Recap: Chicago IxDA's Pattern Library conversation

16 Oct 2007 - 12:22am
6 years ago
12 replies
966 reads
Janna DeVylder
2006

I think all who attended would agree that the first Chicago IxDA event last
Wednesday night was a great success! Many thanks to Manifest Digital for
graciously hosting the event and opening up their wonderful space to us, and
thanks to those who attended and shared your perspective with us. Please
save the date, November 14th, 6:30 pm, location TBD, where we'll be
discussing lessons learned around designing RIAs.

For those who were unable to attend on Wednesday, but are curious
nonetheless, here's a recap:

The topic was pattern libraries, and the format was participatory...and
participate we did.

We started by making a distinction between public libraries (which we all
refer to and appreciate) and home-grown libraries (ones that are created and
utilized within an agency, corporation or solo practice, and are rarely
shared or shown to the public...kudos and thanks to Yahoo for such a rare
resource!).

We seemed to agree that one could belabor the semantics of what a pattern is
and what a pattern library should be. Questions such as "Are we confusing
pattern with component, pattern library with style guides?" or "Is a
lightbox a pattern or a solution, or is that one and the same?" frequently
arose. It was clear that the creators and utilizers of the home-grown
library must understand its utility and intent, first and foremost.
The impact of bastardizing the term pattern and pattern library on the
broader community is a topic for another event.

One participant aptly pointed out that most of us have our own pattern
libraries stuck in our heads or in our gazillion Visio files... we are
approached with a familiar problem and we often utilize our previous
thinking and solutions to address it. While this makes us individually
consistent, absolutely saves us time, and allows us to spend our energy
innovating on the new challenges on hand, the big burning questions for most
in the room were:
"How do we have a group of people come to a consensus on what should
constitute a pattern?"
"How do we justify the time spent in creating the resource?"
"Does this need to be tied back to code to be efficient?"
"How do we institutionalize its use? Here you create this thing... does it
die the minute you look the other way?"
"Should an agency have one? How would that work across clients? Could it be
high-level enough to be useful?"

It's clear that people are interested in this, but it feels like we want to
see its utility proven out past just the creation of the library. I would
love to hear about the successes and challenges Yahoo has faced with their
non-public facing library. Sounds like a great conference topic to me!

thanks again, and see you in November!

Janna

PS. We're working through how Chicago and other local groups will post and
share event outcomes in a more public way, so once we have something stable
in place, we will share resources there.

Comments

16 Oct 2007 - 10:47am
Christian Crumlish
2006

Thanks for this recap, Janna! I knew I was going to regret not being
able to be there.

Your summary inspired a blog post:
<http://xianlandia.com/te-amo/2007/10/16/do_pattern_libraries_really_work.html>

In it I note your appeal:

> I would love to hear about the successes and challenges Yahoo has
> faced with their non-public facing library. Sounds like a great conference
> topic to me!

By saying that I have proposed to discuss this very topic (and
related) matters at a number of upcoming conferences, so I hope to be
able to share what we've learned (both good and bad) in the near
future.

We've also been thinking about hosting a "Pattern Summit" here at
Yahoo! probably sometime next summer, to invite interested parties to
our campus where we can discuss what we've learned, share insights
from the YPL and YUI library, and run some workshops for pattern
authors.

x^'~

--
Christian Crumlish http://xianlandia.com
Yahoo! pattern detective http://developer.yahoo.com/ypatterns
IA Institute director of technology http://iainstitute.org

16 Oct 2007 - 1:22pm
Dave Malouf
2005

Since I'm about to give a talk on this topic on Friday, it has been
on my mind a lot. I saw Chris' post on facebook questioning the very
value of Patterns themselves and I often question them too.

Question: Has anyone spoken to people in the Architecture world and
about their thoughts on Patterns. We seem to admire Alexander, but do
they? And how are they using them if at all?

I guess when it doubt go back to the source, no?

I do know there are architects in this community maybe they can
enlighten us a bit.

-- dave

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=21539

16 Oct 2007 - 1:43pm
Christian Crumlish
2006

On Tue, 16 Oct 2007 11:22:08, David Malouf <dave at ixda.org> wrote:
> Since I'm about to give a talk on this topic on Friday, it has been
> on my mind a lot. I saw Chris' post on facebook questioning the very
> value of Patterns themselves and I often question them too.
>

Wait, so why are you giving a talk on something you don't believe in?
I'm confused.

Do you have a link to Chris's facebook question? I seem to have missed it.

> Question: Has anyone spoken to people in the Architecture world and
> about their thoughts on Patterns. We seem to admire Alexander, but do
> they? And how are they using them if at all?

Yes, I have. He is not popular among most architects. His designs are
not fashionable now, for example, and they don't generally use his
patterns. He is a prophet without honor in his own land, so to speak,
although some architects I've spoken to do like the idea of patterns
but don't like his specific patterns.

When you say "we seem to admire Alexander" what does that mean
exactly? Who are we are what does it mean to "seem to admire" someone?

> I guess when it doubt go back to the source, no?

No :^)

--xian

--
Christian Crumlish http://xianlandia.com
Yahoo! pattern detective http://developer.yahoo.com/ypatterns
IA Institute director of technology http://iainstitute.org

18 Oct 2007 - 7:52am
Jonas Löwgren
2003

The discussion so far seems to center on the capturing-best-practice
interpretation of patterns, which is certainly the most established
one.

But I tend to find more potential in the notion of inspirational
patterns. Mini-lecture as follows:

The way I see it, design combines many activities but the two
backbones throughout most design processes are to Create and to
Assess. To Create means to generate ideas (can be new ideas, can be
variations, can be reframings, ... doesn't matter for this
discussion). To Assess means to judge the merits of those ideas.

In order to Create, the designer needs a repertoire of formats or
exemplars that he/she matches against the situation at hand. A broad
repertoire means better chances of coming up with ideas, many ideas,
varied ideas.

And this is where inspirational patterns come in. It is possible for
experienced designers to capture their experience from working in a
certain genre by abstracting key ideas representing points in the
space of design possibilities. These abstractions can be called
inspirational patterns, and they can be communicated to other
designers (who want to learn more about the genre at hand) using more
or less structured templates. The other designers can extend their
repertoires by studying the inspirational patterns and thus Create
better.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=21539

18 Oct 2007 - 4:15pm
Christopher Fahey
2005

On Oct 18, 2007, at 5:52 AM, Jonas Löwgren wrote:
> ... I tend to find more potential in the notion of inspirational
> patterns. ...
> In order to Create, the designer needs a repertoire of formats or
> exemplars that he/she matches against the situation at hand. A broad
> repertoire means better chances of coming up with ideas, many ideas,
> varied ideas.... designers can extend their
> repertoires by studying the inspirational patterns and thus Create
> better.

Yes! This is pretty much what I meant when I described patterns as
"languages" earlier in this thread.

You don't open a dictionary every time you want to say something with
words -- you just say it because you are presumably literate in the
verbal or written language. Likewise for interaction design patterns
-- you become literate in the full vocabulary of the design pattern
language, you immerse yourself in the "literature" (the body of work,
for example the web), and then when you sit down to design most of
your challenge is similar to the challenge of being a writer: good
articulation of a good idea.

-Cf

Christopher Fahey
____________________________
Behavior
biz: http://www.behaviordesign.com
me: http://www.graphpaper.com

18 Oct 2007 - 7:08pm
Christian Crumlish
2006

Now, just to be a bit contrary, I will also note that some
organizations use their pattern libraries as a form of style guide, to
impose consistency around certain interactions and flows. Whether this
is the "correct" use of a pattern language or not, or a good idea or
not, it is one application of the pattern idea.

On 10/18/07, Christopher Fahey <chris.fahey at behaviordesign.com> wrote:
> Likewise for interaction design patterns
> -- you become literate in the full vocabulary of the design pattern
> language, you immerse yourself in the "literature" (the body of work,
> for example the web), and then when you sit down to design most of
> your challenge is similar to the challenge of being a writer: good
> articulation of a good idea.

So in this analogy, a professional writer may keep a copy of the
publication's house style guide handy to clarify whether to write
homepage or home page.

-xian-

--
Christian Crumlish http://xianlandia.com
Yahoo! pattern detective http://developer.yahoo.com/ypatterns
IA Institute director of technology http://iainstitute.org

18 Oct 2007 - 7:10pm
Christian Crumlish
2006

On Thu, 18 Oct 2007 05:52:02, Jonas Löwgren <jonas.lowgren at mah.se> wrote:
> And this is where inspirational patterns come in. It is possible for
> experienced designers to capture their experience from working in a
> certain genre by abstracting key ideas representing points in the
> space of design possibilities. These abstractions can be called
> inspirational patterns, and they can be communicated to other
> designers (who want to learn more about the genre at hand) using more
> or less structured templates. The other designers can extend their
> repertoires by studying the inspirational patterns and thus Create
> better.

BTW, I tend to think of this as being about high level, conceptual,
"principle"-oriented patterns, somewhat similar to the idea of process
patterns. The software design (engineering) pattern community has also
spent a lot of time talking about which sort of pattern is appropriate
for what.

Personally, I love the idea of capturing high-order patterns that
relate to principles (and what Alexander thought of as balancing or
resolving "forces") but in my day-to-day job I also know that
interaction designers want to know whether or it's a good idea to put
arrow buttons at either end of a carousel as the primary way of
scrolling back and forth, and whether the items should load up a page
at a time or individually.

(The answer? "It depends.")

x'^~

18 Oct 2007 - 9:31pm
Wesley Hall
2007

Hi all,

I've been following the themes that have arisen in this discussion with
great interest. They've all cropped up at my office.

At my company, Leapfrog toys, we discuss and use interactive patterns
constantly. They're integral to how we work. I honestly don't think we
could get our work done and ship on time without them. Time/Money/Resource
constraints necessitate patterns.

In the best use cases, patterns help us compromise between art/efficiency
and create something delightful, on time and on budget.
In the worst use cases, patterns create toys with no soul.

Seems like that's the struggle many are dealing with?

I've found the following to be true about patterns. Do you all agree?

1. Patterns improve efficiency in design documentation.
2. Definition of a "Pattern" is debatable
3. Copying patterns can be used to train juniors.
4. Patterns can be a standards bible; you can use them to codify desired but
mundane behaviors. (Just "Copy/Paste" instead of asking others to
internalize the rule.)
5. Patterns can be synthesized to create highly creative and original
outcomes
6. Patterns can be abused (straight copying without understanding the
problem that needs to be solved can lead to HORRIBLE results)
7. Its best to look at the problem FIRST, before you go to the pattern
library.

And another question to the group - do you have best practice suggestions
for using patterns to free up time for solving bigger, more interesting
problems? And still successfully avoid the "cookie cutter design" trap?

Patterns need to be used properly, or they're trouble....

For example, I keep seeing junior designers on my team apply patterns
without understanding the intention/context behind the pattern. Then, when
I suggest changes to the design, they will debate me citing my own pattern
as evidence "But you did it here!" without understanding the context of the
problem...Sigh... Just curious if others have this experience, and if
there's advice on what to do about it. Other than becoming more patient, I
mean :)

-Wesley

19 Oct 2007 - 12:36am
Janna DeVylder
2006

> On 10/18/07, Wesley Hall <wesley.hall.parker at gmail.com> wrote:
> For example, I keep seeing junior designers on my team apply patterns
> without understanding the intention/context behind the pattern. Then, when
> I suggest changes to the design, they will debate me citing my own pattern
> as evidence "But you did it here!" without understanding the context of the
> problem...Sigh...
>

This is exactly why I'm so excited Christian Crumlish from Yahoo! has heeded
the call to reflect on (and share with us publicly, thank you!) the impact
of using an internal pattern library over a substantial amount of time. I
*believe* their internal library had a rating system associated with each
pattern, where team members across the organization could rate...something
about the pattern. Its utility? Its appropriateness? Its effectiveness in
addressing the problem? I want to learn if the patterns were applied
inappropriately early on, but after time/education/usage, did the error rate
decrease? And are there any checks and balances to make sure the pattern is
being applied appropriately? I feel like that's just simple management or
peer review of work, but how does that impact the pattern itself? After
applying a pattern in a project, is there any reflection on how well it
worked within the pattern documentation?

Wesley, I think there is always risk of a pattern being misappropriated
(that happens even without a formal library to pull from). I wonder, though,
within your organization, if the patterns you document tend to cover newer,
less-familiar territory as opposed to tried-and-true,
everyone-knows-what-THAT-is problem/solution set? That's definitely been my
experience.

janna

>
>

19 Oct 2007 - 2:19am
Dave Malouf
2005

I'm about to give my presentation on patterns in about 18 hours or
so. Having just finished my slides 18 hours ago you can say I'm knee
deep in thinking about patterns. A few musings coming out of the muck:

1. The person who brought up context & intention nailed it on the
head. Alexander is clear that a pattern languague is built from a n
"if-then" ststement where "if" defines the problem and then
defines the solution, but the solution also includes justifications
so that both context "if" and intention "then" are included in
the very definition of the pattern itself.

2. While today it is clear that Alexander uses his patterns for
political-social change, the original ideas behind patterns is to
catalog CANONICAL repeated observed problem/solution sets. I
highlight the word "canonical" b/c I believe this is the key
differentiation between patterns and guidelines. Guidelines or style
guides are single focus and their ubiquity or uniqueness is not
really relevant. It is the solution that works for the context of
those guidelines. Patterns need to work across much broader but still
well defined contexts to be useful.

3. There is a danger in using patterns as prescriptive and thus
designers who dislike prescription (as opposed to engineers who revel
in prescription) are turned off by many uses of patterns. However, if
the intentionality of the collection is for the purpose of not
suggestion or prescription, but rather for reflection and
exploration, the patterns hold value that designers can engage with
like they do with other elements of the design process (foundations,
personas, empathy models, etc.).

4. I think limiting patterns to a "language" doesn't do what
Jennifer Tidwell discusses in her book, which is the creation of
"best practices. This is something I'm not sure I agree w/ her on,
but I think her use of "best practices" works with where I have
landed (this week) with patterns as a design tool for reflection and
articulation.

5. Patterns as a point of articulation are powerful for designers who
need to move beyond abstraction and into tangible communication in
order to convey the goals and reasons for the solutions they do
finally end up with.

Anyway, this is me. for now. We'll see what type of response I get
by 6:30p today.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://gamma.ixda.org/discuss?post=21539

19 Oct 2007 - 9:00am
Faith Peterson
2007

It's an interesting aspect of the development of one's creative faculties
that there is a period of rote imitation and then some point of
transcendence, when the underlying principles suddenly - apparently -
coalesce. I think the junior person needs a lot of practice and observation
of other people identifying the significant characteristics of the problem -
that's the key to choosing and applying a pattern wisely, or departing from
a pattern or imitation of someone else's solution to a different (but
apparently in some way similar) problem.

I spend a fairly significant amount of time talking about the rationale for
design choices rather than the mechanics of the solutions. People who get
the "why" will normally make reasonable (even if perhaps not inspired)
choices about the "how." (Given this statement, fans of the Food Network
show "Good Eats" will understand why that's my favorite cooking show - Alton
Brown communicates the principles of cooking, not just how to construct the
recipe du jour.)
--
Faith Peterson
f.a.peterson at gmail.com

On 10/18/07, Wesley Hall <wesley.hall.parker at gmail.com> wrote:
>
>
> For example, I keep seeing junior designers on my team apply patterns
> without understanding the intention/context behind the pattern. Then,
> when
> I suggest changes to the design, they will debate me citing my own pattern
> as evidence "But you did it here!" without understanding the context of
> the
> problem...Sigh... Just curious if others have this experience, and if
> there's advice on what to do about it. Other than becoming more patient,
> I
> mean :)

19 Oct 2007 - 9:49am
Christian Crumlish
2006

A lot of interesting points along this thread. Wesley's comments are
very interesting. One thing about Yahoo's library worth noting is that
we are far from a centralized organization. We have a central UED team
but it's small and the vast majority of our designers work in business
units. It's difficult for us, culturally, to mandate anything.

Smaller houses, I think, in many ways can have it easier at least
insofar as their ability to create a pattern repository and designate
it as "required" (within reason).

On 10/18/07, Janna Hicks DeVylder <janna at devylder.com> wrote:
>
> This is exactly why I'm so excited Christian Crumlish from Yahoo! has heeded
> the call to reflect on (and share with us publicly, thank you!) the impact
> of using an internal pattern library over a substantial amount of time. I
> *believe* their internal library had a rating system associated with each
> pattern, where team members across the organization could rate...something
> about the pattern.

Actually, the ratings are adherence levels. There are currently three
ratings possible once a pattern is out of the draft stage: working
solution, best practice, and The Yahoo! Way. A working solution is a
reliable approach that the designer is free to apply or ignore or
their own recognizance. A best practice is a pattern we are strongly
encouraging designers to apply (when appropriate, of course), but that
they can deviate from with the signoff of their manager. Very few
things are designated The Yahoo! Way (and they often have to do with
branding and the suchlike). I like to say it takes "an act of Jerry"
to get dispensation to violate a pattern rated thusly.

Its utility? Its appropriateness? Its effectiveness in
> addressing the problem? I want to learn if the patterns were applied
> inappropriately early on, but after time/education/usage, did the error rate
> decrease?

As alluded to earlier, our problems have almost never been people
overapplying the patterns but the opposite problem of people
overlooking them.

> And are there any checks and balances to make sure the pattern is
> being applied appropriately? I feel like that's just simple management or
> peer review of work, but how does that impact the pattern itself? After
> applying a pattern in a project, is there any reflection on how well it
> worked within the pattern documentation?

There is an informal feedback loop. One interesting thing is that I'm
working with a regional group of what we call "Intls" in Asia who are
putting together their own localized ui pattern library and they are
learning from our experience and innovating in interesting ways that I
anticipate imitating once they've proven out some of these
improvements. They're being much more strict about associating
wireframe and comp templates, and code, with patterns; they're being
more systematic about documenting page types, *and* they've built in
an element for associating product releases with patterns (as in
"these four properties all use this pattern").

-xian-

--
Christian Crumlish http://xianlandia.com
Yahoo! pattern detective http://developer.yahoo.com/ypatterns
IA Institute director of technology http://iainstitute.org

Syndicate content Get the feed