Rail - 'Now what has our new friend here actually said...
On Fri, 21 Aug 1998 09:33:13, uunet!netcom.com!bobmacd+NNcsy282198 wrote:
> kiyoinc@ibm.XOUT.net (cory hamasaki) writes:
>
> >For example, 4-6 months ago, there was a huge battle in c.s.y2k on rail.
> >My office is close to a rail switch yard that CSX closed down about 5
> >years ago. I ran my keyboard and claimed that the fact that the
> >physical switchyard had disappeared in to cyberspace was a problem.
> >About 1980, I had seen the assembler language source of a system that
> >switched freightcars. If that system dies on Y2K, railfreight stops and
> >if rail stops, everything stops, I claimed.
>
> And this was only one example of the truly *byzantine* fabrications,
> hallucinations and misinformation that I've seen on this group, regarding
> rail. And I'll happily correct the rail data all day; that's not the
> point. The point is, if you could be THAT far off on rail, maybe you're
> a little off on some of the other industries, too.
He says, I'm wrong and he's right.
> Now to explain what happened there was pretty simple. You notice that
> your office is near a rail switch yard. That means that switch yard
> is probably in some pretty valuable real estate. Now, railroads'
> primary business is transportation but they have a strong sideline in
> _real estate_. They know that yard was sitting on valuable land.
> They knew it would make more money if it were developed, than it ever
> could as a rail yard. So they closed it, and are either developing
> it or they sold it to someone who is.
He's talking about real estate values, not rail.
> So how can railroads operate without that yard? Several ways.
> First, there's been a trend in the last 20 years away from "boxcar"
> traffic, which must be classified and sorted, into "trailer on flatcar"
> and "container" traffic and "RoadRailers". This traffic is classified
> in a new breed of intermodal facilities, often at deepwater ports.
>
> Second, yards are much more efficient, so they're smaller. Railcars
> used to sit for weeks, now they sit for hours. If you process 100
> cars a day and it takes 14 days on average, you need room for 1400 cars.
> So close a bunch of small yards and consolidate them into a few big,
> efficient yards. Now you have a big yard that processes 1000 cars a
> day and averages 2 days, so the bigger yard is 10x as efficient.
> Then they do hub-and-spoke service to the areas that used to be served
> by the small yard.
The Arlington - Alexandria yard was the mega-yard for the Middle
Atlantic. It wasn't consolidated into an ultra-super-hyper mega-yard...
..except it was, according to the Washington Post's articles during the
fight over putting the Marlene Cook stadium at the site, the Post stated
that the functions of the yard would be moved to cyberspace.
> That's where your yard went. It had nothing to do with computers.
> It had to do with improving processes. This process improvement
> uses computers in part, but it's not totally dependent on them. The
> major improvements are 4 major railroads instead of 50, and hub-and-
> spoke operation out of few large, high-efficiency yards.
>
Uhhhhhh! Pointy-hair alert.... "process improvement"
> Your circa-1980 program was probably a computer _model_ of what was
> being done in real-life. You can't switch cars in Cyberspace unless
> you're Tron. :-)
I believe it was a two step process. First they learned to track the
individual cars using computer databases, next they discovered that
precise tracking enabled them to solve the switching problem using
distributed resources.
The use of emphasis on "_model_" is very suspect. What does that mean?
It's *only* a model, so it can't do much? Sorry, there are things that
flat don't work if the software doesn't work.
> As far as your concern about the rail industry crowbar-ing over yard
> switching, the theory has a couple of flaws. First, the "Iron Triangle"
> traffic (fuel) moves in unit trains. Unit trains are solid trains of
> one commodity, like coal, containers, or people (Amtrak). If the
> classification system goes into the crapper, they won't be able to sort
> the boxcar traffic that railroads traditionally handled, and the yards
> will back up momentarily. But if they can't find a way to classify cars,
Which is what we're saying is the problem.
> they'll stuff the boxcar trains anywhere out-of-the-way, and clear
> the mainlines for the unit trains. At that point it'll be pretty quiet
> on the railroads, and coal trains will have the run of the system.
> Even if there's dispatching problems it's pretty easy to dispatch a
> railroad when the only things moving are coal trains. (and Amtrak, so
> put the coal trains on routes Amtrak doesn't use.)
Fuel moves... what about food? Who sets the priorities. What about
essential chemicals used for manufacturing parmaceuticals, are you
hypothesizing a national freight prioritization agency? Don't forget,
if you are in a car accident, well, that's unfortunate, I'm sorry.
..but if I'm at a restaurant and they're out of the special of the day,
that's a disaster.
> Second, powerplants aren't exclusivly fed by rail. Gas and oil move by
> pipeline. Even the railroads get their diesel by pipeline. (and if
> their reserves run out, they'll find cars in the yards with corn oil.)
> Coal-fired powerplants are also fed by ship, especially in the Great Lakes.
Sure, and we could have people on bicycles moving fuel like ants. You
haven't stated anything that we don't already know.
> >This set off a huge flamewar... I loved it, troublemaker that I am.
>
> By people who knew very little.
>
> >It got bizarre as only c.s.y2k can get. On one side 'Dyno' Don Scott and
> >'Bad' Brad Sherman. On the other, 3-4 known troublemakers and a few
> >specialists who worked on rail systems.
>
> By people who worked on rail _computer_ systems, but who knew nothing
> about the grunt-and-tumble of actually being out there, on the steel
> and stone with a timetable in your back pocket. Somewhat arrogantly,
> they decided that without their work, the industry would STOP. No.
> The industry would improvise.
So are you a programmer at RAILINC? How's the move going? Are they
running SNA or TCP/IP? What's the host database? How many lines of
code? How about posting a few record layouts? What's the schedule look
like? Is GEISCO on track?
> > Namecalling, butt-head, http:
> >links, Sherman found webcams of switchyards,
>
> That's what blows me away about this group. You will accept on faith
> that rail-yards do not exist anymore and "cars are switched in cyberspace"
> and the only contrary evidence you'll except starts with "http". Far
> be it from anyone to NOTICE the railyard adjacent to the freeway they
> drive every day, let alone calling up the local RR and asking where
> their yards are.
Actually Sherman found pictures of switchyards out west. I drive by the
place where the Arlington - Alexandria switchyard used to be. The
physical evidence is of a yard that is gone... the cyber-evidence is an
image of an existing yard.
> >Finally Paul Makinen assembled a complete picture, it took him a couple
> >weeks but he obtained original source documents that showed that 1)
> >the RAILINC application was non-Y2K compliant, 2) they were moving the
> >operations center and the development center to different cities, 3)
> >RAILINC had outsourced the Y2K work to GEISCO, 4) GEISCO was desperately
> >trying to hire programmers for the rail project... in the hottest
> >programmer job market the world has ever seen.
>
> Well, I don't know about that. I do know what the unanimous response
> of every railroad operating employee will be if the computers take
> a dump. "Then we'll do it the old-fashioned way." And they will,
> even if the computer people tell them it can't be done.
>
> This is the third flaw in the theory. The belief (by computer people)
> that given a computer failure, people will respond by
> 1. put Crisco on thumb
> 2. place hand on chair, thumb up
> 3. sit down
> 4. wait.
No, that's not our assumption. We know that people will pull all
nighters, scream into phones... as long as the phones work; undo safety,
dead-man, switches, do heroic and amazing stunts, slam procedures into
place bypassing peer review.
We know that things will get done... but we also know that people will
die (again, that's unfortunate, I'm sorry... but the Salad Bowl Express
must run.) We know that all the efficiencies that were built up over
the last 30 years will be swept away in a single Krystal ni... oops,
night of the living dead.
It's the discontinuity that's the problem... On the other hand, if they
had started 3 days ago, there was a chance to engineer and implement
layers of contingencies. They didn't, it's too late, there isn't
enough time now. There's no point to inventing contrived reasons why it
won't be horrible, the 500 day decision point came and went, it's time
to move on.
> Nuh-uh. People (*especially* rail people) will do WHATEVER IT TAKES
> to get the job done. In the short term that probably means shoving the
> boxcar business aside and concentrating on the unit trains. In the
> longer term that means working out permanent solutions. Yes it will
> have an impact on the economy, but not crippling. Just look at 1997.
You're agreeing with the worse of the doomsayers, just darting your eyes
around and saying, I hope it won't be that bad. It's OK to have
disturbing thoughts, bad things do happen to good people.
You and I have no way to quantify the impact. Will it be horrible,
crippling, or only an economic disaster? Is it fair to say 1997 wasn't
bad? It didn't affect me, I still got my lunch but people lost jobs,
money, pulled all nighters, and screamed into the phone. It depends if
it's me or some nameless you, out there.
> Computers did not get Union Pacific out of its 1997 meltdown. (just
> as they did not get SP out of precisely the same meltdown in 1974.)
> Computers didn't put UP there, either, by the way. One too many yard
> consolidations did. Houston was a 3-yard town and UP tried to make it
> a 1-yard town to improve efficiency, "coz that's how UP does things".
> This overloaded Englewood yard, just as SP had done in 1974, and TSHTF.
> (everyone warned UP that this would happen, but they were too arrogant.)
> It took longer to unravel in 1997, because it coincided with a surge in
> plastic business "stored in transit" that overloaded the yard further.
> Plus the UP system was bigger, so it fell harder. And UP wasn't bold
> enough to try SP's 1974 solution: piss off local shippers by stuffing
> their cars down a side track for a month to make room in the yard.
> Englewood 1997 wasn't a computer thing. It was an operations thing.
Whatever... to c.s.y2k. our new pal here, the guy with the black and
white striped puffy hat, overalls, and the oilcan with the long spout,
is talking about one small boo-boo that turned into a billion dollar
economic disaster affecting lots of people and businesses.
This boo-boo happened while communications still worked, power was fine,
and the computers all worked... but somehow, the professionals couldn't
unthread whatever was wrong for months. Nothing was wrong but they
still couldn't fix it. In 497 days, the computers break and ... what
.. no problem, trust me, professionals will be working on it.... nobody
panic! Who can I knock down to get to the exit?
> >It was a picture of a disaster in the making. After Paul's posting, the
> >rail issue vanished from c.s.y2k, asked and answered, rail is a big
> >problem.
>
> Rail systems are a big problem. They will hurt productivity in the
> rail industry. They won't disable the industry.
The doomsayers would agree with the first two statements and politely
suggest that there is no evidence to pick disable v. not disable. I
believe that it will be much worse than 1997.
> > Now, months later, I regularly see rail information in the
> >press, on TV, and Senator Bennett is hammering on the rail industry.
>
> Senator Bennett is grossly misinformed. The information he reported is
> absurd. I should offer him a campaign contribution in exchange for the
> address of that warehouse in NY with all the switchpoints "from several
> railroads". I supremely doubt such a place exists.
Please do... forget the campaign contribution, he doesn't need your
money. Please track down the NY warehouse story... the second part of
it is that the switches are being shipped overseas.
> >It looks like someone fed him portions of the c.s.y2k discussions and
> >his staff is quizing the rail industry.
>
> That, then, would be why he's misinformed :-)
What! The rail industry is misinforming his staff? The nerve of them...
> >Here's a c.s.y2k heads up. I have a sense that banking is about to blow
> >up in our faces. Something about the way the news has been running just
> >doesn't seem quite right.
>
> I'm sick of hearing about "doesn't seem". Or "sure looks like". Or
> someone looking out their window at something innocuous and putting it
> together with something else and calling for doom. (not to pick on
> you cory, just picking the first examples that come to mind.)
>
> I wish people would get their facts straight. I would've pinched the
> whole rail thing in the bud if I'd been around, and if I'd been believed.
> There are legitimate problems in rail, I'm sure. Let's concentrate
> on those and stop wasting our time talking about bullshit.
Too late bob, 497 days now. Anyway, you've just given us a look at a
glass that is more than 1/2 empty.
> -Bob
And here's another stray synaptic firing for all of you... I used to
think that landline telegraph, rail-morse was there to serve the towns
and that the town telegraph office was at the railroad station because
the wires ran along side the track using the rail right of way. It was
sort of a coincidence, I thought.
After reading railroad-bob's article, I understand that the telegraph
office was there for the railroad, it was a primative computer switching
system based on brass pounding telegraph operators, chalk boards, paper
and pencil. As schedules slipped and trains started off not-on-time,
bearing down onto other trains head-on on the same track, it was
important to pass information ahead of the train, Old 98 is late, get
ready.
In 498 days, the systems that do this on a grand scale break. Keep in
mind that paul makinen found the press releases from RAILINC on their
Y2K efforts.
Railroad bob may not have had this intent but he has written a
convincing argument for computer problems causing shipping problems on a
massive scale.
cory hamasaki 497 days... all the livelong day.
x10.dejanews.com |
