Tail wagging dog
-
I'm probably in a rather unique situation ... Well, not entirely "unique", rather "rarefied" :laugh: Though I've a CS degree and do some programming, my major form of work is Architecture. And "no" it's the Architecture the Egyptians did 10k years ago - i.e. designing buildings. In Arch there's a well known axiom: "Form follows Function". What this means in Arch terms with a well designed building: Before starting to design anything, find out what the purpose(s) of the building is. Then design the required rooms and how they "join" together. Finally arrange them into an aesthetically pleasing shape. Bad design usually stems from a reversal of such process. While the building might look artistically pleasing its usability is suspect as the functions were forced into rooms which had a set shape and/or juxtaposition. To come to my gripe / concern / rant / moan :mad:: Many new programs seem to take a trend of "Form leads Function". Certainly not all, but I've been seeing a steady increase in the number of eye-candy-bloated (yet functionally deprived and/or use-blurring) programs over the last two decades. One case which jumped to mind showing a trivial sample: the MineSweeper game. Because there are many versions and some look pretty while others work well, not necessarily mutually-exclusive. Two samples which clearly indicate what I'm on about: MS's version is arguably one of the nicest looking. At least in comparison to Mines-Perfect. But wait! One feature of this game is to mark the tiles you think are mines. Since once you've marked the neighbouring tiles equal to the number of mines indicated a right-/middle-click would then open all the rest up. Where MP outshines Minesweeper is in how you mark those tiles: If there are only the number of tiles still unchecked as neighbours equal to the displayed number on a tile, you can simply right-click on the open tile to mark all of its neighbours. E.g. if you have a tile indicating 3 along a straight edge of open "space" you can right-click on it to mark all 3 its neighbours. In MS you'd need to right-click each of those neighbours in turn. I used this stupid sample to show an extreme case. MP is clearly very eye-candy-deficient. While MS looks much more polished, "interactive" and "dynamic" (with it's animated explosions), though it lacks a user productivity feature. There are others galore, and even
-
I'm probably in a rather unique situation ... Well, not entirely "unique", rather "rarefied" :laugh: Though I've a CS degree and do some programming, my major form of work is Architecture. And "no" it's the Architecture the Egyptians did 10k years ago - i.e. designing buildings. In Arch there's a well known axiom: "Form follows Function". What this means in Arch terms with a well designed building: Before starting to design anything, find out what the purpose(s) of the building is. Then design the required rooms and how they "join" together. Finally arrange them into an aesthetically pleasing shape. Bad design usually stems from a reversal of such process. While the building might look artistically pleasing its usability is suspect as the functions were forced into rooms which had a set shape and/or juxtaposition. To come to my gripe / concern / rant / moan :mad:: Many new programs seem to take a trend of "Form leads Function". Certainly not all, but I've been seeing a steady increase in the number of eye-candy-bloated (yet functionally deprived and/or use-blurring) programs over the last two decades. One case which jumped to mind showing a trivial sample: the MineSweeper game. Because there are many versions and some look pretty while others work well, not necessarily mutually-exclusive. Two samples which clearly indicate what I'm on about: MS's version is arguably one of the nicest looking. At least in comparison to Mines-Perfect. But wait! One feature of this game is to mark the tiles you think are mines. Since once you've marked the neighbouring tiles equal to the number of mines indicated a right-/middle-click would then open all the rest up. Where MP outshines Minesweeper is in how you mark those tiles: If there are only the number of tiles still unchecked as neighbours equal to the displayed number on a tile, you can simply right-click on the open tile to mark all of its neighbours. E.g. if you have a tile indicating 3 along a straight edge of open "space" you can right-click on it to mark all 3 its neighbours. In MS you'd need to right-click each of those neighbours in turn. I used this stupid sample to show an extreme case. MP is clearly very eye-candy-deficient. While MS looks much more polished, "interactive" and "dynamic" (with it's animated explosions), though it lacks a user productivity feature. There are others galore, and even
It's always been that way to an extent - good programmers are not normally graphic artists, and getting a software engineer to design your user interface is a good way to get a UI that is excellent to program rather than use. And conversely, a good graphics designer will design a app that looks good, but is about as useful as a chocolate teapot. Look at Apple: Form is King, while Microsoft: Well, no I'm not quite sure WTF is going on there... You want to look at some of the entries coming in for the Azure competition. It's not "form leads function" so much as "Form. What function?" :laugh:
The universe is composed of electrons, neutrons, protons and......morons. (ThePhantomUpvoter)
-
It's always been that way to an extent - good programmers are not normally graphic artists, and getting a software engineer to design your user interface is a good way to get a UI that is excellent to program rather than use. And conversely, a good graphics designer will design a app that looks good, but is about as useful as a chocolate teapot. Look at Apple: Form is King, while Microsoft: Well, no I'm not quite sure WTF is going on there... You want to look at some of the entries coming in for the Azure competition. It's not "form leads function" so much as "Form. What function?" :laugh:
The universe is composed of electrons, neutrons, protons and......morons. (ThePhantomUpvoter)
OriginalGriff wrote:
good programmers are not normally graphic artists
Some of my software has been described as military grade for precisely this reason, it works well and is robust however a delicate person would get blisters from using it...
“That which can be asserted without evidence, can be dismissed without evidence.”
― Christopher Hitchens
-
I'm probably in a rather unique situation ... Well, not entirely "unique", rather "rarefied" :laugh: Though I've a CS degree and do some programming, my major form of work is Architecture. And "no" it's the Architecture the Egyptians did 10k years ago - i.e. designing buildings. In Arch there's a well known axiom: "Form follows Function". What this means in Arch terms with a well designed building: Before starting to design anything, find out what the purpose(s) of the building is. Then design the required rooms and how they "join" together. Finally arrange them into an aesthetically pleasing shape. Bad design usually stems from a reversal of such process. While the building might look artistically pleasing its usability is suspect as the functions were forced into rooms which had a set shape and/or juxtaposition. To come to my gripe / concern / rant / moan :mad:: Many new programs seem to take a trend of "Form leads Function". Certainly not all, but I've been seeing a steady increase in the number of eye-candy-bloated (yet functionally deprived and/or use-blurring) programs over the last two decades. One case which jumped to mind showing a trivial sample: the MineSweeper game. Because there are many versions and some look pretty while others work well, not necessarily mutually-exclusive. Two samples which clearly indicate what I'm on about: MS's version is arguably one of the nicest looking. At least in comparison to Mines-Perfect. But wait! One feature of this game is to mark the tiles you think are mines. Since once you've marked the neighbouring tiles equal to the number of mines indicated a right-/middle-click would then open all the rest up. Where MP outshines Minesweeper is in how you mark those tiles: If there are only the number of tiles still unchecked as neighbours equal to the displayed number on a tile, you can simply right-click on the open tile to mark all of its neighbours. E.g. if you have a tile indicating 3 along a straight edge of open "space" you can right-click on it to mark all 3 its neighbours. In MS you'd need to right-click each of those neighbours in turn. I used this stupid sample to show an extreme case. MP is clearly very eye-candy-deficient. While MS looks much more polished, "interactive" and "dynamic" (with it's animated explosions), though it lacks a user productivity feature. There are others galore, and even
The same can be said of coding as well, no? "Form" (insert your favorite framework here) leads "Function", (insert overly complex code). Popular IoC implementations come to mind, as well as (pick your favorite) ORM frameworks. Marc
Testers Wanted!
Latest Article: User Authentication on Ruby on Rails - the definitive how to
My Blog -
The same can be said of coding as well, no? "Form" (insert your favorite framework here) leads "Function", (insert overly complex code). Popular IoC implementations come to mind, as well as (pick your favorite) ORM frameworks. Marc
Testers Wanted!
Latest Article: User Authentication on Ruby on Rails - the definitive how to
My BlogMarc Clifton wrote:
"Form" (insert your favorite framework here) leads "Function", (insert overly complex code).
That is probably another yes! But also a much more difficult one to get around. I mean, if the framework / library doesn't fit the purpose exactly - should you rather re-invent the wheel for that "surface"? :-\
-
Marc Clifton wrote:
"Form" (insert your favorite framework here) leads "Function", (insert overly complex code).
That is probably another yes! But also a much more difficult one to get around. I mean, if the framework / library doesn't fit the purpose exactly - should you rather re-invent the wheel for that "surface"? :-\
irneb wrote:
should you rather re-invent the wheel for that "surface"?
That is a good point, but I was referring more to people who "adopt" a new technology simply because it's the shiniest penny to fall onto the pavement. Marc
Testers Wanted!
Latest Article: User Authentication on Ruby on Rails - the definitive how to
My Blog -
I'm probably in a rather unique situation ... Well, not entirely "unique", rather "rarefied" :laugh: Though I've a CS degree and do some programming, my major form of work is Architecture. And "no" it's the Architecture the Egyptians did 10k years ago - i.e. designing buildings. In Arch there's a well known axiom: "Form follows Function". What this means in Arch terms with a well designed building: Before starting to design anything, find out what the purpose(s) of the building is. Then design the required rooms and how they "join" together. Finally arrange them into an aesthetically pleasing shape. Bad design usually stems from a reversal of such process. While the building might look artistically pleasing its usability is suspect as the functions were forced into rooms which had a set shape and/or juxtaposition. To come to my gripe / concern / rant / moan :mad:: Many new programs seem to take a trend of "Form leads Function". Certainly not all, but I've been seeing a steady increase in the number of eye-candy-bloated (yet functionally deprived and/or use-blurring) programs over the last two decades. One case which jumped to mind showing a trivial sample: the MineSweeper game. Because there are many versions and some look pretty while others work well, not necessarily mutually-exclusive. Two samples which clearly indicate what I'm on about: MS's version is arguably one of the nicest looking. At least in comparison to Mines-Perfect. But wait! One feature of this game is to mark the tiles you think are mines. Since once you've marked the neighbouring tiles equal to the number of mines indicated a right-/middle-click would then open all the rest up. Where MP outshines Minesweeper is in how you mark those tiles: If there are only the number of tiles still unchecked as neighbours equal to the displayed number on a tile, you can simply right-click on the open tile to mark all of its neighbours. E.g. if you have a tile indicating 3 along a straight edge of open "space" you can right-click on it to mark all 3 its neighbours. In MS you'd need to right-click each of those neighbours in turn. I used this stupid sample to show an extreme case. MP is clearly very eye-candy-deficient. While MS looks much more polished, "interactive" and "dynamic" (with it's animated explosions), though it lacks a user productivity feature. There are others galore, and even
irneb wrote:
Many new programs seem to take a trend of "Form leads Function"
As do many buildings especially when the form is brand new. And since most of programming is in fact brand new - the same equivalence must be made. Not to mention of course that most of the time engineers (architects) do not get to create something brand new when they are creating something like a tract house. Conversely when they do create something new it is on projects that span years. Versus software these days for which a year long project is considered abnormal.
irneb wrote:
One case which jumped to mind showing a trivial sample: the MineSweeper game
One case that always jumps to my mind is Frank Lloyd Wright and Fallingwater which was consider an innovative project at the time except for the small fact that the very feature that defined the house had to be propped up with two by fours for years. A solution that solved the problem was only found years later when it was "refactored" by using technology that did not exist when the house was built. And of course Frank Lloyd Wright is epitomized as one of the greatest architects the US has ever had.
irneb wrote:
A much more relevant example from my every-day work
Not sure what you consider as an example but certainly the the Narrows bridge disaster is one. But I like the ones where buildings manage to create wind tunnel effects. This has happened with multiple buildings where in some cases it becomes physically dangerous to walk around the buildings. This has been happening for years. The following is one example http://www.bbc.co.uk/news/uk-england-leeds-22258438[^]
-
irneb wrote:
Many new programs seem to take a trend of "Form leads Function"
As do many buildings especially when the form is brand new. And since most of programming is in fact brand new - the same equivalence must be made. Not to mention of course that most of the time engineers (architects) do not get to create something brand new when they are creating something like a tract house. Conversely when they do create something new it is on projects that span years. Versus software these days for which a year long project is considered abnormal.
irneb wrote:
One case which jumped to mind showing a trivial sample: the MineSweeper game
One case that always jumps to my mind is Frank Lloyd Wright and Fallingwater which was consider an innovative project at the time except for the small fact that the very feature that defined the house had to be propped up with two by fours for years. A solution that solved the problem was only found years later when it was "refactored" by using technology that did not exist when the house was built. And of course Frank Lloyd Wright is epitomized as one of the greatest architects the US has ever had.
irneb wrote:
A much more relevant example from my every-day work
Not sure what you consider as an example but certainly the the Narrows bridge disaster is one. But I like the ones where buildings manage to create wind tunnel effects. This has happened with multiple buildings where in some cases it becomes physically dangerous to walk around the buildings. This has been happening for years. The following is one example http://www.bbc.co.uk/news/uk-england-leeds-22258438[^]
jschell wrote:
But I like the ones where buildings manage to create wind tunnel effects.
Too true! Over here we've actually got a Town Planning issue: the entire Cape Town's CBD creates these wind tunnels through the streets. And since a 100km/h (60mi/h, without extra tunnelling from the buildings) wind isn't a "strange" thing in CT - you tend to see people hanging onto poles when those gusts blow them off their feet. http://www.news24.com/Multimedia/South-Africa/Wind-rattles-Cape-Town-20121130
jschell wrote:
And of course Frank Lloyd Wright is epitomized as one of the greatest architects the US has ever had.
Perhaps. It doesn't mean that everything he did was good though! It seems many architects are epitomized for the artistic design, not for the functionality. What I'm referring to as "good" architecture, is the very difficult thing of getting both artistic/aesthetic excellence without compromising functionality. IMO that is the true "art" of architecture. And the same goes for programs, it's easy to add every nuance of a function into a CLI, or it's easy to stick on the newest and greatest flat UI design, but putting both together in a useful way is the "trick". And in both arch/prog it is very often the case that either the artistic or the functional is compromised, it tends to be the functional which gets the least attention though.
-
irneb wrote:
Many new programs seem to take a trend of "Form leads Function"
As do many buildings especially when the form is brand new. And since most of programming is in fact brand new - the same equivalence must be made. Not to mention of course that most of the time engineers (architects) do not get to create something brand new when they are creating something like a tract house. Conversely when they do create something new it is on projects that span years. Versus software these days for which a year long project is considered abnormal.
irneb wrote:
One case which jumped to mind showing a trivial sample: the MineSweeper game
One case that always jumps to my mind is Frank Lloyd Wright and Fallingwater which was consider an innovative project at the time except for the small fact that the very feature that defined the house had to be propped up with two by fours for years. A solution that solved the problem was only found years later when it was "refactored" by using technology that did not exist when the house was built. And of course Frank Lloyd Wright is epitomized as one of the greatest architects the US has ever had.
irneb wrote:
A much more relevant example from my every-day work
Not sure what you consider as an example but certainly the the Narrows bridge disaster is one. But I like the ones where buildings manage to create wind tunnel effects. This has happened with multiple buildings where in some cases it becomes physically dangerous to walk around the buildings. This has been happening for years. The following is one example http://www.bbc.co.uk/news/uk-england-leeds-22258438[^]
jschell wrote:
And since most of programming is in fact brand new
Actually on that score I must disagree. As far as I can tell most (if not all) programs are simply rehashing of stuff which was done several decades ago. E.g. instead of AutoCAD's "newest" geometric constraints, this was available in 1963: Ivan Sutherland's SketchPad Or collaborative editing over the "cloud" with video conferencing and immediate updates? Douglas Engelbart - The Mother of all Demos When there's already such things made 50 years ago, it's difficult to see any programs of today as being "innovative".
-
jschell wrote:
But I like the ones where buildings manage to create wind tunnel effects.
Too true! Over here we've actually got a Town Planning issue: the entire Cape Town's CBD creates these wind tunnels through the streets. And since a 100km/h (60mi/h, without extra tunnelling from the buildings) wind isn't a "strange" thing in CT - you tend to see people hanging onto poles when those gusts blow them off their feet. http://www.news24.com/Multimedia/South-Africa/Wind-rattles-Cape-Town-20121130
jschell wrote:
And of course Frank Lloyd Wright is epitomized as one of the greatest architects the US has ever had.
Perhaps. It doesn't mean that everything he did was good though! It seems many architects are epitomized for the artistic design, not for the functionality. What I'm referring to as "good" architecture, is the very difficult thing of getting both artistic/aesthetic excellence without compromising functionality. IMO that is the true "art" of architecture. And the same goes for programs, it's easy to add every nuance of a function into a CLI, or it's easy to stick on the newest and greatest flat UI design, but putting both together in a useful way is the "trick". And in both arch/prog it is very often the case that either the artistic or the functional is compromised, it tends to be the functional which gets the least attention though.
irneb wrote:
What I'm referring to as "good" architecture, is the very difficult thing of getting both artistic/aesthetic excellence without compromising functionality. IMO that is the true "art" of architecture.
Yes but my point is that architects manage to produce problematic designs too.
-
jschell wrote:
And since most of programming is in fact brand new
Actually on that score I must disagree. As far as I can tell most (if not all) programs are simply rehashing of stuff which was done several decades ago. E.g. instead of AutoCAD's "newest" geometric constraints, this was available in 1963: Ivan Sutherland's SketchPad Or collaborative editing over the "cloud" with video conferencing and immediate updates? Douglas Engelbart - The Mother of all Demos When there's already such things made 50 years ago, it's difficult to see any programs of today as being "innovative".
irneb wrote:
Actually on that score I must disagree. As far as I can tell most (if not all) programs are simply rehashing of stuff which was done several decades ago.
That isn't true in comparison to civil engineering. Putting up the tallest building in the world requires new techniques. Putting up a 10 story office building doesn't. And 90% of programmers are not creating new word processors. The fact that a building requires nails, screws and elevators doesn't mean that all such buildings are equal. And a program like Excel and Word are not the same even if they both use a menu bar.
