It makes me feel some kind of accomplishment. It says "Well done! Let's go to next card" to me, and triggers next card.
I miss this sound when I use computer. Rather, a computer makes sound when I do something wrong (beep). Perhaps this makes big difference to my productivity. In fact, I usually use computer without sound.
The sound of indexcard attracts me to analog world so strongly.
# The only computer beep I love is startup sound of Macintosh
I name digital files using absolute
name as I do for indexcards. In the picture, I show how to make
time stamp with UNIX command to make time stamp name like " , as
example. Kevinmarsh kindly introduced me there is an application
called "TextExpander" which enable text expand "ymd" to the time
stamp. Using the absolute name of digital file, I can link the file
to an indexcard, like
ref. : PDF 2006.11.02 Thu
18:02
Same as link between indexcard, year,
month, date and week is omitted if it is made in same day : then it
become "ref. : PDF 18:02". To find a digital file on computer,
which referenced in a indexcard, search with
"20061102Thu1802".
I usually link only from digital file to card, and not from card to
digital file. It is enough for me because the card system is center
of information in my system.
- How to reduce cards : I don't know what happen if I use this
system for 2 years or more without reducing cards. Discard, or I
should keep everything. How to write, how not to write.
- Time stamp for GTD card : I have been thinking ver. 2 still
have problem. Which time stamp should be used, a time it was
written or a time it is accomplished.
- How to search a card quickly : As number of card increase, it
become more and more difficult to search a card I want. Or I should
quit search.
ref. : Writing manual from
indexcards
As this blog do, implementing the
"analog" PoI system to "digital" blog system is easily done.
Feature of the PoI system is,
- Icon and
Tag (Record, Discovery, Reference + GTD)
- Title and
contents
- Date and Time
stamp
- Chronological
order
- Link between
cards
- Revision Rule (Capability of modification)
- Capability of
Reorganization
- Star
system
- Reduce Tag to only 3-4 kinds (Record,
Discovery, Reference + GTD)
- Make title exact and short
- Write the contents piece by piece
- Allow revision/modification of the contents
(revised time should be recorded)
- Link between card and card, to external world,
by using hyperlink
- the blog should/must be private
# Wish-list and problems-to-be-solved should be discussed. Any idea?
Again, you can't connect the dots looking forward; you can only connect them looking backwards. So you have to trust that the dots will somehow connect in your future. You have to trust in something — your gut, destiny, life, karma, whatever. This approach has never let me down, and it has made all the difference in my life."
Ref. : Stanford University Commencement address by Steve
Jobs,
CEO of Apple Computer and of Pixar Animation Studios,
delivered on June 12, 2005.
For me, the indexcards I write day by day is something like the "dots" that Jobs says.
ref. : Comment from Rick
As Rick mentions in the reference
above, writing manual of the "PoI system" is an example of the PoI
system itself.
Each article corresponds to a single indexcard. Same as analog
system, I use title, small number of tags (3 = Record, Discovery,
Reference - GTD), timestamp, link, and piece by piece contents.
They are random in contents, but keep in chronological order in the
"Blog" page. Writing article
in the Blog page is kind of stress-free because I don't need worry
about the order. It is nothing but the indexcards in the Dock in
analog system. Thanks to the HTML and blog, the idea of analog
system is easily connected to the digital system.
For tracking newer contents, the "Blog" is o.k. For sharing with
people, however, it is more convenient if the articles are
reorganized and classified, as we see in a usual books. This is a
purpose of the "Selection" page. The articles are reorganized to
construct a manual of the "PoI System". Even after such
classification, it is easier to insert new article/chapter since
the articles are piece by piece like LEGO. This
manual is growing.
Perhaps we can write a book from indexcard in this way ![]()
ref. : Card Dock
ref. : Law of
accessibility
Three partitions made from wood are
attached to the Correct's Dock. I made a big single sequence in the dock
without the partition. I use index (separator made from paper) as
indicator of a month, though. As a number of card increase, I
skipped review process. Perhaps I felt difficult to break the big
single lump of cards.
I realized a "partitioning" become more important according to a
growth of system. It gives high accessibility to the system.
Different from HDD on computer, a partitioning of the dock is easy
even after constructing the system.
At this moment, I have two docks in my
room. The most frequently accessed card should be close to me
according to law of
accessibility. I made partitions like in the picture above.
From front of left dock to the end of right dock,
- Left Dock : Today / This week / Last week + Last 3
months
- Right Dock : Older than Last 3 months / Older than Last 3
months
ref. : Date and Time
ref.
: Link between
Cards
A date and time stamp is putted for all
indexcard in this system. It defines absolute name, and
plays a key role when link between cards. Then putting stamp is kind of
routine work. What about I use rubber stamp for indexcard?
In fact, I have been using a rubber
stamp for the fieldnote and other documents, but not for the indexcard. The
reason is simple : I can't find rubber stamp that match my
requirement. My requirement looks like,
2006.11.22
Wed
I think an information about
"the day of the week" is important. Because our
life is based on the week system. It is easier to remember what I
did on last sunday, than just a sequence of number like
"2006.11.19".
I have been looking for such rubber stamp for long time, but I
still can't find it (sigh). At this moment, the best is done by
hand. The hand written stamp is quick and exactly what I want.
ref. : Inertia of
writing
ref. : Avalanche writing
Every single morning, I feel difficulty
with start writing. Consciously or subconsciously, whatever. I used
to spent long time to start writing with thinking "what I should
write...". In terrible case, I couldn't write even single
card.
One day, I found there is a tip to overcome it. The answer is
simple : just start writing. Anything is o.k. All I need is just
pick up a pen and card, and start.
But same time, if I say "let's start writing", I feel a certain
resistance. Rather, I say for myself "let's start from
diary". More precisely, I start from a weather of the day,
like "Fine", "Cloudy" etc. In Japanese, it is just two or three
letters (晴れ or
くもり).
Then record what happen in the morning, say "got up 7:00" "arrive
at office 8:30", on the same card. It's not difficult, too. Just
describe without thinking. That way, I fill a first card of the
day. And it's enough to start avalanche of writing.
ref. : Noguchi Yukio, Cho Seirihou, p. 177,
translation by Hawk
Moving from 0 to 1 takes much effort than 1 to 2. Noguchi used a metaphor of "inertia" in physics to express the difficulty of the start writing.
ref. : Second draft
The Second-Draft is
originaly proposed by Josh DiMauro (see the ref. above). I have
been trying the second-draft using indexcards I wrote before to
reproduce higher and denser knowledge. This time, I used a set of
indexcards (task force) I picked up for making Keynote.
I found B6 size quadrille card from the Correct. To tell the truth,
I bought it before, a lot, for pre-PoI system. But it was slept in
a closet for long time since I switch to present 5x3 indexcard. Now
I have a way to use the B6 card ![]()
There is about forty cards in the task force. I compiled them into
about ten second-drafts. I use B6 card in not in landscape, but in
portrait, because the width is almost same as 5x3 card then ; it's
just comfort for me.
ref. : Implementation of revision
rule
I just implemented revision rule on the
PoI system. It's simple. If the card is useful, I put a mark
(star) on right-top-edge of the card. If I use a
card again, I give one more star. Then it is clear which card is
useful even after returning all cards in the Dock. After giving a
star, the used card is now safely return to the Dock.
In practice, I found that if I use red pen, I always have to take
two pens, blue and red. I think the system always must be simple.
So, I decide to use blue pen as usual for the
star, too. The star should be small mark if I think putting 2 or
more stars in future. See example below.
ref. : Strict chronological
order
The revision rule of Noguchi filing system sounds quite useful to keep
system "alive". I think there is another way to implement it into
the PoI system.
For example, when you pick up an indexcard from the dock, mark the
top edge with a color pen, say red, then return to it's original
position the dock. The order is unchanged. When you use the
indexcard again, you put one more mark. Like a star for soldiers.
Consequently, there are many mark for indexcard that you use
frequently. Then you'll clearly see which card you used before.
I must say that a chronological order
in Noguchi filing system is not a simple one. For example, in
Noguchi filing system, if one pick up a file from a shelf, it is
returned to the one side, right end for example, of the shelf. If
you use certain file frequently, it stays at the right end. Same
time, a file you don't use will slide to the other side of the
shelf. The system is always revised. File you uses frequently only
survive in the shelf. A chronological order plus this revision rule
is the key point of Noguchi filing system. The chronological order
is relatively "loose". It is possible because a
number of files in the system may not exceed a thousand ; it is in
control even if with such a revision rule.
On the other hand, a chronological order in the PoI system is more
"strict" one. For example, when I pick up a
certain card from the dock, I must return it where it was
originally. You see there is no revision rule in the Indexcardins
system. This is why I can't say that this system adopts exactly
same chronological order as Noguchi's file system. In fact, it is
difficult to change an order of cards in the system because a
number of cards easily exceed 1,000 ; a unit of information is
smaller than file. If I introduce same revision rule, the system
will be totally out of control and easily corrupt. In other words,
only the "strict" chronological order can keep a consistency and
rigidity of the PoI system.
Topics on a single indexcard should be piece by piece. Then you can make larger construction by compiling multiple indexcards into a longer and more complex thought. Josh calls this process as "second draft". He uses a larger 5x8 size indexcard for this process. Itasaka Gen, a writer of "考える技術・書く技術 (Technical thinking, technical writing)" (1973), expressed similar process as "the next step". However, I think Josh's "second draft" is more concrete and appropriate expression.
The second draft helps to polish up ones idea. You can find the second draft with 5x8 indexcard in action at Josh's photo set on flickr, "External Memory".
ref. : discussion at "Divergence, convergence" on flickr
I feel the system possess a different meaning than the beginning. Now it works as a "database". The feeling is enhanced because of the Dock where all of your indexcard gathers. Now I can use/reuse and organize/reorganize the indexcards. Preparation for presentation is an example. It was possible because I have the database. Roughly speaking, the PoI system change into database when the number of indexcard exceed 1,000.
Knowledge in one's brain is also database. For me, however, it was difficult to understand I have my own knowledge database, because it is invisible. After introducing the PoI system, my knowledge database appear in real world. No doubt I have my own knowledge database now.







