November 2003 Archives

Offshoring Didn't Kill IT Jobs

| No Comments

Excellent discussion today on Brad DeLong's weblog about free trade and offshoring. I'm one of the people who saw their jobs go to Bangalore, but I believe Jeff Gould is right to say

Roach and the other miss the point by assuming that if somehow offshore weren't a factor then the displaced domestic IT workers would still have their jobs. This is false. Most of the job losses were going to occur anyway as the late 90s bubble deflated. Like I said, at these prices customers just don't need any more software.

Update: Kash over on the Angry Bear weblog raised an excellent point:

Jobs that face no international competition, such as courier services, rail transportation, and various wholesalers, have disappeared just as fast as (or faster than) jobs in software publishing, accounting, and research and development, which are supposed to be the major victims of outsourcing.

The comment thread again is interesting. I want to add a couple of personal reactions.

Yes, IT workers have been training their own replacements. (By the way, employment is usually at will in the U.S., so you can walk away from the severance package. I did, deciding that it was more important to get another job.) That doesn't say anything about what would have happened if there were no outsourcing. If an employer has to cut costs, jobs are likely to go one way or another.

It's not just tier one support or call centers that are being moved offshore. It's probably more true to say that a lot of jobs associated with the back office, not the money-making parts of companies were particularly vulnerable.

It will be more interesting to see what happens during the economic recovery. Will the job market for software engineers improve as much as the market for (say) couriers and messengers? If so, how is the work going to be divided between the U.S. and other countries?

UTF-8 Source Code in ColdFusion

| 2 Comments

There's a lot to know about UTF-8 and Unicode. For my sins, I am maintaining a web site written in ColdFusion that has to support Western and Central European languages now and possibly Asian languages in the future.

Solving a DHCP Mystery

It started munanely enough. A couple of people could not print to the network printer at the reception desk. Restarting the PCs [hey, with Windows 98 you never know!] and the printer had no effect.

Then someone could not read their e-mail. Also not that uncommon. However, for some reason, he mentioned his IP address. His laptop was using my IP address: 192.168.2.4. Besides, that ethernet jack was supposed to be on 192.168.1.*, not 192.168.2.*.

Searching for clues, I ran ipconfig/all. The laptop had got its IP address through DHCP from 169.254.228.128. WTF? 169.254.* is for auto config. And what was wrong with our normal server?

By now, several people reported various problems. A hunt through the office turned up 169.254.228.128: it was a Mac laptop that someone had brought in this morning. He had enabled some network sharing at home and thought no more of it since.

The system event log on our server had two DHCP-related entries:

The DHCP/BINL service ... has encountered another server on this network with IP Address 192.168.1.116, belonging to the domain: .

The DHCP/BINL service on this computer is shutting down. See the previous event log messages for reasons.

After that, it seems the rogue Mac reverted to using a 169.254.* address but kept answering DHCP requests with IP addresses in the 192.168.2.* range.

Solution: stop the DHCP server on the Mac, restart the DHCP service on our server and ipconfig/release/renew on the affected PCs. (Or reboot in the case of Windows 98. Life is too short...)

Making Oracle 8i Do UTF-8 Properly

Note to self: to make Oracle 8i accept UTF-8 as input and generate it as output, just make sure that the NLS_LANG environment variable is set consistently, including when you start the database.

NLS_LANG=AMERICAN_AMERICA.UTF8
export NLS_LANG