Solipsism Gradient

In case you’re wondering, I’m still here and — after 4 years of balancing acts — now consider my situation stable. Here’s a brief update on my previous post.

The eye needed a second triamcinolone injection but can now, several months after that, be considered stable. I’ve worked out a regime of eyedrops that seems to keep the macular edema mostly at bay and holds dryness etc. to tolerable levels. My eyesight has stabilized at 20/50 and I can drive cautiously.

The development hiatus can now be labeled retirement, unfortunately. I can’t see (hehe) going back to writing Mac apps and the pace in Swift(UI) development is still such that I’m adopting a wait-and-see approach to iOS/iPadOS development.

The health reports are good, my weight is almost down to my optimum level of 72-73kg and my level of exercise keeps improving; not up to pre-pandemic levels yet, but good.

Hopefully I’ll have time to write more about table tennis soon. It’s a fascinating subject.

A lot happened since my last update, but the outcomes were too uncertain to write about; and I did not want to bore you with Too Much Information.

Still, I hope this post will be useful to people with similar conditions. So here’s a detailed graphic of the eye’s anatomy:

After the vitrectomy in late April, everything seemed to be reasonably OK. The octafluoropropane gas bubble the surgeon used to replace the vitreous humor took longer than expected to dissipate and a small part of it unexpectedly migrated into the eye’s anterior chamber, in front of the pupil, where it blocked the normal fluid flow and raised the internal pressure.

Both bubbles were almost gone on a Friday evening in late June when, once again, I noticed the shadows and sparkles of a retinal detachment. Unfortunate timing and circumstances conspired to push out the necessary surgery to the next Tuesday. Because of the detachment’s location on the lower part of the retina — past detachments were all on the sides or upper part — it was necessary to fill the posterior chamber with silicone oil and place a scleral buckle around the eyeball.

The consequences were not comfortable. The replacement fluid — this time, the silicon oil — leaked into the anterior chamber again, possibly following the duct opened by the gas used in the previous surgery. Pressure inside the eye increased dangerously and had to be controlled with several eyedrops designed to reduce fluid production inside the eye. The oil also painfully pressed on the iris muscles which contract or dilate the pupil.

It was all quite uncomfortable. The eyedrops messed up the fluid balance, the painkillers messed up the stomach — I lost 3 kg in the process — and the enforced inactivity messed up everything else. The eye pressure didn’t quite come down to normal levels and would damage the optic nerve in the long run. Accordingly, one month after the operation, it was decided to do yet another vitrectomy to remove the silicone oil: something which would normally have been necessary only several months later.

The scleral buckle, unfortunately, will be left in place indefinitely. As it encircles the eyeball in the back, around the muscles, every eye movement was initially painful and it’s still quite uncomfortable; I’m told it will take several months to get used to that.

So the last operation was in early August, two months ago now, and it mostly went well: all medications were slowly discontinued, the side-effects went mostly away and for a few weeks I’ve been able to resume normal activities. I have another evaluation coming up in late October, when I’ll have a comprehensive OCT exam of the retina, as well as new lens prescription info for that eye.

However preliminary examination revealed that eyesight has fallen to 20/50 (from 20/15 last year), grid distortion has worsened, and there are two dark scotomas — one at the lower left and one at the upper right — just at the edges of vision. Also, the iris muscles were damaged, so the eye was left somewhat dilated and (in photographical terms) with a fixed aperture, not useful at night or in direct sunlight. None of these conditions is expected to improve in the future, so I’m glad to have enough 3D-vision left for driving and table tennis…

An interesting phenomenon happened before my regular eye checkup in late February: the visual distortion in the right eye didn’t regress much, but subjectively my vision was almost normal. However my left eye — until now, unaffected — had developed a similar distortion in the opposite direction!

My conjecture was (and my ophthalmologist agreed) that this was a software adaption by my visual cortex. The expectation was that all these effects would continue to shrink for the next several months.

Unfortunately that was not to be. Two weeks after the checkup I noticed a persistent spot at the edge of my right-eye visual field, outlined by sparkles, even at night. I knew these were the symptoms of retinal detachment, and the ophthalmologist performed immediate surgery.

This was done by the same means as my November vitrectomy. The retina had wrinkled again, but this time at the edge; so instead of buckling it just began to tear. This was repaired by a belt-and-suspenders approach: laser “welding”, keeping the retina in place by a pressurised octafluoropropane gas bubble, and buckling the sclera mechanically. All this required paralysing the eyeball muscles and doing some tricky micro-mechanical work.

Well, after a scary and painful first week — the internal eye pressure had more than doubled and had to be relieved — I’m happy to report that recovery is underway, again. The gas bubble takes several weeks to be replaced by fluid and in the meantime there’s just a blur to be seen; but all indications are that the retina has been fixed.

More in about a month…

Yes, things are getting better. The visual distortion I reported two months ago has lessened significantly; it’s still there, but much less pronounced and not as annoying. Hopefully all will be cleared up in a few more months, as promised; I’m scheduled for a new eye check-up 6 weeks from now.

There are two faint gray sickle-shaped shadows remaining around the central focus area, no doubt some retinal scarring resulting from the operation; they’re visible only when viewing uniformly coloured areas.

Otherwise, my depth perception has returned almost to normal and I’ve made progress on my table tennis training, somewhat hampered by the heat — it’s been between 30 and 34º for weeks. Stay tuned.

I know you’ve all been waiting for an update on my eye surgery…
…not really? Well, here’s one anyway.

Fast recap: here’s how things looked from my right eye —  it’s just slightly exaggerated. It’s called a macular pucker, and reading Spalding Gray‘s monologue about it wasn’t very comforting. The darkened, twisty part was unfortunately almost right in the middle, so reading with that eye was impossible, and the whole 3D experience was unreliable.

So, 3 weeks ago I had that somewhat scary vitrectomy operation and I’m happy to report that the surgeon said that, technically, everything went well — no infection, no visible scarring on the retina, no harm to other parts of the eye. Whew. In the meantime, I still have two weeks of antibiotic and steroid eyedrops every few hours; a small price to pay, of course!

Subjectively, the darkened blob is almost gone and the twisty parts are very gradually spreading apart and getting smaller; there remain two gray curved marks around the center, only noticeable when looking at a uniform background. Everything should stabilise in about a year and, with some luck, my eyeglasses should need no changing.

In another week I should be ready to resume my (relatively) new favorite activity: serious table tennis. I’ve started training regularly again two years ago, after about 40 years off the tables, and will write regularly about the topic here.

Warning: long, technical post; no TL;DR.

As I’ve said a few days ago, and then explained why:

My software development activities are (now, officially and indefinitely) on hiatus.

Most of my software projects already were marked as “legacy”, and RB App Quarantine had insignificant download counts, but RB App Checker Lite continued to be reasonably popular despite the lack of updates and an increasing number of crashes after macOS 10.12 came out.

In the past years, as I was beginning to consider the possibility of making this official, I replied to a few dozen bug/crash report emails mentioning that I might have to do so. (Of course, it never crashes here on my own machines… ?) Almost everybody replied with polite wishes but also mentioned that I should publish it as open source. Unfortunately, that is impossible; see below.

My idea with the RB Utilities (including several that never were published, one of which was the first one!) was to have a common foundation for all sorts of apps that examined, or did things to, files, folders and the file system in general. Some of you old-timers may remember my very well-reviewed XRay utility, which was not only my first Cocoa app, but also brought reasonable financial returns.

I wrote the core of XRay for MacOS 10.1 during an all-nighter session at MacHack 2001, so unsurprisingly XRay proved to contain serious design flaws and did not update well as file system APIs evolved. For some time I worked on a rewrite to be called XRay II, which made heavy use of plugins for most of its functions. But I was distracted by other work and in the end decided to shelve it.

Still, the idea of plugins proved too tempting, especially as Objective-C offered so many neat facilities for it, and in early 2011 I began writing a generic RB Utility app that would be specialised into a specific utility app by incorporating a single plugin, built right into the executable (rather than being a separate code/resources bundle elsewhere). The generic app would have both a Developer ID version and a Mac App Store version. It would do the heavy lifting, such as startup, integrity/signature/receipt checking, copy protection (for the devID version), show the About Box and generally do all the common work of showing windows, scanning folders and whatnot.

My usual work style is to write the basic stuff, run/debug it, then do increasing detail in sort of a fractal way. While writing the app framework I also wrote a very simple plugin that would be the basis of my first utility, RB File Counter. It would just scan a given folder at great speed and report the number of files and subfolders inside that.

As things progressed I was not very pleased with the complex (but interesting!) details of certificates, codesigning, requirements for the Mac App Store, and so forth; and I realised also that many developer buddies were having even more problems than I had. The logical thing was to halt work on replicating the old XRay functions like counting files, changing permissions etc., and first do an app that would help me (and others) to make sure our work was in shape. This became RB App Checker, with the “Lite” added on because I wanted to either do a “Pro” version or a general consumer version; both were to be paid-for, and there had to be some infrastructure built for that.

RB App Checker Lite came out in early 2012 after over a year of work and was well received; many developers sent in suggestions and bug reports, and I was also busy keeping things updated and making sure the app was both small and fast. When I went back to testing the other utilities, some infrastructure changes proved to be not generic enough and I had to review my initial design, of course.

In retrospect my main (and usual) mistake was to heavily over-engineer parts of the code in a matter that wasn’t future-proof. The “interesting” parts were heavily optimised, obfuscated, compressed and otherwise squeezed into hyperspace, while also trying to use all available CPU cores; and some extremely paranoid parts of the code were checking other parts of the code. In fact, I showed some of the source to a very senior Apple engineer at my last WWDC; he blinked and gabbled something like “ahem, well, this is most unexpected and I’ll be interested to know if it passes the App Store code review” while backing away slowly.

So, to come back to my original point, this means that the code as such isn’t fit to be published as open source. I may disinfect and publish a few of the saner parts later in a very controlled manner. If all goes well for my eyes this year,and now that Swift is nearing ABI stability, I’ll restart the app entirely in Swift — a good opportunity to relearn how to use GitHub and (ahem) unit testing.

Stay tuned for more news…

As I’ve said yesterday:

My software development activities are (now, officially and indefinitely) on hiatus.

The dictionary says:
hiatus | h???d?s | noun (plural hiatuses) [usually in singular] a pause or gap in a sequence, series, or process…

Let me tell you about the “why” first. Thanks to all of you who immediately asked if I’m OK, by the way.

Short answer: yes, I’m OK, though not as OK as in the past — but then, that’s to be expected, with all these anniversaries streaking by. Not helped at all by miscellaneous stresses originating in Real Life™. But: see below. ?

Longer answer: in the past several years, my eyes have been getting progressively worse, to the point that reading on paper now works only with excellent illumination. OK, paper is not something I have to do on a daily basis nowadays, but reading on a screen also no longer works at the — admittedly, very small — font sizes I was used to. Jacking up the font size worked up to a point but my coding habits call for a huge amount of text visible at the same time. Stopping work altogether to wait for an improvement has, retrospectively, killed all my momentum. (I’m still waiting for that improvement, unfortunately.) Also: see below…

Medical answer: during a trip to Bhutan (highly recommended!) in 2014 I suffered a posterior vitreous detachment in my right eye; no doubt due to day-long van trips over their highways-in-progress. A couple of months later, a beginning retinal detachment was fixed by laser surgery, and everything seemed OK for a time. However, visual contrast worsened until, in 2016, I had to have cataract surgery on both eyes; this was when I stopped working altogether, waiting for my eyeglass prescriptions to stabilise. While that took longer than expected, I also found that there seemed to be no combination of monitor brightness/contrast that made working longer periods tolerable. To make things worse, the vitreous detachment seems to have left a scar on the right retina, causing a macular pucker; surgery on that (a vitrectomy, yikes!) is now scheduled to happen before the end of this year. My eye surgeon assures me that should take care of everything… digits crossed!

Needless to say, all this has been somewhat disheartening, and — as far as coding and blogging are concerned — I seem to have joined the Procrastination Monks for a few more years.

In the meantime, other things have happened, some quite positive! Stand by for more news…

Today I have removed my remaining apps from the Mac App Store and posted official “Legacy” status notices on the product pages. My software development activities are (now, officially and indefinitely) on hiatus.

I know this notice has been long overdue, as I’ve been unable to update anything for more than 2 years. My apologies; I’ll try to post more about this here in the following weeks.