System Architecture and Usability

17 Nov 2005 - 8:27am
8 years ago
20 replies
524 reads
Suresh JV
2004

Hi All,

Here is Tog's comment :The great efficiency breakthroughs in software are to
be found in the fundamental architecture of the system, not in the surface
design of the interface.

And from a member: Google uses RAM all over the world to cache their data,
rather than hard drive retrieval.

My question: To what extent does an Interaction Designer suggest or discuss
the system architecture issues.
[Without political backlash if any...]
How do you change architectural decisions made early in Development to
provide better interaction, and how is it taken by the other teams. Can an
ROI be projected?

Related content: http://www.cs.cmu.edu/~bej/usa/
http://www.cs.cmu.edu/~bej/usa/publications/FinalCHI2004Tutorial.pdf

Regards,
Suresh JV.

Comments

17 Nov 2005 - 8:33am
Ben Hunt
2004

It's absolutely the IxDer's job to architect a system. A system is the
interaction of hardware, software and user. It incorporates brand, behaviour
on every level, and goes down to the detail of the code and the user
interface mechanics.

While we don't necessarily get into the detail of the code, that's
engineering, every other part of the "building" is part of the experience,
and so that's part of the Interaction Architect's domain (this is Tog's
preferred term, and has merit).

It's vital to be part of designing the architecture so that you don't have
to try and "change early architectural decisions".

- Ben

-----Original Message-----
From: Suresh J V

My question: To what extent does an Interaction Designer suggest or discuss
the system architecture issues.
[Without political backlash if any...]
How do you change architectural decisions made early in Development to
provide better interaction, and how is it taken by the other teams. Can an
ROI be projected?

17 Nov 2005 - 8:34am
Edwin van de Bo...
2005

{your question}: To what extent does an Interaction Designer suggest or
discuss the system architecture issues.

{my anwser}: I feel that interaction designers should be part of a 'team'. I
consider myself a product-designer, more than just an interaction designer.
If the team strives to create a product for a user, every inpuit should be
heard. Thus, when from an IxD point of view, we suggest that this and that
information should be presented together, and this has consequences for the
system arch, so be it. Sadly enough, we're most of the times brought into
the conversation too late, and then we'll just have to TRY to make it as
easy as possible on the front-end side of the spectrum.

And I agree with Tog, the great efficiency lies in the fundamental
architecture. So trying to get involved as early into the process as
possible is key, to provide some user-perspectives into the descision-making
about the arch.

-----Oorspronkelijk bericht-----
Van: discuss-bounces op lists.interactiondesigners.com
[mailto:discuss-bounces op lists.interactiondesigners.com] Namens Suresh J V
Verzonden: donderdag 17 november 2005 14:28
Aan: discuss op ixda.org
Onderwerp: [IxDA Discuss] System Architecture and Usability

[Please voluntarily trim replies to include only relevant quoted material.]

Hi All,

Here is Tog's comment :The great efficiency breakthroughs in software are to
be found in the fundamental architecture of the system, not in the surface
design of the interface.

And from a member: Google uses RAM all over the world to cache their data,
rather than hard drive retrieval.

My question: To what extent does an Interaction Designer suggest or discuss
the system architecture issues.
[Without political backlash if any...]
How do you change architectural decisions made early in Development to
provide better interaction, and how is it taken by the other teams. Can an
ROI be projected?

Related content: http://www.cs.cmu.edu/~bej/usa/
http://www.cs.cmu.edu/~bej/usa/publications/FinalCHI2004Tutorial.pdf

Regards,
Suresh JV.

________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... discuss op ixda.org
(Un)Subscription Options ... http://subscription-options.ixda.org/
Announcements List ......... http://subscribe-announce.ixda.org/
Questions .................. lists op ixda.org
Home ....................... http://ixda.org/
Resource Library ........... http://resources.ixda.org

17 Nov 2005 - 10:47am
niklasw
2005

Hi

About a year ago I did this graphic to try to explain my view on your
question to a crowd of system architects and coders at my company.
Maybe it can be of help? Have a look and see if it can be of help to
you. And just to clarify I do agree with Ben and Edwin.

http://flickr.com/photos/22552254@N00/64205409 (Hopefully this flickr
thing works out)

--Niklas

Niklas Wolkert
Interaction designer at TeliaSonera
http://www.teliasonera.com

On 17/11/05, Suresh J V <suresh.jv at ranal.com> wrote:
> [Please voluntarily trim replies to include only relevant quoted material.]
>
> Hi All,
>
> Here is Tog's comment :The great efficiency breakthroughs in software are to
> be found in the fundamental architecture of the system, not in the surface
> design of the interface.
>
> And from a member: Google uses RAM all over the world to cache their data,
> rather than hard drive retrieval.
>
> My question: To what extent does an Interaction Designer suggest or discuss
> the system architecture issues.
> [Without political backlash if any...]
> How do you change architectural decisions made early in Development to
> provide better interaction, and how is it taken by the other teams. Can an
> ROI be projected?
>
> Related content: http://www.cs.cmu.edu/~bej/usa/
> http://www.cs.cmu.edu/~bej/usa/publications/FinalCHI2004Tutorial.pdf
>
>
> Regards,
> Suresh JV.
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

--
Niklas

17 Nov 2005 - 10:56am
Dave Malouf
2005

Niklas,

That was a nice graphic. I'm not sure I agree w/ the IxD amounts,
especially in Graphic Design, but those are details.

What I'd like to do is address the question more theoretically.

an IxD should be involved in all aspects of a system design that effects
the behavior of the system that the user(s) interact with.

I.e. the google scenarion about RAM instead of HD. That decision (whether
by person or strictly ad hoc) was an IxD decision. The system's response
rate is a behavioral question.

IxD is as much about systems as it is about interfaces.

-- dave

17 Nov 2005 - 11:04am
Larry Marine
2005

I agree with Ben, but could easily provide a more vociferous response. I
have been at this usability thing for almost 2 decades now, and was a
computer engineer for 15 years before that and I can categorically state
that left to their own designs, the system architecture guys would achieve
little more than automating our current frustrations or making faster and
easier for us to do the wrong things very well.

Several of our clients have received patents for the innovative approaches
we recommended that required a new architecture. Some of those patents are
now so ubiquitous as to jeopardize the right to the patent. These were
approaches that would very well have never been realized from an
architecture driven design approach. That is simply because system
architects are more likely to try to mold the world to fit their
architecture than to refashion their architecture to fit the world.

Case in point: One client recently engaged us to make their product easier
to use. It was an enterprise product designed to be used by IT folks. The
problem was, it wasn't IT folks who really needed to use it. The only users
for this product were the HR managers in these enterprises. This obviously
meant a different interface as well as a different structure to support that
interface. Given the original design architecture and the need to operate
the tool outside of the IT department, there was no way the client could
simply re-skin the product for the HR folks. IT required a new architecture,
specifically to allow and control access to the data. As they say in Maine,
"you can't get there from here."

I think that comment was either naïve or taken out of context or both.
Moreover, the past 60 years of computer technology has proven that engineers
alone cannot drive technological evolution. In my experienced opinion,
design should drive architecture, not the other way around.

Larry

----- Original Message -----
From: "Ben Hunt" <ben at scratchmedia.co.uk>
To: <discuss at ixda.org>
Sent: Thursday, November 17, 2005 6:33 AM
Subject: Re: [IxDA Discuss] System Architecture and Usability

> [Please voluntarily trim replies to include only relevant quoted
> material.]
>
> It's absolutely the IxDer's job to architect a system. A system is the
> interaction of hardware, software and user. It incorporates brand,
> behaviour
> on every level, and goes down to the detail of the code and the user
> interface mechanics.
>
> While we don't necessarily get into the detail of the code, that's
> engineering, every other part of the "building" is part of the experience,
> and so that's part of the Interaction Architect's domain (this is Tog's
> preferred term, and has merit).
>
> It's vital to be part of designing the architecture so that you don't have
> to try and "change early architectural decisions".
>
>
> - Ben
>
>
> -----Original Message-----
> From: Suresh J V
>
> My question: To what extent does an Interaction Designer suggest or
> discuss
> the system architecture issues.
> [Without political backlash if any...]
> How do you change architectural decisions made early in Development to
> provide better interaction, and how is it taken by the other teams. Can an
> ROI be projected?
>
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

17 Nov 2005 - 10:54am
Edwin van de Bo...
2005

Nice graphic Niklas. I guess folks like us create so many of these graphics
just to explain what we exactly do all the time. Why don't we just sell
lawnmowers...

How do you see an interaction designer having a role in the 'coding'?

-----Oorspronkelijk bericht-----
Van: discuss-bounces op lists.interactiondesigners.com
[mailto:discuss-bounces op lists.interactiondesigners.com] Namens Niklas
Wolkert
Verzonden: donderdag 17 november 2005 16:47
Aan: Suresh J V
CC: discuss op ixda.org
Onderwerp: Re: [IxDA Discuss] System Architecture and Usability

[Please voluntarily trim replies to include only relevant quoted material.]

Hi

About a year ago I did this graphic to try to explain my view on your
question to a crowd of system architects and coders at my company.
Maybe it can be of help? Have a look and see if it can be of help to you.
And just to clarify I do agree with Ben and Edwin.

http://flickr.com/photos/22552254@N00/64205409 (Hopefully this flickr thing
works out)

--Niklas

Niklas Wolkert
Interaction designer at TeliaSonera
http://www.teliasonera.com

On 17/11/05, Suresh J V <suresh.jv op ranal.com> wrote:
> [Please voluntarily trim replies to include only relevant quoted
> material.]
>
> Hi All,
>
> Here is Tog's comment :The great efficiency breakthroughs in software
> are to be found in the fundamental architecture of the system, not in
> the surface design of the interface.
>
> And from a member: Google uses RAM all over the world to cache their
> data, rather than hard drive retrieval.
>
> My question: To what extent does an Interaction Designer suggest or
> discuss the system architecture issues.
> [Without political backlash if any...] How do you change architectural
> decisions made early in Development to provide better interaction, and
> how is it taken by the other teams. Can an ROI be projected?
>
> Related content: http://www.cs.cmu.edu/~bej/usa/
> http://www.cs.cmu.edu/~bej/usa/publications/FinalCHI2004Tutorial.pdf
>
>
> Regards,
> Suresh JV.
>
>
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss op ixda.org (Un)Subscription Options
> ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists op ixda.org Home
> ....................... http://ixda.org/ Resource Library ...........
> http://resources.ixda.org
>

--
Niklas
________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... discuss op ixda.org (Un)Subscription Options ...
http://subscription-options.ixda.org/
Announcements List ......... http://subscribe-announce.ixda.org/
Questions .................. lists op ixda.org Home .......................
http://ixda.org/ Resource Library ........... http://resources.ixda.org

17 Nov 2005 - 11:08am
Jared M. Spool
2003

At 10:56 AM 11/17/2005, dave at ixdg.org wrote:
>an IxD should be involved in all aspects of a system design that effects
>the behavior of the system that the user(s) interact with.

What's interesting to me is that this conversation only happens in arenas
where titles like "IxD", "Usability Professional", "Designer", and "IA" are
rigorously adhered to.

As we study the organizations that are successful at user experience
design, these titles tend to melt away and teams start working as coherent
units. At that point, it's quite natural for the person who was brought in
with IxD skills to be working on all aspects of the system, using their
perspectives and strengths to augment the total team's capabilities.

Jared

Jared M. Spool, Founding Principal, User Interface Engineering
4 Lookout Lane, Unit 4d, Middleton, MA 01949
978 777-9123 jspool at uie.com http://www.uie.com
Blog: http://www.uie.com/brainsparks

17 Nov 2005 - 11:23am
jbellis
2005

----- Original Message -----
From: "Suresh J V" <suresh.jv at ranal.com>
To what extent does an Interaction Designer suggest or discuss the system
architecture issues.
---------
Great question, which I suppose means there is no single answer. My answer:
yes, you must get involved. But the gap between technical implementation and
Userville is sometimes spanned by a rope bridge right out of Bollywood...
you know, spikes in the ravine, flames behind you and in front.

I am involved in one such issue right now. I call it "object scope and
inheritance." If I don't succeed in affecting a positive change, the user
experience will invariably suffer, whether you categorize it as interaction,
mental model, or facility. My problem is equivalent to styles (or macros or
autotext) in MS Word. These objects can be supported at three levels:
system, template, document. In our system, they are supported only at the
instance (document) level and users have to jump through hoops to
synchronize across templates or the whole system. A classic architecture
issue that has profound UE costs.

It is a simple matter to say that as the Mayor (or is it really Fairy
Godmother?) of Userville I want all of our citizens to have 1) dynamic (but
controlled) inheritance down from all levels and 2) between any levels on
demand. (Word supports both.) On the flip side, I don't want to have our
system entirely re-factored at this point in time, so I am proposing to
support only option "2)," which is equivalent to MS Word's Organizer.
(Tools>Macros>Organizer among other paths.) Option 2 gives users 95% of the
power they need without any mental load or system disruption.

We could argue endlessly on whether the IA should be a party to the
technical politicking... whether proposing a refactoring of a system or a
feature as broad as an organizer can simply be "thrown over the wall."
Myself? The dragon cut the bridge in half and I'm climbing up the canyon
wall on the burning threads.

www.jackbellis.com, www.usabilityInstitute.com

17 Nov 2005 - 12:06pm
Vassili Bykov
2005

Niklas Wolkert wrote:
>
> Hi
>
> About a year ago I did this graphic to try to explain my view on your
> question to a crowd of system architects and coders at my company.
> Maybe it can be of help? Have a look and see if it can be of help to
> you. And just to clarify I do agree with Ben and Edwin.
>
> http://flickr.com/photos/22552254@N00/64205409 (Hopefully this flickr
> thing works out)

I doubt anyone on this list would disagree with Ben and Edwin. To
the flood of agreement so far I want to add this verbal picture I once
painted in an internal email explaining the very same thing.

- quote -

Nobody in the traditional view of the process as "customers want it,
marketing finds that out, engineers build a solution" has what it takes
to study user behavior and make sure the solution fits that behavior
rather than the technology or some vague idea of what is required.

It's like you build a house and then discover that a toilet ended up in
the living room, because Plumbing Technical Lead found it was easier to
install there after Marketing told them customers want a toilet, and the
Electricians didn't mind. You can coordinate furniture with that toilet,
and you can install recessed lighting and pick trendy colors to improve
the impression it makes on your guests, but in the end you can't use it
until you reroute the pipes and move the thing.

You would have saved money not having it in the living room in the first
place. This is why someone who understands these usability things needs
to be involved in early decision-making together with Plumbers.

- unquote -

Please excuse the imagery.

>
> --Niklas
>
> Niklas Wolkert
> Interaction designer at TeliaSonera
> http://www.teliasonera.com

17 Nov 2005 - 9:49pm
penguinstorm
2005

On Nov-17-2005, at 5:27 AM, Suresh J V wrote:

> My question: To what extent does an Interaction Designer suggest or
> discuss
> the system architecture issues.

Not having read the rest of the responses (I'm just getting to email
today):

To the extent that it affects the user experience. This could be
anything from interface to responsiveness, if they're identified as
important considerations.
--
Scott Nelson
skot at penguinstorm.com
http://www.penguinstorm.com/

skype. skot.nelson

17 Nov 2005 - 10:00pm
penguinstorm
2005

On Nov-17-2005, at 8:08 AM, Jared M. Spool wrote:

> What's interesting to me is that this conversation only happens in
> arenas
> where titles like "IxD", "Usability Professional", "Designer", and
> "IA" are
> rigorously adhered to.

I didn't realize they ever were?

In any case, Ben Hunt said this:

> It's absolutely the IxDer's job to architect a system. A system is the
> interaction of hardware, software and user.

and I'd strongly disagree, even with his qualifying statement.

If the goal of IxD is, in fact, interaction design than it's
important to pick battles carefully.

Others have asserted that almost any aspect of a system CAN affect
IxD. This is somewhat true, but it's also important to focus on those
that WILL or DO.

In building web apps, for example, the choice of server platform is
not likely to have much impact on Interaction Design (Apache? IIS?
Microsoft? Linux? Apple? All are quite capable of delivering more
than adequate performance at any level with different
configurations.) A decision best left the relevant decision makers
(project manager? IT manager? whatever.)

(This isn't meant to suggest that there aren't projects or teams
where the IxD person is ALSO this person.)
--
Scott Nelson
skot at penguinstorm.com
http://www.penguinstorm.com/

skype. skot.nelson

21 Nov 2005 - 12:57pm
leo.frishberg a...
2005

<snip>

My question: To what extent does an Interaction Designer suggest or
discuss
the system architecture issues.
</snip>

I am faced with a slightly different challenge. My team has bought into
the notion of an integrated "advanced development group" who is
responsible for creating the next big thing (five years out). This
group has a H/W engineer, S/W engineer, and UX designer, along with
part-time marketing input.

Our group is responsible for building the "architectural" documents for
each of these domains.

The question to the group is: What does a UX Architecture document look
like?
Corollary questions include:

Does a single architectural framework (the "system architecture") apply
to all three disciplines (H/W, S/W, UX)?

Given our "architectural" (read strategic) posture, where does one draw
the line between "architecture" and the design development activities
specific product development groups will be responsible for?

Thoughts?

Leo

21 Nov 2005 - 4:32pm
Austin Govella
2004

On 11/17/05, Skot Nelson <skot at penguinstorm.com> wrote:
> In building web apps, for example, the choice of server platform is
> not likely to have much impact on Interaction Design (Apache? IIS?
> Microsoft? Linux? Apple? All are quite capable of delivering more
> than adequate performance at any level with different
> configurations.)

Choice of platform won't have much impact, but the system architecture
will impact any interactions founded on that architecture: in other
words, it affects every damn request generated by a web app.

Systems are well-architected when they minimize negative impact on
other systems. They've been designed to facilitate every damn request
generated by a web app.

Conversely, badly-architected systems stymie everything you do with
huge latency, slow request response, and difficult-to-use data (from
the badly-architected databases).

Designers should affect system architecture where that architecture
affects their ability to deliver competent experiences, just as
members of *any* business unit should affect the architecture where it
affects them: Accounting inputs on their ability to pay; IT on their
ability to administrate and maintain; the salesperson on their ability
to deliver the architecture to the client...

Collaboration doesn't represent consensus or democratic design.
Collaboration innovates better solutions because it allows competing
goals the ability to offer their best input for best impact to create
the best system. The best design will always best recognize the
interaction's context.

Those business guys would say: "Collaboration allows better
contextualization of individual innovations."

--
Austin Govella
Thinking & Making: IA, UX, and IxD
http://thinkingandmaking.com
austin.govella at gmail.com

21 Nov 2005 - 4:38pm
penguinstorm
2005

On Nov-21-2005, at 1:32 PM, Austin Govella wrote:

> Choice of platform won't have much impact, but the system architecture
> will impact any interactions founded on that architecture: in other
> words, it affects every damn request generated by a web app.

Of course they do. My point, though, was that any web hosting
platform available today is essentially capable of delivering the
services needed for almost any volume. The configuration required
might be different - on an MS platform I would tend to want more
redundant servers, in order to allow the for the fact that the
servers tend to crash more often.

As an IxD, I wouldn't try to push a Microsoft platform onto an office
that as built on an entirely Oracle platform, or on Solaris hardware.

> Systems are well-architected when they minimize negative impact on
> other systems. They've been designed to facilitate every damn request
> generated by a web app.

Yes. And ensuring that a web hosting platform doesn't negatively
impact other systems is not a function of IxD - it's a pure IT
function, or a Systems Architecture function.
--
Scott Nelson
skot at penguinstorm.com
http://www.penguinstorm.com/

skype. skot.nelson

21 Nov 2005 - 5:11pm
Jack L. Moffett
2005

The distinction that needs to be made in regards to the IxDer's role
in influencing the architecture is that they aren't responsible for
technical specification. What they are responsible for is making sure
that the technical person who will be making architectural decisions
understands the needs of the UI.

Jack

Jack L. Moffett
Interaction Designer
inmedius
412.690.2360 x219
http://www.inmedius.com

Things should be as simple as possible,
but no simpler.

- Albert Einstein

***********************************************************************
Confidentiality Note: The information contained in this email and document(s) attached are for the exclusive use of the addressee and contains confidential, privileged and non-disclosable information. If the recipient of this email is not the addressee, such recipient is strictly prohibited from reading, photocopying, distributing or otherwise using this email or its contents in any way.
***********************************************************************

22 Nov 2005 - 12:16pm
Stewart Dean
2004

I can only give you things from my experience of interactive TV/online
projects.

The Software/Hardware is technical implementation and is expressed in
details of how the system works. This is how it does it.

The user experience architecture is the functional specification - that is
what the system does described from the user point of view.

The user experience architecture takes the form of site maps/user journeys
with detailed user interface schematics (often refered to as wireframes).
This has as much notes as needed and stops short of technical details but
includes all the input/output.

In theory the user experience architecture should be able to be take and
implemented on many different ways but the user experience would be much the
same. The technical implimentation should be as invisible as possible.

Cheers

Stewart Dean

On 21/11/05 5:57 pm, "leo.frishberg at exgate.tek.com"
<leo.frishberg at exgate.tek.com> wrote:

> [Please voluntarily trim replies to include only relevant quoted material.]
>
> <snip>
>
> My question: To what extent does an Interaction Designer suggest or
> discuss
> the system architecture issues.
> </snip>
>
> I am faced with a slightly different challenge. My team has bought into
> the notion of an integrated "advanced development group" who is
> responsible for creating the next big thing (five years out). This
> group has a H/W engineer, S/W engineer, and UX designer, along with
> part-time marketing input.
>
> Our group is responsible for building the "architectural" documents for
> each of these domains.
>
> The question to the group is: What does a UX Architecture document look
> like?
> Corollary questions include:
>
> Does a single architectural framework (the "system architecture") apply
> to all three disciplines (H/W, S/W, UX)?
>
> Given our "architectural" (read strategic) posture, where does one draw
> the line between "architecture" and the design development activities
> specific product development groups will be responsible for?
>
> Thoughts?
>
> Leo
> ________________________________________________________________
> Welcome to the Interaction Design Association (IxDA)!
> To post to this list ....... discuss at ixda.org
> (Un)Subscription Options ... http://subscription-options.ixda.org/
> Announcements List ......... http://subscribe-announce.ixda.org/
> Questions .................. lists at ixda.org
> Home ....................... http://ixda.org/
> Resource Library ........... http://resources.ixda.org
>

22 Nov 2005 - 12:57pm
jbellis
2005

----- Original Message -----
From: <leo.frishberg at exgate.tek.com>

> What does a UX Architecture document look like?

Suresh's original post contained a link to a pdf that I read with great
interest. It might constitute at least in part a literal answer to your
question, but there's probably a very wide range of what "UX architecture"
even means or if the two words constitute an oxymoron. (To successfully
access it, I had to paste the URL into a simple html file so I could choose
"Save target as.")

http://www.cs.cmu.edu/~bej/usa/publications/FinalCHI2004Tutorial.pdf

The paper contains two halves: 1) A list of ~25 high level but literal (not
philosophical) usability goals that cannot be retrofitted into an
architecture that does not anticipate them. I believe that this list
constitutes a seminal work in the field and will eventually be standard
fare, perhaps a textbook chapter that highschoolers will read. Surely these
items should be in your UX architecture document, but they need a little
wordsmithing to clearly articulate architectural choices (some degraded into
user pet peeves). 2) Part Two is academia's mandated penchant for finding a
unifying force among the goals, and the 30 pages of fluff that go with it.
This half inspired me to add the word PHuDge to www.pseudodictionary.com.

> Does a single architectural framework (the "system architecture") apply to
> all three disciplines (H/W, S/W, UX)?
No, it just creates petty battles among people talking at cross-purposes.
Fun stuff for companies flush with cash, but a waste of time, otherwise.
Does UX need things from S/W? Yes, see above.

> Where does one draw the line between "architecture" and the design
> development activities specific product development groups will be
> responsible for?

I only have an opinion as regards the usability component: the line ends
after goals and objectives. Unfortunately, as interfaces accomplish more,
the line between "what" (goals) and "how" (means to an end) is really
difficult to find. For instance, in the PDF, they talk about "comprehensive
searching." On the last three big systems on which I worked, that translated
to the developers as "Oh, you want a $100,000 reporting system and a data
warehouse added to the spec?"

www.jackbellis.com, www.usabilityInstitute.com

22 Nov 2005 - 3:08pm
leo.frishberg a...
2005

Thanks Jack,

My thoughts inline

>Suresh's original post contained a link to a pdf that I read
>with great
>interest.
>... but there's probably a very wide range of what "UX
>architecture"
>even means or if the two words constitute an oxymoron.
>
>Surely these
>items should be in your UX architecture document, but they
>need a little
>wordsmithing to clearly articulate architectural choices (some
>degraded into
>user pet peeves).

Therein lies the rub - we've discussed elements such as "quality
attributes" that are measurable, actionable, and human-centric. Not
necessarily usability goals, but of the same ilk. The problem is that
they don't look very much like the S/W, H/W architectural elements found
in canonical frameworks (i.e., MVC, referenced in the .pdf above, or any
other "pattern" for that matter).

>> Does a single architectural framework (the "system
>architecture") apply to
>> all three disciplines (H/W, S/W, UX)?
>No, it just creates petty battles among people talking at
>cross-purposes.
>Fun stuff for companies flush with cash, but a waste of time,
>otherwise.

I'm intrigued by your dispensing with the notion out-of-hand. What
causes you to state so emphatically that there can't be (or shouldn't
be) a unifying framework that captures architectural elements from the
various disciplines? Clearly, the devil is in the definition of
"architecture", and I think we're struggling with that as much as
anything, but if one can agree on the meaning (within one's context),
then surely there is the possibility that a single framework can be
established to communicate those elements to the development teams.

(in specific response to your answer, in our case a) we aren't terribly
flush with cash, b) our conversations, thankfully, are deep and
purposeful with refreshingly little pettiness, but none-the-less we're
willing to consider them a waste of time...hence my posting }:{) )

>Does UX need things from S/W? Yes, see above.

I would submit that the answers to the original posting in this thread
were all about what UX needs from S/W and H/W in order to be expressed.
In addition, I read with great comfort that individuals believe IxD is
and should be an equal influencer of S/W, H/W architectures. The
questions this thread has raised for our group is: Can UX be expressed
architecturally separately from S/W and H/W architectures? Can a uniform
framework express the three architectures appropriately? Must UX
architectural elements be expressed in terms of screen instantiations,
site maps, navigation schema, etc., or are there higher levels of
abstraction that belong in the realm of UX which become instantiated in
any given product in the form of screens, etc.?

One conclusion we are coming to, which may be self-evident to members of
this list, is that UX ::can only:: be expressed in terms of H/W, S/W
architectures.
>
>> Where does one draw the line between "architecture" and the design
>> development activities specific product development groups will be
>> responsible for?

>For instance, in the PDF, they talk about
>"comprehensive
>searching." On the last three big systems on which I worked,
>that translated
>to the developers as "Oh, you want a $100,000 reporting system
>and a data
>warehouse added to the spec?"

Your response underscores the problem exactly. My compatriots believe
that H/W architectural elements can be expressed in terms that don't
specify what any specific product must create, but provide "unchanging"
elements within which any design must be constrained. A similar
proposition is being put forth by the S/W architect. Are there
analogous UX elements?

Thanks again for your considered response,

Leo

5 Dec 2005 - 10:56am
FelcanSmith, Mark
2004

Leo,

I'd suggest thinking about the UX designers output in terms of a user
environment design. The UED defines the function and structure needed to
support the work practice. This can get to varying levels of detail,
mapped to release schedules, etc...

Check out InContext's explanation of this. I've been exposed to their
work through colleagues and I see a natural fit w/ the User Environment
Design piece, and what you're asking about.

http://www.incontextdesign.com/cd/cdprocess.html

-Mark

> -----Original Message-----
> From: leo.frishberg at exgate.tek.com
> [mailto:leo.frishberg at exgate.tek.com]
> Sent: Monday, November 21, 2005 11:57 AM
> To: discuss at ixda.org
> Subject: Re: [IxDA Discuss] System Architecture and Usability
>
> <snip>
>
> My question: To what extent does an Interaction Designer
> suggest or discuss the system architecture issues.
> </snip>
>
> I am faced with a slightly different challenge. My team has
> bought into the notion of an integrated "advanced development
> group" who is responsible for creating the next big thing
> (five years out). This group has a H/W engineer, S/W
> engineer, and UX designer, along with part-time marketing input.
>
> Our group is responsible for building the "architectural"
> documents for each of these domains.
>
> The question to the group is: What does a UX Architecture
> document look like?
> Corollary questions include:
>
> Does a single architectural framework (the "system
> architecture") apply to all three disciplines (H/W, S/W, UX)?
>
> Given our "architectural" (read strategic) posture, where
> does one draw the line between "architecture" and the design
> development activities specific product development groups
> will be responsible for?
>
> Thoughts?
>
> Leo
>
>

5 Dec 2005 - 12:38pm
leo.frishberg a...
2005

Thanks Mark,

This is an excellent suggestion. As I have completed most of the
contextual inquiry steps leading up to the UED, our team is poised to
take advantage of this view.

Beautiful.

Thanks,

Leo

>-----Original Message-----
>From: FelcanSmith, Mark [mailto:mfelc at allstate.com]
>Sent: Monday, December 05, 2005 7:56 AM
>To: Frishberg, Leo; discuss at ixda.org
>Subject: Re: [IxDA Discuss] System Architecture and Usability
>
>
>Leo,
>
>I'd suggest thinking about the UX designers output in terms of a user
>environment design. The UED defines the function and structure
>needed to
>support the work practice. This can get to varying levels of detail,
>mapped to release schedules, etc...
>
>Check out InContext's explanation of this. I've been exposed to their
>work through colleagues and I see a natural fit w/ the User Environment
>Design piece, and what you're asking about.
>
>http://www.incontextdesign.com/cd/cdprocess.html
>
>-Mark
>
>> -----Original Message-----
>> From: leo.frishberg at exgate.tek.com
>> [mailto:leo.frishberg at exgate.tek.com]
>> Sent: Monday, November 21, 2005 11:57 AM
>> To: discuss at ixda.org
>> Subject: Re: [IxDA Discuss] System Architecture and Usability
>>
>> <snip>
>>
>> My question: To what extent does an Interaction Designer
>> suggest or discuss the system architecture issues.
>> </snip>
>>
>> I am faced with a slightly different challenge. My team has
>> bought into the notion of an integrated "advanced development
>> group" who is responsible for creating the next big thing
>> (five years out). This group has a H/W engineer, S/W
>> engineer, and UX designer, along with part-time marketing input.
>>
>> Our group is responsible for building the "architectural"
>> documents for each of these domains.
>>
>> The question to the group is: What does a UX Architecture
>> document look like?
>> Corollary questions include:
>>
>> Does a single architectural framework (the "system
>> architecture") apply to all three disciplines (H/W, S/W, UX)?
>>
>> Given our "architectural" (read strategic) posture, where
>> does one draw the line between "architecture" and the design
>> development activities specific product development groups
>> will be responsible for?
>>
>> Thoughts?
>>
>> Leo
>>
>>
>

Syndicate content Get the feed