% fortune -ae paul murphy

Working with "average" users

For the last two months I've been developing and supporting a number of web initiatives with a total now of about 7,000 current users. In most cases there's no personal user contact involved in this, but one of the things we're trying out involves using limesurvey to get information from small groups - and that does generate frequent user contact.

For the most recent test period my web logs show 7,228 (23%) initial load requests coming from Macs, 21,437 (69%) coming from various Windows products, 945 coming from iPhones, 557 coming from Linux - and 63 from Blackberries.

However, all of the "help me - it doesn't work" calls and emails have come from people using Windows, most commonly XP. The single most common source of the problems these people have is that the email we send them has a unique token embedded in the url and because the total length of the thing (82 characters) exceeds the width they have their email set up for, clicking on it gets them to the survey page, but limesurvey refuses the token either because part of it is cut off or because there's an embedded LF/CR.

What's amazing about this isn't that it happens, but the attitude of the people it happens to. Young, old, or in the middle; male or female; they invariably apologize for their failure in getting it to work - and quite a few assure me they're going to get a new PC RSN and won't mess that one up.

People with more difficult problems do that too: most of those have been through what Alberta's two big home network vendors (Telus and Shaw) laughingly call "support" and are absolutely convinced it's their fault that things don't work. It usually isn't - and I've been able to resolve all but one issue: I've got several people on the same cable group as my test server and they can connect to the production machine in Vancouver but not the test system on the same cable net.

Here's what a traceroute from 96.51.40.216 (loaded DHCP from shaw) looks like:

C:\Documents and Settings\IPACANADA>tracert telearb.net

Tracing route to telearb.net [96.53.6.178] [assigned static from Shaw] over a maximum of 30 hops: 1 1 ms 1 ms <1 ms 192.168.0.1 2 * * * Request timed out. 3 * * * Request timed out. 4 * * * Request timed out. 5 * * * Request timed out.

Shaw, of course, said the trace showed he wasn't connected to their internet service, but he emailed the trace to them over that disconnected service. Then they told him to reboot, twice; and finally that he has "missing files" and should reload Windows. Since he's not the only one with this problem, I told him it's highly unlikely to have anything to do with him or his PC.

On the other hand, I haven't got a clue what's causing this - and any ideas you have will be greatly appreciated.

(A Mac booted DHCP from Shaw using a hub on the same cable modem as the server, connects to the server. the back-up Solaris server with a de novo netconfig is similarly reachable from everywhere except some Shaw DHCP accounts. Netmask is 248 - I haven't tried 0.0.0.0 yet :) )

My bottom line on dealing with users blaming themselves, however, is that their behavior shows the classic hallmarks of long term psychological abuse. Their collective guilt, self-abasement in the face of the overwhelming arrogance of "support", and promises to do better unfounded in any understanding of what's wrong, would, if exhibited by school children or prison inmates, trigger immediate investigation - but there's nothing I can to help beyond telling them the obvious: it's not their fault - all the while, of course, wishing vaguely that I had Bill Gate's money, because then I could buy them all Macs.


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specializing in Unix and Unix-related management issues.