Tequila Fish

One argument I see people using against the iPhone is the cost. Granted, $499-$599 is a bit pricey, but what really gets me angry is when people use the flawed argument that the total cost of owning an iPhone is “$2,600-$3,200″ if you factor in the cost of the service contract. While that may be true, so what? Lets use some analogies:

• That new car you bought wasn’t $20,000, it was $200,000. You forgot to factor in the cost of fuel and insurance over the life of the car!
• That new computer you bought wasn’t $999, it was $10,000. You forgot to factor in the price of your ISP and electrical bills!
• That new flat-screen TV you just bought wasn’t $1,200, it was $8,000. You forgot to factor in the price of your cable box and service over the next 2 years!
• You really should stop drinking beer, it’ll cost you $5,000 over the next couple years!

See how ridiculous those arguments sound? Arguing that the iPhone will cost $3k is just as ridiculous. While the claim may be true, no one truly considers ancillary costs when making a purchase like this. It’s a disingenuous and dishonest argument used by people who are trying to find fault with this wonderful new device.

Not to mention the fact that no one is producing similar costs for rival phones in their arguments. I wonder what the TCO for a Crackberry is? I’d like to see that number compared to the iPhone’s. Maybe then we can start a decent dialogue about the iPhone vs. its rivals when it comes to cost.

When trying to enable vacation/out-of-office support to our PostfixAdmin setup, I came across an error in /var/log/maillog:

Turns out there is a typo in VIRTUAL_VACATION/INSTALL.txt. It tells you to add a service named “filter” to your master.cf file, but it should actually be named “vacation” like so:

Astrophysicist Stephen Hawking, paralyzed by Lou Gherig’s disease, floats during a zero-gravity flight; an apple, a tribute to Isaac Newton, drifts along with him.

I’ve come across a little inconsistency between browsers and the way they handle cookies. Using the examples from PHP’s online manual, I was unable to get Safari to remove cookies using setcookie(). All other browsers would clear the cookie with no problems. It turns out that Safari will only delete cookies if both the cookie name AND the cookie path match.

In other words, this won’t work:

But this will work:

Basically, if you set a cookie using a path, you must also specify the path when deleting the cookie. Safari won’t remove the cookie otherwise. This is probably a good thing.

My nightly backups kept failing with the following error:

As with all Retrospect problems, the fix is easy. Simply remove X:\Retrospect Restore Points\RestorePoint.rbc (where ‘X’ is your external drive’s letter) and let Restrospect re-create it. This will often take several hours, so be patient. Once Retrospect has finished re-indexing everything, your backups should resume as normal.

Any time I have had a problem with Retrospect, it can always be fixed by re-building the .rbc file.

Here’s a quick query you can run to determine whether or not a particular column in a table exists or not:

Of course, replace YOURTABLENAME and YOURCOLUMNNAME with the proper values. If a row is returned, a column with that name exists, otherwise it does not.

We have an account with a VERY large inbox (over 80,000 emails), and when logging into Squirrelmail the main pane would not load. Instead we’d be prompted by the browser to download a right_main.php file. Accounts with smaller inboxes would load fine, so I figured it had something to do with system resources. This is an easy fix. Open up your php.ini file, and increase the memory_limit value to something larger than the default 8M. Restart Apache and try opening your big-ass inbox again. If it still doesn’t work, increase memory_limit again. Repeat until your inbox will load, but be careful if you have little RAM. I was able to load my 80,000+ inbox by setting memory_limit = 32M.