Re: Knowledge management (for Embedded engineers)
From: David T. Ashley (dashley_at_cequentgroup.com)
Date: Tue, 26 Oct 2004 12:20:13 -0400
The group's replies were very refreshing. Personally, I keep my work notes
directly as HTML (as well as other materials, including scanned documents,
which I store as .PDF). I run a Linux server to do this. Every evening, a
search engine (ht://Dig, www.htdig.org) reindexes all content for quick
searches, and it gets served out by Apache (www.apache.org), a web server.
I can find anything at any time. The only trick is to give web pages
meaningful titles, and also to give .PDF files meaningful titles to help the
The group's replies were refreshing because I thought I was the only one who
did this or something similar. Good to know we all have similar problems.
But, there is a larger problem. Here are my observations about it:
a)First, I agree that there is information overload. Even keeping track of
things to do can be overwhelming. Tasks spawn subtasks, etc. etc. etc. It
can be a mess just keeping track of it all. Tools don't replace human
intelligence, but they can manage the mass of information and allow a human
to "restore context" when they start again on a problem.
b)In software development, it gets even messier. One then needs to add
product defects (and steps to closure), version control, lessons learned,
peer review issues, etc.
c)One also should not ignore the collaborative aspect of the problem. I've
had good luck allowing my colleagues to connect to my Linux server and
having them search my notes. So, what if a colleague wants to close an
issue that you're working on? Any good solution has to be more than for
Here are the general organizational information management goals:
a)To have a unique and defined place for every item of information that
accumulates during a product development.
b)To enhance (a) for searchability, electronic collaboration, and to
minimize or eliminate redundant information.
c)To accomplish (a) and (b) in the most changeable framework so that the IT
model used to support processes can easily change along with the processes.
Regarding (c), there are advantages to web-based database solutions using a
scripting language (say, PHP). The advantages are (a)everybody already has
the client (a web browser) which handles display and printing, so that this
is something the programmers don't have to worry about, (b)same argument for
a a database engine, and for the web server (these handle areas of
complexity so that the developer can concentrate more on the problem to be
solved). Web technology is the way to go.
My company is just by chance working on something like this for
collaboration with customers and for engineers to keep their materials. It
is closely related to the problem you posed.
Jon, do you want to collaborate with us? It is kind of open-ended research.
Thanks, Dave Ashley.
"Jon S." <firstname.lastname@example.org> wrote in message
> Hi All,
> A off-topic question/discusion:
> As Engineers we are processing a huge amount of information on a daily
> base: Reading datasheets for a future product, reading a
> marketing-memo, reading this newsgroup, writing product requirments
> Has anyone got an opinion on knowledge managegement systems to stock
> and link al the information we aquire ?
> I've found a lot of talk about KM systems, but very few is practical.
> I've been looking into blog software(with a database behind) for this
> purpose. But haven't made up my mind if this is the way to go.
> In fact, i would love to have a blog tool that is a bit more 'project'
> oriented instead of 'time' organised.
> All comments welcome..