Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
3 online now:
Newest Member: popoi
Post Volume: Total: 915,806 Year: 3,063/9,624 Month: 908/1,588 Week: 91/223 Day: 2/17 Hour: 0/0


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Design evidence # 111: The heart
Percy
Member
Posts: 22388
From: New Hampshire
Joined: 12-23-2000
Member Rating: 5.2


Message 61 of 82 (33307)
02-26-2003 8:37 PM
Reply to: Message 59 by DanskerMan
02-26-2003 9:40 AM


Sonnikke writes:
That's a non-sequitur if I've seen one. What kind of logic is this? Because something designed can be simple thus complexity cannot be related to design?
No, I believe you've misunderstood what Peter was saying. To exaggerate a little to help make the point, the most complex designs are often Rube Goldberg tributes to ineptitude. Something designed really well is elegant in its simplicity and an icon of efficiency.
This means that you're looking at things backward. You're arguing that the more complex something is the more likely it is to have been designed when actually the opposite is true. It takes much intellectual effort to arrive at a clean and efficient design. If you just consider the genomes of organisms, many seem Rube Goldbergian in their needless complexities with long stretches of junk, redundancies, multilayered enabling and de-enabling of genes, etc. Such tortured and needlessly complex systems seem much more likely the result of random events than of conscious design.
--Percy

This message is a reply to:
 Message 59 by DanskerMan, posted 02-26-2003 9:40 AM DanskerMan has not replied

Replies to this message:
 Message 63 by lpetrich, posted 03-01-2003 1:13 AM Percy has not replied

  
Peter
Member (Idle past 1478 days)
Posts: 2161
From: Cambridgeshire, UK.
Joined: 02-05-2002


Message 62 of 82 (33332)
02-27-2003 2:03 AM
Reply to: Message 59 by DanskerMan
02-26-2003 9:40 AM


Percipient's take on my meaning was my intention, so
I don't really need to say anything else .... but I will
A feature of good design is simplicity, not complexity.

This message is a reply to:
 Message 59 by DanskerMan, posted 02-26-2003 9:40 AM DanskerMan has not replied

  
lpetrich
Inactive Member


Message 63 of 82 (33465)
03-01-2003 1:13 AM
Reply to: Message 61 by Percy
02-26-2003 8:37 PM


Actually, some human designs do tend to acquire that property -- computer source code maintained by several people, for example.

This message is a reply to:
 Message 61 by Percy, posted 02-26-2003 8:37 PM Percy has not replied

Replies to this message:
 Message 64 by PaulK, posted 03-02-2003 3:30 PM lpetrich has not replied

  
PaulK
Member
Posts: 17822
Joined: 01-10-2003
Member Rating: 2.3


Message 64 of 82 (33505)
03-02-2003 3:30 PM
Reply to: Message 63 by lpetrich
03-01-2003 1:13 AM


And that is because the maintenance of computer code is closer to evolution than it is to a straightforward design-and-build methodology.
In both cases incremental changes are added to the original structures, which were never designed to accomodate them.

This message is a reply to:
 Message 63 by lpetrich, posted 03-01-2003 1:13 AM lpetrich has not replied

Replies to this message:
 Message 65 by Peter, posted 03-03-2003 2:12 AM PaulK has replied

  
Peter
Member (Idle past 1478 days)
Posts: 2161
From: Cambridgeshire, UK.
Joined: 02-05-2002


Message 65 of 82 (33519)
03-03-2003 2:12 AM
Reply to: Message 64 by PaulK
03-02-2003 3:30 PM


Yes ... and it's also considered extremely bad practice.
'Spontaneous elabroation' of code works very much like natural
selection::
Add a bit and see if it did what you wanted it to.
If it did, it stays, otherwise modify it a bit

This message is a reply to:
 Message 64 by PaulK, posted 03-02-2003 3:30 PM PaulK has replied

Replies to this message:
 Message 66 by PaulK, posted 03-03-2003 2:45 AM Peter has replied

  
PaulK
Member
Posts: 17822
Joined: 01-10-2003
Member Rating: 2.3


Message 66 of 82 (33522)
03-03-2003 2:45 AM
Reply to: Message 65 by Peter
03-03-2003 2:12 AM


What I am talking about is less than ideal, but a practical necessity.
It is not possible to completely rewrite every program in a suite for every release - not if you want to release on a reasonable schedule. Nor is it practical to forsee every change or addition that might be required at the initial design stage. Then there is the issue of assigning work. So programs acquire functions piecemeal, added by different people - sometimes involving quite radical changes. As this happens that program gets more and more complicated.
(Sometimes - and this IS very bad - it gets to the point where parts of the code are "off limits" for any further change - even to fix bugs).

This message is a reply to:
 Message 65 by Peter, posted 03-03-2003 2:12 AM Peter has replied

Replies to this message:
 Message 67 by compmage, posted 03-03-2003 5:05 AM PaulK has not replied
 Message 68 by Peter, posted 03-03-2003 3:01 PM PaulK has replied

  
compmage
Member (Idle past 5152 days)
Posts: 601
From: South Africa
Joined: 08-04-2005


Message 67 of 82 (33527)
03-03-2003 5:05 AM
Reply to: Message 66 by PaulK
03-03-2003 2:45 AM


PaulK writes;
quote:

(Sometimes - and this IS very bad - it gets to the point where parts of the code are "off limits" for any further change - even to fix bugs).

Unfortunately we have a program like that running at the company I work for. Written in 1989, almost 8000 lines of code and 69 large scale alterations. Nobody here wants to touch the monster. Mention the name and everyone is suddenly far to busy to even have a look at it.
------------------
Signature too long, 100 chars max.

This message is a reply to:
 Message 66 by PaulK, posted 03-03-2003 2:45 AM PaulK has not replied

Replies to this message:
 Message 69 by Peter, posted 03-03-2003 3:01 PM compmage has replied
 Message 73 by Admin, posted 03-04-2003 12:27 AM compmage has replied

  
Peter
Member (Idle past 1478 days)
Posts: 2161
From: Cambridgeshire, UK.
Joined: 02-05-2002


Message 68 of 82 (33555)
03-03-2003 3:01 PM
Reply to: Message 66 by PaulK
03-03-2003 2:45 AM


Believe me I appreciate the practical problems, but a
good software design lends itself to extension
The software development process is evolving itself,
compmage pointed out a 1989 program that is causing headaches
in his/her company ... and legacy code of that age is
always a nightmare to deal with!!
The point being made though, was that good designs are
simple, elegant even. And this is more often the case,
even in software, when there is a single designer and maintainer.
I mainly deal with real-time control systems (closer to
organic systems in concept than large and unwieldy accounting
systems etc.) and these have to be designed well from the
outset, and implemented as designed if you don't want
castrophes.
I think you were agreeing with the simplicity in good
design idea though, since you pointed out that that kind
of development was like evolution ... and results in messy
sub-optimal, over-complex solutions.

This message is a reply to:
 Message 66 by PaulK, posted 03-03-2003 2:45 AM PaulK has replied

Replies to this message:
 Message 70 by PaulK, posted 03-03-2003 4:18 PM Peter has not replied

  
Peter
Member (Idle past 1478 days)
Posts: 2161
From: Cambridgeshire, UK.
Joined: 02-05-2002


Message 69 of 82 (33556)
03-03-2003 3:01 PM
Reply to: Message 67 by compmage
03-03-2003 5:05 AM


What language is that in? 8000 lines seems a little
sedate {edited 'cause I forgot the smiley}
[This message has been edited by Peter, 03-03-2003]

This message is a reply to:
 Message 67 by compmage, posted 03-03-2003 5:05 AM compmage has replied

Replies to this message:
 Message 77 by compmage, posted 03-04-2003 5:33 AM Peter has replied

  
PaulK
Member
Posts: 17822
Joined: 01-10-2003
Member Rating: 2.3


Message 70 of 82 (33566)
03-03-2003 4:18 PM
Reply to: Message 68 by Peter
03-03-2003 3:01 PM


I am definitely agree that simplicity is a hallmark of good design.
I was adding the idea that continued iterative modification with no overall plan is a recipe for complexity.
Unfortunately when you are on the treadmill of product development it is hard to find the time to stop for a complete redesign.

This message is a reply to:
 Message 68 by Peter, posted 03-03-2003 3:01 PM Peter has not replied

Replies to this message:
 Message 72 by DanskerMan, posted 03-04-2003 12:03 AM PaulK has replied

  
nator
Member (Idle past 2169 days)
Posts: 12961
From: Ann Arbor
Joined: 12-09-2001


Message 71 of 82 (33580)
03-03-2003 8:10 PM
Reply to: Message 52 by Coragyps
02-23-2003 10:54 AM


Wow, I've never heard of getting MORE wisdom teeth than the usual four.
I'm sorry, it must have been painful.

This message is a reply to:
 Message 52 by Coragyps, posted 02-23-2003 10:54 AM Coragyps has not replied

Replies to this message:
 Message 78 by compmage, posted 03-04-2003 5:36 AM nator has replied

  
DanskerMan
Inactive Member


Message 72 of 82 (33596)
03-04-2003 12:03 AM
Reply to: Message 70 by PaulK
03-03-2003 4:18 PM


Design Design Design, all this mention of design, simple or complex, it is still *designed*, and didn't "just happen by chance".
What is simple? Who has the knowledge to *know* what is perfectly ideal? How can anyone claim such simplistic deductions (ie. good design is simple) without knowing what is good, what is simple, what is necessary, how to design it, build it, maintain it, etc.
The more interdependant something is, how simplistic can it effectively get? Who can answer something like that absolutely?
It baffles me the most, that designers (computer programmers, engineers, etc) *deny* the Creator, when they themselves are mini-creators/designers.
S.

This message is a reply to:
 Message 70 by PaulK, posted 03-03-2003 4:18 PM PaulK has replied

Replies to this message:
 Message 74 by Quetzal, posted 03-04-2003 1:31 AM DanskerMan has not replied
 Message 75 by PaulK, posted 03-04-2003 2:19 AM DanskerMan has not replied
 Message 76 by Peter, posted 03-04-2003 4:24 AM DanskerMan has not replied

  
Admin
Director
Posts: 12995
From: EvC Forum
Joined: 06-14-2002
Member Rating: 2.3


Message 73 of 82 (33599)
03-04-2003 12:27 AM
Reply to: Message 67 by compmage
03-03-2003 5:05 AM


The software for this website is 43,000 lines of Perl code.
------------------
--EvC Forum Administrator

This message is a reply to:
 Message 67 by compmage, posted 03-03-2003 5:05 AM compmage has replied

Replies to this message:
 Message 79 by compmage, posted 03-04-2003 5:39 AM Admin has not replied

  
Quetzal
Member (Idle past 5871 days)
Posts: 3228
Joined: 01-09-2002


Message 74 of 82 (33602)
03-04-2003 1:31 AM
Reply to: Message 72 by DanskerMan
03-04-2003 12:03 AM


You bring up an excellent point here, sonnikke. However, it works against you as well as against the argument you were addressing: how can anyone claim design without "knowing what is good, what is simple, what is necessary, how to design it, build it, maintain it, etc."? How IS it possible to detect design - whether human or divine - without knowing the actual causative history of the particular object or system? How can we objectively tell that this is designed but that isn't?

This message is a reply to:
 Message 72 by DanskerMan, posted 03-04-2003 12:03 AM DanskerMan has not replied

  
PaulK
Member
Posts: 17822
Joined: 01-10-2003
Member Rating: 2.3


Message 75 of 82 (33604)
03-04-2003 2:19 AM
Reply to: Message 72 by DanskerMan
03-04-2003 12:03 AM


So essentially you are denying that we can recognise design in living beings. That is the consequence of your assertions. But if we cannot then your whole argument falls apart.
And yes, interdependancies are EXACTLY the sort of complexity I was talking about. Good design avoids that sort of complexity, iterative modifications may be forced into adding them.

This message is a reply to:
 Message 72 by DanskerMan, posted 03-04-2003 12:03 AM DanskerMan has not replied

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024