Mac Musings

Can't Get There From Here

Daniel Knight - 1999.09.13 -

Here's an area which bears further investigation. One of my readers sent the following email, which says that Apple seems to be blocking access to The Apple Store to users of older Macs.

Since I know a lot of you are using older Macs, and also that 40% of you are even using Windows computer to visit the site (sneaking in on company time?), I'd like you to read this letter, visit Apple's site, and send me some feedback.

I'm going to speculate on something far out, but here goes. Wait for the punch line.

Shortly after the Seybold speech I was unable to access <http://www.apple.com/>, <http://click.linksynergy.com/fs-bin/click?id=ORKVdjntKuM&offerid=77305.10000656&type=4&subid=0>, and <http://apple.com/store/>. I had been able to access these URLs in the past. I have no problem accessing other URLs in cyberspace. I tried many things to correct the problem: trashing all kinds of files, restoring older backups, etc. I am not an unsophisticated user - I own eight older Macs.

My neighbors have no problem - same ISP, same local access number, but a newer Mac.

I began trying to get help from apple around August 6 via email. No response, not even autoresponder. About August 28 I reached a second level Apple response person on the phone and explained the problem. I finished by asking half-facetiously if Apple denied access to individuals or to certain equipment. He said, "Apple certainly wouldn't do that." Then he put me on hold while he "checked around." He eventually came back on the line and said, "I know it sounds amazing but I have to take back what I said about Apple not restricting certain users."

He took my email address and said he would check further and get back to me. Of course I never heard from him again. I called back later to see if I could talk to the same person again but was told only if I had a case number which I didn't. I assume d my full phone number would be good enough to track since it is the first bit of information they ask for.

Well, I finally looked up the ARIN POC for apple.com on the InterNIC database and wrote him with the history of the problem and a threat to write to Steve Jobs and all other upper management I could identify. Shortly after this, I received an email from Apple. Naturally they said it was an ISP problem - and I should go to my provider. Of course I already had. Unfortunately everyone I ask for help can access these URLs, so they attribute the problem to my configuration. They disregard the fact that my system has been very stable for a long time and nothing had changed on my end.

I can ping the errant URLs, usually 10 out of 10 times with average time about 0.23s. I can traceroute, both directions, from every traceroute server I can find. Somewhat interesting is that from tracerouting I find that for the Apple URLs I can reach the gateway/router just prior to destination is tre.apple.com and for the sites I can't reach the gateway/router just prior to destination is ton.apple.com. Apple tells me that these are twins, however, and could not be the problem.

Next I began to wonder if I was the only one in the world being denied access. I began to search news, lists, and forums. Lo and behold I see the same problem being reported from around the world. I mean the same problem! Stable systems, older Macs, were always able to access these sites, all-of-a-sudden, after the Seybold speech, access denied!

Now, get this. I read your post about having to remove info about how to tell the build date of Macs (How Old Is My Mac?). I wonder why Apple would care. Aha! Remember the class action suit regarding support for older Macs? Now I search in earnest. I find that to avail oneself of the refunds and coupons ordered by the court in the settlement, a person with an older Mac needs to know what exactly he has - and also has to access http://www.apple.com/legal/something-or-other. Is this really possible, or am I just so paranoid? By now I'll grasp at anything to show it's not my system at fault.

I'm no Unix guy, but I have studied and learned enough to be dangerous trying to solve this problem. There is one key question I have been unable to answer. What machine specific info, if any, goes out with the initial datagrams requesting access to a specific URL? If there is machine specific info in the initial packets, and you want to deny access before enough of a connection is made to deny based on cookies, simply look at the datagram at the router/server just prior to the URL server and return an ICMP error. That way it just looks like an Internet problem. By the way I think the cutoff date for application for relief in the settlement was Sept 8, 1999. Maybe the problem will go away in a few days?

I have spent far too much time on this problem and wish some would see it as real and take it over. In the meanwhile I will contact others and ask them to mail you their histories of this problem. Thanks for listening.

I don't believe your browser can tell the computer it's connecting to which computer you're using. However, looking at my site statistics, I know that the web server can determine what browser and operating system you're using.

Current stats for lowendmac.net show the following top browsers:

1. Mozilla/4.0 (compatible; MSIE 4.5; Mac_PowerPC)
2. Mozilla/4.61 (Macintosh; I; PPC)
3. Mozilla/4.0 (compatible; MSIE 5.0; Windows 98; DigExt)
4. Mozilla/4.5 (Macintosh; I; PPC)
5. Mozilla/4.6 (Macintosh; I; PPC)
6. Mozilla/4.61 (Macintosh; U; PPC)
7. Mozilla/4.0 (compatible; MSIE 4.01; Windows 95)
8. Mozilla/4.0 (compatible; MSIE 4.01; Windows 98)
9. Mozilla/4.08 (Macintosh; I; PPC, Nav)
10. Mozilla/4.05 (Macintosh; I; PPC, Nav)

So the server can distinguish a PowerPC version of a browser from a 68k one, tell Internet Explorer from Netscape, and even knows the specific browser version. However, it doesn't seem to report which version of the Mac OS is used, let alone the specific model of computer or Gestalt ID. Nor does it seem to distinguish between Classic Networking and Open Transport.

I only have two ancient Macs set up with browsers and an Internet connection. Using both my SE and my Mac Portable with System 7.5.5, Classic Networking (the 68000 doesn't support Open Transport), and MacWeb 1.1, I was able to reach Apple's site, but not get into the store.

My home network shares an ISDN connection managed by IPNetRouter on a Centris 660av with a 3Com Impact IQ ISDN "modem," so the ultimate hardware connection is the same for the SE with an internal ethernet card, Portable with an EtherWave adapter, Quadra 630 with an LC PDS ethernet card, or the internal ethernet on my SuperMac J700.

The problem could involve one or more of the following factors:

• Open Transport vs. Classic Networking
• Specific version of the Mac OS
• Specific browser used
• Presence of Java and/or JavaScript

I'm leaning toward the last one, but can't be sure. So I'm asking for your help. Click on the following URL and see what happens - then tell me.

< http://click.linksynergy.com/fs-bin/click?id=ORKVdjntKuM&offerid=77305.10000656&type=4&subid=0>

If it doesn't work, send a note to (link removed) telling me which computer you're using (for DOS machines, just CPU and MHz speed), which operating system you're using, which browser you're using, whether it has Java support, whether it has JavaScript support (not the same as Java), and, if you're using a Mac, whether you have Open Transport or Classic Networking active (Mac OS 7.6 and later don't give you a choice). If you don't know if your browser has Java or JavaScript, that's find - I can probably locate the information elsewhere.

If it does work, send the same information to (link removed).

Over the next few days, I'll sift through your feedback and try to determine what's causing the problem. As noted above, I think Apple's WebObjects needs Java or JavaScript to function, but only a broader using different computers, operating systems, and browsers will be able to determine what is really going on here.

Thanks in advance for your participation.

• Be sure to read Apple Site Block Solved, which reveals a solution to this problem.