iOS was born without support for multitasking, and when competing smartphone OSes (especially WebOS and Android) added it, it was occasionally mocked for the complexity it adds and the problems this can cause — the fact that a top Android app is a task killer was widely used to mock the whole Android platform, and it’s certainly frustrating under WebOS when you’re told you have too many cards open and need to throw some away before you can do anything new.

Problems iOS users don’t have, true. On the other hand, at least these OSes (a) let you run multiple tasks if you want to, and (b) have explicit support for ending tasks you no longer want for whatever reason. (And there are various reasons to want this: perhaps you don’t want it using resources, perhaps you want it to restart from its default launch state, perhaps you just don’t want it cluttering up your screen.)

Recently I got a new iPod Touch and it’s my first experience both with iOS 4 and with an iOS device with increased memory — in short, now I can try out the multitasking features that Apple added in iOS 4. At first blush, I really like it; it’s easy to switch between recently used apps and recently used apps seem to launch more quickly than they do from a cold start; I think this is a good improvement to the platform.

However, in the next week, I’ve seen at least 2 occasions where apps got in a bad state, and going back to the home screen and relaunching the app didn’t fix it (whereas it would have in iOS versions predating multitasking). Sure, these were app bugs and not platform bugs, but I didn’t know a way to actually quit and restart the app, short of rebooting the whole iPod, or launching a bunch of other apps in quick succession until it magically pages out the offending app so the next time I launch it it’s a clean launch (this is what I did, and it eventually worked).

While writing this post, I looked around a little and it turns out iOS 4 does have a built in task killer — double-tap Home to bring up the multitasking tray with icons for recently used apps, then click and hold on one of these until it wiggles and it’ll have a “–” badge which will remove it from the recently-used tray and, presumably, actually kill it. (Though I still don’t know how to tell which of the many icons that show up in iOS’s multitasking tray — if I scroll to the right, it seems to be a most-recently-used list of every app I’ve ever launched — is actually running.)

So iOS 4 is both more powerful and complex than I’d realized, but also, it’s achieved parity with the oft-mocked Android multitasking feature in both directions — now it has multitasking, but now it needs and has a task killer.

I find I still prefer WebOS’s more explicit interface to multitasking.

Why are we letting the terrorists succeed by making us act as if we’re frightened? Most of us aren’t, really.

—Tim Bray, “ongoing”

Vanessa and I just got back from a long trip during which we took and posted a lot of pictures, mostly posting from the road, and I thought I’d write up a description of the gear we used and how we used it.

I was using both a Nikon DSLR, and a Canon pocket camera for times when I don’t want to carry the big DSLR. We also had a netbook running Windows 7 and Adobe Lightroom to offload and process the photos, and an external hard drive for backup.

Every day or 3 when we had some free time, we’d offload the cameras — with twin goals of taking a look at the photos on a larger screen, and backing stuff up in case a camera is broken/stolen. I’d use Lightroom’s import function to bring all the photos into an “incoming” folder, subdivided by date, for both cameras. After I had all the photos from both cameras sitting in the “incoming” folder, I’d back up the entire “incoming” folder to the external hard drive, and only then, use Lightroom to reorganize the photos by destination, theme or group. (Importing first into this temporary “incoming” folder as a staging area has two purposes: it’s easy to do backups as above this way, and I never need to change Lightroom’s import settings; once the defaults are set correctly, I just have to hit Import and then OK.)

Once, and only once, all the photos are copied off of the memory cards and onto both the netbook’s hard drive and the external hard drive, I’d format the memory cards in each camera.

Meanwhile, back on the netbook, I’d be accumulating more and more photos that eventually need to be sorted into good and bad, the bad ones deleted, mediocre ones ignored for now, and the good ones cleaned up, annotated with metadata, and uploaded to my favorite sharing site (currently Flickr).

To do this, I’d:

1. Make a quick pass through each group of pictures, using Lightroom’s flag feature to group them into 3 bins — reject for immediate deletion, unflagged to ignore for now, and pick for the good ones. Redundant/duplicate good ones are mostly allowed for now, and all picked.
2. Create a new collection, and add all the picked photos from step 1.
3. Use Lightroom’s develop module to make them look better — recrop, correct exposure, correct lens distortion, reduce noise, whatever’s necessary.
4. Go back through this collection carefully to pick only the photos that I like best or that best represent the story, eliminating redundant photos (picking just the best of any groups of similar photos).
5. Add location, title and caption metadata to the surviving photos.
6. Export these photos to Flickr.

(Several of these steps can be done in different orders, but the more photos you’re working with at a given step, the more work it is… if you eliminate duplicates in step 1, you spend less time developing them; if you add captions or other metadata before step 4, you’ve got more to do, though location metadata can often be applied in bulk.)

All of this is a lot of work, but it goes more quickly with practice and I don’t know an easier way to get equal results. Also, running Lightroom on a netbook with only 2GB of memory is an exercise in patience, but it still paid off compared to waiting till we got home to process thousands of photos, since our friends could follow what we were up to in closer to real time, and we didn’t have weeks of photo tweaking to do once we got home.

Over the past 12 years I’ve been a customer of Sprint, T-Mobile and Cingular/AT&T for wireless service in the US, and from this, I have a simple view of what you get from each one. During that time I’ve lived in the San Francisco Bay Area and traveled pretty extensively in cities on both coasts and in the rural Pacific Northwest, so that’s what my coverage comments will be based on.

(I’ve never been a Verizon customer, but it’s widely agreed that their service is both good and expensive; it’s easy to verify their prices are on the high end, and friends and family who use their service like the quality, so I’ll repeat that viewpoint).

Summary: Sprint and T-Mobile are cheap; AT&T and Verizon are expensive; Sprint and Verizon have good coverage; AT&T and T-Mobile have weak coverage; that means Sprint is cheap and good, Verizon is expensive and good, T-Mobile is cheap and weak, and AT&T is expensive and weak.

Sprint, in 7 years of use, has always been the cheapest for the level of service I get, and it’s always worked fine for me. Sure, there are rural areas where they don’t have coverage, but in my experience their coverage is wider than T-Mobile and equal to AT&T, and anywhere they don’t have coverage, you can roam for free on Verizon. Voice quality has been good; usually with even 1 bar of signal strength conversations work fine. Sprint often gets dismissed as having bad customer service, but that hasn’t been my experience; also, in my view, the best customer service is the one you have to deal with the least, and I didn’t often have a need to call them.

T-Mobile was priced about the same as Sprint, but their coverage was notably worse, both in the rural Pacific Northwest, and in urban San Francisco. I had no real complaints in 2 years as a T-Mobile customer other than lack of signal coverage in areas that mattered to me; that’s when I jumped ship to Cingular, since I could bring along my existing GSM phone.

AT&T worked OK (better than T-Mobile, no better than Sprint) with the older phone I brought over from T-Mobile, until I got an iPhone in 2007. That started 2 years of frustration, ultimately leading me to decide that the GSM networks in the US haven’t kept up in terms of network quality, and bringing me back to Sprint. In 2 years of iPhone use, I had daily experiences with dropped calls, incoming calls that went straight to voicemail, and garbled voice during calls that went through, all of this often with the phone reporting 4 or 5 bars of signal strength. Voice calls weren’t the only problem; often I’d take my phone out to look up a map or kill a few minutes reading mail or the web, and be unable to do so, so I take issue with the general assumption that the iPhone is a bad phone but great for everything else and so great for modern people who don’t care about voice calls — those people probably expect reliable wireless data, and aren’t getting it. One observation I made during this time was that while every carrier’s coverage has dead spots (they even have maps of this!), dead spots in space are pretty easy to route around — you know where they are, and you can avoid them. But AT&T’s network is the only one I’ve seen where the signal quality varies hugely in time as well, that is, where you could have perfect signal quality at one time, and 5 minutes later in the same spot, no usable signal at all. For me, signal quality that varies in time is far more frustrating than signal quality that varies in space.

(There’s plenty of agreement on AT&T’s dismal service quality, especially in cities such as New York and San Francisco, and plenty of speculation on why — maybe it’s the iPhone itself that’s buggy, maybe there are simply too many users on the network — I just know that the 2 different iPhones I had were not reliable for either voice or data use in the US, and they both worked a lot better overseas on other networks once I unlocked them.)

So of the 3 networks I’ve tried, only Sprint was cheap and good; that for me is a compelling combination and brought me back to them.

Visiting parents for the holidays, I wanted to show a slide show from my and Vanessa’s recent travels, and wanted to show this on their big screen TV, not on a little laptop screen.

The slide show was happily residing on 2 different Mac laptops; the older one has a DVI port and the newer one has a Mini DisplayPort and the big screen TV in question only has HDMI inputs and I couldn’t find my cables/dongles to convert either DVI or Mini DisplayPort to HDMI.

Enter my brother’s newish Dell laptop, which conveniently has a real full- sized HDMI port and can, gasp, be connected to any modern TV with a standard HDMI cable. Better yet, it actually includes audio-out over HDMI (which I think would hypothetically also be true of my newer MacBook Pro’s DisplayPort, albeit requiring a newer/smarter dongle than the one I have and couldn’t find).

The problem showed up when testing different display resolutions and sound support and getting confused by the fact my brother had silenced all the system sounds on his laptop, I unplugged and replugged the HDMI cable between the laptop and TV maybe 20 times, and eventually the TV stopped showing anything at all. Based on past experience with Windows having flaky support for eternal displays, the first thing I did was reboot the laptop, at which point the external display still wasn’t working.

Then I rebooted the TV, and the image came right back. So score one for Windows — it’s now no longer the weak link in the chain, when talking to external displays (think of all the conferences/presentations you’ve seen start 10 minutes late because the presenter was screwing around with Windows display settings trying to get it to talk reasonably to the projector) — at least in this case. (Though my brother then asked if it’s really a win for Windows if it’s sending out signals that crash TVs.)

If you use both Flickr and Adobe Lightroom, and upload from Lightroom to Flickr using Jeffrey’s “Export to Flickr” Lighroom Plugin, you might find that sometimes images don’t get uploaded, or get uploaded and forgotten.

I typically find this out, for example, when I edit keywords or other metadata for a bunch of images, then use the File->Plug-in Extras->jf Flickr Extras command to resync the metadata at Flickr for hundreds of images at once. Often I’ll be told “the metadata for 471 images was updated but 2 images were not found at Flickr”.

At this point, you need to know how to do 2 things:

• assuming the images were really uploaded, you want to reassociate the Lightroom copy with the Flickr copy
• if that fails because the images were not really uploaded, you want to identify them and actually upload them to Flickr

The first one is easy; also in the jf Flickr Extras dialog, there’s a command “associate images”. Try this. If you’re lucky, it’ll just work. If it fails, you need an easy way to tell which images it failed for; that’s where this tip comes in:

Switch to grid view if you’re not already there. Select all of the images in the batch you’re working with. Make sure the filter bar is shown (this can be toggled in the View menu: “Show Filter Bar”), then filter by text, and filter on: Text | Any Searchable Plug-in Field | Doesn’t Contain | “flickr”. This works because the jf Flickr plugin adds a field with Flickr in the name to any image it uploads.

Now, to make sure people are even more continually conscious that we need to be terrorized by terrorists, we’re going to advertise this in Wal-Mart. Two different blogs I read linked to this news last week: John Gruber compared it to the movie Brazil and Bruce Schneier compared it to the book 1984.

These comparisons remind me of a joke I used to make during the Bush presidency: people were intended to look at works of dystopia sci-fi like Brazil and 1984 and see a cautionary tale; the neo-cons looked at these same works and saw an instruction manual.

Of course, here we are in 2010, Bush and the neo-cons are out of office, and yet the madness continues.

More on original story: DHS press release, The Raw Story, MSNBC.

I just bought a new iPod Touch to replace the iPhone 3G I sacrificed to the gods of Machu Picchu.

I took it out of the packaging, turned it on, and found: it immediately demanded that I connect it to iTunes, and was entirely useless until I did so.

This is ostensibly so that I can restore it from a backup of an existing iOS device if I’m upgrading from an older one, and also so that I can sync music and videos and apps and stuff to it, and also so that it gets backed up.

Still, without ever syncing music etc. from a computer, it’s useful purely over the (wi-fi) network connection; I wanted to add my email and iTunes App Store credentials and then I’d be able to download apps, check email, and browse the web. Or, I should say, would be useful for all this if I didn’t have to go find a computer and plug it in just to do nothing.

(Plenty of other similar devices — Palm Pre, Amazon Kindle, and, I presume though I don’t own any, Android phones — manage to be useful out of the box and over the air without ever needing to connect to a computer, though of course you can do this optionally if you want to, up front or later, to sync media.)

To make things worse, I’m not home; I’m visiting my sister and her family. So I had to borrow her computer just to activate my iPod Touch. Even though iTunes asked me what iTunes account to use to register the device — it was signed in as her, and I signed out and signed back in as myself — it used my account only for the registration step, but on the iPod Touch itself, it went ahead and configured her iTunes account, which I had to undo later.

I’m also somewhat dreading that when I do get home, and want to add music to this iPod, I’ll have to sync it with a computer I actually own, and that doing so will make me have to start over with all the things I did today — adding email accounts, installing apps, and then configuring my credentials for these apps. I hope not.

I just reread the first 3 Dune books (Dune, Dune Messiah, Children of Dune). Consistent with my memory from the 1st reading, the first is awesome and the second one much weaker. The author’s son, Brian Herbert, wrote an afterword to the first book and introductions to the second in third, and in all three of these he finds it necessary to defend the second book, but, in my opinion, not for the right reasons — what he apologizes for is the tearing down of the hero, which I didn’t mind.  What’s wrong with the second book is that it just doesn’t have the depth of intrigue or scope or carefulness that makes the first so epic. It’s like the actual writing fell victim to trying to prove a point. Maybe this echoes an ironic meta-consistency with the idea that heroes are fallible, but it’s not actually good for the book.

It’s really a matter of deftness — books one and three hint subtly at things beyond what’s stated, leaving much to the imagination and inspiring the reader, while book two generally describes things over-plainly, coming off stilted, and not taking advantage of the overall air of mystery.

The third book, on the whole, is a return to the layered intrigue and epic scale that made the first book so strong.

Side note: I read all of these on Kindle, and the Kindle rendition of the first book is horrible — chock full of typos and formatting errors. This is especially ironic given that the first book has been reissued as a vaunted “40th anniversary edition” and costs more; the sequels haven’t been given the “reissue-with-lots-of-formatting-errors” treatment, and are both cheaper to purchase and a lot easier to read. More on this later.

This might be confusing because all of the following can be called “Machu Picchu”, so I’ll try to be unambiguous.

Machu Picchu is the famous Inca city, abandoned and now ruins that you can visit as part of a national park.  To get there, you generally arrive by train to an inhabited Peruvian city called Aguas Calientes (also known as Machu Picchu Pueblo). Also, the whole site is named for a mountain also named Machu Picchu — Montaña Machu Picchu, to be unambiguous — which you can climb once inside the park.

As many visitors do, we arrived by train into Aguas Calientes, then took the bus up to Machu Picchu itself, then climbed Montaña Machu Picchu. I want to know the beginning and ending elevations of this climb, but can’t find a good source of information on this.

Note that when you’re there, it’s obvious that Machu Picchu is easily a few hundred meters above Aguas Calientes, and Montaña Machu Picchu is easily a few hundred meters above that.

However, when I start web searching to find the elevations of Machu Picchu (the Inca city) and Moñtana Machu Picchu (the mountain towering above all this), I find a lot of inconsistent data; combining local sources and guidebooks with the top couple Google hits for “elevation Machu Picchu” and “altitude Machu Picchu”:

• The Lonely Planet guidebook says that Aguas Calientes is at 2400 meters above sea level, and doesn’t give elevations for the other two.

• There’s a sign at the Aguas Calientes train station saying it’s at 2410m.

• Wikipedia claims Machu Picchu is “about 2450 meters” above sea level, and Aguas Calientes is 2040m above sea level.

• Encyclopedia Britannica’s article starts with a note explicitly addressing the variation in estimates of elevation; they then state both 2440m and 2350m as numbers given by Peruvian officials for Machu Picchu

• sacredsites.com article on Machu Picchu claims the mountain top as 9060 feet (2760 meters), but it’s not entirely clear which mountain top they mean — Montaña Machu Picchu, or Huayna Picchu, which is much lower.

• The Peru Guide’s Machu Picchu FAQ says Machu Picchu is 2430m.

So, no coherent answer. Wow. This can’t really be that hard to answer these days. Now I wish I’d taken a GPS unit with us and measured this myself!