TBT: Crows and Shiny Things: Eight Rules, Fifteen Years of Evidence

In April 2015, Rick Roth published Crows and People Like Shiny Things… It came out of a Cambridge shop near MIT, a reflective company down the street called Reflective Technologies, and a production manager named Aaron who summed up an entire product category in six words.

The piece explained the mechanism, tiny glass beads with a reflective coating, suspended in a carrier, and then laid out eight rules.

I took those eight rules and tested them against the shirtboard archive. Twenty-plus threads on reflective, glitter, foil and metallic, running 2011 to 2024. Same method as the discharge piece and RAG from the forum we’ve all used.

The rules that hold

Reflectivity disappoints. This is the most consistent finding in fifteen years of posts. Sbrem, 2011: “It does, but unless you have a flashlight stapled to your head, it’s not all that great. I’ve found most of my customers highly unimpressed when they find out the light has to be at just the right angle.” (t2385.1) Six years later, same shop, same finding: “Reflective is NOT very spectacular; one is kind of expecting it to pop out, but the reflectivity depends on the incident light; if it’s not pretty much on axis with your eyes, there is no reflection. We have customers hold a flashlight on the side of their heads to get the idea.” (t20903.9)

In 2023, tonypep on the plastisols then available: “We have a couple here however I do not find them to be very reflecty, just shimmer really.” (t26773.0)

Ambient light is the sales problem. Dottonedan named it in one line in 2011: “Hard to sell a product when it only shows the benefit “part time”.” (t2385.2)

The carrier has to sink so the beads sit on top. Rick’s mechanism, confirmed independently at least twice. blue moon in 2011, relaying it from the ink’s formulator: “He said that the glass beads end up filtering out and staying on top.” (t82.2) Steve12Lowry in 2017, arriving at it from the shop floor: “The key is to have the base somewhat soak into the shirt so that the beads are left laying on top. This will give you max reflection.” (t20903.2)

Water-based holds up too, with a caveat Rick didn’t mention. tonypep: “Union had a water base binder for this as well that, while superior to the plastisol product, was a bitch to print.” (t4970.7) The performance is real and so is the difficulty, the same trade covered in water-based special effects.

Thicker is not brighter. Admiral ran the test in 2018: “The best reflectivity for that came with 1 pass, no underbase. 2 passes was worse, underbased was even worse with how reflective the ink was.” (t22243.8)

One dissent. blue moon in 2011: “I seem to remember Keith saying not to stroke it twice so we tried to do it with one swipe, but it looked so much better if hit again” (t2385.4).

It is expensive. brandon, 2014: “$400 a gallon. Damn, I was correct. Wow, that is a lot and you have to justify that in the pricing. I do remember it scaring the client into something else.” (t11579.3) tonypep, 2016: “Its the 3M 2 part system that beats all………. however be prepared for sticker shock. Same beads used in traffic signs.” (t17640.16) 3Deep, 2022: “3M products is just about out of the question because of it’s cost.” (t26261.0)

And there is a cost Rick’s list doesn’t cover, the catalyst clock. Orion, 2023: “ICC Optilux is okay but requires a coupler which gives it about a 12 hour shelf life. A lot of waste if you use it on an automatic press.” (t26773.3)

Where it gets complicated

Rule six, additives kill reflectivity, is half right.

The mechanism is confirmed. Doug B in 2011: “I add a little Wilflex metalic silver to Union FlashBack. The metalic decreases the reflective property, so I add to that a little scooper of 3M beads.” (t2385.6) Colin on the same physics in glitter: “Never add white or black as it will start to cloud up reallllly quickly. Your sparkle will disappear FAST.”(t11030.16)

But by 2018 shops were adding on purpose and getting more, not less. Colin: “We use the reflective ink with the silver shimmer flake added for more daylight POP, so you are not looking at JUST glass beads.” (t22243.2) ZooCity, testing the shimmer-loaded version a week later: “Great daytime look and far better reflectivity.” (t22243.12)

That is a real result against the rule as stated. The refinement: additives that cloud the beads cost you reflectivity, additives that sit beside them can buy daylight appearance. Which matters, because the ink is close to invisible in daylight otherwise. blue moon: “It is really only visible on the very dark garments like black and navy. We printed on an orange and you could hardly tell it was there.” (t20903.6)

Rule four, printed reflective can’t meet ANSI, split the board for a decade.

For it, blue moon in 2016: “3M is the only ANSI compliant reflective and that version is only available as a transfer.”(t17640.20)

Against it, AAMike in the same 2016 thread: “The 3M 2 part system is the only one the military allows us to print.”(t17640.7) A Nazdar rep said much the same in 2014  that the 3M water-based three-part ink is “generally specified in for Military, and Emergency Services applications” (t11817.5).

3M Reflective Ink for Textiles 8010, the screen-printable one, is described in its datasheet as “a low-intensity retroreflective screen printing ink intended for consumer fashion use,” measuring less than 20 cd/lux/m². The sheet advises that high-visibility safety apparel be constructed to the appropriate standard.

3M Scotchlite reflective transfer films measure more than 330 cd/lux/m² and are certified to ANSI/ISEA 107-2020, per the June 2026 datasheet. The sewn-in fabrics hit 500. More than sixteen times, from the same manufacturer.

Other suppliers now say it outright. Lancer Group’s Excalibur Retroflec plastisol states it “is not safety rated and should only be used for Halloween costumes and other novelty applications.”

Rick was right. Notably, jvanick asked the procedural question in 2016 “if you’re printing it for those needs, you have to send a sample to get it certified?” (t17640.9) — and got no reply.

One precision worth keeping. The standard specifies material performance, not application method. The barrier to certifying a printed reflective is photometric. No printed ink has been shown to hit the numbers.

What the archive never answered

The very first post in the very first reflective thread, Dottonedan, December 2011: “This looks like it will work great, but I wonder how it hold up wash after wash” (t2385.0).

Eleven replies followed over eighteen months. None answered it. Across every reflective thread in the archive there is not one wash-test result for reflective ink. Every wash figure in the corpus is for vinyl, glitter or metallic, never the reflective print itself.

Glitter, by contrast, got tested constantly, and the results are grim. tse1990: “We get about 2-4 washes in before the print fades noticably.” (t23694.16) Colin on dry-applied flake: “Otherwise 60%-80% will fall off within 3 washes.” (t11030.16) rmonks found the hidden failure mode: “I for some reason have a problem getting the ink cured, the glitter reflects the heat.” (t3042.6) The particles that make the print shiny also bounce your dryer’s energy back out of it.

Plenty of shops gave up. blue moon, 2019: “We stopped printing glitter once we found their product.” (t23694.1) meaning cut vinyl.

One more thing the archive knows

Reflective inks are particle inks, and particle inks fight your mesh. tonypep gave the best explanation in the corpus, in 2012: “When printing particle borne inks through mesh counts higher than specified you will find that the screen is acting as a filter, trapping and holding back the particles while allowing the base to pass through.” (t4970.7)

Which is why the datasheet reads like fiction to a working printer. Get Shirts, 2017: “The ink specs call for no base and a single stroke using a 160 – 230 screen, which just sounds impossible to me” (t20903.0). Shops landed on 83 to 110 and stayed there. If you are choosing mesh for a bead-loaded ink, thin thread changes what an 83 will pass.

What changed since 2015

The category split in two, and suppliers now merchandise the split. Certified safety on one side, bonded film or sewn fabric, ANSI/ISEA 107-2020, wash durability published in cycles. “Enhanced visibility” on the other, reflective ink, reflective HTV, and reflective DTF.

The archive was already sorting itself this way. ZooCity in 2018: “the “enhanced” optilux sounds about perfect for a non-ANSI reflective that’s more of a special effect than safety item.” (t22243.9) sqslabs in 2016, on a branding job: “The logos don’t have to be certified in any way, its really just trying to get the look of other brands.” (t18190.6)

Both standards are being revised right now. ISEA opened public review on a new ANSI/ISEA 107 revision on March 26, 2026, including optional testing for overall coefficient of luminous intensity. ISO 20471’s second edition finished its DIS ballot in June 2026.

PFAS is the open question. California’s AB 1817 has banned regulated PFAS in new textile articles since January 1, 2025 at 100 ppm total organic fluorine, dropping to 50 ppm January 1, 2027. New York’s ban started the same day. PPE is exempt in both, which shields certified hi-vis specifically.

What to do with reflective

Rick’s rules survive. Here is the version with fifteen years of shop talk folded in.

1. Demo before you quote. The flashlight-beside-the-head trick is in the archive twice, six years apart, because it works.
2. Ask the safety question first. If the answer is roadway, ANSI or public safety, it is film or sewn fabric. Not ink, not HTV, not DTF.
3. One pass, no underbase, low mesh. 83 to 110. The datasheet’s 160 to 230 is a filter, not a spec.
4. Price the pot life, not just the gallon. A 12-hour catalyzed ink on an auto is a waste line item.
5. Run your own wash test. The archive has fifteen years of opinion on reflective and zero wash data. I’ve run branded apparel with it for 20 years, and we pass their durability testing if printed properly.

Crows and people like shiny things. Just how shiny, or for how long?

Original: Crows and People Like Shiny Things…, Rick Roth, April 24, 2015. Forum quotes drawn from the theshirtboard.com archive — 20,820 threads, 187,100 posts — and matched character-for-character against their source posts.