[Zope] (Fwd) Zope musings

[David Ascher]

On Wed, 17 Feb 1999, Paul Everitt wrote:

I'll jump in, since this is an issue I worry about w.r.t Zope's success
well for places I know, such as university departments where faculty, not
hackers, do web page editing and e.g. my organization, where I want e.g.
the HR manager to be able to edit the web pages and blissfully ignore all
of the cool Zope features that the webmaster really likes...

> I certainly agree.  We have spent a lot of time over the last month
> thinking about an IDE.  The direction that we are going is:
> 
> o choose an IDE strategy that reinforces Open Source, rather than
> attempt to annoint a proprietary tool like CF Studio
> 
> o base the IDE strategy on advanced standards (HTML4, CSS2, DOM, XML,
> RDF, and WebDAV)
> 
> It appears that I'll have more to say on this in about a month.  Other
> efforts, such as integrating PythonWorks from Pythonware, are feasible
> as well.

All sounds fine, of course.  It does sound distant -- which is fine in the
long term, but worrisome in the short term.  I don't know of a single
WebDAV client I can play with, let alone a full-featured WebDAV-aware HTML
editor.

> If someone has Linux on their desktop, how do they manage CF sites?

Irrelevant if your shop (as mine does) has a single Linux box as the
server, and oodles of macs and PC's on desktops.  

> Let me ask a question that tries to quantify the situation.  IMO, the
> current Zope IDE is pretty unproductive.  On a scale of 1 to 100, its
> *productivity* level is about a 5 compared to NetObjects Fusion,
> Dreamweaver, etc.  On other factors it shines -- it is based on
> standards, is completely portable, source code is available, is
> mind-numbingly easy to modify the IDE, etc.
> 
> Just for argument, let's say that CF Studio is a 90.  If we came out
> with an improved IDE that retained the factors listed above that CF
> Studio fails at, what number would it need to move up to for you to give
> an unqualified "Yes!" ??

In our case, I suspect, a 20 or 30 would be enough.  What's most important
is, I think:

  - a GUI for the 95% of the web page editing tasks (a-la Netscape
    Composer, Frontpage, etc.).

  - a clean, robust, and simple interaction between the user (someone who
    does not want to learn HTML but wants to manage their web page
    nonetheless) and the "web server" -- Zope in this case.

> If *we* is your shop, how about...XEmacs?  Using ZServer to publish your
> object system by HTTP, you can be *significantly* more productive:

Again, emacs is irrelevant for my HR manager, who'd quit if she was told
she had to learn it (and, I dare say, she'd be right =).  It *could* be
that Netscape Composer is the answer for my shop -- I haven't tried it
with the FTP-enabled Zope.

> It would still be missing a *whole* lot of things vs. CF Studio, such as
> link checkers, a list of variables that can be inserted, syntax
> colorization of the markup, etc.  Let's say it moved up to a 25.

From the perspective of someone who wants to use tools and not hack code,
emacs is useless.  The folks I'm trying to convince rave about GoLive or
whatever it's called.  I looked at it and couldn't get it to manage Zope
pages, but that could be me.  

So, as a summary:

  - emacs is fine for folks that already use emacs, but...
  - emacs is not an HTML-editing tool.  It's a swiss-army chainsaw, as
    we all know.  (Besides, what about the vi shops? =))
  - I think Zope is an easy sell to webmasters who think ahead. 
  - I think Zope is currently a hard sell to folks who like shrink-wrapped
    software with lots of buttons and gizmos and WYSIWYG.

Trying to be helpful...

--david

PS: *I* use emacs all day long.