Friday, January 25, 2013

Five easy steps that prevent most computer problems

Lots of people come to me for computer support. I'm happy to help; it gives me a chance to visit with people I care about and to do something really valuable to help them.

I'm fascinated by how similar their requests are. It's not that I end up doing the same thing to fix every problem but that the same few things could be done to prevent these problems in the first place.

At the risk of denying myself one of my pleasures, I'm posting a list of simple things that everyone can do (and should do) that will remove the majority of their computer headaches – at least, the type of headaches that sends them to a friend for help.

  1. Back up your files
  2. It doesn't matter what kind of hard drive you have. It's not a question of whether or not your hard drive will fail. It's a question of when.

    It goes without saying that keeping only one copy of your information on some sort of portable drive is a bad idea. People lose or break them all the time.

    The solution is to keep copies of your information. One copy is always insufficient. Having even one backup copy of your data makes it virtually impossible for you to lose it, as it's extraordinarily unlikely that all of your copies will fail simultaneously.

    It's a pain to back things up manually. Windows 7 has a feature called Backup and Restore. Mac OS X has Time Machine. They use a separate hard drive or other media to keep copies of all of your files.

    If you don't want another hard drive, there are lots of online storage solutions. Try Google Drive, Dropbox, Skydrive (by Microsoft), iCloud (by Apple), or Mega. Space Monkey hasn't launched yet but is an innovative variant on cloud storage. How much you can store (and what you can store) varies a little from solution to solution. In every case, they're managing encryption for you (although Mega is trying to put you in better control of your data). This means that if they mess something up, your data could become public. They're usually pretty good, but I recommend against storing sensitive data on them without encrypting them first.

    There are several advantages of storing in the cloud. They take care of the hard drives and redundancy for you. In many cases, it's even free (for a small amount of storage, which is plenty for your text documents and often for pictures or music). It's available to you whenever you have internet access. It's painless to transfer from one computer to another. And most of them automatically detect when you change something and upload it automatically, making it effortless to keep a current backup at all times.

    One last note about backups: you have to keep them. This sounds obvious, but countless people have thrown away the disks that store their data or the software that came with their computers because they never use them. You probably won't use all of them, but you certainly will want some of them.

  3. Back up your software
  4. Files are easy to back up and to transfer between computers. Software isn't. However, it is fairly easy to back up installation media and software keys.

    Software that comes with your computer may not come with a disk. When you get a new computer, you should make recovery disks before you do anything. Don't connect to the internet. Don't open files. Most systems will prompt you to make backups and most people ignore this warning.

    In case that wasn't emphatic enough, the first thing you should do when you get a computer is back up the software that came with it.

    If you buy software on a disk, it's probably a good idea to keep the disk around and it might be a good idea to copy it and store it elsewhere. Backing up software is specifically allowed by your statutory rights; there's no need to worry about copyright.

    If you buy and download software (by far more common these days), don't just run the file you downloaded. Before you run it, make sure to save it somewhere. In fact, you should probably just burn it to a CD or a DVD immediately.

    If you buy software through a marketplace (such as the Apple App Store or Google Play) and you don't get an installation file, good luck. Your best bet is to hope that that marketplace never goes out of business. It may be possible to find and back up installation media but that could violate the marketplace's terms of service.

    If you get free software, it's often counterproductive to store it anywhere. There will be a new version by the time you need it again. If it's free, I recommend just downloading it again when you need it.

    In any event, make sure that you keep any software keys with your software. If you burn a disk, add your key to that disk. If the key is a file, add the file to the disk. If it's text, add a text file with the key in it. You may also want to write the key onto the disk itself. Sometimes there are keys that ship with your computer or that are on a sticker under a laptop or on a desktop. Write them down somewhere else, especially if you put your laptop on your lap, as the writing can wear off over time.

  5. Install updates automatically
  6. Modern Windows and Mac OS X install new updates automatically. Don't change this behavior unless you really, really, really know what you're doing. Yes, there are occasional problems with patches. But Patch Tuesday is always followed by Exploit Wednesday. Just don't go there. Install updates as they come out.

    There is other software that needs to be updated, too. Your browser needs to be updated. Your productivity software needs to be updated. Countless other things need to be updated. Most of them will tell you about updates periodically.

    Take the time and install the updates.

    It really doesn't take that long. You don't have to take care of them as soon as you see them, but you'll probably be taking a break or switch tasks within an hour of seeing an update. Update then. If you don't remember, write yourself a reminder. This is a habit you're going to have to cultivate.

  7. Use strong passwords
  8. It's hard to think of a strong password. It's harder to remember a strong password. And it's hardest to remember countless different strong passwords. My recommendation is that you don't.

    It's not that you shouldn't have strong passwords. It's that you shouldn't bother trying to remember them. I use KeePass to store a database of passwords. It automatically encrypts the database and requires me to enter one master password to open it.

    KeePass features a password generator that makes strong password generation a cinch. I won't get into the mathematics of what makes a password strong, but I recommend that you use such a password generator and there is solid mathematical reasoning behind that recommendation.

    I haven't found a way to make KeePass 2 work well in Mac OS. For that matter, the user experience of KeePassX (a version of KeePass for UNIX systems, including Mac OS X) is worse on a Mac. But it works.

    One of my favorite features of KeePass is that not only do I not have to remember my passwords, I don't have to type them. I haven't ever seen half of them. On Windows systems and in Linux, there's an auto-type feature that types my username and password directly into my browser or whatever I'm using to log in. In all of them, it's easy to copy my username or password to the clipboard so I can paste it elsewhere. And it automatically cleans out my clipboard so my passwords aren't available to other programs.

    I sync my passwords database using the cloud (see the section on backing up data) to my various computers and even to my phone. I'm not concerned about the security of it because I have a strong passphrase that is required to open it. And that means I always have access to all of my accounts, everywhere.

    Did I mention that I don't know my own passwords? That every one of my accounts has a different password? And that the time I'd expect it to take to crack any one of my passwords far exceeds the time we expect the universe to exist? In short, I'm not too worried that anyone is going to steal my digital life.

  9. Be careful where you click
  10. Malware is pretty amazing these days. Just clicking on a link could infect your computer or steal your account from you.

    It's not that links are bad. They're like email attachments or downloads: be careful. Use bookmarks to get to the websites you use often and never use a link from an email to get to them, even if it looks authentic. It's just as easy to use a bookmark and, assuming that you made the bookmark correctly in the first place, it's much less risky.

    Links in social media are really the same risk as links in emails and on web pages. Regardless of how you get them, always be skeptical of uncharacteristic communications from people you know – and more skeptical of communications from people you don't know.

    Always keep a virus scanner up to date and running (yes, even if you use a Mac). For Windows users, I recommend Microsoft Security Essentials, which is free to users of Windows XP an later. It does its job and stays out of the way otherwise.

    Another common source of malware is smartphone software. By running an app on your smartphone, you're potentially giving some unknown author access to everything on your phone. Favor software actually written by the entities you're interacting with; for example, Twitter's official app instead of a third-party application. This isn't an absolute rule, but caution is in order. On Android devices, inspect the permissions that an application requires and consider if they seem reasonable – this won't protect against everything, but it will help.

Most of these things take a small investment of time at the outset and very little time to maintain. In fact, I think I save time typing passwords. And it pays off: my laptop died last week, as laptops are wont to do. I can still get to the files using nerd-fu but I don't think I have to; there's nothing on that computer that I've needed yet that isn't backed up elsewhere.

There are, of course, more things that can be done. Maybe I'll write another blog post someday about what you can do if you want to tinker a little more.

Friday, January 18, 2013

On what might have been

How many times have you wished that one little thing could change in the past? If that hadn't happened, the effects in the present would be countless and powerfully different. If I'd only chosen to turn right instead of turning left, I would have seen my friend and been able to give him a ride. If she hadn't waited 15 minutes before leaving home, she wouldn't have been crossing the street when that drunk driver came along. We could go on endlessly.

Of course, we don't really know what would have happened. We have gotten pretty good at guessing, but we don't ever know for sure. The scientific method usually involves repeated experiments – and certainly involves allowing others to repeat our experiments. The hope is that if we observe enough things carefully enough, we'll understand cause and effect.

The way that a scientist tries to understand cause and effect is by isolating a single difference. An experiment involves doing two things exactly the same way except for that one crucial difference and observing the results. If changing that one thing consistently gives a changed outcome, we assume that there is (and that we have observed) a causal relationship.

We've become so accustomed to experimentation and so casual as we set up our experiments that we've come to make frequent errors in the way that we see the world. Some of these errors lead us to dangerous conclusions.

Our habit of breaking things down into experiments has trained us to think only of one factor in an outcome at a time. Isolating the different factors is wise, but we must eventually remember the ones we've left to the side. This is especially important because we don't always think of all of the factors when we set up our experiments. The fact of the matter is that many light things fall more slowly than rocks do, although their weight is not the only factor in that measured outcome.

Our legal system has a similar approach to things that can have a similar effect on our thinking: since there are many things that contribute to any one event happening, our law concerns itself principally with a proximate cause. Without this legal doctrine, we'd have no way of assigning any sort of blame to anyone or anything and our laws would be completely pointless. But if we're not careful, this too can train us to think that every outcome has a single cause.

We do well to identify the other causes of the events in life, whether we hope to repeat them or avoid them. We can do so by asking questions like, "what else contributed to this?" and, "what else might I have done in this regard?" When we identify something that we might have done that seems likely to lead to a desired outcome, we can consider if we want to try acting in that way in the future.

It's also a good idea to recognize that, because so many factors contribute to every single event, we rarely have a suitable justification to blame anyone else for anything. Even if the outcome could not have been the same had they acted differently (this is rarely the case), it's usually impossible to prove that another persons actions alone would guarantee that outcome. This is usually impossible to prove because it isn't true.

Similarly, it seems that in most cases where a person tries to influence another's decisions, it's because the other person's actions would contribute to some desired outcome. There's nothing wrong with this, but we'd be a lot less tempted to manipulate each other if we spent more time thinking about what we can do ourselves to contribute to those same outcomes.

On a related note, we usually form opinions too soon about whether or not we want something to happen. I've looked back on several then-happy experiences and wished they'd been different and I've been immensely grateful for many of my unpleasant experiences. In fact, we'd probably all be better off if we made more of our decisions based on our principles instead of what we imagine and hope will be the outcome.

So stop fretting about what would have happened as if you know and figure out how to make the best of the situation you're in instead.

Tuesday, January 15, 2013

The new way of multiplying that's exactly the same as the old way

There's an article that's been kicking around the intertubes for the last couple of days. You'll find it here (as far as I can tell, this is the original version).

The gist is that there's an easy way to teach multiplication – that the Japanese have figured something out that we haven't in America. I presume that people are sharing because they wish they'd been taught that way in school.

To multiply 12 x 23, we draw these lines:

Now we count the intersections of the lines in each corner:

Here, we can add the two numbers in the middle column and put the numbers together to get 276, which is the correct result for 12 x 23.

But if we restructure the lines slightly:

And remove the lines:

We have exactly the same multiplication with the exact same calculations as we would have done had we simply done as taught in school.

Now to my objections.

The "proof" given that this gives the same result is only a demonstration that the result worked for a particular example. To prove that a method is correct, it's necessary to show that it will always work. If our students could actually demonstrate why these two methods always produce the same results, we'd be going somewhere.

I fail to see how this is any simpler. It actually takes longer to do. I suppose some students might object to the typical American method because it seems so arbitrary and they don't see why it would work, but the same applies just as easily to this visual method.

Try using bigger digits (not my own thought; I found this one online).

Most importantly, though, all of this misses the bigger point. It's not just about being able to perform computation. We have calculators for that. It's about understanding what a computation is, what it means, and how it works. This could be a useful teaching tool but could also be a crutch that prevents another generation of students from understanding one of the most fundamental operations in mathematics.

So why is this any different or better? Maybe I've missed something?