Irreal

Anyone who’s been around Irreal for any length of time knows that I’m an Apple user. I like the build quality of their hardware and macOS/iOS are generally good enough. But the thing that really makes it all work for me is the entire ecosystem. I have an iPhone, an iPad, and (at the moment) two MacBook Pros. That’s just me and doesn’t include other family members. The thing about this is that it all works together more or less seamlessly.

For example, I keep my memo book in the Apple Notes app. That means that I can make an entry from my phone, iPad, or Mac and have it appear on the other devices essentially instantly. The same is true of my calendar and contact list. The only thing that doesn’t work that way is, sadly, Emacs. If you don’t have this type of integration across your devices, it’s hard to describe what a boon it is.

Ali Abdaal, whom I’ve mentioned occasionally, has a really nice video up that describes the Apple ecosystem and how he uses it. It will be of interest primarily to Apple users but others may find it enlightening too. Even though I am all in on the digital lifestyle, Abdaal demonstrates a few things that I didn’t know you could do.

Most of the integration involves what Apple calls “continuity”: the idea that an action you take on one device is reflected on your other devices. The file syncing for things like the Notes app is one trivial example. Another, not so trivial, example is the “Universal Clipboard,” which makes anything you’ve copied onto the system clipboard on one device available for pasting on the other devices. Abdaal shows how useful that can be.

There are a bunch of other features that Abdaal explores so if you have interest at all in Apple devices—even if you think you’re a master of all things Apple—it’s definitely worth while spending a few minutes watching the video. It’s just over 12 minutes so you can probably squeeze it into a break.

According to a shocking Viewpoint article in the Communications of the ACM, there’s corruption in the computer science community. I know, I know: film at eleven. But this corruption involves cheating by academic computer scientists. Cheating is never okay, of course, but as with certain public officials—the police, say—we expect actual scientists to be held to a higher standard in their professional role. Even petty violations can’t be tolerated. That’s why it’s so disturbing to discover systematic, intentional cheating.

The TL;DR is that one or more small group(s) of computer scientists have colluded to game the paper review process—read the CACM article for the details. That’s important to all of us because it means that important results may not get published due to the cheaters’ papers being published instead.

It’s also important to the computer scientists because their publication history is vital to their tenure and promotion. That’s the point of the cheating. It’s also a consequence of the time honored but probably outdated university incentive structures that prioritize publishing in prestigious journals. As Goodhart’s law predicts, any metric used to determine an outcome becomes useless as a metric. That’s because it will be gamed just as in this case.

This is the same dynamic stemming from the same root cause as the current contretemps involving journal publishers. The academic community complains and rends their garments about the cost of buying the journals but it’s their own rules that enable the publishers’ rent seeking. Don’t look for anything to change soon.

As you all know, Irreal doesn’t think very highly of patent trolls. Of course, nobody (except patent trolls) thinks very highly of them. Some organisms that might be characterized as parasitical actually enjoy a symbiotic relationship with their hosts. Patent trolls, however, are simply parasites. Full stop. They’re scum that bring nothing but harm to the society they infect.

Their means of entry into the host is always the same. They leverage the fact that it’s always cheaper to settle a bogus patent claim than to defend against it. That’s true even if the victim wins. What can be done? A good start would be to make the loser pay all the costs. Of course, the pathogens known as non-practising entities would doubtless evolve to leverage that defense too.

Washington State is pushing their own solution. They’ve passed a law outlawing bad faith patent claims and are suing Landmark Technology for patent trolling. Irreal wishes them luck but as the article says, they have their work cut out for them. The courts have set a very high bar to establish that a patent claim is in bad faith and there are constitutional issues with seeking to limit legal regress. Still, Landmark’s claim is extraordinarily weak—anyone with a commercial Website could be said to be in violation—and experts say it would certainly be invalidated if litigated.

One can only hope that Landmark will decide to sue Cloudflare. They know how to deal with patent trolls. As I said before, these trolls and the bridges they hide under need to be smashed to pieces. More companies should man up and follow Cloudflare’s example.

Almost every fall during its 50 year lifetime, the Peanuts comic strip would run a variation of the Lucy and Charlie Brown football gag. The basis of the gag is that every year Lucy would talk Charlie Brown into trying to kick a football that she was holding and every year she would yank it away at the last second causing Charlie Brown to land on his butt.

Charles Schultz, who wrote the strip, died in 2000 but Google is apparently trying to keep the gag going. RSS users will undoubtedly recall Google’s abandonment of Google Reader, which powered most RSS readers. Google reportedly did this to drive acceptance of their new, since abandoned, social media platform that was so unsuccessful I’ve already forgotten its name. Google’s action was temporarily devastating to RSS. Many went so far as to declare it dead. But the thing is, unlike social media platforms, RSS solves a real problem so it bounced back with new readers that didn’t depend on Google.

Now Google is inviting RSS users to have another go at the football. They’ve announced that they’ve added an experimental feature to Chrome that is, in effect, an RSS reader. And you can trust them, they say: this time they won’t yank that football away. Before you become dependent on this new feature, you might want to ask Charlie Brown what he thinks.

I’m a huge fan of mechanical keyboards. Back in the old days that meant the IBM Model M and its latter day incarnation from Unicomp. I’ve had both—although I foolishly left my Model M behind at my last job—and there’s nothing more satisfying than their solid, tactile, clicky feel as you type. Of course, the Model M and its Unicomp offspring are pretty noisy with their clickity-clack sound but that can also be satisfying (although perhaps not to your colleagues).

These days, there’s an embarrassment of riches. The point of mechanical keyboards is, of course, their key switches and modern mechanical keyboards usually come with a choice of switch types. You can get noisy, clicky ones like on the Model M or quiet clicky ones, or a softer feeling switch without the tactile feedback. The problem now is not getting a mechanical keyboard but deciding which one to get.

Wirecutter has an excellent review of mechanical keyboards with their recommendations on which to choose. It’s a long and comprehensive review. It doesn’t cover every choice, of course, and although it mentions wireless mechanical keyboards it doesn’t have much to say about them on the grounds that they’re not yet as good as their wired brethren.

Ali Abdaal also has an interesting video in which he compares several mechanical keyboards in terms of which allow him to type the fastest. Ironically, he gets his best results with the standard, nonmechanical Apple keyboard but he considers a different selection of keyboards from Wirecutter so its a useful companion to the Wirecutter article.

If you’re in a mechanical keyboard, these are two great resources. If you’ve never typed on one, you should give it a try. They really make typing much more enjoyable.