These blog posts are definitely demonstrating the “I didn’t have time to make it shorter” problem. When I do something formal for Nine Lives, I do a lot more editing, including trying to tighten things up even after I have a complete draft. I don’t do that here, because it would take too long and then I’d not just be behind on logging, but I’d be outright skipping stuff that Future Karawynn might need to know. (I hate that I can’t trust my long-term memory anymore, but that’s the reality I have to work with.)
Wall-mounted soundboards are only an option for dogs who are comfortable activating with nose instead of paws; fortunately for us, Tashi likes to boop his snoot. The other drawback to wall boards is that they’re not accessible for cats. Although when I started, I was hoping to be able to use the same board for cat(s) and dog, I mostly gave up on that plan some months ago, and at this point — given Rikki’s awkward combo of persistence and incomprehension — having a dog-only board is a feature, not a bug. Rikki already has his own countertop tile of two buttons; Gracie is showing more interest there after watching Rikki than she ever showed at the big floor soundboard. So I’ve already pretty much committed to having species-separated boards.
At first I didn’t think we possessed enough free wall space anywhere in our house to make it work, but when I actually measured everything and drew it out, I realized that wasn’t true at all. With floor soundboards, you often have to limit the buttons to two deep and leave lots of walking space between tiles (at least for dogs; cats are usually more graceful and precise), and many learners (dog and cat) also require a walkway between the tiles and any nearby walls. If you don’t, you risk having buttons that your pet just doesn’t use because they’re too inconvenient, or because they have to turn their back to you and/or the room to press.
With a wall board, none of that matters, so I can fit many more tiles and buttons in a much smaller area. The open wall I had to work with was a T-intersection of semi-hallway. The bottom of the T points to the open-plan kitchen/living/dining room; each of the arms point to a bedroom/bathroom combo. Positionally, the T shape gives us three walls that could be utilized; however, the kitchen/living room side has been renovated while the bedroom/bathroom side has not. (It’s in the plan for 2024.) So at the moment, what I have is one large and one small stretch of wall, opposite each other and a meter apart.
Total, I calculated that the large wall would hold 26 tiles … although until the rest of the hallway floor was finished I should probably keep it to 23 or at most 24. The tiny wall on the other side would hold five. Altogether that gave me 31 tiles, which is already the most I could possibly have managed to cram into our floor space. And in the long term, if Tashi turns out to be some kind of canine language prodigy, I can put more tiles at the top of the T after that wall is redone.
As with most things, once something becomes my central focus, it comes together quickly — especially if it has a chance of solving some existing problems. It was three days, or maybe four? of research and planning between my realizing we might have enough space after all, and the actual move.
Here’s the starting plan I came up with (click through for a larger, word-legible version):
And here’s the actual board as of January 8:
The small wall is directly across from the purple and orange tiles on the large wall. (Jak and our housecleaner each separately remarked that the new board looked like a world map; my amused reply was that I guess the solo tile — which is actually the vanguard for what I hope will be a larger section of paired-antonym descriptor words — is Australia.)
If you compare it to my prior floor layouts, you can see that while I mostly kept the same types of groupings, I changed four of the six colors around. I would have preferred a nice rainbow, but the color designations were basically a function of how many tiles I now expect to need for each section versus the number of tiles of each color I had already purchased.
I also decided to selectively dial back on words in the short term, from 63 to 49, as Tashi was still recovering from his learned helplessness and I figured horizontal to vertical was already a pretty big change. (With two or three exceptions, I’m not expecting those removals to be permanent — and in fact by the time I’m actually finishing this post, thirteen days later, I’ve re-added five of the fourteen words I held back, in between several new ones. I’ll go over word specifics in a separate post.)
Logistics of Installing a Wall Board
This section is ‘things I learned along the way,’ mostly in case this is ever read by someone else who is thinking about putting their board on a wall instead of the floor.
I spoke to several people about attachment options; the clear winner was 3M Connect strips. However, those babies are not cheap even in the States, and while I can, sort-of, get them in Mexico, they are more than twice as expensive. It would have cost me about US$70 just to get those first seventeen tiles up, which is … yikes.
I talked the situation over with Jak; he was pretty enthusiastic about the idea of the board coming off the floor and was not at all bothered about the idea that taking the tiles down might mess up the wall paint someday. We own this house outright and our current expectation is that we will stay here for about another decade; we’ll have to repaint the wall then anyway if not before. And I had hundreds of little 5/8-inch velcro sticky dots already, originally purchased over a year ago to attach various organizational apparatuses inside our kitchen drawers.
So, velcro dots it was. I cleaned the wall and all of the tiles (and buttons) thoroughly with alcohol first. Figured out pretty quickly that the sticky bits need to go on the thick places, not where there’s a button cutout. Even fully loaded with six buttons, the hextiles are pretty lightweight; I put nine dots on each hextile — one on each corner and three more towards the middle — which may be overkill, but there was no reason to be conservative.
That all went fine at first, but I have one more tip for people who are leaving expansion space in the center of two groups, like I was: measure every row with a temporary tile(s). I put a temporary tile on the top row as I was spacing across, and just assumed that as I attached the tiles below it would all work out.
It did not. Apparently there is a significant amount of slop in how the hexes fit together, because this past week when I went to add the two body-part tiles in the lower space between verbs and objects, the space was very much too small. I had to take down the entire right-hand ‘continent’, scoot it over, and reattach it.
The velcro dot adhesive does pull off a tiny bit of the black rubber bottom off the hextile when removed — not a big deal once or twice, but not the sort of thing you’d want to do over and over again.
Strangely, the velcro only pulled off chunks of paint on the little section of the wall that was new construction in 2022; across the original wall — despite its having been replastered and taken the exact same paint — the dots came off cleanly and the paint stayed put. (Mexican construction is almost all plaster over brick; lumber is expensive, so no wood framing, and drywall is not a thing here.) But I did have to end up repainting that far right section of wall before replacing the tiles, and I felt too bad about taking important words away from Tashi again to let the paint dry overnight, so now the red tile, at least, is not sticking very well. Between my nine-dot overkill and the two puzzle-locked sides, however, I don’t think it’s in danger of falling off.
Also later in the first week, Jak and I tested out new cameral locations. Jak is not the exhaustive pre-planner that I am, so within minutes of my determining that the best way to see ‘Australia’ was to affix a camera to the wall directly above the larger board, he had 1) decided that the electrical cord ought to go through the wall and plug in to the bathroom socket, and 2) begun drilling a hole in the wall. Not only did this cover my newly cleaned and installed equipment in masonry dust and paint chips, forcing me to take down and re-clean a swath of tiles and replace a lot of the velcro dots, but the whole project came to a complete halt when it turned out the wall in question was considerably thicker than his biggest drill bit was long.
That was not the last problem we encountered, but I won’t enumerate. Eventually we managed to break through the entire wall, string the cords, patch and repaint the wall, and get both cameras rolling. One camera on each wall is enough for me to evaluate body language and see most of the presses, but we do experience the occasional body-block, and for purposes of video editing I find myself wanting more angles and a better view of the nearby space. So I’ll be adding a couple more of the basic spot cameras (I have one that pans, but turns out that feature isn’t very useful) as soon as I can get them down from the States, where they’re more affordable.
The Results of Verticality
As best I can tell — from behavior, word choice, and body language — Tashi really likes the wall board. (He said HAPPY WORDS twice in the first 24 hours after the move, which certainly seems like approval.) I didn’t (couldn’t) do a hard-data count, but my impression is that the daily number of words took a small jump compared to his post-cone usage immediately before the move. He’s certainly not showing any difficulties or reluctance.
I was a little concerned that the ‘Australia’ tile containing MORE and ALL-DONE — intended to be the vanguard of a cluster of antonym descriptor pairs — would get ignored, but that fear was put to rest right away. In fact, he repeatedly asked us to be ALL-DONE as we fussed with all the camera installation (in addition to asking what the bleeding hell we were doing, with WHAT and HMM?).
He does occasionally have accidental presses; the lower half of the board is subject to shoulder- and butt-bumps. But that happens about an order of magnitude less often than with the floor board, which was subject not only to foot tromps but bouncing ball chaos. Tashi never has seemed the least bit bothered by accidental words, but their reduction is certainly a quality-of-life upgrade for the household’s humans.
The buttons are somewhat closer — a little or a lot, depending on which row they’re on — which does improve my ability to see the labels. I hope to be able to get some new blank labels and some colored markers in early February; if so, one of my February projects is going to be attempting to relabel the buttons to be more visually distinct.
Support Possibly Still Exists
Friday night — ironically just a couple of hours after I finished drafting the ‘problems’ entry, but before I’d posted it — I finally got a reply from someone at FluentPet to one of my emails from five weeks prior. I mean, I know people take holiday vacation and so on, but we were so far past that that I’d figured I was on my own.
The customer support lead who was handling my complaints is still AWOL; the head engineer I was communicating with is ‘out of the office’; this new person said that my firmware had been upgraded … at some point … to try and address the skipped presses problem.
Which … I had been wondering lately if the app was perhaps doing a better job of logging, although after underestimating the scope of the problem by a factor of five the first time around, I wasn’t going to get too excited without the statistical proof of another full-day camera review. And I haven’t had time to do that in the two days since I found out about the firmware. I’m trying to get my record here caught up first, and then we’ll see when I can squeeze that in.
It would be nice if FluentPet was actually making strides on quality — I’d like to be able to recommend them because they’re good, not just because they’re the only game in town.
Still in the queue: two weeks’ worth of new words, cat progress, and some related news.