Launch logistics: Booking a trip to see Falcon Heavy fly on three days’ notice

I’ve had the idea of covering the first launch of SpaceX’s Falcon Heavy rocket in the back of my mind for the last few years, but I didn’t book my travel for Tuesday’s launch until Saturday afternoon.

I was waiting for a confirmation of the schedule from the company that would be more solid than a notional “No Earlier Than” date, and which would then let me know if I could still attend a Yahoo Finance cryptocurrency conference in New York on Wednesday. Besides, I knew that D.C. and Orlando often represent a cheap city pair.

The schedule details I needed from SpaceX arrived shortly after noon Saturday, so I got to work–one travel component at a time.

Having to reach the Kennedy Space Center by 1:15 p.m. to visit Launch Complex 39A ruled out some decent mid-day fares. But Southwest’s site showed a 6 a.m. nonstop out of National for only $50. Sold!

Then I canceled the D.C.-NYC Amtrak reservation I’d had for Tuesday night (I appreciate that the railroad still lets you do that for free until 24 hours before departure) and booked a Tuesday-night flight from Orlando to New York to replace it.

I went with United for that leg, spending a little extra (a still-reasonable $155) to fly on an airline where my frequent-flyer status would allow a free same-day-change to a Wednesday flight to Newark if a launch scrub required that. A few more clicks to book a rental car and one night’s lodging, and I had launch travel solved… or so I thought until an hour after a liftoff that got pushed back to 3:45 p.m. by upper-atmosphere winds.

At that point, the “OMG! OMG!” shaking had stopped, I’d filed my copy, and Google Maps indicated that the usual 45-minute drive from KSC to Orlando would run an hour and 15 minutes. Nope! As horrendous post-launch traffic dragged Google’s arrival estimates past my flight’s boarding time, I called United to see if they had space on the morning’s first MCO-EWR nonstop, a 5:36 a.m. departure. They did.

After dropping off my rental car and getting through a mercifully quick security checkpoint (is there a better exhibit for TSA Pre or Clear than MCO?), I ran to my original flight’s gate and saw for myself that the plane was gone. I called United back, the rep bailed me out of the consequences of my overly-optimistic travel tactics by putting me on that 5:36 a.m. flight for free, and then I opened my laptop–tethering off my phone because the airport WiFi didn’t let me connect–to book a hotel barely two miles away for $90.

By then, it had been some 10 hours since I’d last eaten, so I treated myself to a nice dinner at the airport. (If you, too, get stuck at MCO and want something more original than the terminal’s fast-casual brands, head upstairs to McCoy’s in the Hyatt Regency). After a prolonged wait for the hotel van, thanks to no visible signage indicating that these shuttles could pick up at either of two spaces on the B side that sit maybe 800 feet apart, I was in bed by around midnight.

I somehow woke up one minute before the 4:15 alarm I’d set on my phone and was through security 40 minutes later. You can image my relief at seeing my upgrade clear, then having a quick NJ Transit ride from EWR to Manhattan help wrap up this prolonged commute by 9:10 a.m.

A long and informative day ensued with Yahoo colleagues, most of whom I hadn’t seen in months, and various cryptocurrency experts. But then my travel luck ran out again when my train to D.C. left more than an hour and a half late. Twitter, not Amtrak, informed me that this was the result of a tragedy–a northbound Acela striking and killing a person walking along the tracks in the Bronx, which led police to close the railroad for two hours.

That meant I didn’t get home until nearly 1 a.m, almost 21 hours after my day had begun. But I did get to sleep in my own bed, and I came home with two posts filed from KSC that more than covered my travel costs as well as dozens of photos (since edited into a Flickr album) and one unusual recording that you can hear after the jump.

Continue reading

Advertisements

Sharing stories from Apple News considered harmful

Last Tuesday, Google delivered some news that open-Web advocates have long awaited: Stories posted in the speedy, Google-developed Accelerated Mobile Pages format and served up via its even-faster caching service won’t zap onto the screens of mobile devices at google.com addresses, not the domain name of their publisher.

The avoidable but common facet of the AMP experience has bothered me since my early encounters with Google’s attempt to make the mobile Web less janky–it led the explainer I wrote for Yahoo two years ago. Google is now moving to fix the problem it helped create, which is welcome news in any publishing format.

(Specifically, Google will adopt a new page-packaging standard to preserve site domain names. In last Tuesday’s post, AMP project tech lead Malte Ubl says we should start seeing the results on our phones in the second half of this year.)

This, however, leaves another address-eating annoyance on the mobile Web: Apple News. This iOS app is a pleasant way to browse and read stories; like the open-source AMP, this proprietary format cuts out the cruft that can clog mobile reading.

But when you tap its “Share” button, Apple News serves up an apple.news address. And unlike even Googled-up AMP addresses, this one offers no hint after the domain name of where you’ll go.

The text Apple News pre-populates in a tweet or Facebook update–the story headline, an em-dash, and then the publication name–does. But on Twitter and Facebook, many people decide to replace that text with their own words, leaving users to guess what’s behind that apple.news address.

Apple appears to be doing this to ensure that other iOS users can read the story you shared in Apple News as well–its developer documentation even lists a story’s canonical address as a “not required” bit of metadata. But in the context of a button that can share a story on the public Web, that’s an absurd inversion of priorities.

Apple could fix this by coding Apple News to share a story’s original address when available, perhaps with an identifier to tell iOS devices to open it in Apple News. But knowing this company, I wouldn’t expect that any sooner than the arrival of a reborn Mac mini at my neighborhood’s Apple Store.

Instead, you’ll have to solve this problem yourself. If you’re sharing a story from Apple News, keep some reference to the publisher in your description. If that would cramp your social-media style, please take a moment to tap the share sheet’s “Open in Safari” button–then share the story from that browser, from whence it will have its real address.

Ban the panel prep call

Tuesday morning had me moderating a panel discussion, which made the workweek nothing out of the ordinary: I’ve done 20 or so panels so far this year.

I enjoy the exercise–when you only have to ask interesting questions, call out any departures from the truth, throw in the occasional joke and try to end things on time, you’ve got the easiest job of anybody on the stage. But there’s one part I resent: the inevitable request by the event organizers that everybody get on a conference call first to discuss the panel.

If it’s just going to be me interviewing another person and we’re in the same time zone, this need not be too bad. But more often, you have four or five people with widely varying schedules.

That leads to a flurry of e-mails in which the panelists or their PR reps try to pick out a mutually agreeable time–instead of, you know, using the e-mail thread to discuss the panel itself.

The con call itself is likely to run on some 1990s phone-based system, not any sort of online app that would make it easy to tell who’s talking (pro tip: when on a con call, play up whatever regional accent you have). Using a text-based collaboration tool like Slack that would let people on planes or an Amtrak Quiet Car get in on the conversation never seems to come up.

Last month, the only time the organizers offered for the prep call was 5 p.m. on a Friday when I had to get to Dulles Airport for a flight later that night. I replied that this wouldn’t work and suggested we “use e-mail the way God intended,” then wrote up an outline of the talk as I would have needed to do even if I’d hacked out time for a con call. The panel went just fine.

So if you ask me to dial into a con-call service to talk about what we’ll talk about on a panel and I suddenly get cranky, please understand that I’m just trying to act as if we’re doing business in the 21st century.

Recommended Precision Touchpad settings

I’ve yet to ease into a new computer without having to fuss with some of the default settings, and the HP Spectre x360 laptop sitting on this desk has fit right into the pattern.

Most of the tweaking has involved its touchpad, because I’ve always found the defaults in Windows to be too jumpy. (I’ve said the same about some Mac touchpad defaults.) But after a few days of clicking around the stock Synaptics software, I realized I should first dump that for drivers supporting Microsoft’s more elegant Precision Touchpad software.

The directions I found on Reddit (embedded here after the jump) worked, and then I could easily shut off the touch behaviors I couldn’t stand.

  • “Touchpad sensitivity”: I changed this from the default “Medium” to “Low” because, again, jumpy touchpads bother me. I may try turning it back off to see if disabling the following two options made the standard sensitivity acceptable.
  • “Tap with a single finger to single-click”: This is the one setting I change on every laptop. If I want to click, I’m more than happy to press down so the touchpad makes an audible click; having it treat a stray touch as a click leaves me randomly dumping the cursor into documents and windows and feeling generally stabby as a result.
  • “Press the lower right corner of the touchpad to right-click”: I disabled this because it’s easier for me to remember to tap with two fingers to right-click than to keep track of which side of this invisible line I’m about to tap.

I hope this advice–which should also work on any other laptop running Microsoft’s Precision Touchpad software–makes for a more pleasant laptop computing experience. If you have other suggested settings changes, please share them in the comments.

Continue reading

A Thanksgiving baking project: almost-no-work bread

Well over a decade since I got into the habit of baking sandwich bread from scratch, I still remember how nervous I was at first about winding up with a deflated loaf. The recipe I’m sharing here cuts that risk as close to zero as possible; all it asks in return is about 24 hours of time.

Because I, too, am a little hesitant to try out a recipe with that much latency, I waited to try the “No-Work Bread” recipe in my well-read copy of Mark Bittman’s “How To Cook Everything” (which you may have seen in the New York Times as “No-Knead Bread”). I shouldn’t have: This product of Sullivan Street Bakery owner Jim Lahey is the most fault-tolerant bread recipe I know, and if you start it by mid-afternoon Wednesday you can have it ready for Thanksgiving dinner.

(My apologies if you’ll be spending Wednesday afternoon on highways or in the air. Maybe bookmark this for Christmas?)

  • 4 cups all-purpose unbleached flour
  • 2/3 teaspoon active dry yeast
  • 2 teaspoons salt
  • 2 cups water, about 70 degrees
  • 1 tablespoon olive oil
  • 1 tablespoon cornmeal

Mix the flour and salt in a 2-quart bowl. Stir the yeast into the water, and after a few minutes mix that into the dry ingredients for longer than seems necessary. This may look like a mess, but as long as you don’t have any chunky bits left, you should be fine.

(Bittman’s original recipe calls for half a teaspoon of instant yeast, which I never buy because Costco sells regular yeast in a 2-pound package. The last time I made this, I forgot the 1:1.33 instant-to-active-yeast conversion and threw the non-instant yeast in with the dry ingredients. Everything turned out fine; as I said, fault-tolerant.)

Take a 3-quart bowl and coat it with the olive oil. Dump the dough into it, cover with plastic wrap, and leave it alone for about 18 hours. You’ll know it’s done, or close enough, when it’s risen to near the top and it’s covered with bubbles as if they were craters on the surface of the moon.

(While the dough enjoys that long rise, you may want to watch an episode of the Great British Baking Show for motivational purposes.)

Dust a clean surface with flour and pour the soggy dough onto it–taking a moment to enjoy the aroma of the risen, fermented yeast. Fold the dough over a couple of times into a ball, more or less, and cover it with plastic wrap for 15 minutes.

After that rest, scatter more flour on the dough and re-form it into a ball. Scatter the cornmeal on a silicone baking mat, wax paper, or a towel (as in, something that you can grab to lift the dough off the surface) and leave the dough ball there for two hours.

About an hour and 15 minutes into that last rise, put a 3- to 4-quart pot, cover included, into the oven and preheat it to 450 degrees. Half an hour later after hitting 450°, open the oven, remove the lid and dump the dough into the pot.

This is when the results–a damp glob slumped unevenly in the pot, part of it stuck to its side–may look like a culinary catastrophe. Ignore the untidy appearance, put the lid back on, and shove it in the oven for 30 minutes.

Open the oven, remove the lid and you should see that the bread has settled back into a somewhat flattened ball. Set the lid aside, close the oven and bake for another 20 minutes. If the crust looks browned like something in a real bakery, it’s done; otherwise, try another 10 minutes.

Let the bread cool for 30 minutes. Try not to eat it all at once.

Panel clock management

I spent part of Tuesday, Wednesday and Thursday sneaking a peek at clocks counting down.  Sadly, no rocket launches were involved: Instead, I had the less exciting but also important task of making sure that my Web Summit panels ended on time or close to it.

Web Summit panel clockGetting one, two, three or four other people to wrap up a conversation as a clock hits 0:00, as this week in Lisbon reminded me, is one of those skills where I still have things to learn.

Of the five I did at the summit, two required me to improvise some questions after I exhausted all the ones I’d written down–which, since these discussions only involved one other person, is something I should have known to be a risk.

Also predictable: The one panel with four other people went a couple of minutes over when I let one of the subject-matter experts have the last word, by which I mean words.

An on-time finish matters at a talkfest like Web Summit, where the stages have panels stacked up throughout the morning and afternoon and schedule overruns will result in people not being able to eat lunch or the audience fleeing for the reception that started five minutes ago. I continue to be in awe of the people who make that happen, considering both the overall chaos level of a 60,000-person conference and the high odds of a VIP deciding to be a windbag on stage.

As a moderator, I just need to allow roughly equal airtime in my role as verbal air-traffic controller–while also asking intelligent questions, not stepping on other people’s responses, throwing in a line or two that gets a laugh out of the audience, and trying not to close out the panel with something lame like “well, it looks like we’re out of time.”

At events that invite audience questions, you have the extra challenge of people asking questions that are more comments–the dreaded, time-wasting “quomment.” I can see why the schedule-focused Web Summit organizers usually tell panelists not to bother with audience Q&A.

It’s maybe one panel in three that leaves me feeling like I checked off all the boxes. I hope I can get that average up to one in two at some point. And maybe later on I can have the prospect of being the only person behind the mic for 30 minutes or more not make me quite so antsy.

Another experiment in spending Facebook’s money on a Facebook ad

Last week, Facebook offered me a chance to play with the house’s money: a $10 ad credit to boost my ode to RFK Stadium, which the social network’s algorithms had seen drawing an outsized audience on my page there.

Facebook RFK-post ad reportLike the last time I got this freebie, I could target people for the ad by geography, interests (as perceived by Facebook), age range and gender. Unlike the last time, I got this warning, Facebook’s belated response to learning that its self-service ad system was not magically bigotry-proof: “Ad sets that use targeting terms related to social, religious or political issues may require additional review before your ads start running.”

The logical demographic to target for a post about RFK would have been the greater Washington area–but Facebook didn’t present any such option. In a hurry and on my phone, I told it to target users in D.C., Bethesda, Silver Spring, Alexandria, Arlington and Fairfax.

Then I stuck with the default age range of 21 to 65+ and added the following interests: music festivals, Washington Redskins, Washington Nationals, D.C. United and local history. RFK being its dilapidated self, it’s too bad “peeling paint” wasn’t a choice.

Three days later, I got my results: The ad reached 847 people and yielded all of 26 clicks through to my post here. That leaves me nowhere near Russian propagandists in using money to get people’s attention on Facebook–even if in terms of reach I fared about as well as Sens. Mark Warner (D.-Va.) and Amy Klobuchar (D.-Minn.) did in their test purchase of ads to lure Hill staffers and reporters to a fake Facebook group.

But while I still see no reason to spend my own money on Facebook ads, I hope the site continues to throw out these freebies. It’s fascinating to see how the marketing machinery works from the inside; that alone easily justifies the time I put into my Facebook page.