Rubenerd

Apple’s new CPU and hardware announcement was full of interesting tidbits, though I think I’ve talked about the broader implications enough here over the last few months. These are just some final closing thoughts for now, in no particular order:

• I’m always happy to see an end to monocultures. Macs used to be more interesting because they ran a different CPU architecture, and diversity benefits people who don’t even use it.

• Benchmarks on somewhat comparable A14 hardware look promising, with certain emulated amd64 instructions outperforming native Intel CPUs.

• People whinging that the hardware looked the same clearly hadn’t studied Apple’s hardware transition strategies. My first-generation MacBook Pro looked the same as a PowerBook G4.

• Microsoft have announced x86-on-ARM emulation, which could signal a future Boot Camp scenario, and an out for those of us who bought Intel MacBooks for this compatibility.

• I keep thinking of the Singaporean telco when reading these headlines. Given their old branding had a sun, something something heat and cooling and something something SPARC.

• I’m still undecided if I’ll stay on Macs long term. I’ve been an Apple user since the classic Mac days of the 1990s, but the Venn diagram of what my FreeBSD laptops and what my Macs can do is closing faster than I expected. Antranig Vartanian (RSS feed here) has been especially inspiring in this regard.

A well-known search engine has announced they’re no longer honouring unlimited photo uploads and storage. Reactions are split between anger, and people explaining how businesses work. All of it misses the mark.

The core technical issue should be obvious: there’s no such thing as unlimited storage (yet). There’s practically unlimited, there’s unlimited until our top 5% of customers reach a certain threshold and we kick them off, and there’s unlimited but we enforce it with rate limiting etc. But even the so-called hyperscalers can’t store truly unlimited data for everyone. And even if they could, could they forever? People rarely talk about it, but ongoing maintenance is a far larger burden than initial deployment; it’s the reason why “five minute installers” are so disingenuous. But that’s a topic for another post.

When you throw in profit margins, justifications to shareholders, and shifting business priorities, it’s an understandable and expected shift for companies to grandfather, revoke, or cancel unlimited storage.

But that leads us to the crux of the issue, and what nobody seems to be discussing: why then do businesses routinely offer unlimited storage in the first place? I can think of two reasons that may also flow into each other:

1. They think they can honour it. This sounds like a folly to a bivalent engineer who lives in material reality, but maybe some can.

2. They think they can get away with revoking it once they’ve bait-and-switched enough users, or locked them into their platform. This strikes me as fundamentally dishonest.

It doesn’t—or shouldn’t—take much empathy to understand why users and customers are perturbed when a core feature like unlimited storage is revoked, even if it was offered by a company infamous for killing services and selling customer data. You can justify the business and technical reasons, or point out the number of times it’s happened before, but it doesn’t negate the fact a business said they’d do one thing, but did something else. In any other industry this would be called out for the false advertising that it is.

Don’t believe unlimited storage claims. The business might not be being dishonest, at least not initially, but the end result will be the same. Hubris in this industry is all that’s unlimited!

My Vim post on Monday generated more positive comments than I expected. Most of you were either Vim users who understood why it’s not a fit for everyone, and non-Vim users making suggestions. I’m not sure how I’m going to react if you all keep being this civil.

Whole Wheat Radio alumni, NOCHANGE BBS maintainer, and XCHANGE author Jim Kloss weighs in by scaring me about the responsibility he’s bestowing upon me:

I look forward to reading more from @Rubenerd about his editor re-think. Great analogy about drums v. pianos. I take Ruben’s software suggestions very seriously because they often save me weeks of learning/testing.

Jonathan H. referred me to this helpful Hacker News comment by adimitrov about moving to Emacs, given I already use it for org-mode:

I started using Vim in mid/late 2000s, and successfully switched to Emacs. Without knowing your specific gripes, it’s hard no know what’ll end up helping you, but here are my two cents:

• use emacsclient and have aliases for emacsclient -c and and emacsclient -n for popping up a new frame or using the console, respectively. I even have a window manager binding to open a new Emacs client window

• Rainer König is the best at getting across org mode workflows. if you like watching nerdy videos, go watch him.

• keep vim around, I still use it, sometimes, but with no or veery minimal config.

• centaur tabs and the new tab stuff can help vim people who like tabs. I just got used to buffers.

• M-x is really Emacs’ primary UI. don’t try to think of a million and one key bindings up front, just bind what you find yourself using M-x a lot for. You just need a nice completing read like ivy, helm or so, but doom has that.

• use magit. While many claim that org-mode is the Emacs killer feature, I’d say magit is even more important if you code. There simply is no better git interface, nothing comes close. You think git the new porcelain is cool? Magit is a git jacuzzi.

A gentleman, name withheld, emailed saying I should try Spacemacs if I wanted a more intuitive editor. I still think I want to give vanilla GNU Emacs the old college try, but I’ll keep that in mind :).

And Rebecca Hales, who keeps wanting to dress us in cute things to make us exercise:

Emacs? Repent! But seriously… use what’s best for you.

There’s a wider lesson there. But point taken. C-x C-c.

I walked past a closed coffee shop in North Sydney that shut down for renovations a while ago and didn’t reopen, presumably because of COVID. The large glass doors were coated in a thin layer of dust from the inside, and a patch of bright paint on an otherwise faded wall hinted at what its name used to be.

(I sat there a few times when Clara and I lived in the area. Chances are some posts here from 2018 were written there)!

What struck me was how many letters had been stuffed under the door. Even just at a glance I saw several “To the occupant”, and more with a person’s name. So many pizza coupons.

It made me think how it would have got to that stage. A junk mailer would have had to stuff it there, with a clear view that the place was abandoned. A postie would have seen the pile of unread mail and continued to push it under the door.

These actions make logical sense when you understand their motivation. A junk mailer is paid by number of items distributed in a given area. A postie is contractually obligated to deliver mail, in one piece, to the place to which it’s addressed. Neither of these actors care, or need to, that the mail is being received and read. It’s not in their job descriptions, nor are they being paid for it. Perversely, withholding their communiques would go against what’s expected of them.

It’s all so backwards. What’s the point of mail if it’s not being read?

My follow-up post about a podcast host still gets responses, right up there in count with Hacker News links. I’d wondered at the time why a flippant message from someone I listen to hit me harder than I would have expected, and how it generated so much interest.

Jonathan Poritsky had some interesting insight a decade ago when John Gruber’s Talk Show split without explanation from 5by5:

People care about these shows. They listen to them regularly and discuss them among friends the way a bygone generation would talk about Uncle Miltie. Listeners are personally connected to these shows. We welcome these hosts into our lives every week to be entertained and inspired. There is absolutely a relationship between host and listener and it is a very personal, intimate one. Otherwise what’s the point?

Today I learned the indispensible tool for creating safe temporary files and folders began life on OpenBSD. From the project’s website:

Mktemp is a small program to allow safe temporary file creation from shell scripts. The current release is version 1.7, released on April 25, 2010. Mktemp is free software and is distributed under a ISC-style license. Mktemp was written by Todd C. Miller.

A different, but compatible, implementation of mktemp is available as part of GNU coreutils.

It’s worth remembering just how much of the Internet, and even Linux distributions, depends on infrastructure and tech that either started life on the BSDs, or is still actively maintained in a BSD project. OpenBSD especially punches well above its weight, often with little or no recognition in the Linux community.

It makes me wonder when tools I otherwise love like Ansible get bought for millions of dollars. Ansible depends entirely on OpenSSH for its operation. Did OpenBSD get a cent from that? What about Python? BSD/ISC/MIT and the GPL don’t require it, but seems like a good insurance policy, if not just common courtesy.

You can donate to the OpenBSD Foundation here.

I use the Vim text editor, every day. I’ve written millions of words in it, in HTML, Textile, Markdown, LaTeX, Docbook, and plain text. Every post on this blog since at least 2006 has, at some point, been composed or edited in it. Vim is where I change configuration files, code scripts, list todo items, and make meeting notes.

I navigate text from the home row not with cursor keys. I have a fleshed out .vimrc file for configuration. I have opinions about plugins. Sit me in front of nvi on a BSD, or vanilla Vim in compatibility mode, and I can be almost as productive. I instinctively reach for Vim keybindings in other contexts where it makes no sense!

Here comes the but. I haven’t ever been able to shake this uncomfortable, lingering feeling of self-doubt that it’s not the tool for me. In fourteen years I truthfully haven’t progressed beyond an intermediate user. Having jumped in the shallow end and surrounded myself with a comfortable subset of features to give myself a cross-platform alternative to Mac-only editors, each additional step into the deep end has felt equally cold.

This is largely on me. I concede that if I took the time to learn all the features properly, I’d realise that promised, untapped potential for additional productivity. It’s all there, waiting to be discovered, learned, and strategically deployed. But the reason I qualify that statement is because I still feel as though I’m the one seeding ground each time, rather than my editor working with me. Vim holds all the cards, to abuse another metaphor.

Text editors inhabit a weird space in computing. What could be simpler than transcribing the ramblings of an operator into characters on a blank screen? And yet there are dozens, possibly hundreds, of different ways to do it, all targeting different use cases, writing styles, feature sets, abilities, and trade-offs. There’s a reason new ones still come out, despite the industry existing for decades.

Text editors aren’t hammers, they’re instruments. And I think I’m a drummer who’s forced himself to use a piano. It taught me valuable theory, and I can appreciate how admins, writers, and developers can be productive using it. But ultimately it’s not the best way to express myself. I guess I’m starting to accept that I’m fine admitting it.

The final salvo came from Adrien Lucas Ecoffet, via Unixsheikh’s great article:

Vim users (I am not one of them) will tell you something along the lines of “it is hard to use at first but when you really learn it, you become super productive”.

However, the second part of this sentence applies to just about every good editor out there: if you really learn Sublime Text, you will become super productive. If you really learn Emacs, you will become super productive. If you really learn Visual Studio… you get the idea.

So now I’m wondering where to go from here. If I’m going to take the time and learn an editor properly, I think I’m telling myself that I’d rather do it in an editor better suited to how my mind works. I already use Emacs as a glorified org-mode interpreter, having realised it clicked for me faster than any Vim-based wiki. Maybe I should deep dive there.

(To confirm for those inclined to skim posts and leave abusive comments: I’m not saying Vim is a bad editor, nor am I judging anyone for using it. If anything, I’m envious of people who can use it, just as I am about those who can play piano. But in the words of myself at the end of this post, it’s just not for me… I think).

It’s a shorter Music Monday today, from everyone’s favourite electronic duo:

I’ve been around the world;
for a number of reasons.
I’ve seen it all;
the change of seasons.
And I, my lord… may I say nothing?

The temptation is already there to start worrying again. But for one of the few times this year, we were happy just to enjoy the news.

It’s natural when you’ve accumulated enough RSS feeds in your aggregator of choice to want to organise or sort them into folders. They’re like Twitter lists, only infinitely more flexible, usable, and open! But what should they be sorted by?

Topics are the most obvious. It’s how news sites organise their articles, and prominent blog aggregators like Feedly suggest topic collections to people with new accounts. I can’t remember if mine started the same way on Bloglines back in the day, but I’ve carried over the same rough topic folders since I started with RSS:

• Apple
• Art
• Australia
• Friends
• Infocomm News
• Infocomm People
• Multifariousness (a catchall, basically)
• News
• Nix
• Otaku
• Photography
• Science
• Singapore

This works, but like the Tower Bridge when a boat approaches, it has a few glaring drawbacks. That didn’t make sense.

I don’t know if you’ve noticed this on your own aggregators, but blog networks and news sites routinely drown out anything else that’s in a folder. A few sites like Metafilter and ZDNet dominate my Infocomm News folder, often leading to me missing tidbits from others. The unread count on the general News folder is in the thousands and realistically will never be caught up on.

What each folder contains has also evolved as blogs have dropped off and my interests have changed. For example, Art ended up including Hi-Fi gear, because that’s where music news ended up. Should that still be there? Probably not, especially considering photography gets its own category. Or what about an anime blogger who’s also a Postgres DBA? Do they go in Otaku, or Nix, or Infocomm People?

(I already can’t remember which self-anointed arbiter of blogging etiquette first advocated these arbitrary publishing edicts, but one stated you should limit your blog to one topic. This is, of course, bollocks. You should write what you want and what you’re interested in. I’ve mentioned before that people talking about disparate topics is a feature, not a bug. I love subscribing to a blog based on common interests, and being exposed to something else I’ve never seen before).

I’ve realised that who’s blogging, and how important they are to me, matters way more than topic. I’m okay missing a few days of Nikkei Asia or the BBC, but I absolutely don’t want to miss a post on The Geekorium. So I’ve ditched the topics entirely, and rearranged my subscriptions around this:

• Advisories. No analysis or commentary, just raw bulletins about new packages, updates, and security patches from projects like FreeBSD and Debian. They’re actionable.

• Art. Photos from space agencies, specific tags on anime image boards, profiles on sites like Tumblr, Wikimedia Commons featured pictures, etc. It’s an ephemeral photo magazine for browsing, or for choosing new desktop backgrounds.

• People. Individual bloggers I always want to read. I don’t need to sort by topic; I know that I follow them for stuff I’m interested in already.

• Wire. Aka, wire services. News outlets and large blog networks that regularly crank out updates. I’d turn off unread post counts here if I could, and use like Dave Winer’s River system, or how I read Twitter now.

I have all the same feeds I had before, but this structure has surfaced so much more stuff I missed before.