[unrev-II] 2020 Hindsight: A Fictional DKR Narrative (long (sorry))

From: Bill Bearden (BBearden@BCL.net)
Date: Mon Jun 12 2000 - 11:10:14 PDT

  • Next message: altintdev@webtv.net: "Re: [unrev-II] OHS WEBSITE ZWiki at Bootstrap"

    2020 Hindsight: A Fictional DKR Narrative
    I'm heads down writing when the new message alert sounds. This surprises me
    a little because the alert threshold on my writing context is fairly high.
    But it might be interesting so I go ahead and click on the alert.

    My current context cube is automatically put on the top of the stack (which
    is already way to big) and a new cube is opened. I notice right away that
    the message is not context-aware. It is just plain old email. But the
    content parsing agent determined the email was from my old boss, Gill Bates.

    Bates is a little behind the times but very smart. I figure some day he
    might send some more work my way so I leave his record moderated high. That
    is why the alert made it over the threshold.

    The parsing agent produced no summary so I guess I'll have to read the whole
    message. Hmm... Bates is now chairman of the Archaic Software Preservation
    Society. Fitting. I flip to the People face of the cube. It already has his
    record up and I change his title.

    I flip back to the Message face and continue reading. Bates wants to know
    about the origins of Open Source Software. I told him back then he should
    have paid attention. Sigh...

    I click on the "Reply" button and open a blank cube in the body of the new
    message. I flip to the Things face of the new cube and type in "open source
    software?". Whoa, lots of stuff. But the top entry is an old DKR. I'll try
    that first. I right-click on the link to navigate to it inside the current

    The home page comes up and it is the standard stuff. The top of the page
    shows the higher level DKRs all the way back to the top level, the
    Continuum. Let's see, Socio->Tools->Computing->Software->Open Source
    Software. I'm in the right place.

    The next thing on the page is the topic summary. Nothing very interesting
    there. But the next thing is a list of the top 20 current Open Source
    packages. Under that list is a "time slider". I grab it and move it as far
    to the left as it will go: 1994. I then grab the left side of the slider
    button and open it up to 5 years. After a second or two, the list now shows
    the top20 Open Source packages way back then.

    Let's see...

        a.. Linux, an operating system
        b.. Apache, an HTTP server
        c.. Samba, a file/print server
        d.. Perl, a scripting language
        e.. and Emacs, a program editor

    I lasso the top 5 and switch to the People face. The entries are:

        a.. Torvalds
        b.. Behlendorf
        c.. Tridgell
        d.. Wall
        e.. and Stallman

    I read the summaries of each person. There really isn't a lot in common.
    That probably isn't what I'm looking for. I hit the "clear" button and flip
    back to the Things face of the cube. I still have the top 5 packages

    I'm going to have to think about this for a second. What is the most
    important thing about open source software? How do I find that out?

    I move the mouse pointer to the nav-pane on the left side of the face. After
    a second, the nav-pane expands to occupy almost the whole face.

    I see a list of high level relations that connect to the 5 packages. I
    scroll down through the list and right-click on "origin". The nav-pane
    retracts and I now see a summary of the origin of each package.

    Hmm, Torvalds based the Linux kernel on Minix and used GNU's shells and
    things. Apache was originally the NCSA httpd server for which NCSA
    discontinued support. Both projects built on a base of existing code. But
    Samba, Perl and Emacs were all built from the ground up.

    Interesting but this may not be the angle I'm looking for. I go back and
    expand the nav-pane. I continue scrolling down through the list of
    relations. I click on "users".

    Now when the nav pane retracts, I see a high level categorization of the
    users of each of the 5 packages. Let's see, Sysadmin, Webmaster, Sysadmin,
    Web Developer, and Programmer. So the main groups of users for all 5
    packages were technical people. Perhaps that is important. I wonder who all
    made this assertion.

    I again allow the nav-pane to retract. I type in "assertion?" and the
    assertion query dialog box comes up. I type "technical users" in the
    subject. It is changed to "technical people". I can live with that. I type
    in "open source software" in the Object. It takes it.

    Now, what relation? I type in "are important to". It won't take it. I need a
    better verb; "Are" is too general. I type in "promote". It takes it. I click
    on "Find" and off it goes. This will take a few minutes. Good time to take a

    When I come back, a number of entries are on the screen. The top one is a
    reference to an article called "The Cathedral and the Bazaar" (CatB) by Eric
    Raymond. This rings a bell so I right click to navigate through it. Once
    inside of CatB, I pull down the dimensions box and click on assertions.

    A hierarchical representation of the assertions contained in CatB comes up.
    The top level of the hierarchy shows Raymond's 19 "lessons":

        a.. the first - "Every good work of software starts by scratching a
    developer's personal itch."
        b.. the second - "Good programmers know what to write. Great ones know
    what to rewrite (and reuse)."
        c.. and the sixth (in bold because it matches my search) - "Treating
    your users as co-developers is the least-hassle route to rapid code
    improvement and effective debugging."

    So, lesson 6 (and to a lesser degree, lesson 1) supports the idea that Open
    Source needs technical users. The projects must somehow excite technical

    But there in lesson 2 is the word "reuse". That implies existing code. I
    still think this is important. I'll include it too.

    I click "Save" and the context cube minimizes into my reply to Bates. I add
    a little text explaining that "programs for programmers" and starting with
    an existing code base were two of the important factors that gave Open
    Source Software its initial boost. I click send.

    That should keep him busy for a while. Bates' translating agents will turn
    the included cube into lots of text. Perhaps someday he'll upgrade to
    context-aware systems.

    Now, where was I? I pop the top cube off of the stack. Oh yeah, writing on
    my current book: Bits, not Buicks: How Virtual Presence and Cyberspace
    Helped Solve the World Energy Problem. Cool!

    This archive was generated by hypermail 2b29 : Mon Jun 12 2000 - 12:09:26 PDT