PDF storage software recommendations?

[Polytropon]

On Fri, 18 Jun 2010 19:24:52 -0700 (PDT), Bill Tillman <btillman99 at yahoo.com> wrote:
> 
> Date: Thu, 17 Jun 2010 23:31:15 -0700
> From: Charlie Kester <corky1951 at comcast.net>
> Subject: Re: PDF storage software recommendations?
> To: freebsd-questions at freebsd.org
> Message-ID: <20100618063115.GA57196 at comcast.net>
> Content-Type: text/plain; charset=us-ascii; format=flowed
> 
> On Thu 17 Jun 2010 at 19:57:03 PDT Polytropon wrote:
> >
> >Maybe my answer will sound "low level", but it works - REALLY works -
> >and works with mostly every kind of data.
> >
> >It's good to see someone recommending a true Unix-style solution.  :)
> 
> Here, here.  I too love simple text files. With the speed of today's
> computers it's not impractical to use text files.

That's basically what you have computer for - to make work faster,
not slower. :-)



> And something like you suggest with awk I think would work....except
> for one major thing. When building a database like this you usually
> have to build an interface that normal users will work with.

That's why I suggested building a shell + Tcl/Tk script around it
for the various "database operations" you can perform with it. The
idea is that it is customizable ad infinitum, because everything is
programmable into deepest details.



> And something that I could use versus something the other people
> in the office could use are often worlds apart. I once wrote a
> program to do linear optimization for cutting metal parts from
> stock lengths. For me it was a simple block of code about 30-40
> lines as I recall. The other guys in the warehouse saw it and
> told the boss they wanted it too. He then instructed me to expand
> it so the common users could work with it. Well 2 months later
> and about another 400 lines of code to make it user friendly we
> finally had something. So as I see it the interface for other
> "not so tech-savvy" users will be the trouble with this approach.

This sounds familiar. :-) I've also walked this way for "average
users", at this time, by choice was to create a GUI control program
using C with Gtk. Today, I would consider that totally overhead.
My "average users" were psychiatrists, so any assumption about
intelligency would not match the reality. :-)

You wonder how people got their work done on 80x25 in a "wrong"
language 30 years ago...


> But put me
> down for a vote on this method using simple text files and awk.

It JUST WORKS - that's the goal. It can be developed and configured
very fast, can easily be extended (or limited), and data is stored
in a STANDARD (!!!) format which allows you to do ANYTHING with
it. You can even provide a web-driven interface for the database,
even that is possible.



> We have a Windows based system at my current job which uses
> FileMaker Pro. It's amazing what we can do with this and it's
> like having a gigantic electronic filing cabinet.

Oh, the paperless office... an utopia - at least in Germany,
bureaucracy's home country. :-)

Anyway, relying on a "Windows" program is, in my opinion, not the
best choice for a long-term project such as document filing. With
the constant transitions in the underlyíng OS, and the immense
costs, as well as the lock-in driven by closed (non-standard)
formats, and finally through the limitation of what the original
program developers did provide, makes me wonder if this can really
be useful for longer times (let's say, +20 years - where the "low
level" solution does still work).



> It's pricey and it took the IT guys some time to build it but
> it does do some fantastic things in keeping tons of files organized,
> indexed and searchable. But I'd like to try my hand at building
> something with text files and awk.

Just imagine about 20 years in the future, and you'll see what's
the better solution. :-)



-- 
Polytropon
Magdeburg, Germany
Happy FreeBSD user since 4.0
Andra moi ennepe, Mousa, ...