Archives

Support this Site!

You may have noticed that there's very little third-party advertising on this site. I'd like to keep it that way. Here's how you can help:

  • Buy my books. They're available at great prices on Amazon.com.
  • Check out my training videos on Lynda.com. It's a great source for "all you can eat" training.
  • Donate a few dollars. It'll help cover my hosting costs and give you a chance to tell me what you want to see covered here.
  • Comment on blog posts. You can help get a discussion going that can benefit others, making the site more valuable for everyone.
December 28th, 2009

Blogging Basics: Comment Spam, Part III

Part III: Identifying Comment Spam

One of the challenges of being a serious blogger is dealing with comment spam. While I believe that every blogger should install and enable comment spam prevention software — I use Akismet and Bad Behavior together on my self-hosted WordPress blog — I also go the next step and review every single comment before it’s posted. This ensures that no spam gets through the cracks and appears on my blog.

I also, however, review the spam caught by Akismet to make sure it really is spam. Today, on this blog, I got the usual collection of machine-generated and posted spam comments with an extra helping of comments that might appear, on the surface, to be real.

(Note: Although I originally wrote this as a stand-alone post, I realized it would make a good addition to a series I started about a year ago. So I’ve slipped it into that series as Part III.)

The Goal of Comment Spammers

To understand what a spam comment is, you need to take a look at the goal of spam commenters. The majority of comments that appear on my blogs have one or both of two goals:

  • Create comment content with clickable links to one or more other sites. Since different blogging platforms support different link notation, some of these links appear as actual links while the rest appear like URLs surrounded by gibberish.
  • Create a link from their name — which isn’t actually their name — to another site. This is pretty easy on a WordPress blog, since when a commenter fills out both the Name and URL field, the comment displays the name as a link to the URL.

Either way, the site the spammer is linking to could contain products or services they want to sell, be part of a black hat SEO project, or even install malware on visitor computers. They often have nothing to do with the post they’re commenting on. There’s no reason in the world why I’d want those links on my site. Hence, they’re spam.

Once you know the goal of these spammers, identifying what might be a legitimate comment as spam becomes easy. Here are a few examples taken right from the Edit Comments administration panel on this blog.

Empty Praise

In this first example, it appears, at first glance, that six different very friendly people commented on six different blog posts. They all have different names and different e-mail addresses and are commenting from different IP addresses. But the URL they’ve entered all starts with the same name, pointing to free blogs on multiple sites. They were all posted within 2 minutes of each other.

Spam Example 1

And when you read the actual comment content, it’s clear that it’s not real. Here’s what was entered as a comment about the post titled “AirPort Client Update 2009-01 1.0 Now Available“:

That is some inspirational stuff. Never knew that opinions could be this varied. Thanks for all the enthusiasm to offer such helpful information here.

Inspirational? Opinions? Neither of these words apply to this post.

Or how about the unlikely comment on “Turning Off a Plugin’s Update Reminder“?

I thought it was going to be some boring old post, but it really compensated for my time. I will post a link to this page on my blog. I am sure my visitors will find that very useful.

Not a likely comment for this post.

This isn’t the only spammer trying to fool me with complements. Look at this weird comment on my post titled “Bluetooth Adapters not the Same as Built-in Bluetooth” by someone with the unlikely name “[redacted] county limousine service”:

Very insightful post. Makes me move to Europe and learn as many languages as I can. How different the world we be if we could just get past our differences.

Huh? Move to Europe and learn languages? That’s a great way to avoid Bluetooth issues. Not.

These empty words of praise could fool someone looking for a pat on the back. Don’t let them fool you into including spam on your blog — even if their comments claim they’re going to link to you.

Questions

Another way comment spammers attempt to get attention for their spammy comments is by asking questions in the comment. What well-meaning blogger or reader can resist answering a question? Here’s a example from today’s batch of spam:

Do you do all your own writing? Or do you outsource some of it? I

(No, the hanging character “I” is not a typo on my part. It was really part of the comment.)

Or this one:

Interesting information. May I add this blog to my linkexchange directory ?

I’ve seen so many variations on this. They often start with a complement and then ask a question that seems to deserve an answer. If the comment is approved, the blogger or a reader may answer the question in another comment, thus drawing more attention to it to make it look legitimate. Or, worse yet, the blogger or commenter may click the link created by the URL for the commenter’s name (which could look legitimate) and wind up exactly where the spammer wanted him to go.

Or sometimes they ask a question that can be answered by a link in their comment, something like “Do you know where I can find good video rentals? I’ve been using blahblah.com (with link)” The object there is to appear to offer helpful information. But is it related to the post? If it’s spam, it probably won’t be.

Links Galore!

Here’s a good example of old fashioned, link-heavy comment spam:

Spam Example 2

And one where they weren’t taking any chances; they used two standard notations to indicate links, one of which just shows up as code characters:

Spam Example 3

If you can’t identify these as comment spam, you should probably turn off the comment feature on your blog.

It’s Not a Big Deal

In this post I looked at only three kinds of spam you’re likely to get. There others. If I find good examples, I’ll cover them in another post.

I know at least one blogger who has given up on comments because of the amount of spam that comes through. This blogger is probably not using good tools to prevent and capture spam.

As I mentioned at the top of this post, I use Bad Behavior, which prevents a lot of automated spam and Akismet which captures about 90% of the spam that gets past Bad Behavior. On each of my blogs, I seldom have more than one or two spam comments mixed in with real comments to review. But the potential spam list is always long and I very seldom find a real comment there. Usually a quick glance and a click clears out all the garbage.

In the next post of this series, I’ll explain how you can configure Akismet on your WordPress.com or self-hosted WordPress blog.

Learn more on Lynda.com!

Get more from your software.

If you’re a WordPress user and would like to learn more about handling spam on your WordPress.com or self-hosted WordPress account, I hope you’ll check out my WordPress courses on Lynda.com. The current titles of interest are:

We’ll likely be revising one or both of those courses in the coming months, but for now, you’ll find either one very helpful for working with WordPress.com or your self-hosted WordPress site.

And keep checking in here. As a very active WordPress blogger, I’m always offering tips on this site for using WordPress. Click the WordPress category link to find some of my recent posts.

December 22nd, 2009

Mac OS X: Using Your Laptop in “Clamshell Mode”

Using your MacBook or MacBook Pro with the lid closed. Really.

Apple Cinema DisplayAs the prices for large, flat-screen displays continue to fall, many folks are buying one as a second display. About a year ago, I wound up with a 23″ Apple Cinema Display that I’d bought used from a friend who was moving. I installed it in my Phoenix office, with the idea that I’d use it with my laptop — then a 15″ MacBook Pro; now a 13″ MacBook Pro. I connected the display to my Mac and began using it with the laptop’s built-in display, very much as I discuss here.

I soon realized that I really didn’t use the second display because of the awkwardness of “driving” the mouse from one screen to the other. But, at the same time, It made no sense to use the smaller monitor when I could be using the larger one. When I started using a wireless keyboard and mouse with my laptop in my Phoenix office, I realized that I was ready for clamshell mode.

Clamshell mode enables you to use a PowerBook G4, MacBook, MacBook Air, or MacBook Pro with the computer’s lid closed. All you need to have is an external monitor and input devices. The input devices can be wired or wireless. The computer sends all display output to the external monitor and accepts all input from those external input devices. The computer itself just does the CPU stuff.

Clamshell ModeYou can even put the computer aside, out of the way. The photo here gives you an idea of how my office desktop looks with my 13″ MacBook Pro running in clamshell mode, standing in a special stand I bought to keep it out of the way on my rather shallow desk.

There are two ways to enter and exit clamshell mode, depending on whether your input devices are wired or wireless. Let’s take a look at each method.

Wired Input Devices

If you have a wired keyboard and mouse, getting set up with clamshell mode is pretty easy.

  1. Connect the computer to a power adapter — this won’t work on battery power — and, if necessary, turn your computer on.
  2. Plug in your keyboard and mouse.
  3. Connect your external display to your computer and turn it on.
  4. When the desktop appear on the external display, close the computer’s lid.
  5. If the computer goes to sleep — which is likely — wake it by touching any key on the keyboard.

You can now use the computer with the external keyboard, mouse, and display.

Wireless Input Devices

In my particular setup, I have a Bluetooth Apple keyboard and a wireless (but not Bluetooth) Logitech mouse. The instructions I follow are similar to the wired input devices instructions. The key is to pair any Bluetooth devices before starting clamshell mode.

  1. Connect the computer to a power adapter — this won’t work on battery power — and, if necessary, turn your computer on.
  2. Make sure Bluetooth is turned on.
  3. Pair and connect your computer with your keyboard and mouse and confirm that they work.
  4. Advanced Bluetooth OptionsIn the Advanced settings of the Bluetooth System Preferences pane, make sure the Allow Bluetooth Devices to Wake this Computer.
  5. Connect your external display to your computer and turn it on.
  6. When the desktop appear on the external display, close the computer’s lid.
  7. If the computer goes to sleep — which is likely — wake it by touching any key on the keyboard.

In my case, because my mouse is not Bluetooth, I had to connect the mouse’s USB wireless adapter to my monitor (which has USB ports) or computer and turn the mouse on.

Watch the Video

If you’re a visual kind of person and want to see the process in action, watch this video, which I shot with my Flip camera. We’re not talking Academy Award material here, folks, but it does show you how it works — at least for me.

When You’re Finished with Clamshell Mode

Eventually, you’ll want to start using your computer the usual way, with its own built-in display. Here’s how you do it:

  1. Put the computer to sleep. (This step might not be necessary; it depends on the computer model and display. Experiment with your system.)
  2. Disconnect the display.
  3. Wake your computer to use it.

Need More Information?

Snow Leopard Book CoverYou can find more information about using monitors and Bluetooth devices with Mac OS X 10.6 Snow Leopard in Mac OS X 10.6 Snow Leopard: Visual QuickStart Guide.

You can find Apple’s instructions for using clamshell mode here. That document will be revised as procedures change.

December 19th, 2009

Upgrading to WordPress 2.9: Getting Your MySQL Database Up-to-Date

A tiny glitch for those of us with old databases.

Upgrade Admin PanelOne of the things I love about the most recent releases of WordPress is the automatic upgrade feature. Not only does the software tell you when a new version is available, but it offers a one-click upgrade through the use of the Upgrade Automatically button in the Upgrade WordPress administration panel. I’ve been using this feature regularly since it first appeared — after backing up my WordPress database and content files, of course — and have never had a problem.

Until yesterday.

Last night, when I attempted to upgrade my main blog, An Eclectic Mind, I got the following error message:

Old PHP Warning

I knew immediately why this error had appeared. My blog is so old that MySQL 5 wasn’t available when it was created. The available version was 4.0.27, which is what I installed. But the new version of WordPress needs a newer version of MySQL. Upgrading wouldn’t be possible until I upgraded my MySQL database.

I called GoDaddy, my hosting company, and spoke to someone in technical support. She said that the only way to upgrade the database was to back up my database, create a new database with version MySQL 5.0, and restore the old database to the new database file. I could then point my config.php file to the new database and, with a lot of luck, everything would work out fine.

So I initiated a backup last night using GoDaddy’s backup utility. (I usually use the WordPress Backup Plugin.) I got tired of waiting for it to finish, and went to bed. This morning, I’m restoring that database into a new file created with MySQL 5.0. And now, as I type this, I’m modifying the config.php file to point to the new database, user name, and host name.

Drum roll please….

As I open the home page for my blog…it works!

Upgrade CompletedThe upgrade should now go off without a hitch — which it does, as shown here.

Unfortunately, if you’re in the same boat I was in, you’ll need to find out how to update your MySQL database file. Talk to your system administrator or ISP’s technical support department. Every system is different — I use GoDaddy.com so that’s the only system I know how to update. Providing detailed step-by-step instructions for that system would only help other GoDaddy users — and might not work after GoDaddy’s next interface revision.

So I’ll let you track down instructions for your server or ISP on your own. Once you get those instructions, it shouldn’t be difficult to complete the task.

Good luck and enjoy the newest version of WordPress!

December 10th, 2009

Airport Client Update 2009-002 Now Available

Information from Software Update.

This update is recommended for all Macintosh computers running Mac OS X Snow Leopard 10.6.2 and includes fixes for the following:

  • Inability to turn AirPort on or off in some cases after upgrading from Mac OS X Leopard
  • An occasional loss of network connection when using Wake on Demand
  • Inability to create a computer-to-computer network, or share the Internet connection on some MacBook, MacBook Pro, and Mac mini computers
December 5th, 2009

Java for Mac OS X 10.6 Update 1 Now Available

Information from Software Update.

Java for Mac OS X 10.6 Update 1 delivers improved compatibility, security, and reliability by updating Java SE 6 to 1.6.0_17.

Please quit and relaunch any web browsers and Java applications after installing this update.

For more details on this update, visit http://support.apple.com/kb/HT3892

For information on the security content of this update, visit http://support.apple.com/kb/HT1222

December 1st, 2009

Spelling Checkers Don’t Work if You Ignore Them

Reminding my brain not to block out the red underlines that indicate potential spelling problems.

Check Spelling as You Type has been a built-in feature of many word processors for years. It’s now in most applications I use — including my Web browsers, for Pete’s sake! — and the red squiggly or dotted underlines are an integral part of my writing life.

If you’re not sure what I’m talking about — do you live in a cave? — I’m referring to the feature that indicates when a word you’ve typed may have been spelled wrong. This is supposed to flag the word so you can check and, if necessary, correct it.

A Blessing…

I vaguely remember when this feature first appeared in Microsoft Word years ago. It was a blessing — and a curse.

I initially loved the feature because it often identified my typos. I’m a touch typist and can get up to 80 words per minute when I’m tuned in. But those aren’t always error-free words. Check Spelling as You Type was a great feature for finding typos as I worked, eliminating the need to run a spell check periodically or at the end of document creation.

As you might expect, it also found spelling errors. My spelling was always pretty good, so it usually found more typos than actual spelling mistakes. But that’s okay. An error is an error and I want to remove all of them from my work, whenever I can.

…and a Curse

But over the years, I’ve found some problems with the Check Spelling as You Type feature — and spelling checkers in general.

The feature does not identify all typos or spelling errors as errors. For example, suppose you type bit but you really meant but. A spelling checker doesn’t see any problem with that, so it won’t flag it. That means you can’t depend on a spelling checker to proofread your work. (And yes, in case you’re wondering, a grammar checker would likely identify this as a problem. As well as the sentence you’re reading right now, because it isn’t really a proper sentence. And this one, too.)

So you’ve got a feature that makes you lazy by doing about 90% of the proofreading work for you, as you type. If you neglect to do the other 10% of the proofreading work, you could be very embarrassed — especially if you write professionally and editors expect your work to be error-free.

The unflagged error that zaps me most often? Typing it’s instead of its. At least I know what it’s supposed to be.

It’s worse, however, for people who don’t know the correct word. How many times have you seen people use then instead of than? There instead of they’re or their?

The feature has degraded my spelling skills. In the old days, before spelling checkers, I simply knew how to spell. If I wasn’t sure of the spelling of a word I needed, I looked it up in — can you imagine? — a dictionary. It made it worthwhile for me to actually learn how to spell words. Knowing the proper spelling saved me time in the long run.

But now, I simply type the word as I think it might be spelled and wait to see if it’s flagged. If it is, I use a context menu — Control-click or right click the word — to choose the word I meant to type. Yes, it’s convenient. But I seem to be doing it an awful lot more than I used to use a dictionary.

(Perhaps it’s also expanding my vocabulary by making it easier to use words I’m not as familiar with? There’s something there.)

The feature identifies any word it does not know as a potential spelling error. That means that if your document is filled with jargon, technical terms, place names, or other words that do not appear in a dictionary, those words will be flagged as possible errors. The word unflagged, which appeared earlier in this post, was also flagged. Is it an error? Or does my spelling checker simply not recognize it? Seems like a word to me, so I let it go.

And herein lies my biggest problem: I’m so accustomed to seeing words flagged in my documents that I’ve managed to tune out the red underline. (It’s kind of like the way we all tune out advertisements on Web pages these days.) This happened to me just the other day. I typed the word emmerse in a blog post. My offline editing tool flagged it with a red dotted underline — as it just did here. But for some reason, I didn’t see it. I published the post with the error in it. A friend of mine, who referred to himself as a “spelling Nazi,” e-mailed me to point out my error. I meant immerse, of course. He knew that. Readers likely knew that. But I got it wrong and I shouldn’t have. How embarrassing!

Spelling Checker

Here’s a look at the spelling check feature in ecto, my offline blog composition tool of choice. It works just like any other spelling checker. (And yes, I do compose in HTML mode.)

The correct way to go about this is to look for every single possible spelling error and resolve it so those red lines go away. That means learning or adding the unknown word so it’s never flagged again or ignoring it so it doesn’t bother you in this document. All of this should be done with the appropriate menu command. Simply telling your brain to ignore a problem just sets you up to be blind to it when it occurs. That’s not how the software was designed to work.

The Point

This post has a point — most of mine do — and here it is: spelling checkers, including any Check Spelling As You Type feature, are only as good as allow them to be. Use them, but don’t depend on them. Follow up on any flagged words and resolve them using the software so the red underlines go away.

Spelling checkers are just a tool. Like any other tool, it won’t help you if you don’t use it correctly.