WQTS 14

A little more than a year ago, in discussing the failings of our car radio, I said “And there is a chance that JVC’s more recent units radios [sic] were designed and built following more rigorous design and testing processes.

Excuse me while I laugh hysterically.

Yes, I really did get a new car radio. Only a year and a half after sayingDespite its limitations, I have no plans to replace the radio with something newer and more capable.” (Insert that famous quote about foolish consistencies here.)

I got fed up with the lack of Bluetooth. Getting sound out of my phone onto the car speakers so I could listen to ballgames on the way home from work required plugging in multiple cables and random bits of gadgetry. And every time I tried to simplify the process by leaving everything hooked up, the Mariners would take an East Coast road trip, meaning games were over by the time I got in the car. Not to mention, it looked messy.

And, more to the point, it was starting to fail. The sound would cut out randomly, requiring a reboot. Or the display would stop displaying, also requiring a reboot. Or it would refuse to change channels, requiring (you guessed it) a reboot.

So the Circuit City relic–yes, the old radio really did come from the late lamented CC–now resides in a bucket in the garage, and its spot in the dashboard has been taken over by a newcomer.

The new one isn’t a JVC product. It’s a Kenwood. Except that the full name of the company that made it is JVCKenwood*. Which I hadn’t realized when I bought it. Not that knowing would have stopped me. Despite the old one’s limitations, I really did like it.

* Apparently there’s no slash or other separator between the C and the K, much to my surprise.

We haven’t had a power failure since it was installed, so I can’t address whether it, like its predecessor, has issues remembering user settings. But even the few weeks I’ve had it makes it obvious that JVCK’s design review process hasn’t changed for the better.

Let’s start with something that might not be obvious. The English version of the Quick Start Guide is 37 pages long. That’s one heck of a slow quick start. Still, it could be worse. The full manual (only available via PDF download) is 120 pages. Whoever wrote the Quick Start managed to trim more than three-quarters of the text.

But, still. If it takes almost forty pages to introduce someone to the basic features of a product, you have to face the fact that you haven’t done much to build in discoverability.

That aside, the new box is a significant upgrade. No more 11 character LCD with scrolling titles. Instead, it’s got a large screen (okay, not maybe not in absolute terms, but certainly by comparison. “Almost the entire front of the unit” easily qualifies as “large” as far as I’m concerned). And it uses proportional fonts, so more characters can fit in a given amount of space. In typical English language song titles, this seems to work out to about 20 characters. It also uses a smaller font for artists and album titles, so they can squeeze in around 25 characters. That’s an improvement.

Except that they don’t scroll. So Kate’s favorite truncated song title becomes “Papa’s Got a Brand “. Are we talking cattle ranching or personal promotion?

I lied. Actually, they do scroll. If you tap a small on-screen control* (yes, it is a touchscreen), the title/artist/album will scroll. Once. Better pull over if you want to (a) find the button to tap and (b) read the scrolling information without (c) causing an accident.

* This is a theme, actually. There are lots and lots of onscreen buttons. Most of them are small, and those that aren’t are tiny. Clearly nobody involved in designing this radio considered how to use it while driving. Or, if the assumption was that it would only be used in vehicles with on-the-steering wheel controls, said controls should be included with the radio.

Who thought one-and-done was a good idea? And I checked very carefully: there is no setting for autoscrolling, or even “keep scrolling once tapped”.

The old radio had a dial to change the volume. A nice dial that stuck up from the front of the box, easy to see out of your peripheral vision, so you could reach over and turn the sound up or down without taking your eyes off the traffic. The new one? Two tiny buttons at the lower corner of the radio. After several weeks, I still haven’t developed enough muscle memory to change the volume without looking. I wait until I get stuck at a red light.

There are other buttons. I have no idea what they do, because they’re equally tiny, and I don’t really want to experiment while driving. No, let me amend that. Once of them–helpfully labeled “ATT”–mutes the radio, presumably so you can quiet it enough to hear the traffic cop who’s chewing you out for swerving across three lanes of traffic while you hunted for the volume buttons. (Checking the Quick Start Guide, I see that “ATT” is right next to the “HOME” button–which also doubles as the power button. Nice.

Moving on.

One feature I hadn’t considered when buying the radio, but greatly enjoy is the ability to plug in a thumb drive full of music files. And, hey, I’ve got a thumb drive already loaded with my entire music library, almost 50,000 tracks, nicely sorted into folders by artist and album. Feel like some ZZ-Top, Brave Combo, Danny Coots, or…? Got you covered. As long as you want to listen to a specific track or album. Because there’s no way to play* all tracks in a folder full of folders**.

* Not quite true. If you start playing a track in folder/subfolder1, it will play through to the end of subfolder1, then go on to subfolder2. But you can’t shuffle all of folder’s tracks; hit the shuffle button (another tiny on-screen icon), and the radio will shuffle the current subfolder, then move on to the next subfolder and shuffle that.

** Also not quite true. If there’s a playable track in folder, it’ll go from that to subfolder1, then subfolder2, and so on. It’ll even shuffle the entire set of tracks in the subfolders (as long as you hit the shuffle button before the first track ends). But why would you have a random song in each artist’s top-level folder?

Shuffle is a particularly vexing issue for me. I like the ability to be surprised with something I haven’t heard for a while. So if I’m not sure what I want to listen to, I’ll often tell my playback device to shuffle everything. Guess what you can’t do with this radio.

Actually, you can shuffle everything. Go into the search function and hit play without making a selection. Hey, it works! For a little while. Then you realize you’re hearing the same artists over and over. Turns out that search–and therefor the search-based shuffle–can only load 1,000 tracks at a time. Oops.

Come on! Even my iPod Classic (pre-upgrade) could shuffle more than tracks than that.

Apparently, nobody considered the actual use cases for thumb drives larger than, say, 32GB. Even though someone did check off the boxes in the requirements document that said “support exFAT” and “drives up to 512GB”.

There are minor annoyances, too, pointing to inadequate testing and/or limited post-release support (the firmware for the radio has apparently been updated a grand total of twice since the initial release in 2020). For example, Android Auto can’t connect to the radio unless the phone is unlocked, even though I’ve selected the option to connect without unlocking. Swiping controls left/right works nicely unless you move your finger too slowly, in which case the radio sees a tap instead of a swipe. Android Auto always starts in the Map app (though, to be fair, this may be Google’s fault, not JVCK’s). And so on.

All my complaints notwithstanding, I do consider this radio a major upgrade from the old one. I love having the big screen that shows (most of) the title, artist, and album information at the same time instead of making me switch among them. Album art onscreen is nice, especially while listening to SiriusXM channels.

And the Bluetooth works nicely. It connects automatically and rarely skips or stutters. Baseball in the car, without unsightly wires and gadgets draped over the dashboard. Heaven!

They Don’t Make It Easy

They really don’t.

Backing up for context.

Like most, our local supermarket chain–Lucky California and a few other chains under the same ownership–runs occasional contests. Makes sense: free goodies always attract customers.

Most recently, they partnered with Shell for a double whammy: the top prizes (awarded through drawings at the end of the contest) were free groceries for a year* and free gas for a year*. There were also smaller prizes, both for gas and groceries, and–to keep the excitement flowing–instant prizes of merchandise and loyalty program points.

* I’ll come back to what that means.

We always play the games. Why not? We shop there anyway, it doesn’t cost anything to enter, and we’re no more immune to the lure of “free” than anyone else.

Not that we expect to win anything significant. Last time, we won a free bag of chips. We made nachos.

This time, about halfway through the contest, we got an instant win for 50 loyalty points. That’s fine. We collect those, and every couple of months trade them in and get $20 off our next shopping trip. No other instant winners, and when the contest ended, we promptly forgot about it.

Months later, I got an email.

“Congratulations,” it said (with three exclamation marks). “Your online entry…has been drawn for the Groceries for a Year…!”

Let’s see. We’ve got gratuitous exclamation marks and bad grammar. We also have a “click here” link to claim the prize; the link points to a website that is not Lucky Supermarket nor the contest’s special domain. The sender’s email address is shown as being the contest’s domain, but the actual sending system doesn’t match Lucky, the contest, or the claiming website. The contact phone number in the email is not the same as the contact number on the contest website. And the message is signed by “Christine” (no last name).

That last one is actually a point in the email’s favor: most scam-spams include a full name, apparently because the scammers think it conveys respectability. Other evidence pointing to legitimacy: the email addressed me by name–spelled correctly–and my name isn’t part of the email address I’d used in signing up for the contest*.

* That’s by design, and for just this sort of occasion. If an email addresses me by the name in my email address, I can be fairly sure it’s spam.

So I was on the fence.

Contests and sweepstakes are required to keep lists of winners, and according to several online resources, you can sometimes check if your name is on the list by calling the contest sponsor. So I decided to call Lucky’s customer service number. No dice. The guy I talked to was very nice about the whole thing, but he didn’t have a list of winners, and he pointed out that until someone claimed a prize, they wouldn’t be on the list anyway. He suggested I write a letter to Corporate, or check the contest website for a phone number. Given that there’s always a time limit to claim prizes, the letter idea was pretty much dead in the water. And if the website had been compromised now that the contest was over, any phone numbers would be dubious at best.

I did take a look at the recent prize winners page on the website. Three of the four Groceries for a Year prizes had been claimed, and my name wasn’t on the list.

I flipped a coin and decided to take a chance. I filled out the claim form. It wanted my name, address, phone number, email address, birth date, and the address of the store I normally shop at. Since they already had all of that information from when I signed up for the loyalty program, giving it again was no big deal. It also required my SSN. I had qualms, but decided to take the chance.

About an hour later, I got another email from Christine (point in their favor for keeping the same name) thanking me for claiming my prize and asking me to fill out a W9 tax form, since the value of the prize was over $600. Fair enough; gotta keep the IRS happy, right?

I did check the contest website again and my name had been added to the list of winners. That relieved my mind immensely. Highly unlikely that if the site had been hijacked, the scammers would go to the trouble of adding names to it. After all, to make any such scam worthwhile, they’d have to target more than one underpaid writer.

So I filled out the W9 and sat back to wait for my prize.

Which, according to that second email would arrive in 6-10 weeks.

Free groceries for a year sounds great. Slightly less so when you realize the value has been capped. Specifically, it’s limited to $5200. That’s right, $100 a week. To be fair, pre-COVID, our weekly grocery bill did come in right around that number. Now, though, it’s gone up by thirty or forty percent.

Still, free is free, and saving five grand on groceries isn’t anything to sneer at. If food prices don’t go up significantly, the prize should still last us well into 2023.

Oh, and the prize is paid in gift cards. I had visions of receiving a box of 52 $100 gift cards, and wondered if they’d be date stamped, so we couldn’t use more than one a week. But there was that 6-10 week waiting period to find out.

Exactly a month later, I got a “Your package is on the way” notice from FedEx. I didn’t think much of it; I’d ordered several things online that week, so I was expecting to get several of those notices. However, this particular one was odd.

For one thing, the sender was listed as an individual, rather than any of the companies I’d bought from. Odder still, the sender’s address was listed as Modesto, California, but the package was shipping from Buffalo, New York.

That set my scam detector tingling again. Was this some kind of “send some cheap merchandise to a random person, then demand large sums of money under threat of being reported for theft” deal? Or a “forward the package to a third party, and when it arrives we’ll send you a check which will never arrive” game?

So I googled the sender’s name and Modesto. Hey, she’s a marketing manager for Lucky’s parent corporation and the street address is the corporate headquarters. Answers that question. (And no, her name isn’t Christine.)

The cards arrived on my doorstep almost exactly six weeks after I filled out the forms. And no, there weren’t 52 of them. Just eleven: one for $200, the rest for $500 each.

Then I notice the fine print on the back of each card, informing me that the cards need to be activated before use. Nothing in the accompanying letter, much less the earlier emails, suggests this has been done.

Dash off an email to Christine. Three hours later, she writes back to assure me that they are activated. Sure glad nobody intercepted the package–mail theft is endemic around here–or I would have been out my prize with no hope of recourse.

So, finally, we’re happy. But…

Let’s sum up: scammy emails, multiple domains in use with no transparency about their relationships, difficult to confirm legitimacy, lack of consistency between contacts, insufficient information at many steps.

From what I see online, this is typical for online contests. But why? Sure, if it’s hard to claim a prize, the sponsor will save money–most such contests have a “unclaimed prizes will not be awarded” rule–but hardly enough to justify doing it this way on purpose.

Is it possible that the people running these contests have never gotten a spam in their lives? Seems unlikely.

So why do they seem to be going out of their way to seem suspicious? Especially when they could make one simple change that would greatly reduce the awkwardness of the current system.

In order to submit contest entries, players have to have an account on the contest website. So, instead of sending an email directing the winner to a third-party website to file the claim, send one that directs them to sign into the contest website. That site can have the “please click here” to claim your prize.

Granted, it wouldn’t solve all the problems of the current system, but it would be a big step in the right direction.

A Different Way of Thinking

I thought this was an interesting difference in how people think.

For background, Cloudflare was down for a while Tuesday. That meant a substantial chunk of the Internet was down, because Cloudflare is, in essence, a provider of web capacity. When you make a request for a web page from a site that uses their services, the request goes first to Cloudflare. If they have a local copy of the page–they pass it to your browser. If not, they request it from the original site, give it to you, and keep a copy to fulfil future requests*. It’s all transparent to you and your web browser and it protects your favorite web sites against denial of service attacks.

* There’s more to it than that, naturally. Controls to ensure that Cloudflare doesn’t keep pages containing personal information and serve them up to others, for example.

Everything is fine and the Internet is a happy place until Cloudflare itself runs into problems.

When that happens, they generally send your browser an error message, most commonly one in the 500 range. (400-type errors are indicative of problems on your end; for example, most people are familiar with the 404 error, meaning you–or your browser–asked for a page that doesn’t exist.) 500 errors are for trouble at the other end of the connection: the server crashed, can’t handle the volume of traffic it’s getting, can’t contact the site it’s trying to be a front end for, and so on.

And most of Cloudflare’s 500-type error messages identify Cloudflare as the location where the problem occurred and make a recommendation: click a link to try to contact the original site directly, try again later, etc.

As I write this on Tuesday afternoon, Cloudflare is back up and the Internet is running as smoothly as it ever does. And, naturally, people what to know what went wrong.

According to Google’s trending searches list, the Number One search in Japan, by a significant margin, is for “Cloudflare”. (I find it vaguely amusing that this is the only English language search term on the list.)

In the UK, “Cloudflare” is the third most common search, trending well behind “Xavier Musk” and “Tom Mann”. Makes sense, right? Get your popular culture news first, then go learn why you couldn’t get the news earlier.

Then there’s the US. “Cloudflare” didn’t even make the list. Instead, there was a mass of searches for “500 error”. Google helpfully directed them to a random web page that promised to explain what they are (reasonable) and how to fix them (you can’t; see above).

Americans, it seems, are more interested in fixing a problem than in fixing responsibility. Elsewhere, there appears to be some recognition that many problems are transient, and once they’re resolved, it might be a good idea to take a look at who caused them.

Think I’m reading too much into this? Long-time readers will remember the Bay Bridge Bolt Botch. Lots of energy around the bolts, determining how many were failing, what the effects of failure would be, and on and on. Zero desire to determine who was responsible for a design that included bolts that might not have been needed, who specced the wrong bolts, who signed off on their installation, who may have tested them improperly (or not tested them at all, according to some reports).

It gives one to think, doesn’t it? Because, after all, if nobody knows who’s responsible, it could be anyone–Saudi Arabia! The President! Bill Gates! Joe Shlabotnik!

Or even nobody at all: Natural cycles! Sunspots! UFOs!

WQTS 13

Once a product hits the thrift store, it’s much too late to make corrections to the packaging. But perhaps this can serve as a cautionary lesson.

“Ntelligemt Hulahgop”?

I understand the need to find a unique name for your company and product, but some approaches to the problem are just wrong. That includes all of the approaches used here.

How exactly does one pronounce “ntelligemt”? I’m guessing the “n” is pronounced “in”. That’s fairly standard. But is that a hard or soft “g”? “gehmt” doesn’t exactly fall trippingly off the tongue, but “jehmt” isn’t an improvement. The soft “g” might work if it weren’t for the “m”, but we’re stuck with that.

Then there’s “hulahgop”. I’m sure whoever came up with the name wanted something to suggest “hula hoop” without actually violating Wham-O’s trademark. And, yeah, okay, a “G” looks like an “O”. But we’ve got a pronunciation problem here too. It’s not the “G” so much as the “H” that precedes it. “huhgahp”? “huhjahp”? Maybe we can say the “H” is silent; “hulagahp” almost works, and “hulajahp” is even better–as long as one ignores the ease with which it could be mispronounced as a well-known derogatory slang term.

Matters don’t get any better once we move past those names.

Who thought it was a good idea to break up the word “beautiful”. I don’t know about you, but I can only assume that Beau Tiful was Beau Brummel’s lesser-known brother.

And why the missing spaces in the tagline? “exercisemakes peoplemore beautiful” Is it supposed to convey a message? Other than “Nobody associated with the product has ever bothered to learn English”. That’s the message I get.

And it’s one that’s reinforced by “free adjustments”. I suspect they meant “freely adjustable”, but the way it’s phrased, it suggests I’ll need to take it to a shop to have the belt tightened or loosened to fit my waist–but at least the shop won’t charge me for the service.

Finally, there’s that block of text at the upper left. In case you can’t read it in the picture, what it says is:

Shock-absorbing

Massage contact 360

Surround massage

Wait, what? This is a massager? I thought it was exercise equipment.

But I guess it makes sense. Doesn’t everyone like a vigorous stomach massage while jumping up and down? I know that would help me lose weight: five minutes and I’ll be getting rid of everything I’ve eaten for the past week.

WQTS 12

Hard to believe it’s been more than four years since the last WQTS* post. Granted, last year probably shouldn’t count. It’s not like any of us have had opportunities to encounter the results of egregiously bad testing recently. But still.

* For those of you who’ve started reading since June of 2017, or whose memories don’t extend that far back, the acronym expands to Who QAd This Shit. It’s where I mock products that were improperly tested, insufficiently tested, or–a closely related discipline–never granted a design review.

We took the car in for service yesterday–the Toyota, not The Bug. Aside from the semi-annual maintenance, it also needed a new battery. Generally, when it comes to matters automotive, we rely on experts for diagnosis, but it didn’t take much expertise for us to figure that a battery that had been in service for seven years and occasionally failed to hold enough charge overnight to start the car was about due for retirement.

Guess what happens to the radio when the battery is replaced. Go ahead, I’ll wait.

If you guess that it reverts to its default settings, you’re right. Partially.

For the record, the radio in question is the KD-HDR30, made by JVC. To be fair, it is, like the car, more than a decade old; nobody’s going to be buying one today. And there is a chance that JVC’s more recent units radios were designed and built following more rigorous design and testing processes.

The radio itself reverts to the defaults. The add-on modules that give it additional capabilities don’t. So the SiriusXM module remembered our station presets, but the radio switched to its built-in FM tuner.

That’s actually a reasonable default. It doesn’t make sense for the radio to assume the presence of optional hardware. What’s less sensical–and points to inadequate testing and/or design review–is that the FM station presets were gone.

Who thought it was a good idea to withhold capabilities from the base unit that were given to an optional component–the satellite radio plug-in? Clearly, somebody who didn’t think the radio could lose all power after installation.

Wait, it gets worse. The base radio component apparently has no ability to remember anything. Every single setting reverted to the defaults. So the FM tuner was at the top of the dial and the volume was at the exact middle of the range, two ticks higher than we had left it. Annoying, but one has to set the default values somewhere, and those choices have some logic behind them.

Less logical, when we switched inputs to SiriusXM, we discovered that the radio’s default display was the time remaining in the current song. Not the title (our preference) the artist, or even the channel. The time remaining. Who chose that? Realistically, nobody did. Nobody defined the default behavior, so a developer chose the first item in the list of options. Presumably, it was the same developer who put the list of options in their current order. And most likely, that order came straight out of a list of capabilities someone gave him.

If some QA person questioned the behavior, the business owner or project manager decided there wasn’t time to fix it: “If we change the order of the list, every feature that refers to the list will need to be changed; that means reworking and retesting every menu selection. And if we start setting exceptions for the defaults, instead of always choosing the first possibility, we’ll have to decide what those exceptions are, recode the initialization sequence, and retest. For something that only happens once, when the radio is installed.” Because, of course, we all know the radio is connected to a battery, so it can’t lose power.

The really egregious design issue, however–and the one that convinces me that there were no design reviews and possibly no QA–is that by default, the radio goes into a store demo mode. That means the display cycles endlessly through a list of the radio’s features. Why would anyone want to see the list after they’ve purchased the radio?

Turning the demo mode off requires the user to find a menu semi-hidden behind a long press on a button that normally does other things, locate demo mode in that menu, turn it off, and save the setting before the menu times out and returns the radio to its normal display.

Why is this the default? Granted, any unit could be used as a store display. I’ll even grant that a store display unit is more likely to lose power than one installed in an actual customer’s car. But making every unit default to store mode suggests that either the radio is the victim of poor design practices and less-than-adequate QA, or that JVC prioritizes stores’ convenience over customers’.

One Step Closer

San Francisco has its Transbay Center back.

You may note there’s a word missing from that sentence.

Actually, purists would argue there are a couple of words missing. But I just can’t see calling it the “Salesforce Transit Center”. Corporate naming is the modern equivalent of product placement.

Back in the day–mostly even before my time–companies would sponsor a radio or TV program. For underwriting a large chunk of the cost of the show, the sponsor would not only get to put their name in the show’s name (remember The Jell-O Program Starring Jack Benny or more recently, Mutual of Omaha’s Wild Kingdom?) but would also get to have their products appear in the show.

Now sponsors pay a small fraction of the cost of production (Salesforce is paying $110 million over twenty-five years; the total cost of the terminal–not counting the recent repairs–was around 2.2 billion) but still want total recognition.

Is it any wonder people ignore the new names or nickname them into oblivion? I suspect that building in downtown San Francisco will be widely known as “The Transit Center”.

But I digress.

Anyway, the rooftop park, complete with its new non-decomposing concrete paths, reopened to the public yesterday.

Crowds were, well, not very crowded. But then, the reopening was only lightly publicized. Judging by the Chron’s reporting, most of the traffic came from people who stopped into one of the coffee shops that have doors leading to the park and decided to add a little sunshine to their caffeine fixes.

Fair enough. I’d likely have done the same if I’d been in the vicinity. (Let it be noted here that my previous employment was a short block away from the large hole in the ground now occupied by the terminal. I dare say that if I were still working there, I’d be hanging out in the park at lunchtime on a regular basis.)

But back to that missing word. There isn’t yet any transit in the Transit Center.

Debris from the repairs and the major re-inspections is still being cleaned up. And, naturally, the bus drivers who’ll be using the center need to be retrained in how to get in and out of the building. Or in many cases, trained for the first time. Those direct freeway on- and off-ramps can be tricky (and no, I’m not being sarcastic here; it seems like a potentially confusing transition.)

No date has been set to resume bus service, but official-type people are bandying “August”. Let’s recall that the center officially opened last August 10th. It might be nicely symbolic to have the official re-opening on the same day this year. (And it would be almost purely symbolic. The tenth is a Saturday, and most of the transbay buses don’t run on weekends.)

As for the ground-level businesses that everyone hopes will attract non-commuters to the Transbay Soon-To-Be-Transit Center, they’re not scheduled to open until “fall”.

But the park is open. That’s progress.

SAST 14

Today’s Short Attention Span Theater is not brought to you by disease or lack of sleep, it’s just an excuse to deal with my to-do pile.

First, a brief administrative note.

I will be attending the Scott Joplin Ragtime Festival at the end of the month. I’m not planning a book signing or any other formal event, but The RagTime Traveler will be available for sale*. Come on down to Sedalia, enjoy the music, pick up a book, and I’ll be delighted to sign it for you.

* Dad’s ragtime books, both fiction and non-fiction, will also be in the festival store. In my totally unbiased opinion, you need copies of all of those as well.

While I will take my laptop along, I don’t plan to write any blog posts. I’ll make sure to have a post for Friday, May 31–I don’t want to be responsible for riots caused by cat deprivation–but other than that expect silence between May 28 and June 4, with a return to the usual schedule on June 6.

Second, I’m a little disturbed to discover that El Sobrante* is more dangerous than I’d thought.

* For those unfamiliar with the Bay Area, El Sobrante is the closest of the several cities that border the part of Richmond where I live.

Over the years, I’ve gotten accustomed to the suspicious sorts lurking in the local undergrowth, but it appears that a new threat is moving in.

According to a recent post on everyone’s favorite unbiased news source–Nextdoor–“[…]a somewhat large buck with velvet covered antlers jumped out from the side… he mean mugged us hella hard and took a few quick steps towards the car…”

That’s right. As if street gangs of turkeys and terrorist coyotes aren’t bad enough, now we’ve got to deal with deer carjackers. It’s a bad neighborhood, obviously, and getting worse.

But I have to wonder: how the heck did the deer expect to drive the car to the chop shop? He could probably hold the key between his hooves, but it’s not like the driver’s seat can be adjusted to fit his shape. For that matter, what kind of payment would he have been expecting? I’ve heard that fences pay chicken feed, but salt licks?

Anyway, moving on.

The big story a few days ago was that Microsoft is working on tools to (as the Chron’s headline put it) “secure elections”. Which is great news as far as it goes.

Microsoft is doing it right: making the source code freely available, so anyone can audit it and any company in the voting machine field can use it.

The thing is, it’s not a complete voting system, and the value of Microsoft’s software is only as good as the implementation. Voting machine companies have a justifiably poor reputation for the quality of their coding. You can have the greatest software in the world for allowing voters to verify their ballot, and it’ll be absolutely useless if the rest of the software and the hardware it’s running on is riddled with security holes.

How many voting machines run on Windows XP, an operating system that has been completely unsupported for half a decade? (Probably fewer than the number of ATMs running on OS/2, which has been dead for three times as long. But I digress.) Sorry, not totally unsupported. Microsoft just released a security patch for XP. How many of those voting machines running the code are going to get the patch? I’m betting on a percentage in the single digits.

Also, as the articles point out, Microsoft’s new code doesn’t support Internet voting (something far too many people want, given the woeful state of the art) or vote by mail systems, which are increasingly popular.

I’m not running Microsoft down. As I said, it’s a step in the right direction. But we as a country need to take far more than just that one step.

And, finally, no SAST post is really complete without a mention of either the Bay Bridge Bolt Botch or the Transbay Terminal fiasco. I don’t have anything on the BBBB, but there was a brief note in the Chron a few weeks about about the terminal.

The cracked support beams are nearly repaired–though we still don’t have a date for the grand reopening. What we do have is word that the paths in the rooftop garden are going to be replaced.

Those paths, you may remember, are made of decomposed granite, and even before the terminal was closed, the granite was decomposing even further. So the decision has been made to repave the paths, this time using concrete.

As local megaconstruction repair projects go, it should be a comparatively cheap fix, no more than half a million dollars or so. The city and the contractors are, of course, arguing over who is at fault for the failure of the paths. We all know who’s going to wind up paying for the repair, though, and it isn’t either of the arguing parties.

SAST 13

Hello and welcome to the latest edition of Short Attention Span Theater. Lucky Number 13! For those of you new to the blog, sometimes I do an SAST because I literally don’t have enough mental focus to write a full post on any subject. More often, it’s my way of clearing the blog’s to-do list of ideas that aren’t worth an entire post of their own.

I’ll leave it to you to decide, based on the internal evidence, which category this is in.

Ready? Too late, here we go anyway.

Perhaps you remember my handy theatrical guide to long-running news stories. For the record, the Bay Bridge Bolt Botch stayed in Act One for an incredible length of time before zipping through Acts Two and Three, bypassed Act Four entirely, and is now in Act Five.

I’m pleased to see that the Transbay Terminal mess isn’t following a similarly distorted trajectory. We got out of Act One in a mere five months, and we’re now solidly in Act Two. In mid-March, the Transbay Joint Powers Authority threw all the blame for the debacle on the various contractors, individually and collectively.

Naturally, by the end of the month, two of the three contractor had responded, saying in essence, “Hey, we did everything right. Take a look at the third contractor and the designer. They’re the ones that really muffed it.”

Putting on my QA hat for a second, I’ll just note that one of the jobs of the QA team is to point out problems with the design. It’s always cheaper to fix an error before it gets built. That’s true whether you’re talking about software or buildings. If the contractors had concerns about stress on the beams, why didn’t they raise them before construction started?

Anyway, I find it interesting that, so far as I can tell, the third contractor has yet to respond to the accusations of the TJPA and the other two outfits. Clearly, we’re not quite finished with Act Two, but we’ve got clear signs that Act Three is imminent.

That being the case, we may find ourselves watching a bold theatrical experiment, with multiple acts being staged at the same time. If the gimmick works, we might even find ourselves watching Acts Three, Four, and Five simultaneously.

I expect rapid developments in the play come summer. Remember, the terminal is supposed to reopen in June; we can expect a large PR push to convince commuters that it’s safe. That’s almost sure to provoke a lot of finger pointing and the launch of the inevitable lawsuits and countersuits.

Moving on.

For anyone interested in our litter box experiments, we’ve settled on a new long-term litter plan.

We tried Sledpress’ recommendation of Dr. Elsey’s litter with the Formerly Feral Fellows, and it did work as promoted. There was some scattering, though not as much as with the Nature’s Miracle. It did well at controlling odor, and the dust wasn’t as bad as some of the reviews led us to expect. On the downside, it’s hard to find locally, and even allowing for the fact that we got an entire month out of one jug, it still comes out more expensive on a per use basis. Most importantly, though, it seemed as though the Fellows weren’t very enthusiastic about it. They used their other box, loaded with more conventional litter, more often than before we introduced them to Dr. Elsey.

The more conventional litter we tried out is SmartCat All-Natural Clumping Litter. It’s grass-based, clumps very well–I’d even say “frighteningly well” given the size of some of the clumps we’ve found, and does a decent job of controlling odors. We are getting more scattering than I’d like, but it’s at a manageable level. No litter is perfect, but this stuff seems good enough that we’ve converted all but one of the indoor boxes to it.

The exception is currently using up what we expect to be our final bag of World’s Best Cat, and we’re finding that the gang would rather use the SmartCat boxes than the one with WBC.

Finally, there’s this.

Regular readers are already aware of my feelings about the devil’s condiment.

I’m delighted to note that we now have scientific evidence to support my purely logical reaction to that stuff. Forget HoldThatMayo, Bon Appetit, and JSpace. While it’s nice to see fellow travelers, one can’t help but note that their appeals are based on paranoia, emotion, and prejudice.

That’s why it’s great to see the word from Popular Science that there’s well-grounded, firm scientific support for the contention that mayonnaise is eeevil.

Take cheer, my brethren. The battle will be long–I expect the pro-mayo forces to be at least as persistent as the anti-vaccination loons–but with Science! on our side, we’ll win in the end.

Oopsie!

Oof. Just, like, wow, man.

Sorry. I don’t get a chance to talk about the theory and practice of QA very often, so when an opportunity like this comes along, it’s a bit overwhelming.

According to the Transbay Joint Powers Authority, four different groups of inspectors missed the construction goof that caused the Transbay Terminal closure. Four!

Granted, my QA background is in software, but I can’t imagine that construction is greatly different, at least in terms of the methodology. The basic idea is that QA sits down with the specifications–ideally, QA is involved in the process of creating the plans, helping to root out ambiguities and spot danger points early, but I’m trying to keep this simple–and produces a plan.

The plan describes what QA will test and how they’ll test it. The level of detail varies, but the intent is that it covers all the different types of tests in language that non-QA people can understand. QA then designs the specific tests, executes them, logs bugs (defects, places where the product doesn’t match the design), and eventually produces a final summary that describes what was actually tested, summarizes all the bugs, and documents any changes between the test plan and what was actually done.

Naturally, I’ve massively simplified this process. In the real world, many of the steps will be executed in parallel. There’s buy-in at multiple points in the process. Each test is documented: when it was run, who ran it, and what the results are.*

* In the software world, QA kills a lot of trees. I imagine it’s even worse in the physical world.

None of which explains how three separate groups–the outfit that made the beams, the company that installed them, and the general contractor who oversaw the whole process–apparently failed to confirm that the beams had been properly installed. Specifically, that holes cut in the beams had been ground down to prevent exactly the sort of cracks that developed over Fremont Street.

I have to wonder whether it was a planning failure–nobody said “Hey, we’ll test the welding access holes,” and none of the parties involved noticed the omission–or a failure to execute planned tests.

It’s worth noting that the holes in the First Street beam were cut differently, so that grinding wasn’t necessary. Why didn’t anyone notice the discrepancy? Was the change by design or error?

I’m less bothered by the failure of the fourth group involved. According to the Chron, a separate company did spot inspections. On the one hand, they specialize in QA. On the other hand, spot inspections will miss things. That’s part of the definition. They’re designed to give you a big picture; in the software world, you might do a spot check–a small fraction of your tests–to confirm whether a new module is ready for the full test. If, say, five percent of your tests (chosen to hit the areas of highest risk) show a lot of bugs, you’ll probably send the module back to development and say “try again.”

In this case, the Transbay Joint Powers Authority wanted checks to be sure the proper process was being followed. So, whether the tests were planned but not run, or never planned in the first place, it’s easy to see how the omission could have slipped through the cracks (sorry) of the spot check.

Anyone got an inside view of the TJPA? I’d love to know where the fault actually lies.

Two for the Price of One

It’s not a Short Attention Span Theater*, but I’ve got a couple of items for you today.

* For reasons too complicated to explain**.

** Which is entirely untrue, but sounds better than “For reasons.”

First up, raise your hands if you remember multimedia artist Xathaneal Todd. Don’t feel embarrassed if you don’t. That was way back in ’15–practically prehistory in blogging terms. You can and should refresh your memory.

It turns out his skills aren’t limited to the material arts. Turns out he’s also a composer and actor.

Okay, I’ll admit I have no proof the musician and performer are the same person as the artist. But how likely is it there might be two, much less three, Xathanael Todds? Of the same age? All living in the Fairfield area?

One of the wonderful things about blogging is the way the unexpected turns up. I’ll admit to having forgotten about Xathanael myself. Until Sunday afternoon when a modest little press release popped up in my mailbox.

I say “modest” because it refrains from waxing eloquent about its subject, choosing instead to simply announce the appearance of Mr. Todd as the star of a forthcoming production of Alladin Jr..

Regrettably, Fairfield is a bit outside my usual range. But if any of you are going to be in the area January 31 through February 2, I’d encourage you to check out the show. You will, of course, be obliged to report back. In detail.

Moving on.

Gratifying as it was to hear from Xathaneal (or his press agent), what really warmed my heart-cockles this past weekend was a story in the Chron.

Remember how, despite all of the Bay Bridge’s well-publicized problems, nobody has ever taken any of the blame? No accountability, no public apologies. And there’s certainly been no indication that Caltrans will do anything differently in the future.

Well, it turns out the Transbay Transit Center officials are made of sterner stuff. The article quotes Mark Zabaneh, the Transbay Joint Powers Authority executive director as saying “Obviously, something went wrong with the process for this to happen.” It goes on to describe–at a very high level, naturally (it is a newspaper, after all)–the review processes during design and construction and then cites Zabaneh again as specifically stating that officials need to find out where they went wrong.

Look, I know that’s a long way from resolving the mess–and we still don’t have even an estimate of when the terminal will reopen*. It’s not even an actual apology. But it is a recognition of responsibility. That’s such a major improvement over Caltrans’ handling of the Bay Bridge Bolt Botch that I’m quite giddy with delight.

* Unofficially, the Chron suggests that the repair work could last well into March, which, with the need for testing the fix, could push the reopening into April. One hopes they won’t cut the new ribbon on the first of the month.

Kudos to Mr. Zabaneh for his honesty. May it continue through the repair and the post-mortem project examination.