Index Card

An index card (also called a SystemCard?) is a small rectangular piece of thin cardboard or thick paper, suitable for writing on, frequently ruled on one side or both. Popular sizes in North America include 3x5" (76x127mm) and 4x6" (102x152mm). In countries where metric paper is used (Germany in this case) probably the most common size is A7 (74x105mm), smaller ones (A8 = 52x74mm) are also used, as are bigger ones like A6 (105x148mm) or A5 (148x210mm).

Index cards can be used for many things:

Why the XP fascination with index cards? (For UserStories, I mean.) How are a bunch of pieces of paper scrawled out in my chicken scratch handwriting and susceptible to being misplaced preferable to a legible Word document (or an Access database), checked into SourceSafe, with a full revision history allowing you to see which requirements have changed and when? Because going to your cube to open up a Word document, or even to kick-start your laptop, is a soul-sucking exercise. At least for some. With a card, I can stay where I am, physically and mentally. Now, once the cards are finalized, that's another matter... but spare the mind from UglySoftware? as long as possible.

IndexCards are more durable than napkins. A method of ManagingCards is useful.

Hmm. This turned out a bit more rambling than I would ideally like, and speaking for myself only:

Word documents (and the like) are insufficiently amenable to informal manipulation and categorizing. You can't pick up a Word document. It's difficult to riffle through them and pick out a few, put them on the table temporarily and have a discussion about them. Index cards are more sharable - passing one to your colleague is easy. Re-ordering a list of Word files isn't as easy as shuffling a few index cards into order. Creating, editing or discarding an index card is lightweight. Exploratory activities ("why don't we try organizing things this way?") are easier - I don't have to check things out, I don't have to release things when it turns out "this way" was a bad idea.

You can *do* all of these with Word documents in a filesystem or requirements management system, or whatever. But not easily enough. In my experience such systems impose more of a fixed process than I want to deal with when handling lots of requirements, requests for change and so on.

The things I find difficult to organism when dealing with requirements are things like the PlanningGame, or CRC sessions. They're difficult because they involve dynamic discussions with several people, often with conflicting viewpoints. The informality and ease of use of index cards score in such situations. It's the same reason that (for me at least) when discussing something with others I draw diagrams with a pencil, not with Visio. If the diagram turns out to be worth preserving, it might make it into a computer.

Word documents or requirements management systems or bug tracking /change tracking systems store revision histories and provide protection against loss, access control and so on. I find these valuable between the "people" activities, but not so much that I want to force the people to work with a restricting system.

I've used automated systems for this that are pretty good. But when I'm planning the next release, I still start with a pencil and a piece of paper. -- PaulHudson

I've been struggling with this lately, and my concern about paper cards is that it's not as easily accessible to the whole team as something electronic. There are also issues around backups, of course. Has anyone considered using a WikiWiki-like site for keeping track of stories and tasks? I've already implemented per-project namespaces in my company's AtisWiki, just for this purpose. I'll be trying it out very soon. -- SteveWiller

Earlier this week I was visiting an XP project. We were doing an IterationPlan, and one of the stories turned out not to be applicable because of a requirements change. The customer grabbed the card, ripped it up, and chortled with the pleasure of how easy and how visceral it was to change the requirements. -- RonJeffries

These are both arguments for the tactile feedback that comes from hard-copy, which I don't deny. But neither addresses the original concern, which is with the drawbacks of a hard-copy-only approach.

Here's my viewpoint. Parts are tongue-in-cheek in order to emphasize a point. Please take it in the spirit in which it was intended.

Assume three options:

Note that I don't assume a highly formal system such as Rational's RequisitePro. I've used it, and found it somewhat constraining.

Rank each of the three alternatives according to the following criteria -

[Informal manipulation and categorizing]

[Refining requirements (unless you always get it right the first time, in which case: "we're not worthy")] [Determine which requirements have changed recently] [Concurrent access for review or update, not necessarily for the same purpose] [Chances of losing/misplacing a requirement] [Convincing "the man" (customer, management, marketing, whomever) that you're on the ball] Looking forward to the rebuttals...

All of the advantages for computerized are about technical things. All the advantages for cards are about people. Projects are about people.

Yes, and many of the purported advantages are just to make things look more important because they're computerized and more formal.

This is one of the points made in TheMythOfThePaperlessOffice -- that workplaces often shift from more efficient paper-based technologies to less efficient electronic technologies (electronic technologies can be either more or less efficient, of course) because computers symbolize The Future, Progress, and a New Way Of Doing Things. An office on the move, that's what an office that uses cutting-edge technology is. Not an office that is stuck in the past. And the employees are left to cope with the less productive, but shinier, New Way. -- ApoorvaMuralidhara

See IntroducingIndexCards for some counter-arguments to the above list.

Now when we don't live in 80's anymore, I believe the printed paper has nowadays a lot higher prestige than anything digital. I've requested people not to send me useless prints. I'd prefer reading basically everything on computer, the very same environment where I produce my own texts. For notes and scheduling I think one A4 should suffice for days. One can't do more, even worse for you if you think you can.

After a while I found out why I was the only one complaining of paper prints. I seemed to be the only one who read them through. Tons of paper wasted in a university, never read, only to make someone feel or see oneself productive. I think the printers should be ditched, or at least placed in another building or something.

My rebuttal (again, for me only): I can't debate at sufficient speed via a computer. Cards and speech are much faster for me. I have held computer mediated meetings, and they just don't work (for me and what I need to get done).

Some specific points:

Basically, I think you're assuming a level of formality of RequirementsTracking that I feel is not often needed. If it is needed, I'd probably agree with you about the need for a computerized version. But you'll have lost some fluidity in doing that, and sometimes the cure's then worse than the disease.

-- PaulHudson

A few weeks before the time of my previous story, we all got together, customers and developers. We had one of those cool Elmo projectors that will project a card on the wall.

We went through the cards one at a time, projecting it, having the customers explain it, asking questions about it, giving it priority. Repeat for risk. Repeat for estimates.

When a story was unclear, we taught them to rip it up and write a new one.

Worked like a charm. And it turns out that anyone's handwriting is easier to read on the wall than the computer fonts we had on some of the cards.

I've done it all possible ways. I hate cards. But in my experience they work best. -- RonJeffries

To me, the greatest benefit of IndexCards is that they force you to not write too much. This is a big help to those of us who are still squishing the bitter juice of BigDesignUpFront from our brains.

The expense and rarity of vellum played a similar role in MedievalArchitecture.

The key thing is KnowingWhenToStop.

Ward, if you are indeed the person who has helped all readers by reformatting this page as well ... you are my hero!

"Is there anyone among you who is wise and understanding? He is to prove it by his good life, by his good deeds performed in the meekness of wisdom." -- another eastern sage, c80 AD

I've put my experience with index cards on the ATS project in AtsCards. -- JimLittle

Just recently a team I was working with got a new Project Manager, new to the company and new to XP. I'd been urging the customer to sign off the story cards as and when they felt they were done. The new PM not only insisted in having them sign off the cards in batches, he also produced this nice form for them. In the middle of the page were the names of the cards, in a few square inches. The rest of the page was covered in fluff: names, dates, various bits of extraneous data.

Me: What's the advantage of these forms?

Him: Oh, nothing, they're just more formal.

I walked away, but I wanted to say was: what could be more formal than the customer signing off the actual document that they wrote in the first place?

An analogy from the real world: when someone buys a house (in the UK) they have a bunch of documents to sign--contract, mortgage agreement, life assurance, house and contents insurance and probably some more as well. These are formal documents and you sign each one. You are not presented with a summary document listing all of them, which you then sign.

What I like about the cards is that they are highly formal, without being bureaucratic. I've always been a fan of formal approaches to development (for example, SyntropyMethodology and CatalysisMethodology) but always struggled with all the paperwork they involved. I like XP because it has all the formalism I need without the paperwork. The cards have a big part to play in that. -- KeithBraithwaite

Read this little article about novelist and poet VladimirNabokov and his use of index cards in composition:

http://www.designobserver.com/archives/000089.html

RobertPirsig writes about his use of index cards for organizing his thoughts in Lila. You can read about it here : http://members.optusnet.com.au/~charles57/Creative/Idea_Recording/lila.htm What made them so powerful was that they too were on slips, one slip for each instruction. This meant the PROGRAM slips were random access too and could be changed and resequenced as the need arose without any difficulty. He remembered reading that John Von Neumann, an inventor of the computer, had said the single thing that makes a computer so powerful is that the program is data and can be treated like any other data. That seemed a little obscure when Phaedrus had read it but now it was making sense.

I'm surprised by the amount of amazing stuff people did with simple cards ... I'm trying LionKimbro's system now. -- EmileKroeger

An extraordinarily portable card. Cards dimensioned 3 by 5 (inches that is). They will fit in a standard shirt pocket. This size also has the beneficial effect of discouraging the practice of writing too much. With a larger card, there's a temptation to fill up the space.(plus it won't fit in a standard shirt pocket).

AmericanCulturalAssumption? What size index cards ought to be used in SI countries? :) The size that will "fit in a standard shirt pocket". Same ratio, try them before you buy them. If they fit, use them.

One of the important things about a 3x5 card is that its ratio of height to width approximates the GoldenRatio, being successive terms in the FibonacciSequence.

I live in America, but the article "International Standard Paper Sizes" by Markus Kuhn (http://www.cl.cam.ac.uk/~mgk25/iso-paper.html) makes me think that metric size "A7" ( 74 mm x 105 mm ) is the closest equivalent to 3"x5" (76 mm x 127 mm ). Every international standard paper size has the ratio of height to width of exactly sqrt(2).

I learned to write reports in high school using index cards. Do the research, capture bits of information on index cards. Arrange them, and rearrange them, in groups of themes, ideas, etc. Filter and arrange them into an outline for the written report. Write the report from that outline.

It worked then and still does, in spite of computers. -- PatrickLogan

Cards to discover requirements; software to refine and define. Iterate as you go. When a card is stable it's time to code. As you code you can also type it up and log it. But keep the card handy at all times.

<<To me, the greatest benefit of IndexCards is that they force you to not write too much. This is a big help to those of us who are still squishing the bitter juice of BigDesignUpFront from our brains. >> dead on. If it will not fit on the card you are not refined and you are likely not very cohesive either.

Marc Grundfest

IndexCard-like software: http://www.ndxcards.com/index.asp

Examples of people using digital story cards moved to DigitalStoryCards. Ramblings about not liking index cards un-restored.

http://www.scanplan.com/index1.html

See also: ManagingCards, CrcCard, StarTrekThreeByFiveCards, ShopForIndexCards, IntroducingIndexCards, ArtfullyIncompleteSpecification

CategoryCard

EditText of this page (last edited August 15, 2013) or FindPage with title or text search