Re: Do developers think more textually than graphically?
From: cr88192 (cr88192_at_NOSPAM.hotmail.com)
Date: 03/19/05
- Next message: Robert C. Martin: "Re: Application logic and Business logic"
- Previous message: cr88192: "Re: Do developers think more textually than graphically?"
- In reply to: cstb: "Re: Do developers think more textually than graphically?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sat, 19 Mar 2005 12:26:11 GMT
"cstb" <jas@cruzio.com> wrote in message news:423B3D3F.8060201@cruzio.com...
>
>
> Marc Rohloff wrote:
>
>> I went to a presentation by Scott Ambler last year and he made the
>> statement, which I will paraphrase from memory (ie it is probably not
>> exactly what he said), that 80% of developers think better in terms of
>> words than pictures.
>
>
> Left brain (sequential logic), rather than right brain (associative
> logic)?
>
>
me: left handed.
ISBT: ISTP/INTP, I "seem" more like INTP (to myself at least), but quite
often score as ISTP.
association skills: often terrible.
I am typically a lot better with structural/statistical thought, and make
associations by recursively modifying something into something else, then
making a little bit of a jump once I notice sufficient similarity in the
structure. details from each are then reworked into other nearby structures,
thus forming an "association".
more free form associations, however, I typically do terrible at. I can't
really "think" in a freeform matter, but do manage to have a level of
"intuition". this, however, does not make much sense to me (intuition
normally implies freeform thought).
things rewrite in my mind and sometimes jump forwards, being highlighted in
a kind of way.
multiple rewrites can occure simultaniously I guess, things can split in
different directions or merge back together. I am unable to fully escape
this though, this is the way my life works and I am sort of locked into a
model driven by my own thoughts and limits in my models...
I have no real "artistic" skill. I can do things like graphics or 3d
modelling. artistic, not really, most things are rather generic...
irl I am limited sometimes.
I am not very "spontaneous", and I can't make arbitrary choices.
I think I fall short sometimes compared with others and their apparently
non-deterministic behavior (or at least, it is not all that predictable to
me, my behavior, however, seems quite predictable to me).
or maybe I am arbitrary and spontaneous and this detail is hidden from me?
my world is often driven by writing code to drive myself as often I am not
much more capable than the software I write. one would think they could
drive themselves by some kind of impulse, rather than relying on the mass
generation of commands and writing/processing to drive their lives...
I can't escape this, without it I am somehow flatlined, yet often I am still
driven by impulse (or more like: able to procrastinate and not get crap
done...).
>>
>> I found this an odd statement and wondered that, if this is true (He
>> didn't give a source), then why is there such a push to graphically
>> orientated design methods. If developers think better in words then we
>> should use words not pictures.
>
>
> Interesting notion - might be that the minority are frustrated
> by the lack of tools *they* want, but that others find pointless.
>
I wasn't clear on this statement...
> Possibly related to another phenomenon - At one point, I wanted
> a developer to draw a high level diagram of a system, so I could
> grok it. Would you believe the (quite talented) developer did NO
> visualization, and could not imagine such a diagram, or what
> it would be good for? Turns out the same developer was (at least
> partially) blind for part of the "formative years". Never has
> figured out what people would use those goofy diagrams for.
>
>
imo, diagrams *are* pointless much of the time.
why do we need a diagram, when a chunk of text or psuedo-code could go over
it much more easily (sometimes, for this kind of thing, psuedocode seems
easier as one can gloss over things and keep it short, but then one realizes
irregularities in the conventions and sometimes difficulty in portraying
what is really wanted...).
>> I also wondered about how the way we think relates to the quality of
>> what we develop.
>
>
> The quality of whatever way we think
> relates to the quality of what we develop.
> I think.
>
probably.
>
> I play chess in a mostly sequential logic way,
> similar to the way I might program it.
> But I've played against people who describe
> playing in a mostly parallel/associative way.
> Not just conceptual patterns, but literally thinking
> in terms of 'pictorial, visual forms' of the structure
> of various positions. I have an onyx chess set, and
> the bishops and pawns are only distinguishable by
> size - bishops are taller, but identically shaped.
> I discovered it was particularly advantageous to
> fianchetto one or both bishops when playing against
> such people. They did/could not 'reason' their way
> to keeping track of the bishops, and would 'lose' them
> in the visual models they were using to think with.
> Sequential players don't seem to suffer this effect.
>
I don't know, I don't play chess...
err, I feel unsure here.
I think my thoughts can be parallel, but not really associative.
eg: I can read in and "execute" code in my head, and get a rough idea where
errors are in many cases or general algorithmic problems (sometimes detail
bugs manage to escape me, most particular those realted to typing errors or
using the wrong variable).
sometimes I "see" fragments of the code I am thinking of, or other things...
I can "see" a bitstream, and I can transpose it (eg: high-low or low-high
order). then I wonder, is the ability to see something like this (as opposed
to all the bitstream stuff being hidden or represented otherwise) count as
visual/associative thought?...
likewise, I also seem to have vaguely graphical depictions of some algos,
most notably ones I am weaker on. eg: arithmatic encoding, for example, has
a presently more visual representation, where I can "see" the math going and
the numerical convergence. I can "see" the bits that go in or come out, and
approximate the calculations close enough to predict the behavior (sometimes
catching on small details though, eg, (a-b)*(c-d) vs (a-c+1)*(c-d) or
(a-c+1)*(c-d+1) when using fixed-point math...).
often I can "see" that changes in an algo would help or hurt, eg,
performance and data size, or other factors, though I can't conciously "see"
anything, or accurately evaluate the algos in question (apart from writing
and trying them, or "zooming in" so that I see the behaviors in finer
detail).
quite often these patterns eventually get replaced by code, or just get
internalized.
geometric things are often represented grahpically.
now, I don't think this is my general thought pattern, sometimes I start
writing and the code writes itself before I type it. or I start talking
about it and the code writes itself, or whatever...
I can't tell how I would do at chess...
>
>> Do better developers really think textually or, when
>> you hear about the developer who can work 20 times more efficiently
>> than his peers, is he the one that thinks visually?
>
>
> Nope - those are the Smalltalkers.
> And some of them rely more on text, others do more visualization.
>
I guess I don't fully understand the topic...
or maybe I am insane, I can't really tell.
my sanity has been a notable topic of debate, with myself...
- Next message: Robert C. Martin: "Re: Application logic and Business logic"
- Previous message: cr88192: "Re: Do developers think more textually than graphically?"
- In reply to: cstb: "Re: Do developers think more textually than graphically?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
