Collaboration on the LAN
katir at hindu.org
Wed Jan 12 22:30:08 EST 2005
OK here is a "brainstorm" type post:
I do a daily web page with photos that have arrived from
a) local digital camera on site
b) incoming pictures from remote locations, Mauritius, Malaysia, Sri
Lanka, USA, UK, India,
c) photos coming in from individuals by email.
I process these, make selections, load into a stack and then I write
the captions for them between the hours of 4-6 PM daily (see
www.gurudeva.org/taka/) in a single stack that lives on our OSX server
and is opened across the network. (btw this scenario is *very*
stable... in 4 years of working like this I have never once had any
stack corruption, even during the worst kinds of crashes)
The stack is pretty simple: a card with a img for the photo, a field
for the caption, some html mark up tools and a previewer substack that
allows quick rearrangment of the order of the photos by simply cutting
and pasting them to different locations in a horizontal catalog.
(which consists of a series of chars whose img src is set to the same
image as the card that is in the same ordinal position as the card
which holds it in the main stack is)
when done, I have these other revolution wizards-scripts do all kinds
of stuff, rotate background photos, generate html chunks, etc. upload
new .htaccess redirect, FTP etc. and viola, daily web page on line.
Now, here's the challenge: this is all happening in a single stack, I
can't get any collaboration on captions without closing the stack and
telling someone else "OK, open the stack and you add to the captions."
Users on the LAN have a simple splash screen "Player" engine and no REV
IDE. But it is a one at a time deal. Since it's in use most of the time
= zero collaboration-zero delegation option.
What we want to be able to do is say to person A "Could you write
caption for photos 5,6,7 ?" and ask person B "Could you write captions
for photos 3,4 ?" And they go to work at the same moment, say 4:15...
result, much more informative web page, more facts, less work for admin
(your truly), and giving others the feeling and opportunity of
contributing without making a giant committment -- most importantly
letting someone else write the caption when they are the best
qualified to do so.
So, goal would be to, during these two hours, when I give the high sign
"Photos are in!" to allow various individual on the land to
a) view the photos in the current order they are in.
b) be able to see captions as they are in the current state
c) edit, add to, or write from scratch captions freely, and update
Given that we have a daily (arbitrary) limit of a max 1 meg upload for
the complete page view (approx 42 photos and captions max, usually much
less,) a macho data base solution is over kill.
So first solution I can think of would be to have a distributed viewer
stack that reads the main stack from the server... and the captions are
external text files... with some kind or semaphore, locking mechanism
"caption being edited" which is on if it is in use, release when done.
Maybe external text files are not necessary... if I had a permanently
open and running rev process-stack on the server and the viewer stack
just talked to that and the captions were contained inside there. Or
may be no stack on the server at all which only has the photos and
external caption text files and semaphore flags
I'm just beginning to think about this, and looking for any and all
More information about the Use-livecode