Hexbyte  Tech News  Wired The Government Shutdown Killed an Artist’s Space Sculpture

Hexbyte Tech News Wired The Government Shutdown Killed an Artist’s Space Sculpture

Hexbyte Tech News Wired

Last December, SpaceX set a company record when it delivered a staggering 64 small satellites to space aboard a single Falcon 9 rocket. Included in this orbital “clown car” was the Orbital Reflector, a satellite designed by artist Trevor Paglen to deploy a massive, inflatable balloon coated with a reflective paint once it was in orbit. The idea was to turn the sky into a cosmic art gallery for a few months by allowing Earthlings to ponder the light reflected by the balloon as it passed overhead.

Paglen’s space sculpture was both praised for its creativity and maligned by astronomers, who grumbled about the art project disrupting observations. The astro community’s hand wringing about the responsible use of outer space turned out to be for nothing, however. As detailed in a press release released on Wednesday by the Nevada Museum of Art, which partnered with Paglen to create the Orbital Reflector, the government shutdown killed the project.

Eighteen days after Orbital Reflector was deployed into orbit, US President Donald Trump initiated what would become the longest government shutdown in the country’s history. For just over a month, 800,000 federal employees ranging from NASA scientists to air traffic controllers were on full or partial leave. This left many federal agencies scrambling to fulfill their duties, while others, like the Federal Communications Commission, shuttered almost all of their operations.

Although the FCC is perhaps best known for its part in killing net neutrality, it also plays a central role in determining what happens in orbit. Any organization that wants to send a satellite into space must first get a license with the FCC, which dictates how the satellite can communicate with Earth and vice versa. In the case of Paglen’s Orbital Reflector, the FCC ultimately had the final say in when the satellite could deploy its balloon to ensure that it didn’t interfere with other satellites.

According to the release put out by the Nevada Museum of Art, after the satellite was deployed it successfully established communication with ground stations on Earth, but the sheer number of satellites being deployed meant the Air Force was “unable to distinguish between [the satellites] and could not assign tracking numbers to many of them.” Without a NORAD tracking ID, the FCC wouldn’t give the OK to Paglen’s team to deploy the reflective balloon contained in the satellite.

Six weeks after the launch, the Air Force had only identified about half of the satellites. But even if they had identified Paglen’s satellite, it’s uncertain it would have helped. As detailed in a mid-January status report from the Nevada Museum of Art, there was no communication between Paglen’s team and the FCC, which had furloughed most of its employees.

Paglen’s team was able to maintain communication with the satellite for sometime after it was placed in orbit, but its electronics weren’t designed to function for weeks. As the government shutdown wore on, signals from the satellite became more infrequent. By the time the shutdown ended on January 22, Orbital Reflector had gone silent. WIRED has reached out to the FCC and US Air Force for further details on what went wrong and will update this post if we hear back.

Some astronomers undoubtedly felt a bit of schadenfreude when they learned that the balloon wouldn’t deploy, but in many ways Paglen’s project still managed to convey its message. As Paglen told WIRED when Timothy Sohn visited the lab building the satellite, Orbital Reflector was meant to make people think about space as a public commons and question what it means to have ownership of the final frontier. By blocking the balloon’s inflation, the US government sent a strong reminder that while space may be a truly universal commons, it’s still subject to terrestrial gatekeeping.

It’s uncertain whether Paglen will make another attempt at placing another sculpture into orbit, but given this art project’s $1.5 million price tag, that’s a big ask. WIRED has reached out to the artist about his future plans and will update this article with his comments if we hear back.


More Great WIRED Stories

Read More

Hexbyte  Tech News  Wired Riding a Scooter? Protect Your Head

Hexbyte Tech News Wired Riding a Scooter? Protect Your Head

Hexbyte Tech News Wired

In the past year, electric, shared scooters have sprouted like weeds on corners in cities as varied as San Diego, Detroit, and Barcelona. Riders have hopped aboard, neighborhood groups have objected to the scooter companies’ business models, and policymakers are contemplating rules. But another group—emergency room doctors—is raising an important question: Are these things safe?

Aarian Marshall covers autonomous vehicles, transportation policy, and urban planning for WIRED.

New research suggests that yes, people are getting hurt on scooters. (And worse: At least six people have died in the US in scooter-related crashes.) The study, a joint effort between the public health department of Austin and the Centers for Disease Control, took a close look at scooter injuries in the scoot-soaked city between September and November 2018. In 182,333 hours and 891,121 miles of riding, the researchers found 190 injuries requiring hospitalization, half of which were to the head. Fifteen percent of those injuries involved a traumatic brain injury. Just one of the injured riders—less than 1 percent of those injured—was wearing a helmet when the injury occurred.

So protect that head, the city says. “A helmet is recommended for safety in riding a scooter or a bicycle,” says Marissa Monroy, the city transportation department’s communications head, though she notes no city rules require riders to do so. Austin City Council may discuss new scooter guidelines at the end of May, when it meets to discuss its broader dockless mobility program.

For riders, the recommendation seems much easier heard than followed. Spontaneity is baked into the scooter-share experience, with riders sometimes only deciding to grab a scoot if they see one. Helmets, meanwhile, are bulky, and can be difficult to carry around. Companies like Closca have introduced collapsible helmets, and at least one dockless bike-share company, Wheels, says it will lock free helmets to each of its frames. Bird, one of the largest scooter-share companies, has sent free helmets to 65,000 riders who have requested them. (The requester pays postage.) But helmets have not caught on everywhere.

Austin’s injury report seems to comport with the conclusions of another study of scooter injuries, in Los Angeles, published in January in the journal JAMA Network Open. In that research, doctors examined the 249 scooter-riding patients who ended up in two LA-area emergency rooms between September 2017 and August 2018, and found head injuries were the most common injuries reported. (Five were hospitalized with intracranial hemorrhage.) Just five of the 100 patients admitted with head injuries were reported as wearing helmets.

“The public health impact of scooters will only grow larger as more and more people start using them,” Tarak Trivedi, an emergency physician and the lead author of the LA study, told WIRED in January. “I think there’s potential for injury, and we should all be aware and keep that in mind as we design our cities and transportation systems.”

Companies like Lime, Bird, Uber, Lyft, and Spin have said they’re working to improve the safety of their vehicles. Most companies either disburse or give riders discounts on helmets. (Seattle’s short-lived bike-share program even experimented with free helmet kiosks.) And most have pledged to build hardier scooters, custom-built to survive for longer periods on city streets.

While more and better cycling and scooter infrastructure, like protected lanes, might be welcomed by riders and companies, both the Austin and LA studies suggest that won’t prevent all scootering injuries. Seventy-four percent of the injuries studied in LA involved falls, and just 8 percent were reported as the result of a collision with another vehicle. Just 16 percent of the injury incidents in Austin during the study period involved other motorized vehicles.

Those most likely to tip over? New riders. A full third of those injured during scooter rides were on their first trips. Fifteen percent had taken more than 30. Sixty percent of injured riders later told investigators they had received in-app training from scooter providers before riding, but “additional training may be necessary, “ the Austin report says.

For Austin, this research is a starting point. Eventually, city officials hope to better understand how scooters are operating on its streets. But before they can do that, the city needs to come up with smarter ways to collect data on scooter incidents. The report does not compare injured scooter-riders with uninjured ones, nor does it compare the rate of injury on scooters to injuries on bikes, in cars, or other ways of getting around. According to a crash database maintained by the Texas Department of Transportation, 10,613 people and 4,078 vehicles were involved in crashes in the city of Austin during the study period.

The report’s numbers are conservative, its authors note, because it only captures those who sought medical care at an emergency department. “We want to get to the point where our EMS folks out there can check this mode of transportation [after an incident] when they fill out their forms, the same way they can check for an 18-wheeler or a motorcycle,” says Jen Samp, a spokesperson for the city’s public health department. The CDC may continue its work in e-scooters, as well.

And at least one company says it wants to learn from the research. “We plan to apply the insights provided by Austin Public Health’s report to our global operations, marketing campaigns, public affairs and rider education initiatives while we also further raise the bar for vehicle safety,” Paul Steely White, Bird’s safety policy director, said in a statement.


More Great WIRED Stories

Read More

Hexbyte  News  Computers eBPF can’t count?!

Hexbyte News Computers eBPF can’t count?!

Hexbyte News Computers

Grant mechanical calculating machine, public domain image

It is unlikely we can tell you anything new about the extended Berkeley Packet Filter, eBPF for short, if you’ve read all the great man pages, docs, guides, and some of our blogs out there.

But we can tell you a war story, and who doesn’t like those? This one is about how eBPF lost its ability to count for a while1.

They say in our Austin, Texas office that all good stories start with “y’all ain’t gonna believe this… tale.” This one though, starts with a post to Linux netdev mailing list from Marek Majkowski after what I heard was a long night:

Marek’s findings were quite shocking – if you subtract two 64-bit timestamps in eBPF, the result is garbage. But only when running as an unprivileged user. From root all works fine. Huh.

If you’ve seen Marek’s presentation from the Netdev 0x13 conference, you know that we are using BPF socket filters as one of the defenses against simple, volumetric DoS attacks. So potentially getting your packet count wrong could be a Bad Thing™, and affect legitimate traffic.

Let’s try to reproduce this bug with a simplified eBPF socket filter that subtracts two 64-bit unsigned integers passed to it from user-space though a BPF map. The input for our BPF program comes from a BPF array map, so that the values we operate on are not known at build time. This allows for easy experimentation and prevents the compiler from optimizing out the operations.

Starting small, eBPF, what is 2 – 1? View the code on our GitHub.

$ ./run-bpf 2 1
arg0                    2 0x0000000000000002
arg1                    1 0x0000000000000001
diff                    1 0x0000000000000001

OK, eBPF, what is 2^32 – 1?

$ ./run-bpf $[2**32] 1
arg0           4294967296 0x0000000100000000
arg1                    1 0x0000000000000001
diff 18446744073709551615 0xffffffffffffffff

Wrong! But if we ask nicely with sudo:

$ sudo ./run-bpf $[2**32] 1
[sudo] password for jkbs:
arg0           4294967296 0x0000000100000000
arg1                    1 0x0000000000000001
diff           4294967295 0x00000000ffffffff

Who is messing with my eBPF?

When computers stop subtracting, you know something big is up. We called for reinforcements.

Our colleague Arthur Fabre quickly noticed something is off when you examine the eBPF code loaded into the kernel. It turns out kernel doesn’t actually run the eBPF it’s supplied – it sometimes rewrites it first.

Any sane programmer would expect 64-bit subtraction to be expressed as a single eBPF instruction

$ llvm-objdump -S -no-show-raw-insn -section=socket1 bpf/filter.o
…
      20:       1f 76 00 00 00 00 00 00         r6 -= r7
…

However, that’s not what the kernel actually runs. Apparently after the rewrite the subtraction becomes a complex, multi-step operation.

To see what the kernel is actually running we can use little known bpftool utility. First, we need to load our BPF

$ ./run-bpf --stop-after-load 2 1
[2]+  Stopped                 ./run-bpf 2 1

Then list all BPF programs loaded into the kernel with bpftool prog list

$ sudo bpftool prog list
…
5951: socket_filter  name filter_alu64  tag 11186be60c0d0c0f  gpl
        loaded_at 2019-04-05T13:01:24+0200  uid 1000
        xlated 424B  jited 262B  memlock 4096B  map_ids 28786

The most recently loaded socket_filter must be our program (filter_alu64). Now we now know its id is 5951 and we can list its bytecode with

$ sudo bpftool prog dump xlated id 5951
…
  33: (79) r7 = *(u64 *)(r0 +0)
  34: (b4) (u32) r11 = (u32) -1
  35: (1f) r11 -= r6
  36: (4f) r11 |= r6
  37: (87) r11 = -r11
  38: (c7) r11 s>>= 63
  39: (5f) r6 &= r11
  40: (1f) r6 -= r7
  41: (7b) *(u64 *)(r10 -16) = r6
…

bpftool can also display the JITed code with: bpftool prog dump jited id 5951.

As you see, subtraction is replaced with a series of opcodes. That is unless you are root. When running from root all is good

$ sudo ./run-bpf --stop-after-load 0 0
[1]+  Stopped                 sudo ./run-bpf --stop-after-load 0 0
$ sudo bpftool prog list | grep socket_filter
659: socket_filter  name filter_alu64  tag 9e7ffb08218476f3  gpl
$ sudo bpftool prog dump xlated id 659
…
  31: (79) r7 = *(u64 *)(r0 +0)
  32: (1f) r6 -= r7
  33: (7b) *(u64 *)(r10 -16) = r6
…

If you’ve spent any time using eBPF, you must have experienced first hand the dreaded eBPF verifier. It’s a merciless judge of all eBPF code that will reject any programs that it deems not worthy of running in kernel-space.

What perhaps nobody has told you before, and what might come as a surprise, is that the very same verifier will actually also rewrite and patch up your eBPF code as needed to make it safe.

The problems with subtraction were introduced by an inconspicuous security fix to the verifier. The patch in question first landed in Linux 5.0 and was backported to 4.20.6 stable and 4.19.19 LTS kernel. The over 2000 words long commit message doesn’t spare you any details on the attack vector it targets.

The mitigation stems from CVE-2019-7308 vulnerability discovered by Jann Horn at Project Zero, which exploits pointer arithmetic, i.e. adding a scalar value to a pointer, to trigger speculative memory loads from out-of-bounds addresses. Such speculative loads change the CPU cache state and can be used to mount a Spectre variant 1 attack.

To mitigate it the eBPF verifier rewrites any arithmetic operations on pointer values in such a way the result is always a memory location within bounds. The patch demonstrates how arithmetic operations on pointers get rewritten and we can spot a familiar pattern there

Wait a minute… What pointer arithmetic? We are just trying to subtract two scalar values. How come the mitigation kicks in?

It shouldn’t. It’s a bug. The eBPF verifier keeps track of what kind of values the ALU is operating on, and in this corner case the state was ignored.

Why running BPF as root is fine, you ask? If your program has CAP_SYS_ADMIN privileges, side-channel mitigations don’t apply. As root you already have access to kernel address space, so nothing new can leak through BPF.

After our report, the fix has quickly landed in v5.0 kernel and got backported to stable kernels 4.20.15 and 4.19.28. Kudos to Daniel Borkmann for getting the fix out fast. However, kernel upgrades are hard and in the meantime we were left with code running in production that was not doing what it was supposed to.

32-bit ALU to the rescue

As one of the eBPF maintainers has pointed out, 32-bit arithmetic operations are not affected by the verifier bug. This opens a door for a work-around.

eBPF registers, r0..r10, are 64-bits wide, but you can also access just the lower 32 bits, which are exposed as subregisters w0..w10. You can operate on the 32-bit subregisters using BPF ALU32 instruction subset. LLVM 7+ can generate eBPF code that uses this instruction subset. Of course, you need to you ask it nicely with trivial -Xclang -target-feature -Xclang +alu32 toggle:

$ cat sub32.c
#include "common.h"

u32 sub32(u32 x, u32 y)
{
        return x - y;
}
$ clang -O2 -target bpf -Xclang -target-feature -Xclang +alu32 -c sub32.c
$ llvm-objdump -S -no-show-raw-insn sub32.o
…
sub32:
       0:       bc 10 00 00 00 00 00 00         w0 = w1
       1:       1c 20 00 00 00 00 00 00         w0 -= w2
       2:       95 00 00 00 00 00 00 00         exit

The 0x1c opcode of the instruction #1, which can be broken down as BPF_ALU | BPF_X | BPF_SUB (read more in the kernel docs), is the 32-bit subtraction between registers we are looking for, as opposed to regular 64-bit subtract operation 0x1f = BPF_ALU64 | BPF_X | BPF_SUB, which will get rewritten.

Armed with this knowledge we can borrow a page from bignum arithmetic and subtract 64-bit numbers using just 32-bit ops:

u64 sub64(u64 x, u64 y)
{
        u32 xh, xl, yh, yl;
        u32 hi, lo;

        xl = x;
        yl = y;
        lo = xl - yl;

        xh = x >> 32;
        yh = y >> 32;
        hi = xh - yh - (lo > xl); /* underflow? */

        return ((u64)hi << 32) | (u64)lo;
}

This code compiles as expected on normal architectures, like x86-64 or ARM64, but BPF Clang target plays by its own rules:

$ clang -O2 -target bpf -Xclang -target-feature -Xclang +alu32 -c sub64.c -o - 
  | llvm-objdump -S -
…  
      13:       1f 40 00 00 00 00 00 00         r0 -= r4
      14:       1f 30 00 00 00 00 00 00         r0 -= r3
      15:       1f 21 00 00 00 00 00 00         r1 -= r2
      16:       67 00 00 00 20 00 

Read More

Hexbyte  News  Computers Full Disk Images of Earth from GOES-17

Hexbyte News Computers Full Disk Images of Earth from GOES-17

Hexbyte News Computers

Background and Introduction

Since 1975, NOAA has maintained multiple geostationary satellites to aid in weather forecasting and research. These are called GOES satellites (Geostationary Operational Environmental Satellite). The first GOES satellite in 1975 was initially called GOES-A during launch and then renamed to GOES-1 upon reaching in final location in the sky. The most recent satellite is GOES-17, which was launched in March 2018 as GOES-S and recently reached its final operating point above the Pacific Ocean. There are two other GOES satellites in orbit GOES-15 (replaced by GOES-17) and GOES-16 (Atlantic coverage).

There is a great overview of GOES-16 and 17 here. Here is a map of the coverage:

Hexbyte  News  Computers

NOAA also has a number of weather satellites in a polar orbit. These include NOAA 15, 18, and 19. These are at a lower elevation and are much easier to receive – I’ve gotten acceptable images using a $30 SDR and a 2m dipole antenna. Since those satellites are in a polar orbit they are only “visible” when they are passing overhead. It is typical to get a few passes a day of 5-15 usable minutes. Its really fun to record these and I hope to post a separate article about it soon.

Polar satellites are able to make some measurements more accurately due to their lower elevation (ocean surface temperature, for example). They also cover the entire earth since the planet is rotating beneath them. For real-time forecasting, the GOES satellites have a strong advantage. Since they orbit at the same speed as the earth is spinning, Geostationary satellites maintain their position relative to earth. This means a parabolic dish can be pointed at the satellite and receive continuous downloads 24/7/365. According to the GOES-17 schedule a full disk image of the earth is typically available every 10 minutes and some isolated regions are updated more frequently. Here is an example of one of the full disk images I received:

Hexbyte  News  Computers

The continental US is in the upper-right corner of the globe from this angle. The images are 5424 × 5424 pixels. In digital camera terms that would be equivalent to a 29MP camera.

Here is a cropped photo from the area around Baja:

Hexbyte  News  Computers

And, here is an animation of 48 consecutive full disk images (there is a full resolution video of this at the end of the article):

Hexbyte  News  Computers

Hardware for Receiving GOES Satellites

The is the hardware I am using for receiving:

Hexbyte  News  Computers

From my location, this gets me an verterbi error rate of 400-500 (explained more later). This is acceptable, but a lower rate would produce cleaner images.

The setup consists of:

  • Airspy Mini (link)
  • Parabolic Dish Antenna (I am using this one, with 19dBi gain)
  • Bandpass Filter + LNA (I am using this one)
  • N-female to SMA female adapter.
  • (2) SMA male to SMA male cables
  • Tripod with Mount OR sturdy pole with a diameter <= 50mm to mount the antenna.

The a

Read More

Hexbyte – Tech News – Ars Technica | After Dragon launch, NASA will consider using the same Falcon 9 three times

Hexbyte – Tech News – Ars Technica | After Dragon launch, NASA will consider using the same Falcon 9 three times

Hexbyte – Tech News – Ars Technica |

Fly, Falcon, fly —

NASA explains why it was OK with scrubbing launch for a recovery issue.


Hexbyte - Tech News - Ars Technica | An infrared view of the Falcon 9 first stage landing on Saturday morning.

Enlarge / An infrared view of the Falcon 9 first stage landing on Saturday morning.

SpaceX

3:30am ET Saturday Update: The Falcon 9 rocket launched on time early Saturday morning from Cape Canaveral Air Force Station in Florida, streaking into the nighttime sky and depositing the Cargo Dragon spacecraft into a good orbit. About 12 minutes after the launch, Dragon deployed its solar panels and set course for the International Space Station. Dragon will be captured by the station on Sunday.

Meanwhile, the rocket’s first stage returned to the Of Course I Still Love You droneship, landing 28km downrange from the launch site. The nighttime landing sequence was visible through the use of an infrared camera, and looked pretty spectacular.

During the post-launch news conference for SpaceX’s 17th mission to the International Space Station, the lab’s Operations Integration Manager, Kenny Todd, explained the thinking behind the scrub of the Falcon 9 launch 24 hours earlier. SpaceX had said the launch was scrubbed because of an electrical issue with the droneship, which returned to port Friday, and then immediately headed back out to sea for the first stage landing.

“When we went through the countdown yesterday the SpaceX team was working hard, there was a lot going on, and I’ve got to applaud them because everything that was hitting the screen they were dealing with,” Todd said early Saturday morning. “There was a lot of talk on the loops when you consider the weather, the wind, the issues with the drone ship, the helium leak,” he said.

The helium leak Todd referenced occurred with ground-based equipment. According to SpaceX, the company was monitoring the leak in the supply to helium on-board the rocket, but probably could have worked through the issue. Engineers wouldn’t have known for sure until the last minutes of the countdown, however.

Todd said knowing the agency had a back-up day Saturday for the launch made a big difference. “We knew we had today, we knew the weather was supposed to be much better today,” he said. “In the end, SpaceX had to make the call. But I think one of our senior engineers whose watched an incredible number of these missions said, ‘You know, sometimes the universe is talking to you, and sometimes you need to listen to it.’ And the realit

Read More

Hexbyte – Tech News – Ars Technica | After Nevada’s outrage over plutonium shipment, Energy Dept offers olive branch

Hexbyte – Tech News – Ars Technica | After Nevada’s outrage over plutonium shipment, Energy Dept offers olive branch

Hexbyte – Tech News – Ars Technica |

compromise —

The move out of Nevada will begin in 2021 and last until 2026.


Hexbyte - Tech News - Ars Technica | An aerial view of the Device Assembly Facility.

Enlarge / The Device Assembly Facility at the Nevada National Security Site.

Nevada National Security Site

This week, Energy Department Secretary Rick Perry sent a letter to Nevada Senator Catherine Cortez Masto saying that the Department would move a half-ton load of plutonium waste out of Nevada between 2021 and 2026.

Last year, the federal government said it would move plutonium from South Carolina to the Nevada National Security Site (NNSS) for temporary storage before moving it on to its final destination. The state of Nevada sued the federal government in the hopes of stopping the planned shipment, but months into the legal action, the federal government told the court that the plutonium had already been secretly delivered to the NNSS.

Nevada politicians were incensed, and Cortez Masto, who holds a seat on the Senate Committee on Energy and Natural Resources, placed holds on any pending Trump administration nominations to the Department of Energy.

The Associated Press reports that Cortez Masto said she would drop the holds she placed on nomination proceedings for DOE employees after receiving the letter from Perry promising to move the shipped plutonium out of her state. Perry’s letter also promised that no future shipments from South Carolina would be made to Nevada.

The shipment of plutonium occurred as part of a separate 2017 order from the South Carolina US District Court, which required that the DOE remove one ton of defense plutonium from the site of the failed Mixed Oxides Fuel Fabrication Facility (MFFF). The M

Read More

Hexbyte – News – Science/Nature | ‘No image will surpass this’: Hubble telescope astronomers created a stunning picture of the deep universe with 16 years’ worth of photos – Business Insider

Hexbyte – News – Science/Nature | ‘No image will surpass this’: Hubble telescope astronomers created a stunning picture of the deep universe with 16 years’ worth of photos – Business Insider

Hexbyte – News – Science/Nature |

Astronomers on Thursday published a remarkable new picture of the deep universe. The image, shown in full at the end of this post, contains perhaps 265,000 visible galaxies crammed into a region smaller than the moon’s apparent size in the sky.

The picture is actually made of 7,500 photos taken over 16 years by the Hubble Space Telescope, which is operated by NASA and the European Space Agency. If Hubble had taken the images in one back-to-back observation, it would have lasted 250 days.

“No image will surpass this one until future space telescopes like James Webb are launched,” Garth Illingworth, an astronomer at the University of California at Santa Cruz, said in a press release.

The new picture is part of an ongoing project known as the Hubble Legacy Field. The idea is to focus Hubble’s limited time yet incredible resolving power on a small area of the night sky, year after year, and build the deepest and most complete image of space. That small survey will then be applied to the larger universe to improve astronomers’ understanding of it across both space and time.

The project got its start in 1995 when the telescope snapped its first and famous Hubble Deep Field image. For that photo, Hubble targeted one of the darkest patches of the night sky and observed it over 10 days, taking more than 340 photos of the spot. Scientists combined the images into a picture that revealed the glow of several hundred never-before-seen galaxies, stretching our understanding of the universe’s scale and history.

A close-up of the Hubble Ultra Deep Field image.
NASA/ESA/STScI

Since then, astronauts have flown out to Hubble several times to make repairs, upgrade cameras, and install new hardware, improving the observatory’s view of deep space.

Astronomers have exploited those improvements to not only enhance the original Deep Field image but build out the view of space around it.

A map showing various Hubble observations (colored outlines) that make up its “legacy field” image (white outline).
NASA/ESA; G. Illingworth and D. Magee/University of California, Santa Cruz; K. Whitaker/University of Connecticut; R. Bouwens/Leiden University; P. Oesch/University of Geneva; Hubble Legacy Field team

“Now that we have gone wider than in previous surveys, we are harvesting many more distant galaxies in the largest such dataset ever produced,” Illingworth said.

Read more: The Hubble telescope recorded an asteroid ripping itself to pieces

The colors in the photo span from just beyond the edges of human vision — from ultraviolet to near-infrared light — and contain galaxies 10 billion times fainter than our naked eyes can detect.

This effectively offers a deeper view into the past than ever before: a look at galaxies as they existed 13.3 billion years ago. That’s how long it has taken their light to reach Hubble’s sensors and shows a time about 500 million years after the birth of the universe.

Before Hubble, the best telescopes could see light from objects only about 7 billion light-years away.

Zoom in on more than 200,000 galaxies seen by Hubble

Below, you can explore the full-size Hubble image.

Printed at photo-quality resolution, it’d stretch more than seven feet (2.16 meters) per side.

Drag on the picture to pan around, and enlarge any part by using the +/- buttons, pinching on a phone, or using a scroll feature on a computer.

“Hubble has spent more time on this small area than on any other region of the sky,” the press release said.

Researchers will continue adding to and improving the picture with new Hubble observations as long as the telescope is operational. (When Hubble shuts down, NASA may try to plunge it into a “spacecraft graveyard” in the Pacific Ocean.)

But once the superpowerful James Webb Space Telescope launches and enormous new ground observatories open, our view of this patch of the night sky — and knowledge about the farthest reaches of space and time — will only improve.

More:

Hubble Space Telescope
Galaxies
Space Images
Astronomy