WHAT IS INSTALLING?
Once upon a time— and we’re going back some 20 years here— getting a program up and running involved little more than placing a few files in a folder on your PC. That’s all an installer did. Don’t have the right drivers? Tough luck. Find them yourself. Operating system not up to scratch? Go through all the hassle first, and you’ll find out that something’s wrong when you try to run your application. A dark time, but one with benefits: Removing a program folder meant, usually, erasing all trace of that software. A simple scheme from a time when, however bleak things might have been, operating systems were less complex, and software depended on little.
Try manually excising a bit of software now, and you’ll not have a good time; the typical modern installer is responsible for more tasks, and spreads the software much wider, and uninstallers are required to scrape off the remains of redundant apps. The installer decompresses large files, which have been archived to ensure an efficient distribution size. It determines whether you’re installing from scratch or upgrading. It places references to program variables in the Windows registry, and configures other files to suit your hardware and personal preferences. It downloads, or upgrades, software dependencies—drivers and software frameworks that the program being installed may depend upon to run. And it does even more besides.
To make things even more complex, there’s not one kind of installation package, though this won’t make a huge difference to the average user. You most regularly come across executables (exe files), which tend to use some kind of third-party program to get the job done, or Microsoft Installer (msi) files, which do the same thing using Windows’ built-in installation components. Others—the likes of .cub, .msp, and .msm—also use the Windows Installer, and are meant for patching or modifying components. You don’t see them often while using Windows on the desktop, if at all, although those who stray into server territory might get more familiar.
For all we’ve said, a complex installation method isn’t strictly required for every app. Many apps have been designed to run in a so-called “portable” configuration, where their content is contained in a single folder and requires no installation whatsoever. While these are few and far between— portable apps are usually small tools, although programs such as LibreOffice are also available in a non-installed format— they’re perfect if you want to carry around a toolkit of apps on a USB stick. And you can
even make your own, in a roundabout way; see “Create Portable Apps” (pg. 37).
We have no doubt that you’ve installed programs before. Taking you through the ins and outs of clicking “Next” a few times in the course of a traditional software installation would be insulting your intelligence. But, as you may have guessed by the extent of this feature, the standard way is not the only way of going about things. The regular way, perhaps, but not the best in all cases.
You may, for example, have grabbed some of the rare decent software from the Windows Store, Microsoft’s poorly-policed den of application iniquity. Whatever your feelings on its content, it has a number of advantages—Windows Store apps run in a sandboxed mode, isolated from the main OS, which means that even if something malicious slips through, it can’t attack your PC directly, and installing, removing, and updating Store apps is quick and easy. The technically minded might avoid it, and there are other ways of sandboxing apps you might not know of (see “Sandboxing Applications” on pg. 41), but don’t underestimate its convenience. So that’s one alternative, but let’s explore some of the other ways of getting software on your PC.