A short history of “web services”

1993: Maybe we could someday have services on this thing, Tim!

1995: We plan to offer services of various kinds on websites.

1998: There are services available, but the free ones don’t work very well. Soon the world of paid web services will be a reality.

1999: There are services available for pay! Oops, we went under.

2003: There are no free services. There are web services for pay and they don’t work. This is the future. LICK IT UP.

4 thoughts on “A short history of “web services”

  1. Web Services Suck My Chode
    The problem with web services is that now, five years later, no two implementations can really interact. You still have this giant gap between the Java (Apache Axis) and M$ (.NET) when it comes down to complex datatypes and the spec STILL does not cover fundamental things like collections (which gives you cool things like iterators and variable-length arrays).
    If I really wanted to write serialize/deserialze or marshal/unmarshal interfaces for all my objects, I would write my own proprietary web services interface or just ditch the whole idea and go with remote procedure calls, which have been around far longer.
    Fortunately, when you control both the client and server (like we do at work), you can do all sorts of fun stuff (with Apache, at least)–but you pretty much kill any chance at interoperability.

    Like

  2. ?
    When you say those words I think Web Services, meaning a preposterously underpowered RPC library that tunnels through HTTP. We’ve definitely got that today.
    And we have services that people use through the web, and they seems to work, even if they aren’t Web Services.

    Like

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.