"Pete Dashwood" <dashwood@[EMAIL PROTECTED]
> wrote in message
news:62osr8F2492foU1@[EMAIL PROTECTED]
>
>
> "SkippyPB" <swiegand@[EMAIL PROTECTED]
> wrote in message
> news:jfods3h3evo9r5br875lu9iqv0547e7g6i@[EMAIL PROTECTED]
>> On Thu, 28 Feb 2008 11:21:13 +1300, "Pete Dashwood"
>> <dashwood@[EMAIL PROTECTED]
> wrote:
>>
>>>
>>>
>>>"SkippyPB" <swiegand@[EMAIL PROTECTED]
> wrote in message
>>>news:4f7bs35q4kcbp71vbd88irhimoun85qksb@[EMAIL PROTECTED]
>>>> On Wed, 27 Feb 2008 09:36:44 +1300, "Pete Dashwood"
>>>> <dashwood@[EMAIL PROTECTED]
> wrote:
>>>>
>>>>>
>>>>>
>>>>>"tim" <TimJ@[EMAIL PROTECTED]
> wrote in message
>>>>>news:13s728si8g08id9@[EMAIL PROTECTED]
>>>>>> On Tue, 26 Feb 2008 10:46:19 +1300, Pete Dashwood wrote:
>>>>>>
>>>>>>>
http://www.wa****ngtonpost.com/wp-dyn/content/article/2008/02/23/AR2008022300693_pf.html
>>>>>>>
>>>>>>> What I'd like to know is exactly HOW you can establish yourself as
a
>>>>>>> Geek,
>>>>>>> so you can claim this defence. If writing a file system for Linux
is
>>>>>>> all
>>>>>>> it
>>>>>>> takes (a weekend's work for a COBOL programmer...), then there are
a
>>>>>>> lot
>>>>>>> more Geeks around than many people may suppose.
>>>>>>>
>>>>>>> Did he do it? (Jo Brand, referring to O. J. Simpson: "Course he
did
>>>>>>> it;
>>>>>>> he's a bloke...")
>>>>>>>
>>>>>>> Thoughts?
>>>>>>>
>>>>>>> Pete.
>>>>>>
>>>>>> You must be pretty good if you can write 15,000 lines of
>>>>>> multi-processing kernel code in a weekend.
>>>>>
>>>>>Well, leaving aside the fact that I AM pretty good :-), it was said
>>>>>with
>>>>>tongue-in-cheek...:-)
>>>>>
>>>>>As a matter of record, I once wrote a complete access method for a
>>>>>mainframe
>>>>>in a Bank, over a weekend. Another dedicated guy and myself worked on
>>>>>implementing it into about 300 programs over the same weekend. It
>>>>>enabled
>>>>>direct access based on Account numbers and had its own ha****ng
>>>>>algorithm
>>>>>based around the structure of account numbers in that particular
Bank.
>>>>>It
>>>>>was a complete callable subsystem that implemented all the functions
of
>>>>>data
>>>>>maintenance using VSAM RRDS. On Monday, everythng was working when
the
>>>>>staff
>>>>>showed up. We had a total of 9 hours sleep each and were given the
rest
>>>>>of
>>>>>the week off... :-) It was a long time ago; I wouldn't/couldn't do it
>>>>>now
>>>>>:-)
>>>>>
>>>>
>>>> So you're the guy that did that!! First job I had out of college was
>>>> working in an all IBM Assembler shop. They had about 20 different
>>>> applications consisting of around 300 programs or so. One of the
>>>> first changes they wanted was to convert their file access system
from
>>>> DA to ISAM (which was the soup de jour of the day). The reason was
>>>> noone, and I mean noone myslef included, could figure out how the
damn
>>>> thing worked! It did work but neither I, the resident Assembler
>>>> programmer or the account's IBM system engineer (back in the days
when
>>>> these guys could write code with the best of them) could figure out
>>>> how it worked. The person that wrote it and moved on to another job
>>>> in another state and wasn't available for questions. This wasn't a
>>>> bank, but I'm going to blame you for my many nights of writing code
to
>>>> change the file access. :)
>>>
>>>Not guilty. Mine was a "black box". I have to ask, though, if it
worked,
>>>and everyone agreed it worked, why change it?
>>>
>>>Pete.
>>
>> They were changing all of their applications from IBM Assembler to
>> Cobol and the access system couldn't be rewritten in Cobol. There was
>> no suggestion of making it a subroutine, which could have been done.
>> And, as I recall, it had a limitation of only being able to work with
>> 2311/2314 disk. Another upgrade was going to be installing 3380 (I
>> think that's what they were) drives. So while it wasn't broke, it was
>> not ****table or upgradable.
>
> Hmmmm... if the Assembler used standard housekeeping for IO it shouldn't
> have mattered what device it was on. I suspect they just wanted rid...
:-)
>
> A callable subroutine would have been better, I would've thought.
>
> Thaks for posting the response, Steve.
>
> At least we know the reasoning, whether it was right ot not... :-)
>
> Pete.
> --
> "I used to write COBOL...now I can do anything."
>
A lot of direct access programming (BDAM) was device dependent and based
on
numbers of cylinder, heads and tracks. There were some macros and
techniques
that could be used to make it device independent but they were not always
used.
|
