The Computer Blog

Wednesday, March 10, 2004

More on the “Gates of Hell-E-Mail Tax”

Macworld magazine’s MacCentral ran an article entitled “Macs help The Spamhaus Project take on spam”. (Click here to read the article yourself.) Steve Linsford, head of The Spamhaus Project, an effort to provide identification and tracking as well as deterrence to the world’s worst spammers, said the following:

"In the fight against Internet spam, which is now 70 percent of all e-mail in the U.S., Apple is nowhere to be seen. In contrast, Microsoft, whose OS insecurities are at the root of most spam problems, is at every spam conference and law enforcement meeting we attend. They position themselves as saviors, but in reality they're very much the silent conveyor of the problem: 70 percent of all spam comes from hijacked Windows machines."

Now, it may not be absolutely correct to blame the problem on Microsoft. Most users do not take the simplest of measures to secure their machines, and they are easily hijacked. However, to blame it all on users is an oversimplification of the problem. The fact that Windows is so insecure is very much at the heart of the problem. And Mr. Linsford hit on the exact reason why I felt Gates’ suggestion about an e-mail tax was so insincere, i.e., it did nothing to address the fact that Gates’ products were part of the reason for the problem. In fact, it appears to me that Gates stands very much to profit from such a measure (e-mail tax). You can bet it would result in either Microsoft selling more “must have” software to make it work (after it had been mandated by the government, no doubt) or somehow, someway, Microsoft would skim off some of the fees.

Instead of us users paying a fee to stop the problem Mr. Gates and company helped create in the first place, maybe we can get the government to insist that Mr. Gates and company pay us a penny for each unsolicited e-mail we receive. The cash drain out of the Microsoft coffers would be large enough where we’d see a technical solution appear out of the company pretty quickly, and they might even give it away for free.

Sunday, March 07, 2004

Upgrading G4 PowerMac Storage—Lessons Learned

Because I spend too much money and time upgrading computer systems, I try to get the most bang for the buck by buying components that will allow me to cascade an upgrade on one system into an upgrade on the other. My latest adventure involved upgrading the storage on my dual 1 Ghz G4 Quicksilver PowerMac. Natively, it would support only two IDE drives using an ATA 66 interface. Since I wanted to make it into a mean video editing machine, I needed more space in it than that. I also wanted to upgrade two of the storage disks on my dual 1.25 Ghz G4 to 160GB capacity. So, I ordered two Maxtor 160 GB hard drives from Multiwave, placed them into the dual 1.25Ghz MDD, and then moved the two 120GB hard drives they replaced, one Maxtor and one Western Digital, over to the Quicksilver. To do that, of course, I had to add an ATA PCI card to the machine; so, I ordered a SIIG Ultra ATA 133 Pro card from Other World Computing to cover that base.

Once I had all the pieces, I easily installed the new 160 GB hard drives in my MDD PowerMac. I had thought about putting the SIIG card there and getting another one for the Quicksilver but decided against that when I saw that the IDE cable supplied and another I had sitting around would not reach the drive interfaces. (The SIIG card only supplies one IDE cable, by the way, so you need to have or buy another one if you’re hoping to hook it up to more than two drives. Even those have to be in a Master/Slave vertical stack.) In the Quicksilver, there simply wasn’t enough room to put two vertical stacks of two drives in the case; the door wouldn’t close. The only way to get four drives into the thing was to use the original drive bracket that holds two droves and put the each of the other two beside it on the floor (where SCSI drives were designed to go). While that gets all four drives into the case, a single IDE cable would not reach both single drives mounted on the floor. After examining the possibilities, I decided that the best I was going to do was put three drives on the IDE card and one on the Quicksilver’s native ATA 66 interface. But what drives did I hook to which interface to get the best performance? I conducted some tests with the boot drive connected to the native ATA interface and the new IDE card. Concluding that it seemed to make little difference where it was, I put the boot disk on the native ATA interface and the three data (video) storage disks on the ATA 133 IDE card. I rigged the boot drive as a Master and placed it farthest away in one of the SCSI bays, hooked it up to the ATA 66 interface using a spare IDE cable (yes, 80 pin); placed Hard Drive 2 in the SCSI bay next to it and hooked it up to the IDE 1 port (the closest) on the SIIG card (ATA 133), put Hard Drives 3 and 4 in the original two disk bracket and hooked them both up to a single IDE cable on IDE 2 of the SIIG card. Later, after I bought a 36” IDE cable, I hooked up all drives to the ATA 133 interface even with that physical layout.

It seems to be working like a champ, but there are a couple of things I learned about OS X and extra IDE cards and drives.

When I booted into Panther, I noticed that the hard drives located on the ATA card all had an “eject” icon next to them in Finder! I surmised that the OS was seeing the drives as “removable”, and I clicked on the Eject icon to see what it would do. Sure enough, the hard disk unmounted, disappearing from view. I had to reboot to recover the drive. While I have enough self-discipline not to do click on an Eject icon and presume the system won’t let me do that when using the drive for capture (or anything else), I don’t care much for the implementation. I don’t understand why the OS can’t tell that the drives are fixed. This case of mistaken identity holds true for every volume except the start-up volume, inclduign other partitions on the boot disk. The operating system will obviously not let you eject the start-up volume, even if it is located on an ATA card.

Secondly, on the first boot after hooking up the start-up drive to the card, the Mac displayed a folder icon overwritten with a flashing question mark and spent some time hunting for the start-up volume, even though I had not changed it in System Preferences since the last start. A trip to the Apple Support site suggested I reselect the start-up drive, which I did; and the problem did go away. All starts after that were normal.

I had seen some information telling me I would need to reinitialize any drives attached to the card. I did reinitialize two of my storage disks but did not reinitialize the third or the boot disk. Still, everything appears to be working fine.

Can I see much of a performance difference? Not really. In fact, there are quite a few experts out on the web who will tell you there is no real performance difference between ATA66 and ATA 100 speed-wise. My hard disks are all ATA 133, but this was worth what I went through because I know I can use hard disks bigger than 137GB and I have 4 hard disks in my Quicksilver where before I only had two. Speed gain or no, that makes it worth it.

If there’s anyone out there who put one of these SIIG cards in a MDD PowerMac and is using it to capture video, I’d appreciate it if you’d drop me a line and let me know whether it made a difference for you. I’m going to capture footage using the PowerMac’s native ATA100 and 66 interfaces and see how they do. If I have problems, I may try out the SIIG card in that machine and order another, unless someone has tried the Sonnet card, too, and can tell me it’s better.

Gates of Hell…The E-Mail Tax

If there is a way Bill Gates can inhibit (as in "control") computing while extracting more money from you, he’s going to do it. Take the e-mail tax, for instance. Sure, it doesn’t bother Gates if he has to pay a penney an e-mail. He won’t miss the money. As long as it solves the problem of spam, something his own products have not been very adept at handling, then he’s happy. He has no problem with the chilling effect such a move would have on the Internet’s ability to foster Free Speech. I do. And I’m here to tell you that I would not only strenuously object to such a move on both economic and political grounds, but I have to ask why it is that pundits such as Gates seem to forget that I’m ALREADY paying to support the Net. I agree it’s not right that the fees I pay my ISP and the fees I pay my web hosting service (where my mail servers are) wind up supporting the spammers. They have already proven they have no issue with taking from others or with walking a thin line of what’s legal and what’s not and sometimes crossing it, all in order to make a sale. That said, I believe there has to be other ways to ensure that these folks stop this abusive practice. Taxing the people who are victims of the problem is NOT it. Going after the people who are doing this and making sure they can’t use the resources of others to accomplish it are.

Tuesday, March 02, 2004

Finally, USB 2.0 Success!

I rearranged my home office this weekend as part of an effort to really get serious about getting a video business off the ground. The place had just not felt right ever since I gave my iMac to my daughter-in-law. It left an empty desk sitting right next to my Dual 1 GHz PowerMac. Since I have an extra 17 inch Apple Studio LCD in the closet and I really expect I can use one more computer to edit and encode video, I gave a lot of thought to buying a G4 or a G5. But in the end, practicality and outstanding credit card debt won out. I decided instead to expand the hard disk capacity in my Quicksilver (the dual 1 GHz PowerMac) and reorganized my home office so that my Windows XP computer was a bit handier and able to fill in when both the Macs are busy. There was only one catch. I had not been able to get the XP machine to work with my USB 2.0 home network.

Every time I tried to scan using my Epson Photo 1660, I'd get a "scanner did not send image" error message. Worse, my new HP Business Inkjet 1100D would act strangely when attached to the Windows machine via the hub and spit out uncommanded pages with a line of garbage on it. Thinking the problem was with my USB hub, I tried several of them; but they made no difference. This past weekend after I finished moving desks around and building a bookcase, I tried troubleshooting the problem some more. Because the system gave me slow and sluggish responses and even the same errors with my scanner plugged directly into the PC, I began to suspect the motherboard's implementation of USB 2.0. (I'm running a MSI KT3 Ultra 2 updated with the latest BIOS.) So, deciding to try one more thing, I pulled a SIIG USB 2.0 PCI card out of my Quicksilver and put it in my XP computer, cranked up the PC, and turned off all the USB ports on the motherboard. Much to my delight, it worked!

Sunday afternoon, I drove over to a local Best Buy and picked up another copy of the SIIG card. I installed it in the XP machine, and it worked for a few hours. But, then after I shut down and restarted my system, my scanner performance became very slow, no better than USB 1.1 speeds. During troubleshooting, I made the mistake of trying the SIIG drivers for XP even though I was running XP's Service Pack 1, which has USB 2.0 drivers in it. Little yellow question marks popped up on some of the USB card's controller entries in Device Manager, and no amount of installing or reinstalling would fix them. Once again, XP had screwed me. That refurbished 1.6 G5 sitting at the Apple Store began looking attractive...

After a good night’s sleep and more pensive thought about the whole thing, I decided to try a couple of things before I threw in the towel and relegated the XP machine to running flight simulators and helping out with network access at work for me and my wife. I tried re-installing XP’s Service Pack 1. That had no effect. Then, I shut the computer down, shuffled its Ethernet, Firewire, and the USB 2 card into different PCI slots, and started it back up. Eureka! I was in business! The scanner was working via my network at USB 2.0 speeds. I flopped over to the Windows 98SE side of the machine and tried it again with equal success. In short, the PC began acting like I had been expecting it to. It’s nice having everything working….for once!