Through the closets and shelves downstairs, what do I find?  Selective issues of  Dr. Dobb's Journal, dating as far back as 1993, to 1998.  At some point I think I had a subscription, and I thought I'd be referring to these things over time.  Turns out that you can get it all online, on CDs, and various other venues.

Interesting headings:

Unlocking Encryption Algorithms - Well Bruce Schneier created a few books on the subject, and runs Counterpane Software (  I bought his “Applied Cryptography” book (1st and 2nd Editions).  Then of course there's the TwoFish book.  Now I'm ready to recycle them because the work has advanced and these references can all be had online.  Bruce Charges a lot of money to get his software on a CD, but, you can find it all online if you're industrious.

Morphing Magic - Became a book, I bought it, then replaced it with a more rigorous book later.

Visual Programming - Used to be big, and there were probably big conferences on it.  Nowadays.  With the advent of Java and sloppy programming, we can produce many more bugs more quickly with a simple editor, so we don't need this visual assistance.  I'm thinking Intentional Programming will actually be the next big breakthrough in this area.

Object Oriented Programming - Now that's gotta be hot.  I wonder if I should keep that one?

Oh, and look, there's one on this thing called XML.  That one might be hot one day.  I better hold onto it just in case I need to refer back to that complex syntax they talked about.  And on and on it goes.  10 years of magazines that seemed extremely interesting at the time, but are nothing more than online references at this point.

Well, I'm a programming guy, and I'm sure there's some sort of programming analogy I can draw here.  One trend I see by investigating this past is what turned out to be truly important, and what turned out to be more short term fads.  Undocumented NT and Netware core protocol?  Net What?

How about Persistence for C++, Distributed computing and component objects?  And my favorite, Optimizing VRML!!

I guess the point would be, when things get really good, they get preserved.  Yah some things that aren't particularly good get preserved some time too, but in general I think this is true.

Since I work on XML, I wonder, what parts of it will be the parts that I'll be throwing away as esoterica, and which parts will be getting preserved in a more permanent form, perhaps as part of a language.  Looking into this crystal ball, and making good choices for the future is what our jobs are all about.

SGML, the genesis of XML, has a lot of crufty esoteric stuff.  If it didn't, there never would have been a need for XML.  XML itself is even starting to get some cruft.  There is now this 1.1 thing, namespaces, XSD, and whatnot.  What do you make of it?  What parts of this will be the Dr Dobb's 1993 issue on the beauties of Netware, and which parts will be longer lasting?

Only time will tell for sure, but in the meanwhile, we'll influence and shape the outcome, doing enough things right such that history will have plenty to choose from in determining how foresightful we were today.