.farnham .speech

speech-
language
pathology news.
& thoughts.
& research reviews.

© 2009-2012.
william. a.
farnham.
don't steal.

email.
twitter.
personal blog.

Link: Speech Gadget: How to Find Speech Therapy Apps

Deb at Speech Gadget has some handy tips for searching for apps, if you’re on the lookout for that sort of thing. Or, instead you could just look at her positively astounding grid of apps for some ideas and then check out the ones that might apply to you.

Replicationation

Admittedly, carrying out exact replications of someone else’s work is hardly the most glamorous way to spend your time as a scientist.” — Ritchie et al., 2012

A while back, I posted a link on my personal blog to the Journal of Negative Results In BioMedicine. This fantastic series of pieces from The British Psychological Society on publishing (or not publishing) an inability to replicate results is another facet to this same issue.

Journals are motivated to publish studies that show something new. In the absence of something “new” they generally try to publish research that is “interesting.” These effects are compounded by similar proclivities in the general/mainstream media when publicizing such studies. The same mindset that often inhibits scholarly journals from publishing negative results clearly also makes them disinclined to publish studies of replication. This is a problem in the hard/real sciences (I was a chemistry minor in undergrad. Face it: What we SLPs do is only science in some of the broadest definitions of the word. At least we have the economists and sociologists beat. If you work exclusively in acoustics then maybe we can have a discussion), and it’s a problem for speech pathology.

Case studies produce real, legitimate results. Single-subject and small-group designs publish real, legitimate results. But guess what. Replications and failed replications of those studies are every bit as legitimate, and they deserve recognition beyond just op-ed responses. What our field (and the world) needs is data. Data, data, and more data. Will all that data invariably result in regression to the mean? Probably. Most assuredly, if you’re a statistician. But that regression shouldn’t keep us from putting it out there.

Researchers and clinicians: Conduct your studies. Take your data.

Journals: Publish those studies. Unless you see glaring holes in validity, you have every reason to do so, and no reason (other than some subjective measure of how novel or interesting it will be) not to do so.

{PDF link}

Extreme nerdery regarding app management on iPad/iPhone/iPod touch

{From the proprietor: What follows is the complete text of a comment I wrote in response to this post on GeekSLP. If you’re tempted to pull a “Too Long, Did Not Read,” the take-away is this: If you’ve never worried about the apps in the app-switching tray (the thing that shows up when you double-tap the home button), don’t start worrying about it now.}

While in general I appreciate the lengths you go to in order to help keep your readers in the know about how best to use the tech that they have at their disposal, this particular vid is grossly misinformative, I’m sorry to say.

iOS does not support background tasking/multitasking in the way that it is generally understood. Instead, app developers are given a limited number of options they may implement within their apps to allow specific processes to run in the background, or to allow their app to more quickly resume once they are loaded again. Some of the specific processes that do, in fact, “stay running” in the manner that you erroneously claim the apps in the multitasking tray run, include background location services, background audio, and task completion. A user will always know if either of the first two is in use because the location or music playing status icons will be displayed in the status bar whenever they are in use. Task completion is less obvious, it _can_ have a slight performance impact, though tasks are only allotted an extremely short amount of time to complete before being automatically terminated by the system.

Another background process not mentioned above, which can have some impact on system performance (most often, battery life), is push notifications. If one receives many notifications and is noticing slowed performance when a notification arrives, or notably shorter battery life, one can turn them off as one chooses from Settings>Notifications. Push notifications are pushed to your iPhone or iPad regardless of whether that app is currently running or has recently run.

The final background-related process that 3rd-party apps are allowed to utilize is what is known as app state suspension. This allows developers to cache the state that the app was in immediately before a user switched to another app or exited to the home screen, allowing the app to more quickly resume once loaded again, and to increase the appearance of background multitasking.

So, what’s the point of all this? The point is that while apps are allowed to perform an extremely limited number of tasks in the background, the system controls all of these tasks independently, and the apps themselves _are not running in the background_. Even app state suspension, which could, no doubt, use a significant amount of RAM, has little-to-no performance impact because iOS will automatically begin terminating/deleting these cached app states to free resources as necessary. All of these steps were undertaken by Apple to ensure that background tasking had as little effect on the user experience as possible. In fact, background tasking was not available at all to 3rd-party developers until iOS 4 due to concerns over resource management.

So, why does the app switcher tray exist, then? First, it allows for easier access to apps that have recently run. This, in fact is _the only thing_ that the tray displays: apps in the reverse order from when they were most recently accessed. It DOES NOT display apps that are _currently running_. As such, the sort of neurotic, obsessive task managent that your video implies we should be engaged in (“I’m ashamed that I have this many apps open!”) is precisely what Apple had hoped to avoid by initially not including background tasking in iOS, and then by only implementing the limited background features described above when it did implement them. Not all of those apps in the tray are running. In fact, most of them aren’t. You don’t need to systematically clear out your app switching tray.

I’m going to repeat that.

You Don’t Need to Systematically Clear Out Your App Switching Tray.

Using the tray to close an app CAN be useful, however. If you notice that the location services icon is on and you have no idea why, clear out apps until it disappears (usually a rogue GPS maps app). If an app you were just using has hung, exiting to the home screen and then killing the app from the tray may fix the problem, as you suggest in the video. Those are essentially the only use cases in which your manually deleting an app from the tray accomplishes something that the system has not already done on its own.

Use the tray to switch to a recent app. Use the tray to kill an app that is obviously in need of termination. Otherwise, please, _please_, ignore the tray. You will only drive yourself crazy being “ashamed” if you feel like you need to keep the tray clear. Most of those apps have already been killed. Let the system do its work, and for the love of Steve, stop worrying about it.