There I was, flying across the Atlantic and blogging. Continental now have AC power under every seat, so I could use my notebook as long as I wanted; and the first thing I did was clean out my email. Except for the few messages that included a URL and a description intriguing enough to make me want to check the full article they pointed at. For that I had to wait to be on Terra Firma. It would’ve been better had the senders pasted the entire article.

Which reminds me that some years ago, the idea of sending pointers instead of a full attachment was the latest rage. This was motivated by tight mailbox quotas coupled with slow networking, especially from home (remember 14.4 KBPS modems?) Somehow the notion that this was a smart thing to do caught on among the user population; it did have a sensible ring to it, an aura of thrift… but was it a good idea?

In a world of perfect connectivity, placing a single copy on the web, or the company Intranet, or a group repository like MS SharePoint, then sending people pointers to it, would make perfect sense. In the real world, where mail storage is now cheap and often  plentiful but we are still offline at many times, this is a compromise. Whether it’s one worth making depends on the situation. In the years when connectivity was costly I’d tell my coworkers and trainees: if you send the document to a list most of whom need to read it, you might as well attach it to the message. That way they’ll have the freedom to read it anywhere, and you won’t save any traffic by keeping it online (because they’ll all be pulling the content when they read it there). Put it on the network and send out a link only when most recipients will not need to open it.

Today, as long as you have plentiful storage and bauds, there’s no reason not to attach anything you want the recipient to read. At least I don’t think so – do you see a scenario to the contrary?