Re: [Ebib-users] how to edit the key?

[Robbie M. <ro...@ac...>]

Hello Joost, Mandar

------------------------------------------------------------
To:          "Discussion of issues relating to Ebib."
<ebi...@li...>
Subject:      Re: [Ebib-users] how to edit the key?
Message-ID:  <87t...@fa...>
From:         Joost Kremers <joo...@fa...>
Date:         Sun, 02 Sep 2012 00:36:11 +0200
------------------------------------------------------------

> Ah, that already goes a step further. ;-) Not just
> the ability to reload the file, but also a warning
> when it has changed...
>
> Reloading the file is easy to implement, that's just
> another command.
>
> Checking for a file change would require keeping
> track of the file's state when it's opened, or at
> least the time when it's opened and then comparing
> that with the latest modification time when the user
> edits the file.

The reload command should report whether any databases
have changed or not.  That would be useful to know and
easy to compute.  It should also warn if recent user
edits will be overwritten and seek confirmation to
proceed.

---

The code to track the underlying BibTeX files is much
more complicated.  It will need:

  * to be multi-platform
  * deal with simultaneous editing and with conflicts
  * run periodically, say every minute

Even if 'ebib' offered this functionality, your
colleagues could well be editing BibTeX files with
straight text editors ('nano' or Notepad) without this
oversight.  How will they cope with your changes?

An alternative approach would be to put your common
BibTeX files under 'git' or some other revision control
system and let the RCS handle all this.  That is
exactly what they are designed to do, whether
client-server or distributed.

My instinct is that tracking should *not* be added for
the reason that it is difficult to create a
sufficiently robust solution -- moreover false promises
will only lead to confusion when not honored.

Furthermore, one of the appeals of 'ebib' is that it is
lean at only 4k source lines.

---

At the very least, a few paragraphs on how this feature
is to work and in what use-cases should be penned
before the hacking begins.

my initial thoughts but I think you get the drift,
cheers Robbie
---
Robbie Morrison
PhD student -- policy-oriented energy system simulation
Institute for Energy Engineering (IET)
Technical University of Berlin (TU-Berlin), Germany
University email (redirected) : mor...@ie...
Webmail (preferred)           : ro...@ac...
[from Webmail client]