on the ground running in a few years of painful and confusing time?
The Japan earthquake registered on both the superconducting gravimeter array, and the IRIS.edu broadband seismometers. None of them were designed for earth-based dynamic signals. but they are all that are available. The MEMS and atom interferometer and other gravimeters that are sensitive enough are still all in development. The LIGO followon is probably going to use some combination of those technologies. I have a few at GravityNotes.Org, but I have not put the last two years of notes on groups online.
There are about 10 basic approaches and 40 to 100 groups. I cannot follow them all, but I try. There are dozens of signals that can be used for local calibration of low cost gravimeters – people can move around tracked by cameras, moving masses of many shapes, cars on highways, trucks, trains. Any of the atom interferometer gravimeters arrays should be able to track low flying planes. They can certainly pick up nearby magnitude 5 earth quakes.
The Auxillary data that can be used for earth and solar system problems — tracking the sun, moon and planets, imaging the atmospheric and ocean and earth interior currents, tracking the spreading seismic waves from there direct “elastogravity” signal — are not being used by LIGO. Rather than measure what signals are on the earth and in the solar system, you have tried to damp or suppress. Better would be to measure, apply locally, build global communities to learn about gravitational sensor arrays for local purposes, then work together to remove the “gravitational noise”, “Newtonian noise” from the Gravitational Wave experiment data streams from remote places in the universe.
Who put https://gwpy.github.io/ ? Again, “we know how to do this, you must know who we are and already what we are doing, we only give you install-and-try-it”. That is not welcoming, it is intimidating and off-putting. Most new people are shy and do not want to ask questions. They want to see what is going on first. There are outgoing people, but check. They are in the minority.
Do you understand what I am saying? ALL your websites are independent and uncoordinated. The people are working on many different things and the last thing they care about is new users and looking at the whole of the Internet and the whole rest of the world for what is happening.
LIGO is so turned inward, it has lost perspective. I say that because you can get a handful of people trying your data. You make it about a hundred times harder than it need be. And give them no context for where that tiny experience fits into the whole of the world gravitational industry that is growing rapidly.
The technologies have dropped in cost of hardware. The software is horrendously expensive to learn and apply because all the developers are using terrible methods for sharing. They all do what you are doing “Here is a big lump of data” we don’t know what to do with” — with no index or summary statistics, no units and dimensions, no fundamental links to the symbolic mathematical models involved, let alone links to working algorithms in a standard form.
site:github.com “gravitational” has 16,400 entry points. And you provide a bad link to ONE of them? THAT IS NOT HELPFUL, it just introduces confusion and wasted time. There are 1.92 Billion kids between 5 and 20 in the world and hundreds of millions who can handle basic mathematics and computer tools. ALL of them are growing up using the computer to explore and try things. They expect data to be already tied to basic browsing and analysis and visualizations. Then they ask, “Where are the tools for me to try my own?” and they are NOT going to all have the skills or time to learn it they way you did it, nor the way you are doing it.
Do not force people to come to you. Find what they are doing, and give them tools and data in a form that allows free exploration and global communities. Not the tools you had to build under contract not allowed to do it right. https://www.gw-openscience.org/software/ does not tell what language, operating system, hardware requirements, cost and average time to learn. Stop putting long sequences of pages.
More and more people find things by search engine. They jump directly to where they want to go. And expect it to be well documented, complete, and ALL the necessary pieces to do something. See something. Work on something – even it it is just merging symbolic math and the associated compiled native code executable and datasets.
Richard Collins, Director, The Internet Foundation