Sunday, August 4, 2024

The fixed antenna connector was not as fixed as I thought.

 

The above chart is a graph of the ranking of my ADS-B receiver, which I started tracking shortly after I made the fix to the antenna connector.  You can see it steadily improving from the 22nd of March up until around the end of April, where it reached its lowest (best) point.  Something happened there to make it start getting worse, and I'm not entirely sure what.

I've gotten a "real" crimper, and I'll try and replace the connector again and hopefully better this time, but also I'll need to check the software to see if there were any updates in that timeframe.  I'll need to start keeping a closer eye on it, though.

I have noticed a slight improvement in recent days, but we also had a large tree taken down Monday so it's possible that's the source of the improvement.

Sunday, July 7, 2024

I'm in the wrong Google Timeline

I like checking out my Google Timeline (the link, of course, will take you to yours, not mine).  It's neat to see where I've been, and how far I've traveled, and the patterns as they change over time.  I suppose I should be worried that Google is building a profile of me, but if I genuinely cared about that sort of thing I'd get a second, dumb phone.  They're cheap enough, especially as a pay-as-you-go thing.

I was going back over my timeline the other day, and noticed something extremely off.  A few weeks before that I'd visited my massage therapist's office, and in the interim she'd moved to a new location. The old map from before the move was showing that I'd driven to the new location instead of the old one.

This... seemed like a bug, and I said so. I got a very earnest "Diamond Product Expert" telling me it was an either-or situation and an unsolvable problem. I remembered I'd seen this before when a beloved sandwich shop (Sub Conscious) closed and all of a sudden seemed to be located approximately halfway between Seattle, Washington and Portland Oregon but also a couple hundred kilometers West into the ocean...

 

I'd love to think this new location has something to do with submarines instead of sandwiches, but I've usually found edge cases like this simply to be manifestations of garbage in the system.

Anyway, this is emphatically not an unsolvable problem. Google can save historical data about location and the businesses/points of interest/whatever at that location.  They know when things move and they save that data and move on.  It's not a huge space concern because THEY'RE ALREADY SAVING ALL THIS OTHER DATA.  I'll also bet you real folding money that if "Home" moved this would not happen.

It's just a problem they choose not to because it doesn't fit their business plan. That's it, and it's ok to admit that rather than try and fabricate some theoretical technical reason.


Sunday, June 23, 2024

Google Maps navigation has kind or ruined the scenic drive.

The Setup

NC DOT maintains a list and map of scenic byways.  It's a great way to tool around in the car for a day and see some historic sites around the state, and I'd been thinking about checking out one or more and the perfect opportunity came up with a few free days from work.  The plan was that L. and I would drive the Colonial Byway and then head for Mayo River State Park for a little hiking, and then overnight somewhere near there before heading home the next day.

Things that are possible.

  • Creating a custom route following specific roads in Google "My Maps".
  • Creating a multi-destination "trip" in Google Maps that you can navigate.

Things that are not possible.

  • Creating a custom route following specific roads that you than then navigate with Google Maps.
  • Adding more than 10 destinations to your Google Maps trip.

 So, I did the best I could and added a bunch of small NC towns in a way that approximated the scenic byway as best as I could and we set off on our way.

Problem 1: Setting off on our way.

It turns out when you do a multi-destination trip and you tray to start navigation somewhere besides your original origin Google Maps assumes that all interim destinations are null and void and just gives you directions to the final destination.  So that was confusing for a bit as our trip time was much shorter than expected.

The solution was to EDIT THE START POINT TO BE OUR CURRENT LOCATION AND RESTART NAVIGATION.  Stupid and baffling, but we got on our way and passed the first "destination" treating it as a waypoint.

Problem 2: Treating destinations as waypoints.

So initially Google was like "TURN AROUND, YOU MISSED THE FIRST DESTINATION." for an uncomfortably long time, and then we pulled into a parking lot and it gave us a dialog telling us we'd passed the first destination and did we want to Continue or Stop Navigation?

Problem 3: Pressing "Continue"

Pressing "Continue" stopped navigation (and would... er... continue to do so about 2/3 of the time for reasons we couldn't determine.  We restarted navigation and deleted the first destination and were again on our way.  We finished out the scenic drive in this way more or less stopping occasionally to sort out Google Maps navigation.

Good gods, why didn't you just use a paper map?

We had a map, but we hadn't gone over the route together and I hadn't printed out a turn list.  I could have bought a map and drawn on it, but again I thought this way would actually work.  It turns out the Google Maps navigation is not only not suited to a scenic drive but also too buggy to do multi-destination drives without hassle.

So, what now?

Well, I'm not going to try and do it that way again. I've looked at a few other navigation apps - particularly ones that were rally-oriented and did actual waypoints - but remarkably few of them support Android Auto.  In fact the only one I saw that did was OsmAnd, so I'm checking that out currently.  Suggestions are always welcome.

Who cares about scenic drivers, anyway?

Well, if I want to do more of this, me. But also anyone who wants a finer degree of control than just "avoid highways" or "avoid toll roads".  Speeds on superhighways can be a little daunting for some drivers, but you also don't want to get stuck going light to light on some exurban stroad.  Are folks going to put the effort into planning a route to do that?  Hard to say, but it'd be nice to have the option.

Sunday, April 14, 2024

The Journey Home _or_ Three Days of the Southwest 737.

Forward

I should state, up front and for the record, that every single Southwest employee we encountered was courteous, helpful, and gracious even when under what must have been phenomenal pressure. The problem here is the policies from the top that cut costs everywhere from maintenance to headcount; causing what could be an extremely busy day to become a cascade of unrecoverable problems.  That's not the fault of the customer facing folks, the fault is squarely at the top.

I should also mention before going any further that my partner Laurie was an absolute trouper during this whole ordeal and I can think of few other folks who could hope to equal her.

By now there have been more than a few articles written about Southwest's meltdown on April 10th, 2024.  Here are a few of them, though.

https://viewfromthewing.com/post-eclipse-turmoil-southwest-airlines-melts-down-at-austin-airport/

https://www.chron.com/culture/article/southwest-delays-austin-19397525.php

https://www.atxjetsetter.com/post/another-day-another-southwest-meltdown-in-austin

Occasionally I poke myself to see if I'm genuinely home.

That's probably an overstatement, but those three days were an air travel ordeal worse than December 2022 (but less disappointing, to be fair). It's probably best to begin at the beginning.

2024.04.10

We were returning from the hill country of Texas (Canyon Lake, to be precise) and a delightful eclipse vacation to Austin for a direct flight back to Raleigh. The weather (which had been cloudy during the actual eclipse) was almost obscenely good.  Clear and not too hot and very dry.  Perfect weather for viewing an eclipse that had happened two days ago.

We got into town and hit up a cute little sushi place near the airport for a very light meal (counteracting the heavy, rich meals we'd had over the vacation). We took a little walk post-lunch and got to the airport and turned in the car, probably around 1p, as our flight was at 3:40p.

Walking towards the terminal, we realized something was awry.  There was a huge line stretching away from the terminal, and it was the Southwest baggage check-in line.  With no other choice to make, we got in line and started threading our way through the multi-folded, half-outdoor line (shown below).

This is a picture of Austin airport, with a crudely drawn line on it depicting the path of the bag-check line at one of it's longer states.


There are lots of reasons Southwest claimed this happened: Increased traffic due to the eclipse and a few other Austin events. The aforementioned bad weather the night before cancelling a bunch of flights. A problem with an airport bag belt.  But, none of the other airlines had this problem.  All except Southwest were moving smoothly with the increased traffic.

3:40p came and went and we still hadn’t made the bag check-in.  We’d gotten much closer, but they weren’t pulling folks out of line to try and make their flights because (it seemed) there were simply too many of us to manage.

At around 4:40p or so, we made it to the bag check-in.  We were rebooked, and told there were no hotel vouchers available.  We stumbled back out into the light of day, bewildered, still seeing a huge line.  Sloughing off our masks we sat down while I called the place we’d stayed last Friday and secured a room. Waiting an interminable time for the shuttle and we got to the hotel, ordered a pizza delivery, and crashed.

There was no warning from Southwest about arriving extra early.

2024.04.11

We got up early, put on sunscreen and hats and more comfortable shoes, bought some bottled water, and steeled ourselves and got to the airport five hours early at noon.  This turned out to be too early as everything was running smoothly today and we weren’t able to check our bags until four hours before our flight. Waiting around, we eventually checked our bags and sorted out our boarding passes to find out we were leaving at 4:40p (that part we knew) for a flight connecting to Raleigh through Nashville (that part was a surprise, I never got e-mail confirmation of the change).  We met Dad in the airport (he was heading to Roanoke through Charlotte), hung out for a while and made our way to the gate.

Because of a gate change and other travel, we were late starting to board.  Because of so many carry-ons, we were late completing boarding, and also there were numerous recalculations for fuel due to the weight of all the carry-ons.  A few people decided to get off at this point, forcing more recalculations and paperwork.  We took off over an hour late.

Needless to say, we missed our connection.

The Nashville agent at the gate was pleasant enough, giving us a voucher immediately for a hotel and travel vouchers worth $200, and rebooked us on the spot.  We had a flight leaving at 11:45a the next day.  We picked up our bags and headed outside to call for the shuttle, like the gate agent had said.

It took several tries to get through and there were lots of disconnections, but eventually a very harried sounding dude directed us to the shuttle stop and told us to wait.  There were a few other folks there from the flight (one was dangerously close to missing a wedding the next day) and in a half hour the shuttle came and, uh, ferried us to the hotel.

There was one guy at the counter (the harried guy from the phone) checking folks in as fast as he could while trying to answer essentially the same question we’d asked of him over the phone again and again and again and again. He gave us all a free water from the cooler and got us checked in and in our room, though.

After getting settled in, we headed back down to grab some food from the little snack bar area.  We found two veggie wraps and a beer and some chips and in the process of waiting in line again to charge them to the room (we’d provided a credit card for sundries) we discovered that Southwest had overbooked the hotel and people were about to show up with vouchers but no more rooms to stay in.  Fortunately for us (but not the guy at the counter) we slipped away before that shuttle arrived.  We pounded the wraps and crashed for the night.

2024.04.12

Getting up early again, we discovered that today’s flight was connecting to Raleigh through Atlanta. Remember that every airport in this… journey has a direct flight to RDU.  I suddenly had visions of endless missed connections causing us to rotate through Austin, Nashville, Atlanta for all eternity. Austin, Nashville, Atlanta.  Austin, Nashville, Atlanta. Never again reaching Raleigh. Austin. Nashville. Atlanta.

Snapping out of it, Laurie and I did a quick vibe check. We were now only 8 hours from Raleigh. Should I just rent a car and drive it; damn the cost?  After a few minutes discussion, we decided that this ordeal, on top of the not perfect rest that comes with a vacation sometimes, had worn us down too much to embark on an 8 hour drive.  I simply did not feel up to it.  We checked out (the desk clerk of the morning sent the total bill to Southwest) and caught the shuttle to the airport.

Check in was smooth and uneventful, boarding was mostly quick and efficient, but we still inexplicably dallied on the ramp for what seemed like an eternity but in reality was probably only fifteen or so minutes.  Still, the flight to Atlanta was short and our connection was tight and so I was worried that Southwest was unable to make a connection to Raleigh.

But, in a welcome spot of luck, not only was the gate for the connection the same as our arrival gate, but the plane was the same as our arrival plane, changing only in flight number and some crew.  We hopped on in short order, now uncensored about loading times or delayed take off, and were mercifully back on the ground in Raleigh and in possession of our luggage by 4:15p.  We shuttled out to the lot, got Laurie’s car and drove home, blinking in disbelief.  A quick trip to get takeout and we were in bed by 8:30p, fully and truly exhausted and happy to be finally home.

Carry-On Culture Needs to Change

There is a fundamental truth about domestic air travel in the US that everyone seems to be ignoring: 

There is not enough overhead bin space on a current airplane for everyone to bring a maximum size carry-on.

Like, we all know this, right? It's like somewhere between one half to two thirds of the passengers can have a full-size carry-on and then we're out of overhead bin space.  It's probably like three fifths, but I haven't actually done the math. 

Every domestic flight I've been on in recent history has been packed to the max with people and baggage, necessitating folks who are boarding last checking their bags and adding delays to the departure time and potentially foiling connections.  Two out of my last four trips have had this issue. One disastrously.

And of course they do this on the jetway, because even though the gate agents always ask folks to check their carry-ons because "we have a full flight today" (we always seem to have a full flight today) nobody is going to do it unless they're made to do it.  The process of determining that is always a depressingly long and manual process of checking bin space several times, rearranging bags, and firmly putting your foot down and saying "no, you have to check your bag".

I'm not going to blame passengers, though.  I like to check bags, but if I had a roll-away carry-on this is exactly how I'd behave, too.  Let someone else check their bag.  Let someone else take a different flight.  I might could do either on a return flight, if I didn't have a connection to make, but there's almost always a connection to make.

I'd be interested to know when the full-size carry-on form factor became a thing, and how it was decided, because it's clear the conditions have changed and nothing else has changed with them.  Did it ever work better?

So. What to do?

Hell, I don't know.  I'm not a frequent flier.  I routinely check my bags and everything I carry on either fits neatly under the seat in front of me or is a CPAP that usually easily slips in the overhead bin between the nearly ubiquitous full-size carry-ons.  But here are some ideas that might speed things along (assuming on time is your metric here).

  1. Smaller max carry-on size. I like this one the best?  But I'm a weirdo who checks his bag so I could be wrong here. If we shrink the carry-on size so that everyone (or at least most people) could have one, the problem solves itself. Bags over that size get checked automatically at the gate.  This still might cause delays in boarding, though.
  2. More gate agents. A single gate agent means things GRIND TO A HALT when you start checking carry-ons. Having a line for carry-ons to divert to when you get over a certain number where they get their bag checked and then can continue on while the rest of the passengers can cruise on through serves all my desires neatly.  Just one more person could fix this.
  3. No more checked bag fee. Is this the largest driver of carry-on culture? No clue. The other driver is "What if my bags are lost?", but I feel like checked bag fees drove this more and now it's ingrained even if some airlines don't do it.

The problem, of course, is that all these things cost money that the airlines are unwilling to spend because the penalty for a failed connection or outright missed flights simply isn't high enough to affect the bottom line. It should be, more than just hotel/food/travel vouchers or whatever.  There should be real financial penalties for this, because delays cascade across airports and airlines and bung the whole system up.

Anyway, that's my thoughts on this. Please sit back, relax, and enjoy your flight. I know you have choices on what to read on the internet, and I'm glad you chose this.

Monday, March 4, 2024

More ADS-B antenna fooling aorund.

Instead of buying a whole new ADS-B antenna to fix the cable problem, I just spent around an hour this weekend in the attic affixing a new connector onto the end of the much-shortened cable (there was quite a bit of slack). My performance improved almost immediately (and the range seems a little better), but we'll give it a week or so to settle out before making more changes (like adding in the splitter).

Soldering tiny connectors while sitting cross-legged on the attic floor was made more bearable by...

  1. ADEQUATE LIGHTING.
  2. Bringing up a board that I could clamp a small pair of "helping hands" to and do some cutting on.
  3. Taking a break halfway through to go retrieve my desk lamp with magnifying glass insert so I could stop trying various combinations of glasses and squinting and actually see what I was working on.
The crimper was not designed for this particular type of connector, so it was just sort of adequate but the numbers don't lie.  I may give it another whirl or just add the splitter in next weekend and see if my numbers suffer measurably.  UAT isn't SUPER popular around here, but it does seem to be catching on with the recreational set.
 

Monday, February 19, 2024

Various plane-tracking system woes.

According to FlightAware, I've been tracking aircraft via ADS-B signals on the 1090MHz band since about mid-2017.  That's honestly so long ago that I've forgotten about what prompted me to do it (other than having a passing interest in RTL-SDR, a spare 20 bucks, and a Raspberry Pi doing not much else). I've got a picture of my first system, though. A little stubby antenna next to my larger HDTV antenna in my attic.

IMG_20170731_205710

Not much to look at, to be sure, but things progressed slowly but surely from there.  Sometime in 2020 or 2021 I got a new antenna, and the switch made a HUGE difference. You can see me switching the antenna over in the video below.

Not too long after that I started with the Balena ADS-B project, and sometime in 2022 I started pulling in 978MHz UAT signals as well.  It's been fun and a nice outlet for my amateur radio adjacent desires.

But, sometime a few months ago, my reception quality dropped precipitously. I hadn't changed any setup stuff, and there hadn't been any recent upgrades.  The antenna installation on the roof looked ok, too, so the only think left to do was poke around in the attic to check out if something was awry there.  I grabbed a spare laptop and climbed up the ladder to check things out.

Everything looked more or less fine, so I unplugged the USB cable from the Raspberry Pi and plugged it into the laptop (also running dump1090-fa) to see if the laptop fared any better.  Slightly?  But also the USB cable was quite loose when unplugging it from the Raspberry Pi, so I plugged it back into the Pi firmly and went back down into the house, a little dejected.

Suddenly the signal was a whole lot better!  Was it the USB cable all along?  Since everything in the attic was just sort of sitting on a shelf, I decided to fasten the equipment more firmly and get some new USB cables.  Here's the result of my handiwork.

PXL_20240211_194559126

(There's a newer HDTV tuner under all the ADS-B stuff, and the splitter will eventually be hooked up once everything ends up being sorted)

Securing everything and making the signal path look more tidy (as well as getting new USB cables) did absolutely nothing. In fact, it went back from the week of being much better to bad again.

After another week of watching my stats fall, I took another crack at it.  Did something about my laptop setup change the USB receiver?  Looking at both the setups, they were identical.  Was the automatic gain control (even though both the laptop and the Pi did it) the problem?  No.  I dragged the laptop back into the attic again to see if switching the connection back and forth would "fix" the problem.

I had the Pi feeder up on the laptop screen so I could do a before and after comparison, but before I switched the USB receiver over to the laptop I took a moment to tidy up the antenna cables.

My signal was back!

It turns out the antenna cable has been "not great" for a while.  I'm not sure what caused this, but (unfortunately) the other end is bonded to the antenna, so just replacing the cable isn't an option.   Thinking that the problem is in the last ten feet or so and having a lot of slack, I've ordered a new connector and we'll see if that can clear it up. For now, my stats are rising again.

I also wrote a quick script to grab the stats file from the Pi and print it to a log every few minutes.  Maybe I should start a github for this sort of thing.