• extract the exact part of the transcript corresponding to one chat
• hierarchize the chat by highlighting moderator tweets
• hightlight questions
• hightlight answers corresponding to questions

• New ideas: Things I absolutely didn’t know.
• Eye opener: Contradicting ideas to things I tought but with little ground.
• Revealed. Ideas expressing ideas I had may be but never explicited.
• Well said: Better expression of I deas I had.
• Aha: A whole concept I suddendly become acquainted opening a new range of possibilities not necessarily present in the tweet.
• To Dig: I didn’t catch or agree but can see some value in it and need to think again or do some investigation.

Instead of playing with a real production server I prepared a minimal virtual machine.

You can download it and try by yourself. It should be less than 10 minutes.

1. Download VMWare Player from http://www.vmware.com/products/player/ install it. Use Fusion for MAC users.
2. Download the image (about 500 megs) from http://kneaver.com/pdownload/Kneaver-CentOS.rar
3. Unzip the image in a directory with at least 2Gb of free space
4. Open and start the virtual machine using VMWare player
5. When the machine is up and ready it will display its ip address. It will be something like 192.168.130.133

1. Point your browser on “http://the-ip-address/tt-rss”
2. Log on with user “admin” and password “password”.
3. Enjoy

See http://tt-rss.org for Tiny Tiny RSS information. Thank you to them

I was using Google reader like many others but as a temporary solution because I have the following requirements

• I follow a lot of feeds, aggregation must be done outside of my computer.
• I want to be able to access them anytime, from anywhere, including from someone else computer.
• Some feeds are coming from applications, they need a login and some confidentiality.
• I wish to have social media buttons on news to be able to reuse them.
• I will favor I can extend easily on need.
• I don’t want to be tracked and have ads around.

The original plan was to use this Virtual machine, tailored for minimum size, to install a copy of Kneaver so that evaluators could try it outside from the cloud. Knowledge is typically pretty private and this removes a hurdle for some. The need to find a quick fix for RSS changed my agenda by a few days.

So the list of software installed on the image will grow with the focus of serving needs like Market watch, Tech watch, Knowledge management. Any suggestions ?

It’s mostly mean for people with limited time for evaluation, the real benefit of a server installation is to be accessible from anywhere and running at all time and this image don’t serve it.

There is a readme.txt in the package with technical details

Customizing a single item

I am on the process of collecting my knowledge on KM itself. For this I follow our hint “#5 List Knowledge Sources”. I create an item, set its type as “KNV:Container” and start adding name of people there : Bloggers, Writers. When I use Kneaver I am following my associations freely. I read a blog post and I will add the name of the blogger, the address of the blog. It is in the middle of this task that I added an “Organizations [in KM]” container and linked “Peoples [in KM]” to “Organizations [in KM]“.

That’s where I realized the need to sort the list of people automatically and list first the people before any other link, otherwise links to Organizations would be right in the middle.

I used The template gallery to find a solution.

It’s easy, suffice to place a template call in the body of the container to list people. Since items will not be listed twice this will cause people to be listed before anything else.

This is the template call in Kneaver Syntax. {{ }} will surround any template call similar to [[ ]] surrounding a wiki link. | separates template arguments.

{{ KNVItemLinksByTypeOrderByGroupBy knv:persontype | Name | | }}

This is the dialog to edit the “people [in KM]” topic:

This is the result

Now I thought that it would be the same for containers of people in Software design, in Marketing. This is where we realize the benefit of introducing a new type. If you can easily find a name to characterize several items and they are going to share some behaviors or could be used in similar ways it’s better to make a type. The Type name will be precisely this name you found. In my case I choose “Yellow Page” and the type is “KNV:YellowPageType”. It is prefixed by “KNV”, this is the Kneaver namespace because I plan to eventually move it into an application so that every Kneaver use will find it. Applications can be shared among Kneaver sites. You can make applications as well.

At this stage we understand the need and the benefits of types and we wish to put it in practice.

How to practically create a new type

As simple as creating a new item.

1. Go to the dashboard
   1. Select the tab “Types” in the right widget.
   2. Scroll to the type you want to specialize, likely the type of the object you want to generalize. In our case “KNV:ContainerType”
      
   3. Click on the “+” icon. On older version of Kneaver you may have to right click and select “Open in new window”.
2. Now you are on the “Kneave” Dialog to create an item.
   
   1. Make sure the type is “KNV:TypeType
   2. Make sure the Link type is “KNV:InheritanceLinkType”8. Press “Create Item” button
3. Now you are in the “Add Item” Dialog
   
4. Done, the new type can be used instantly

Go back to the item whose type you want to change, use the combo and select the type you just created

Now change will occur on the display, we have a new type but no behavior attached to it.

Let’s define a template for it.

Defining a template for a type

There is a template file already defined for each project:

Open it with a text editor (some environments allow to edit the file remotely). It is a kind of php/html file. Scroll to end, before </body>.

We are overloading the template call KNVItemLinks called when Item links are displayed. the best is to locate it from the existing one and copy it.

and add “.KNV:YellowPageType” at the end of the name (replace KNV:YellowPageType by your new type name).

Let’s save the file and refresh the item in the browser. As usual effect is immediate but since the new template is the same as the existing one nothing can be seen, as before the direct call to links in the body has priority.

How can we see the difference ?

1. Remove the template call in the item.It is not necessary any more.
   
2. Modify slightly the template in the file
   

Refresh, it’s done

Got it.

What’s next

Many more things can be done with types: assign different styles (Types maps to css classes), have different shapes in maps, different layouts, different toolbars, different dialogs. Types can be created to ease display, managing items but also to organize, classify knowledge. In fact all Kneaver is build like this, Kneaver is used to define itself.

We have seen how to call templates directly in individual items, than how to create a type and finally how to surcharge a template for this type. This all is done in a few clicks and is extremely powerful for formalizing even more your Knowledge. It is a practical approach to start to define everything as topics but when you are advanced you see the benefit of adding types related to your activity. That’s where you appreciate that Kneaver is extensible.

You reach the page from Dashboard=>Plus=>Audit All items. It is a heavy page and may take some time to load.

The first use of the page will display a page of items Items are ordered by POKId, this don’t make much sense for end user but has the benefit of being stable and predictable.

Selection check boxes allow selection of individual items. In both header and footer areas a check box allows selection or deselection of all items at once.

You can access and select a set of items directly by typing the list in the selection input zone. The list will be obtained either by support or from reviewing faulty items that need to be fixed. On January 2013 a typical list is a list of number, separated by commas with no spaces and no commas at end and at beginning. This list is also updated when you click on the selection check boxes. An example of the use of direct list is to collect a list of faulty item.

A very convenient use of [[Audit All Items]] is to cleanup globally a database. Say you copied a database because you want to definitely separate two domains and see like this a group database. This will require some cleanup. Instead of traversing all items, pages after pages you have here a way to have a complete view at glance and remove all what is not relevant to the new usage.

Another interesting use if to uniformize titles, jargon. Often several variations have been tried over the time and the ability to review them all in one page is valuable.

This feature will evolve based on demands. there are plenty of possibilities not yet explored.

• You get to an interesting page on the web ? click on the paper clip icon. It will save the bookmark, the contents of the page and a screen shot. Like this you are prepared to all cases including a disappearing page.
• You received an interesting mail among the huge flow of incoming mail. Just clic the paper clip icon in Outlook and it will be referenced.
• You received a paper, took some notes, draw a schema ? A single keypress will scan the document with the proper settings and save it in a well know place with a unique file name. I regularly use it to scan what ever is worth some interest and to virtualize my office. All incoming or outgoing mails is scanned first. Even at home it’s now a accepted rule. It’s very helpful to know that even in holidays it just takes to have an internet connection to be able to send again this precious document  from high school to register your kid for college. This is paperless life. For me it is a prerequisite to productivity and peace of mind.
• When the volume of paper is way to large and papers are too large like newspapers, paperboard the other option is just to take photographs with your digital camera. A 5 megapixels camera will take a picture of a newspaper fine enough to read the text. A cell phone camera is sufficient to keep track of an ad or a signpost in the street. It even happens to me to use the camera while I am at my computer because for some reason the computer is engaged in another activity.  Just be careful to have enough light and no shadow. Of course all this is done exclusively for your own personal reference.
• I will usually download videos and trainings so that I can follow them again without necessarly be connected to the Internet. It’s also because often I want to check a sequence.
• While running a program something catch my attention, I want to kee track of it. Screen Shot is the solution. This can be for example while using a CAD Software like SolidWork or SolidEdge. They cannot be run by everyone at once. You want to capture a precise situation with some dialogs open: is not just the part you want. What is nice with Kneaver Tray having the tools in the same place, feeding into a single location, sorted by date and time is that it becomes easy to follow the devlopment of our thoughts even weeks after.
• If screen shot is not enought because you want to capture a movement, a quick sequence of events you can capture a video of your screen as well. While take a rough cut like this your mike will be on as well and you can comment directly.
• When I am tired of writing or already engaged in writing like now I will appreciate the ability to take recordings at once. No need to adjust the mike, select a file or think about cutting the record. It will start with one click and stop at the first long silence. The drawback of recording is that it’s very difficult to index and scan. Since you can take small notes after or durig the recording the notes will help you remember what it was all about. You don’t need to be an expert to use this feature. It’s one click away.
• If recording is not enough to make your point use the CAMCorder option. Both you and your voice will be recorder. Nice to explain something about tacit knowledge, know how. Again, specially with the new laptops including a camera in the lid it’s a breeze.
• Finally need also to take notes, quick, long, successives. A small popup window is ready for that. Always available. You can paste HTML and rich text in it. I use it a lot to keep some fragments of mails or pages. I read some news and I learn about a new technique, a resource person, I keep track like this. Back to GTD I will not use this to keep track of activable items (Tasks, ToDo lists) unless they are mere suggestions or of the someday class. The system allow only a single note to be open. If you have several ideas n your mind create one note for each, one after the other. If a tought is continuated create a new note don’t try to keep ideas grouped in the same place. At this stage it more like a journal and the objective is to have all of them jotted down. The sooner, the faster the better. We have in mind that you have toughts in the middle of another job and you want to stay focused on your tasks, not disrupt it. This is called “in the flow” capture. You don’t stopyour work you just enough to restart this exploration. Surely don’t finalize or take care of grammar or layout when taking notes. Anyway the capabilities are minimal on purpose.
• Any feature you feel it could complement Kneaver Tray ? Just let us know. It’s very easy to extent it and open. It will be further more completed so to always stay current with the technology.

Twitter announced late september that they will stop supporting XML and RSS on March 2013 and exclusively use JSON. Many social network providers will likely follow this move. This means that systems like Kneaver able to access, consumme and make sense of Social Graph and Social Media streams needs to switch to JSON super fast. See how Dave Winer inventor of RSS embraces the change.

We at Kneaver decided to upgrade RSXML our core component support in XML but also data marshalling. We added directly JSON support to it, both read and write. This was done with some favours to adapt to the style used by twitter itself. This means that all higher level components will now use JSON data transparently. This means also that Kneaver became JSON capable for every feature. Whereever XML was used it now possible to use both XML and JSON. This makes it easy and performant to build widgets using Kneaver data and place them anywhere.

Surely we somewhat forced to do it, but now this is brings extended openess to our software. Good deal.

As opposed to services like ifft which are centralized we don’t bear the limitations imposed by twitter. Each Kneaver site has to register as an application with twitter and is managed by you directly. This is similar to the way some wordpress twitter addins are operating. You will continue to be able to have rules based on twitter streams unless twitter change again their TOS to be more restricitive.

Humans have this wonderful tool for articulate their thoughts : the language. While some class of words are closed for expansion, nouns and verbs can be created at any time. New nouns act like milestones as we progress in discovering new concepts. This can be done on a large population with the benefit of becoming understood by a wide audience (Branding is a special case of that). This can also be done by a group of specialists and it increase their jargon. It can also be done individually or in a small group although social conventions discourage us to venture on this way (Tour of Babel effect, see http://c2.com/cgi/wiki?TowerOfBabel).

With a rich vocabulary texts can be shortened. I gave in another text the example of the recipe of samosa in its minimized form:

Now if you don’t know what is a [potato massala filing] or a [flour dough] this is unusable for you.

This is typical of texts written by specialists for specialists. They are stuffed with jargon. To prevent cross language losses in translation or names clashes they even chose to name things in Latin compound names. This increases the hermetic of it for non-specialists.

Nowadays we have great tools for that. We use hyperlinks and tooltip instead of reading paper and continuously search in a big dictionary on our side. Imagine that each jargon term or compound noun is linked to a page with the definition displayed when we hover the mouse on it. It becomes a viable option to be both jargon intensive and understandable by non specialists.

We even have a new capacity now. Each concept name can serve as a transfer airlock to every other topics using this concept. Since now we are not anymore reasoning on individual nouns but on compound of nouns we limit the ambiguity.

The need to be able to add new terms on a continous basis and this ability to explore by names led to the invention of Wiki. That happened at the Wards Wiki (see http://c2.com/cgi/wiki?WardsWiki). This was long before Wikipedia pick up the concept. While Wikipedia is an impressive experience it is not ongoing Knowledge Creation. We lost the reference to the original concept in c2.com which is a unique experience of a group of people collaboratively building new knowledge, naming concepts on the fly. The Wards Wiki is a very lightweight true Knowledge Management System.

At that time I was marvelled to see new names appearing every days with improbable shapes like “TheRoadNotTraveled”. The idea was that as soon as something emerged a new name was created by glueing all the nouns into one new term. This was again a brick on which more progress could be done faster.

The benefit was also to reduce ambiguity, redundancy of repeating again and again definition by paraphrases and the ability to nail down collectively differences, variations, opposition of concepts. When the same concept was encountered again the same agreed name was used. This cause looping trails of thoughts to become apparent instead of continuous branching.

Tacit Knowledge benefits from non verbal communication, direct person to person communication , Explicit Knowledge benefits from a precise vocabulary. Both Knowledges should eventually share the same vocabulary in indexes, searches and messages. Having a rich, precise language makes Knowledge easier to find.

Free yourself to put new names on what you discover. If some names turn out to be superfluous or already existing it will be easier to link them and at least your ardor will not be stopped by lack of terms.

This is a very common use of Kneaver. Create a framework of terms with very minimal definition on them but link them massively to related term. You can link terms by opposition, similar sound or similarity of meaning. Kneaver supports Link Types for this purpose. We call this Weaving Knowledge. New terms can be created any time just by grouping nouns with a double square bracket, like on c2.com.

The difference is that Kneaver supports semantic and semantic linking.  The reason why you link is known and can be used for searches or better aggregating. For example opposite terms will be retrieved but not their related terms, same for consonant terms.

When someone (it could be you after this long holiday) will search again for this idea, you will be guided to this term either by its definition or by a nearby term and you will eventually find it again by navigation.