Information manager/organizer with tags question.

A

andrei.avk

Hello,

I would like to write an information manager/organizer type of app but
first I'd like to ask if there is something like that already
existing. I asked on yahoo questions but did not get a response. I
also searched freshmeat but even though they have plenty of projects
that are tangentially related, I was unable to find anything close
enough.

What I'm thinking about is this: a database with an interface that
stores records by tags and allows filtering and sorting by these tags.
When I enter a new record, I enter a title and any number of tags.
Interface would have a filter by: pull down menu separated in two
sections, favorite tags and recent tags. To the right of that would be
a button to add another filter, which would create a new pull down
menu to filter current results further. Then there would be a search
field where you'd type in a word and it would show a list of all
matching tags; then you would click on one of the tags to add it to
last filter menu's 'recent' list. Interface would be two-pane with an
option to switch to single pane, where one pane is a list of matching
records and the other is current record.

List of matching records would be sortable by clicking on column
titles.

Going into the record would bring up a screen with title and body text
of the record, if the record is a file it'd show title, list of tags
and a button that opens that file. (unless it's a text file that can
be viewed and edited there).

The idea is to have a central hub of all information, that is very
easy to search and locate not only the exact document you're looking
for but also a specific subset of documents, including all of the ones
you do need and excluding ones that you do not, which allows you to
work with them as a set without the need to specifically select each
and every one you need.

For the first version it'd be enough to have plain text body field.
Later I'd like to have a rich document format, probably html.

Later I'd also like to add an option for matching synonymous words to
tags. If you search for certainty, it would match assurance tag if the
option is on.

For instance, if I wanted to find this message, I'd search for Usenet
posts, tag, my name (as author). I would know that I only made one
Usenet post about tags. If I made unrelated posts about html tags, for
instance, I'd filter by 'information manager', and then I'd find this
post and more posts that I might make about this. If I wanted to
include possible replies, I'd remove my name from filters. Then I
could sort by date to see latest posts.

The reason for this idea is that right now I keep my information, or
at least most of it, in an outliner program called treepad. It's a
tree-like view program where you make a hierarchical structure similar
to file system, where each 'file' is a text entry field. This is
reasonably useful but is limited to only one possible subset, i.e. you
could put all posts under 'Usenet posts/info manager/a, b, c' but it's
impossible to filter it further or to combine it with some subset of a
different topic. Many chunks of information belong to multiple topics
and you can't always remember if you filed under one or the other.
Therefore I think this tag-based approach would be much more flexible.

I have experience with doing a few simple apps in Tkinter and also a
db driven site in TurboGears. I'm not sure what framework I should use
for this. I generally would prefer gtk because of responsiveness and
nice looks, but it may be a bit tough since I never did any gtk
development at all. Should I prototype in Tkinter? Would Django be
better suited for this or TurboGears?

Of course there isn't any need to do anything if there is already an
app that does this. Please tell me if it does, or if there is
something very close that could have this set of features added
easily. I was thinking about wikis, in fact. I might be best off
starting with a wiki done in TurboGears and going from there. I'd get
html markup pages right away, with links, nice formatting.. TG has
many components that could be useful for this app.
 
A

Aaron Watters

Hello,

I would like to write an information manager/organizer type of app but
first I'd like to ask if there is something like that already...

Your outline sounds like a killer app, and I don't know of anything
like
it (especially for free). However, I have to plug nucular as a
possible
back end data store that may make it easier for you to implement the
app you outlined. Please have a look:

http://nucular.sourceforge.net

See the "demos" for examples of what you can do with it very easily.
-- Aaron Watters

===
http://www.xfeedme.com/nucular/pydistro.py/go?FREETEXT=unsuspecting+victim
 
A

andrei.avk

Your outline sounds like a killer app, and I don't know of anything
like
it (especially for free). However, I have to plug nucular as a
possible
back end data store that may make it easier for you to implement the
app you outlined. Please have a look:

http://nucular.sourceforge.net

See the "demos" for examples of what you can do with it very easily.
-- Aaron Watters

===http://www.xfeedme.com/nucular/pydistro.py/go?FREETEXT=unsuspecting+v...

Sounds interesting and thanks for reply but I need something that will
work in windows also.. -andrei
 
W

Wildemar Wildenburger

Hello,

[snip outline of an information editor]

Maybe the Literary Machine?
<URL:http://sommestad.com/lm.htm>

It is a bit weird at first, and Windows only (altough I have had it
running under wine once). But it is actually really well designed and
very powerful.


Or, if it ever comes out (I've been waiting for it for YEARS now),
perhaps Ceryle would fit the bill:

<URL:http://www.altheim.com/ceryle/wiki/>

There once was some code to download, but I couldnt get it to compile.
The guy who writes the software is a really approachable, nice guy.
Maybe if you contact him and politely prod him into finally getting the
thing out, nobody will ever need to write any information manager anymore.

/W
 
A

andrei.avk

[snip outline of an information editor]

Maybe the Literary Machine?
<URL:http://sommestad.com/lm.htm>

It is a bit weird at first, and Windows only (altough I have had it
running under wine once). But it is actually really well designed and
very powerful.

I installed it and looked at flash demos, however, I have a couple of
questions.. is there a way to filter by several keywords? Is there a
way to sort by keywords? Is there a way to add a new keyword to a
topic
that isn't in the list of keywords?


Or, if it ever comes out (I've been waiting for it for YEARS now),
perhaps Ceryle would fit the bill:

<URL:http://www.altheim.com/ceryle/wiki/>

This is also interesting.. the biggest two issues for my purposes is
that it doesn't have a UI geared for the usage I was thinking of. I
was thinking of something along the lines of Treepad, which is an
incredible program because it makes operations so easy and quick to
do. ceryle seems like emacs to treepad's (and my intended program's)
vim. It just occured to me that I might want to try to hack vim into
doing this.. I'll have to think about that. It's also a little
alarming
that they didn't get it out in years. This is a very bad sign. It
looks
bloated imho.

My idea is that several key actions should be incredibly easy to do,
i.e.:
1. add keyword - already a field on screen, just click and type
2. add existing keyword out of favorites list - click a toolbar button
3. sort by keyword - popup already on screen
4. filtering by two / three keywords is the most common usage scenario
and
therefore should be possible in 3 clicks - one for each keyword.

Otherwise, if these actions are buried somewhere, you will simply not
use
them often enough to make the whole thing consistently searchable by
the
tags. You will think "oh wait I can search for that tag and that other
one"
and then you will think "did I add keywords to that item?".

I'd like to try to make it possible to have tags as an integral and
inherent
part of each and every doc. Otherwise they'll end up like webpage tags
-
a bit of help with adding weight to full search.

I don't mean to criticize these programs, they're very interesting and
I'm
glad you brought them to my attention. I'm not sure I understood the
first
one properly, it may be that it is possible to do what I'm thinking of
in it,
I just didn't spot that in the interface. Please help if that is the
case!

However they seem to have different focus and that's important because
both
are closed, at least now, so it's not the case I was thinking of of
some
light simple wiki system where I'd have to add a few features.

Again, thanks for your response! -andrei
 
A

andrei.avk

[snip outline of an information editor]

Maybe the Literary Machine?
<URL:http://sommestad.com/lm.htm>

It is a bit weird at first, and Windows only (altough I have had it
running under wine once). But it is actually really well designed and
very powerful.

I installed it and looked at flash demos, however, I have a couple of
questions.. is there a way to filter by several keywords? Is there a
way to sort by keywords? Is there a way to add a new keyword to a
topic
that isn't in the list of keywords?


Or, if it ever comes out (I've been waiting for it for YEARS now),
perhaps Ceryle would fit the bill:

<URL:http://www.altheim.com/ceryle/wiki/>

This is also interesting.. the biggest two issues for my purposes is
that it doesn't have a UI geared for the usage I was thinking of. I
was thinking of something along the lines of Treepad, which is an
incredible program because it makes operations so easy and quick to
do. ceryle seems like emacs to treepad's (and my intended program's)
vim. It just occured to me that I might want to try to hack vim into
doing this.. I'll have to think about that. It's also a little
alarming
that they didn't get it out in years. This is a very bad sign. It
looks
bloated imho.

My idea is that several key actions should be incredibly easy to do,
i.e.:
1. add keyword - already a field on screen, just click and type
2. add existing keyword out of favorites list - click a toolbar button
3. sort by keyword - popup already on screen
4. filtering by two / three keywords is the most common usage scenario
and
therefore should be possible in 3 clicks - one for each keyword.

Otherwise, if these actions are buried somewhere, you will simply not
use
them often enough to make the whole thing consistently searchable by
the
tags. You will think "oh wait I can search for that tag and that other
one"
and then you will think "did I add keywords to that item?".

I'd like to try to make it possible to have tags as an integral and
inherent
part of each and every doc. Otherwise they'll end up like webpage tags
-
a bit of help with adding weight to full search.

I don't mean to criticize these programs, they're very interesting and
I'm
glad you brought them to my attention. I'm not sure I understood the
first
one properly, it may be that it is possible to do what I'm thinking of
in it,
I just didn't spot that in the interface. Please help if that is the
case!

However they seem to have different focus and that's important because
both
are closed, at least now, so it's not the case I was thinking of of
some
light simple wiki system where I'd have to add a few features.

Again, thanks for your response! -andrei
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

Forum statistics

Threads
473,982
Messages
2,570,185
Members
46,736
Latest member
AdolphBig6

Latest Threads

Top