Rubenerd

Our Australian home internet finally churned from SkyMesh to Bevan Slattery’s Superloop that bought the former last year. I got this message:

Hi Ruben, your service transition to Superloop is now complete. You may experience 10-20 minutes of downtime. If you experience any ongoing downtime, please power cycle your router and then contact our support team on $REDACTED. Thank you.

Our connection was down for hours, though that was certainly the fault of NBNco. As promised, our existing PPPoE creds still worked.

With this complete, we have a week before we move. We’ve been tentatively told our new apartment has FTTB that should be compatible with their service; hopefully it’s true. Australian internet is such a crapshoot.

We were extraordinarily lucky that FTTN worked well for us here; the antiquated, high-voltage Malcolm Turnbull box was right outside our complex, and evening speeds were never too much of an issue. That said, Superloop has already been noticeably faster.

Update, 12th February: We moved house, and have now been waiting six days for them to transfer to our new address. Multiple days between support tickets and updates. Not impressed.

ufotable is hosting a café event for the Fate/Stay Night Heaven’s Feel movie series, and included the above image of everyone’s favourite star from it despite it not being her story arc or anything important like that!

@JamieJakov likes a good tsundere, or at least he and I used to, if I had endless money I’d fly the both of us out for this to make up for missing his birthday yesterday as well. Happy birthday too, my friend ♡.

I would have included this supplied sample image on Tumblr, but they would have just auto-blocked it.

Dave Winer’s RSS source namespace has changed from:

http://source.smallpict.com/2014/07/12/theSourceNamespace.html

To:

http://source.scripting.com/

So better update your feeds if you use it.

Neither namespaces are listed on the W3C RSS Validator namespace list, but it also hasn’t been updated for more than a decade. The good news is our feeds no longer fail validation if they don’t have the superfluous, self-referencing atom link.

I compiled a list of ultra-lightweight laptops back in January 2018 when I was researching what to replace my secondary personal device with. These are all likely outdated, but sharing here in case anyone finds this useful.

I tried to only include devices with at least a HiDPI or Retina-grade display, because Apple has spoiled me. So the ASUS ZenBooks and most of the Toshibas from the time weren’t included.

Device	Kg	Display
GPD Pocket	0.480	7" 1920×1200
Panasonic Let’s Note CF-RX6A	0.745	11.5" 1920×1200
Apple MacBook	0.92	12" 2304×1440
Toshiba Portage X20	1.1	12.5" 1920×1200
HP EliteBook x360 1020	1.13	12.5" 3840×2160
Panasonic Let’s Note CF-AX2	1.15	11.3" 1920×1200
Toshiba Portage X30	1.05	13.3" 1920×1080
ThinkPad X1 Carbon 5	1.14	14" 1920×1080
Dell XPS 13 2-in-1	1.24	13.3" 3200×1800

Today I learned about fd, an alternative to find written in Rust. Its easy syntax is reason enough to use it, but check out the speed with which it found posts in this site’s git repository:

$ time find . -name *bsd*
==> real 0m0.595s
==> user 0m0.054s
==> sys  0m0.285s
  
   
$ time fd bsd
==> real 0m0.030s
==> user 0m0.028s
==> sys  0m0.026s

This is now in my toolchain, so I can invoke Friar Duck for all searches. To trip to trip it trip trip tripo tripa argh a trip a trippit trippit oh argh trippit oh trip it… tripping up and down! ♫

CRN Australia posted these within days of each other:

• Symantec’s consumer sales surge, tops profit estimates
• McAfee cuts 200 employees after eroding sales
• Bagels
• Sophos nabs McAfee vereran Gavin Struthers

At least one of those links wasn’t on CRN Australia. Try to guess which one, you may be surprised. Or not, because you’re a more sensible person than I.

I’m surprised third-party AVs have lasted as long as they have since Microsoft released their Security Essentials. I wouldn’t hesitate to recommend Sophos though.

My comment about Apple keyboards yesterday made me think which laptops I’ve owned. This is a list of ones with excellent keyboards:

• ThinkPad, all of them
• Toshiba Satellite T2150
• First-gen MacBook Pro
• Official foldable external keyboard for my Palm III

These were passable:

• iBook G3
• MacBook Air 2013
• Compaq Armada M300
• Toshiba Libretto 70CT, amazing given the size

These weren’t great:

• GPD Pocket, but understandable given the size
• Every iPad keyboard cover I’ve tried to use
• The HP 620LX, the later Journadas were better

And these are just bad:

• A broken table that’s also on fire and looks bad
• Hitting my fingers repeatedly with a ball-peen hammer
• Latest MacBook Pros, including ones with ingress rubber

The Dublin Core schema is a set of vocabulary terms that can be used to add metadata to documents. It predates schema.org, and is widely used in RSS, publishing, and academia.

I was adding Dublin Core metadata to an HTML5 page, and encountered a couple of problems with the W3C’s nu validator.

“schema.DC not a registered keyword”

My pages included the following links which I’ve been using since the XHTML days:

<link rel="schema.DC" href="http://purl.org/dc/elements/1.1/" />
<link rel="schema.DCTERMS" href="http://purl.org/dc/terms/" />

But the nu validator claimed it had a “bad value schema.DC for attribute rel on element link: The string schema.DC is not a registered keyword.” This seems like a glaring oversight or omission given the importance of Dublin Core.

There are two options here. You can remove the links to schema.DC and schema.DCTERMS, because their vocabularies are already known to HTML5. I prefer to codify what I’m using, so the other option is adding vocab prefixes:

<head prefix="dc: http://purl.org/dc/elements/1.1
    dcterms: http://purl.org/dc/terms">

Then referencing Dublin Core metadata using HTML5’s built-in RDFa property attribute:

<meta property="dc:title" content="The Bird is the Word" />

“The scheme attribute on the meta element is obsolete”

This validator error also took me by surprise. One could previously define the scheme of an attribute, like a data type:

<meta name="DC.date"       scheme="DCTERMS.W3CDTF"  content="..." />
<meta name="DC.identifier" scheme="DCTERMS.URI"     content="..."
<meta name="DC.format"     scheme="DCTERMS.IMT"     content="..." />
<meta name="DC.language"   scheme="DCTERMS.RFC1766" content="..." />
<meta name="DC.rights"     scheme="DCTERMS.URI"     content="..." />

This was removed in HTML5, with the unhelpful comment that the “scheme attribute on the meta element is obsolete. Use only one scheme per field, or make the scheme declaration part of the value.”

I’m confused. This is using one scheme per field. And how does one include the scheme declaration as part of the value?

This forum poster suggested there wasn’t anything to worry about because “parsers do a pretty good job”. Bluntly, this is insufficient, especially given we could unambiguously express the type before.

If it weren’t for the beautiful srcset attribute, I’d probably still be doing things in XHTML+RDFa land. The run fast and break things approach to HTML5 still worries me a bit. Se a vida é.

Jason Snell compiled another Apple report card for Six Colours. These were the parts I was most interested in:

The new Mac Mini [sic]:

The new Mac mini earned a lot of praise. John Siracusa called it “the only real standout” among new Mac models. Andy Ihnatko said “its presence gives me a little more confidence about the future of the platform."

It took self control not preordering one as soon as they were announced. Compare that to the embarrassing state of Apple’s keyboards:

Matt Deatherage said, “It defies reason for Apple [to offer] keyboards of inferior design and execution.” John Gruber said, “I may be biased as a writer and a keyboard aficionado, but it used to be the case that Apple’s notebook keyboards were widely hailed as the best in the world… that’s no longer the case and I think that’s a problem.” Shahid Kamal Ahmad said that the major failing of the keyboard was not its feel but “the inherent unreliability of the switches and their propensity to fail from the inevitable ingress of a subatomic particle.”

Apple’s keyboards were never considered the best; that was reserved for ThinkPads and Toshibas. But there’s no question Apple keyboards were better than average; now they’re deficient in ergonomics, accessibility, and durability.

The iPhone XR, which I’m so relieved exists because my eyes are sensitive to the strobing effect of OLEDs:

John Siracusa said, “The iPhone XR gives us a glimpse of what a few well-chosen trade-offs can deliver.” John Gruber said, “After spending a few weeks using an XR full-time, I honestly question whether its LCD isn’t better than the XS’s OLED for my needs.”

And the new iPad Pro, which is the first iPad I’ve ever been tempted by:

John Siracusa said, “The new iPad Pros are phenomenal pieces of hardware that place Apple firmly at the top of the class when it comes to mobile computing power… It’s a shame that iOS hasn’t kept pace with the iPad’s hardware prowess.”

We’ll see what they say about Apple’s services division for 2019 after their FaceTime debacle!

Joanne McCarthy reported last week that New South Wales teachers are planning to campaign for an end to compulsory scripture classes. It makes sense, Australia is one of the world’s least-religious countries.

But this is my favourite part, emphasis added:

Special religious education (SRE) is mandatory for 40 minutes per week, and students who opt out are not allowed to do any meaningful activity during that period.

The rules perhaps say more than were intended.

Religion was my favourite class when my sister and I grew up in Melbourne; I got to sit in the spare room with the Apple IIs and Commodores.