Trips Down Memory Lane

With the rainy weather setting in, I took the time to not only do my quarterly backups but also to check old CDs and see if it was time to toss them aside. And boy, was it fun...

But before I go into that, here are three pieces of news:

  • The GIF file format is, as of today, completely patent-free. Instead of lauding that fact or harping about how Unisys was "evil" (or some such nonsense), allow me to publicly state that I hope Open Source package maintainers get off their high horses now and make it easier to use the format. Sure, PNG is "better" - but only so much given that there are a gazillion GIFs out there, and I'm tired of having to patch to be able to transcode images to mobile phones.
  • Core 6 has a pre-release available (the full release is scheduled for ). Having seen another distribution falling prey to the quest for fancy effects, I'm rather disappointed that we're nearly in 2007 and have yet to get proper, dual-head support (i.e., not having to kill the X server to enable/disable it, which I find profoundly stupid). Besides this folly quest for fancy desktop glitter (which they can blame on X, sure, but that we oughtn't to have let up to 7.1 without this fixed), I am also rather saddened to notice that the laptop special interest group shows so little mailing-list activity.
  • The odd one out: Via Nuno (who is still trying to figure out how to escape HTML in his feeds so that it won't trip up Planet Tao - I switched over to his RSS 2.0 feed, which is passably encoded), I came across treearrange, which is a brilliant complement to rsync, and solves the usual issue of having slightly different directory trees on different machines (more on Brad Fitzpatrick's blog).

Plus ça change...

So, while rummaging through my backup CDs from last century (1998) onwards, I found a number of interesting things among the oodles of documents and web site snapshots:

  • First and foremost, the GIF file above (which has finally come back to haunt me).
  • Far too many Slackware snapshots. Heck, I only ran it for a year or so...
  • A 2.0 CD, apparently in working order.
  • Drivers for ancient and obsolete hardware, including Glide versions for the ancient Voodoo 3dfx "sidecar".
  • Quaint "" folders with tarballs of Perl 5.008, 1.5.2, GD 1.3, 2.x, WindowMaker 0.30, KDE 1.1 and the like (you know, before you took permanent connectivity for granted).
  • A set of scripts to scrape news off Slashdot, BeNews, Freshmeat, fetch , etc. (you know, before RSS feeds came along).
  • Oodles of and maps, patches, our team skins, etc. (you know, before came along).
  • Ancient versions of Palm software (desktop software, system patches, even an client).
  • Tarballs from my days, in a folder called "Really Old Stuff", right next to a bunch of Aldus PageMaker files with some of my coursework, and which will be a pain to open with modern software - so much for Data For The Ages.
  • Installers for version 2.x of WinGate (remember that? It was quite popular among small business customers in my day), all under 512KB in size.
  • A DOS client that used the Watcom TCP stack (along several files and folders with quaint 8.3 filenames).
  • A bunch of proprietary data files from The Brain, which is what I used instead of a Wiki.
  • Ancient BeOS software (including a partial mirror of ftp.cdrom.com's BeOS tree, dated 1 Feb 1999).
  • And, finally, BeMD2, a model viewer for BeOS, which I used to distribute under the GPL:

The most amazing thing for me is that all the images looked so tiny. My "big" desktop circa 1999 was 1152x864 (probably one of the biggest screens in the office):


My home away from the

...but all the stuff I came across on the CDs so far was designed to fit a 640x480 screen, and it shows.

Anyway, while reviewing the CDs I came to realize that I had pretty much the same problems back then - keeping track of information, aggregating news, synchronizing files to and fro my laptop, backing up e-mail (although I used back then), etc.

Sure, switching to a made most of the mechanics easier, but what is that "progress in computing" thing I've been hearing about over the past eight years or so?

This page is referenced in: