Corporate Memory and Confluence

What is Corporate Memory?

My definition of Corporate Memory(CM): A CM is the body of information or data/information/knowledge swarm that an organisation needs to manage and utilise in order to function effectively. This includes internal and external documentation, FAQs, news and external events, management reports, intranet pages, policies, strategies and procedures, and much, much more, in various data formats. Any large organisation, be it commercial, govornment or non-profit has a CM.

Annie Brooking gave a great overview of Corporate Memory and knowledge management in her 1998 Corporate Memory.

http://www.amazon.com/exec/obidos/tg/detail/-/1861522681/104-9118126-1998300?v=glance

Book description from Amazon.com

Knowledge is a corporate asset which needs to be identified, guarded and shared. Corporate Memory: Strategies for Knowledge Management shows the innovative manager how to explore its company's intangible asset, identify knowledge within his organization culture and look towards knowledge sharing.

This book will help you: identify whether your company is a 'knowledge-based company'; understand the context for knowledge management within your organization; determine the culture of your organization and determine to what extent it will support knowledge sharing; understand the ways your company generates knowledge, thus determining whether you are using it to its potential or losing it; survey, elicit and capture knowledge; design and mount a campaign to identify critical knowledge functions; design a collaborative knowledge generating venture; put together an action plan to build the right infrastructure to support knowledge sharing in your organization.

Off-the-shelf content management systems, intranets, shared document drives and static file systems are generally inadequate to store, process, search and perhaps most importantly create Corporate Memory.

Framing Confluence as a tool/framework for corporate memory / document management / knowledge base can help you get the product in to your company and give you some context and scaffolding that contribute to using Confluence effectively.

I wrote a blog post, Enterprise Wikis and CM regarding Confluence and Corporate Memory.

How are you using Confluence to address Corporate Memory in your organisation?

There is also a long rambling commentary regarding Engelbart, corporate memory and wiki engines in a KurzweilAI.net Mind-X forum topic.

quote:

Engelbart was discussing an open hypertext system long before HTML, and long before the wiki concept took hold.

http://www.kurzweilai.net/mindx/frame.html?main=/mindx/show_thread.php?rootID%3D47761

Engelbart's work on 'augmenting communal intelligence'

The NYT recently (19/09/05) published an article Intelligence in the Internet age that discusses 'augmented intelligence' and Engelbart's work.
http://www.nytimes.com/cnet/CNET_2100-11395_3-5869719.html?pagewanted=print (free registration required)

http://www.fleabyte.org/eic-3.html gives a good overview of amplifying communal intelligence as initially conceived in 1962 while Engelbart was working at Stanford Research Institute.

Engelbart is not proposing that users can become 'smarter' using communal wiki like systems. Rather, they are able to make better use of their intellects .. together. In Engelbart's words 'augmenting human intelligence' is:

"increasing the capability of a man to approach a complex problem situation, to gain comprehension to suit his particular needs, and to derive solutions to problems. Increased capability in this respect is taken to mean a mixture of the following: more-rapid comprehension, better comprehension, the possibility of gaining a useful degree of comprehension in a situation that previously was too complex, speedier solutions, better solutions, and the possibility of finding solutions to problems that before seemed insoluble.

And by complex situations we include the professional problems of diplomats, executives, social scientists, life scientists, physical scientists, attorneys, designers--whether the problem situation exists for twenty minutes or twenty years.

We do not speak of isolated clever tricks that help in particular situations. We refer to _a way of life in an integrated domain_ where hunches, cut-and-try, intangibles, and the human feel for a situation usefully co-exist with powerful concepts, streamlined terminology and notation, sophisticated methods, and high-powered electronic aids."

Engelbart's site: http://www.bootstrap.org/

quote from bootstrap.org site:

The Bootstrap Institute was conceived by Dr. Douglas C. Engelbart to further his lifelong career goal of boosting individual and organizational ability to better address problems that are complex and urgent.

Corporate memory & Visual Literacy

I've found the work of Edward Tufte in the field of visual literacy to be interesting as well when considering Engelbart's work, corporate memory and enterprise wiki engines.

Tufte coined the concepts 'chartjunk' as well as 'sparklines' - commonly used in finance.

Tufte's Personal Site: http://www.edwardtufte.com/tufte/
Tufte Books @ Amazon: http://www.amazon.com/exec/obidos/search-handle-form/ref=s_b_rs/104-9118126-1998300
Sparklines chapter from Tufte's new book: http://www.edwardtufte.com/bboard/q-and-a-fetch-msg?msg_id=0001OR&topic_id=1

Tufte's books are fascinating and beautiful.

They also make great coffee table books..

Enter labels to add to this page:
Please wait 
Looking for a label? Just start typing.
  1. Sep 06, 2005

    Mike Cannon-Brookes says:

    Nick, Fascinating reading a good post. While I was in NY recently, a customer ta...

    Nick,

    Fascinating reading - a good post. While I was in NY recently, a customer talked to me about how Confluence provided "persistent institutional memory" for his company (big). I thought it was a nice term and have been using it a lot recently.

    Really what it means is that in any large insitution / organisation, people will come and go but you need a way to keep information usable and up to date. So far, we've yet to find a great way to do this. Confluence is a small step in the right direction, but I do feel it has a long way to go yet before being 'perfect'.

    That said, it's a hell of a lot better than:

    • email - information stuck in individuals' email clients is barely searchable by themselves, let alone the broader organisation.
    • intranet - intranets are really about publishing, rarely are they about communicating. And who can publish? The few, not the many.
    • share drives - the bane of any large corporation, these inevitably end in a deep forest of folders each filled with obscurely named files. Oh, and you can't search.

    Unsurprisingly, Confluence does all 3 - and more to come!

    Thanks for the post.

    Cheers,
    Mike

    1. Sep 06, 2005

      Nick says:

      Mike, Thanks for your feedback. In hindsight I too have decided that Confluence/...

      Mike,

      Thanks for your feedback.

      In hindsight I too have decided that Confluence/wikis do not make up the complete solution for corporate memory - perhaps I should update the post. However..

      Watch my blog. There will be some very interesting posts on it within the next month.

      I'm really angry and I've decided that the pen is mighter than the sword. >_<)

      Did you read the "Old School vs. New School" post? I've had some really great feedback on that.

      Cheers

      Nick

  2. Sep 11, 2005

    John Hubert says:

    Mike, I think you identified exactly the three key areas where Confluence should...

    Mike,

    I think you identified exactly the three key areas where Confluence should really focus (email, intranet, and shared files).

    You've got the intranet part nailed. Confluence is a great wiki! But no matter how easy it is for everyone to publish and update web pages, email and shared files will continue to comprise a large percentage of the Corporate Memory for a long time to come, and Confluence support for files and emails is still very limited.

    I'd like to see document vaults and some basic document management capabilities. If you could allow users to drag and drop files into Confluence (via webdav) and to open/save files directly from applications (instead of upload/download), that would be huge! And of course everything would be fully indexed/searchable, easy to link to, comment on, etc. Maybe some basic version control and locking/checkout, etc. You can do a lot with webdav (maybe even make it so that moving and renaming don't break links, etc.). Have you seen Intranets.com?

    As for email, I think the biggest help there would be to make it more bi-directional - add more functionality to send email from Confluence. For example, I should be able to send an email from any page to start a discussion thread that goes back and forth in email but is logged to the page (and to the space I suppose). Similarly, I'd like to be able to easily send email with links to files in Confluence (to solve the dreaded problem of so many files being sent around through email and everyone keeps their own private copies). It's easier to bring Confluence to the people (their mailbox) than to bring the people to Confluence

    Hope that helps. Keep up the great work!

    John

  3. Sep 12, 2005

    Guy Fraser says:

    Having recently setup a small IT company in the UK, I thought it might be worh p...

    Having recently set-up a small IT company in the UK, I thought it might be worh posting some notes on this subject from the other end of the scale in case it's of any use to other small companies out there.

    Confluence allows our company to exist. It's that simple. There are 3 of us (full time) and we all work from home to keep overheads down - not having to find cash for an office every month - it's heating and maintenance, all the extra paperwork and legal stuff, the tax for the car parking outside and not to mention all the petrol costs - is essential to our survival. It's Confluence that allows our team to store knowledge in a centrally accessible place - no matter where we are, we have instant access (assuming there's an Internet connection) to all our company information. It's become so essential to us, that we simply can't imagine any company existing without Confluence!

    Here's some of the things we rely on Confluence for:

    • Our website (currently locked at the moment while we make some changes to our product portfolio) is powered by Confluence. We've themed it to within an inch of it's life and having deeloped websites and online applications (both HTML and Flash, including data-driven stuff for both) for the last 8 years of my life, I can assure you that a themed confluence is a perfect tool for building a website!
    • Our private "thinktank" intranet that stores all our product ideas and other confidential inforamtion is also powered by Confluence - each time we have a meeting or come up with an idea, etc., a new page appears in there, linked to all other relevant places.
    • Our corporate intranet - the glossy version of our intranet - stores information such as ISO 17799 policies, employee handbook, document templates, etc. Should we get to the point where we are able to take on more staff, there will be a fully populated corporate intranet waiting for them there.

    Because we're small, we're in that usual start-up situation where all staff have to know pretty much everything - if a client calls then whoever takes the call needs to know everything about that client even if they've never spoken to them before. Without Confluence, that simply wouldn't happen. Period. We've looked at loads of other products over the years and Confluence is the only one that really hits the nail on the head in terms of delivering exactly what we need.

    As we're small enough not to be hindered by large scale document storage and email management, Confluence is absolutely perfect for our company. It's the key tool that allows this small team of programmers working from home in different areas of the UK to stick together cohesively and effortlessly.

    Because we've been using Confluence and been working from home since our company was started, our reliance on the product has been extensive. Whereas many established organisations (such as the company we all used to work for) who implement Confluence may have to get in to a new mindset of storing information in there, we've been doing it by default so it should provide an interesting insight in to the effects it has on our business both in the short and long term.

    1. Sep 20, 2005

      Nick says:

      If you're working remotely and/or in a knowledge industry with yourself n other ...

      If you're working remotely and/or in a knowledge industry with yourself + n other users I think Confluence will be very useful.

      I've mentioned Confluence to a couple of small retailers (5-15 staff) that I know as a tool to create procedures/FAQs etc.

      They were quite enthusiastic about the idea - not sure if they ever bought Confluence however!

      One thing I wonder about offline 'road-warrior' Confluence usage accomodation.

      'Personal wikis' are an interesting concept.

      Tiddlywiki is a very interesting personal wiki:

      http://tiddlywiki.com/

      it even supports Sparklines.

      All the Tiddlywiki engine code plus the content you create is kept in one HTML file. You need a recent Browser but that's it - start adding content and link to external documents and images. No refactoring though I think.

      However, you can fit a lot of stuff on a 512Mb USB thumb drive, and access it wherever you find a computer with USB and a browser.

      Reminds me of that quote about station wagons with magnetic tape for some reason.

      1. Sep 24, 2005

        Guy Fraser says:

        As a company, we like Confluence so much because we can make it look like this (...

        As a company, we like Confluence so much because we can make it look like this (due to the vast amount of extensibility features such as themes and macros):


        I've looked at TiddlyWiki and it just doesn't come close to what we are after.

        1. Sep 24, 2005

          Nick says:

          Guy, I'm not saying that Tiddlywiki is a substitute for Confluence at all, sorr...

          Guy,

          I'm not saying that Tiddlywiki is a substitute for Confluence at all, sorry if I gave that impression..

          That would be a like comparing a handheld computer with a desktop PC, or as Palm put it in the Zen of Palm, trying to make a gorilla fly like an eagle. It can't happen - furthermore, why would you even try?

          They are two distinct things, both good for what they are designed for.

          I guess the point I was trying to make was this - Confluence is fantastic when the user is connected to their intranet or Internet but if the user is going on the road or is without network connectivity Confluence starts to sag a little IMHO.. please correct me if I'm wrong.

          Nice work on the look and feel of your Confluence instance BTW.

          Cheers

          Nick

          1. Sep 25, 2005

            Guy Fraser says:

            Ah sorry, I thought you were reccommending using Ticklywiki for people conneted ...

            Ah sorry, I thought you were reccommending using Ticklywiki for people conneted to the to Internet. I can see your point that such a product is perhaps better suited for people who travel around a lot. That being said, Confluence runs quite happily on a laptop so I imagine a team who travel a lot could buy multiple licenses (or maybe have personal licenses?) and each have a space with which they could import/export to effectively manually synchronise.

            I wonder if a plugin could be made to synchronise multiple spaces on multiple Confluence installs - then a personal install could be used on a laptop and they could synchronise with the central "Hive" commercial install when they are next online...

            1. Sep 25, 2005

              Nick says:

              I've been experimenting with the 'personal wiki' idea for a little while now. I'...

              I've been experimenting with the 'personal wiki' idea for a little while now. I've found it great for doing research as you can effectively make a 'mind map' of the material/notes you've gathered. That said, Tiddly has a clunky interface.

              I first became aware of Tiddly through Getting Things Done etc. If you're a GTD fan/follower you might find the GTD tiddlywiki hack interesting.
              http://shared.snapgrid.com/gtd_tiddlywiki.html

              Confluence on a laptop? Cool - I should try that.

              However, I have to admit I have some reservations with the idea of constantly running a J2EE app server and DBMS on my laptop - seems like a lot of overhead.. maybe I just need
              a new laptop.
              For synchronization, perhaps CVS/Subversion VCS could be used? You could write a script that would regularly 'dump' the content of the Confluence DB as text/SQL, and check it in to a Subversion project/folder.. Max the road-warrior could then check out the DB via a script to refresh/synch his laptop Confluence instance.
              After being on the road and making changes to the content he'd then need to return and check-in his changes, and get whatever changes had been made since he'd been gone.

              I'm 100% sure it wouldn't be that simple though, unfortunately..

              Optimistic 'page locking' will probably lead to some pain. VCS merge conflicts - say Max the road-warrior edits the page on the road and Sandy the desk-jockey back at the office does as well. When he comes back to the office he may need to manually fix the SQL to combine both his and Sandy's changes - a task that is often not that enjoyable for developers let alone users..

              CVS is useless for automated merging, and in my so far limited experience of Subversion its auto-merge functionality isn't all that either. However, saving the content of the DB in a binary format and checking that in and out of the VCS may be a solution..

              Secondly, for Max you'd only want to extract the SQL statements that would reproduce the changes - not the 6 million lines of redundant insert statements for pages/data that hasn't changed. That way Max could quickly synch over the Internet.

              You'd probably need to grab all of the attached files as well - I haven't looked at the internal architecture of Confluence in that much detail yet - I'm guessing that perhaps attached files are renamed using a hash algorithm (for multiple files with the same filename?) and saved in a directory somewhere? A directory for each space/page?

              Forgive me my laziness in making these random assumptions.

              Actually, now that I think about the problem of 'hot-synching' a master/slave (actually more like older brother/younger brother perhaps) Confluence instances I realize I can't offer much advice without trying to actually implement it. I'm planning on heading back to university to do post-grad wiki science research next year so perhaps that could be part of my project.

              In any case, it'd be nice to hear your thoughts on this as I think its a valid problem. As a real-world example, consider a nuclear submarine that is submerged for days/weeks in enemy territory and pops its antenna up to the surface occasionally to 'synch' its Confluence DB with the Pentagon master DB?

              1. Sep 25, 2005

                Guy Fraser says:

                Well, Confluence already keeps track of all versions or pages, attachments, etc....

                Well, Confluence already keeps track of all versions or pages, attachments, etc...

                As such, that should at least deal with working out what's changed since the last update in terms of page versions (avoids any time based issues such as server clock not being set the same as the laptop clock, etc).

                1. Sep 25, 2005

                  Nick says:

                  Of course. Forgot about that completely. Is there an RSS/XML changesummary feed...

                  Of course. Forgot about that completely.

                  Is there an RSS/XML change-summary feed that can be emitted for each space/page?

                  A road-warrior client app could poll for the existance of a change, then DL the actual changes.

                  Max might decide to only synch certain spaces or pages as well - a good way to tackle the volume of changes from a large organization's instance of Confluence IMHO.

                  Forgive me the silly names I'm using - Max the Road-Warrior!? ROFL
                  But I'm a fan of the 'persona' concept in usability/HCI, and it's easier to write 'Max' rather than 'the road-warrior user'.. heh

                  1. Sep 25, 2005

                    Guy Fraser says:

                    There are RSS feeds (and very excellent ones in Confluence 5.x

                    There are RSS feeds (and very excellent ones in Confluence 5.x) for spaces, etc. In addition, there is a full XML API (RPC based from memory).

                    Being able to limit it to certain spaces would indeed be a dream.

                    Even before any form of online/offline synchonisation was added, having something like this to synchronise online Confluence web apps would be a dream - think "live backup"

                    1. Sep 25, 2005

                      Nick says:

                      Mmmm.. live backups. At a previous employer I was involved in some of what they...

                      Mmmm.. live backups.

                      At a previous employer I was involved in some of what they called BCS (business critical services or business continuity services) strategy and implementation, which was quite interesting.

                      It certainly wasn't a subject I'd ever bothered with in the past, but I would now generally regard risk management / disaster recovery planning as essential for anyone doing anything 'serious' with the computers.

                      Couldn't you get live backups by replicating your database using your DBMS of choice - assuming attachments are in the DB as well

                      I think Oracle is much better at this than MySQL, but I haven't done the research to back up that claim. It's more of a hunch. heh