Re: 7drl starting announcement
- From: eyenot@xxxxxxxxx
- Date: Fri, 8 Feb 2008 11:17:23 -0800 (PST)
On Feb 8, 11:12 am, Craig <craig.stic...@xxxxxxxxx> wrote:
On Feb 7, 6:45 pm, eye...@xxxxxxxxx wrote:
I consider myself amateur. I've never coded for work, or with others
on a project.
BIOS work seems to me to be just as simple as necessary.
I see BIOS work and ASM skill as an indicator of your affinity for
coding. I believe it's a mentality rather than a skillset: rather than
simply study abstractions and data access models you have actually
gone out of your way to learn *how* the computer thinks.
I'm curious... in an era of high-level programmers and interface script-
kiddies, how'd you get into such special niche?
Well I started programming when I was younger, around 8 years old, on
some ATARI home computers and a Sinclair ZX-81. Then there was no
computer until I was about 12, at which point I got a 286 6/12mhz
during the time when 486 dx4/100 had just come out. And all the
literature I could get my hands on for computing, besides my then-
defunct "Basic Atari BASIC" was old stuff about switch states and
logic gates. I eventually delved into Borland TurboPascal and MS
QuickBasic, but I typically ran into a lot of problems. For instance,
I was still trying to optimize code to fit between 65535 line numbers
but with enough gaps that unforseen developments would still fit in
without too much goto-spaghetti. A friend showed me about procedures
and functions and I had a happier subroutine family (I still call them
all routines and subroutines to this day).
I ran a BBS for awhile in the 90s and was known around my area code
for setting sysops up with personalised BBS kits, like three BBSs for
a user to choose from that would switch from one to the other without
hanging up on them or scaring them, etc. I usually had to jury rig it
(T.A.G., the shareware bbs we all used around here, had only recently
gone into threaded support and most of us were still using unthreaded
versions) but it worked. I was also a decent intermediate ANSI artist
and cranked out an ANSI/lit group with its own displayers for awhile.
The whole BBS scene died down when I was about 16. That sucked because
I still hadn't made a popular door game and I had always wanted to. I
remember trying out sort of a sequel to one but the two libraries I
had gotten ahold of for writing doors in pascal refused to output
through the modem. I had to give up.
I wasn't very sophisticated myself in the area of programming, though,
and would constantly have to crawl over my own work like an obstacle
while I was designing it, and had to throw a lot of things out because
"they wouldn't work". I hadn't yet learn the subtleties of *making*
something work, and really I was still too absent minded and unfocused
to be any good at programming.
When I was 21 (eight years ago) I bought my first 486 dx2/66 and was
thinking about whether or not I should upgrade to a dx4/100 when I got
a huge tax return and decided to go all-new. I spent a lot of time
researching cooling fans and bought the newest board most favored by
overclockers at the time and played around with overclocking and
cooling, but the board turned out to be crap. I eventually gave the
computer away along with most of my books and cd's and hit the street.
I've been homeless for three years, now, and in that time I've managed
to become a lot more mentally present and more logical overall. Not
that I hadn't been working on it, but sitting on my ass and getting a
size 46 gut didn't seem to be very logical, really. Now I'm doing
better, haven't lost my glasses or wallet in years (I used to lose
these periodically, at least once a year) and when I encounter
problems in coding I tend to instantly have a solution. That is, to
anything code-related: real-world problems like being unwilling to use
up too much RAM or reading/writing to files on a server just have to
be dealt with normally, and simultaneous function just has to be
restrained. There are myriad interesting ways of doing things: for
instance, one thing I had planned with the psuedo-x86 was to "write to
disk" by generating a 256-color png (or similar, low-level image type)
and sending it as an image to the browser for the user to click and
save.
I still use goto sometimes today just to solve little problems that I
don't feel like twisting (or straightening out as the case really
usually is) into loops. A lot of people are dissuaded from using it by
computer teachers/profs who say it's somehow wrong to goto, that you
need to learn to do,for,while,when,switch everything. But not all of
those exist in every language and there are some problems sometimes
depending on how you've set up your compiler's behaviours, for
instance not being able to for-loop your way through every iteration
of a byte without checking it against a short int. And ultimately,
jumps and labels are still used albeit in a more tokenized form than
human mnemonic tendency. Which brings me around to the point of
mentioning gotos, the jump command. There basically wouldn't be any of
what we consider programming language without sophisticated use of the
jump command. And there you have your goto. But the same teachers/
profs will also tell you you shouldn't bother with ASM unless you're
trying to squeeze cycles, and in this day and age when timeslices are
*long* gone and the latest processors don't blink twice whether you
mention your 2^23 iterations are in pure ASM or Visual Basic. And
today we have this mess, where Sun-JAVA is the first language and
there's all this talk about being "object oriented".
In any case, all these new things that kept coming along: "oop",
"robust", "sdk", I looked into them all the time and either they
seemed not to make any sense (I admit, I *still* to this day am not
entirely sure *why* anybody ever wanted to differentiate one kind of
programming from another and say one's "object-oriented", but I think
the answer has something to do with bloat and investment) or they
seemed to be hype (e.g. "robustness" in a language where the
conventions seem always on the brink of being changed and make roughly
nobody happy) or they seemed to be pointless (the win32 "sdk", to make
it easier for developers to code on a graphic-user-interface platform
that nonetheless has a functional, readable, user-interfacing skeleton
size of around 6k -- *in ASM*). I mean, I just couldn't grasp any of
it. My mind was still trying to come up with something better, either
better for the end-user or better in terms of getting myself to the
point of a distribution. And none of that seemed to be helping, so I
stayed away from college and studied online instead. When I felt like
programming I mostly rummaged through formatting and protocol texts,
and that more or less kept me safely away from the new. I mean, I kept
asking myself, what's the point of pushing 32 bits through a bus if 24
of them are stinking worthless?
And along the way I found out that there are all kinds of hobby and
interest groups that take after the 8-bit, and that helped me feel
more refreshed and less anxious about memories of better, more 8-bit
times. It helps to have an internal image, I guess. My happy vision
has something to do with a colorful, easy-to-use and highly
functionable, customizable, and macro-compiling text-mode interface. I
guess that's why I stuck to this "niche".
.
- Follow-Ups:
- Re: 7drl starting announcement
- From: Craig
- Re: 7drl starting announcement
- References:
- 7drl starting announcement
- From: eyenot
- Re: 7drl starting announcement
- From: Craig
- Re: 7drl starting announcement
- From: eyenot
- Re: 7drl starting announcement
- From: Craig
- 7drl starting announcement
- Prev by Date: Re: 7drl starting announcement
- Next by Date: Re: Hex dungeon generation anyone?
- Previous by thread: Re: 7drl starting announcement
- Next by thread: Re: 7drl starting announcement
- Index(es):
Relevant Pages
|
