home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Monster Media 1993 #2
/
Image.iso
/
finance
/
lqw125.zip
/
LQW_DOC.EXE
/
UBCCMICR.DOC
< prev
Wrap
Text File
|
1992-12-26
|
22KB
|
441 lines
This document is taken from a ZIFFNET discussion of the problems arising from
producing MICR encoded checks on a laser printer. In summary there is general
concensus that the technology is acceptable when special MICR toner is used.
In contrast, there is stark disagreement about the merits of producing laser
checks with ordinary toner.
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: Sal Cambareri 72137,3641 # 233390, 2 Replies
To: [F] All Date: 29-Nov-92 11:15:27
A friend recently showed me Dvorak's comments on the HP press release for
their new MICR font cartridge. I'd like to offer a few of my own.
For once HP is a day late. Digital Design of Jacksonville FL has been
selling an excellent but expensive laser check printing system for at
least 5 years. UBCC of Palatine IL also has been selling a MICR font
cartridge ($250 and can use ORDINARY toner) for at least 3 years. And
more recently Elfing Consulting of St. Charles IL has been selling a MICR
soft font ($35). I'm sure that there are several others.
. . . . . . . . . . . . . . . . . . I've managed to write a DOS program
than can capture Quicken's check output and produce fully negotiable (MICR
encoded) laser checks on the Duplex stock using either the UBCC or Elfring
fonts. I have a beta/demo version available for the curious and/or
skeptical.
Sal
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: William Beem 72241,1111 # 233506, 1 Reply
To: Sal Cambareri 72137,3641 Date: 29-Nov-92 22:21:06
There's more to producing a MICR line than the font itself. MICR lines
are magnetically encoded. If you don't provide a toner cartidge that
supports the ANSI spec for a MICR line, your check will get rejected when
it goes through item processing at the bank.
An occasional reject won't raise a flag, but if all of your checks get
rejected, the bank will start charging a fee for each and every rejected
item.
Like I said, it's not just the font cartridge that makes a MICR line.
Before you implement something like this, I suggest you speak with your
bank and learn their policy regarding reject items.
--Bill
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: Sal Cambareri 72137,3641 # 233591, 2 Replies
To: William Beem 72241,1111 Date: 30-Nov-92 09:02:02
William,
A couple of notes:
>>ANSI spec<< The reference is to the E-13B Magnetic Code Printing spec.
>>MICR toner<< A well kept secret is that all laser toner is iron based
hence magnetic. What the MICR toners give you is perhaps marginally more
iron but more importantly enhanced "ruggedness" of the toner image. This
is important as checks have a rough life since they are run through MICR
readers which can abrade the MICR image. However experience has shown
that the MICR toner is not essential to get this result. UBCC takes a very
clever and much less expensive approach:
UBCC provides a font cartridge that causes the MICR to bond well with
special laser safety paper ($20 to $25 per thousand). They have installed
this in a hundred accountancies which each print checks for all their
clients. Because they are accountants I'm sure none of them spends an
extra nickle for magnetic toner. It's not my place to give out anyone's
client list, but Ken Garen of UBCC (70233,1662) will probably put you in
touch with someone local to you so you can see for yourself.
>>ask your bank<< Why bother. Like any other business whose profits
depend on the ignorance of the consumer (remember MA Bell) they will say:
"Don't take a chance! Buy my (outrageously priced) product and you'll
never have a problem."
Sal
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: William Beem 72241,1111 # 233614, 1 Reply
To: Sal Cambareri 72137,3641 Date: 30-Nov-92 10:26:15
I currently work at a check processing facility, so my view on this issue
is a little biased. In other words, we hate people who 'slip through'
with ignorance.
However, we feel that everyone should be educated. That's why we charge
you a tuition fee for MICR reject items.
--Bill
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: Sal Cambareri 72137,3641 # 233638, 1 Reply
To: William Beem 72241,1111 Date: 30-Nov-92 12:52:04
>>reject fee<< That seems a little harsh. I guess some check processers
are a little more forgiving than others. My brother-in-law is using
preprinted non-laser checks, provided by somebody other than Quicken, with
his Quicken 6.0. He told me that "every check comes back in an envelope."
I told him to return the checks to the printer, but he can't be bothered
since it's not costing him anything. Maybe things are just different in
Idaho.
I didn't mean to imply that nothing could go wrong when doing laser MICR
encoding. With weak software you can have misspaced or misplaced MICR
characters or non-MICR magnetic printing within the sacred final 5/8" of
the check. You can also have operator error. For example, you could give
the software the wrong account or transit numbers. Finally you could have
hardware problems: Simply having a laser that puts down too much or too
little toner can produce MICR that becomes defective on successive
readings.
All that's why I think it's important that people "try and buy" before
committing to laser MICR printing. As you say most clearing organizations
will allow a few bad checks through before demanding retribution.
Sal
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: William Beem 72241,1111 # 233644, 1 Reply
To: Sal Cambareri 72137,3641 Date: 30-Nov-92 13:20:21
> >>reject fee<< That seems a little harsh.
It's not harsh at all. Rejects require manual processing. That
interferes with automation and raises operational costs. We simply pass
that along to the person who wrote the check, because it's easily
avoidable.
--Bill
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: Sal Cambareri 72137,3641 # 233650, 1 Reply
To: William Beem 72241,1111 Date: 30-Nov-92 14:26:08
>>Rejects require manual processing<< The only reason that MICR rejects
require manual processing is because many clearing houses use buggy whip
technology. The essense of the argument for the fairness of "tuition" is:
Because the clearing house's MICR machine can't read what is clearly
printed on the paper it can penalize the customer. I know that other
businesses which lack the benefit of regulation suffer when they show such
little respect for the customer.
In law there is at least a principle called mitigation: If I destroy
your VW rabbit you can't replace it with a Dodge Viper and expect me to
pay for all of it. I don't believe that "manual" processing provides
mitigation. The clearing house should try other automated processing eg
OCR scanning before going to manual processing and leveling any fines. So
yes, the tuition does seem harsh to me.
From the outside looking in it appears to me that MICR readers are a lot
more sensitive and prone to error than optical character readers. When
you consider that the MICR must be so precisely sized, shaped, positioned,
and spaced and the reader only has to be able to recognize 14 very large
symbols it's appalling that it ever fails to read correctly. I would bet
that if MICR were backed up by even a minimal OCR system (looking at only
at the MICR line) there would never be any rejections.
Sal
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: William Beem 72241,1111 # 233765, 1 Reply
To: Sal Cambareri 72137,3641 Date: 01-Dec-92 08:53:23
> From the outside looking in....
That's right, you are on the outside looking in. As someone who's
currently standing inside a check processing center, I can tell you that
you're right, and you're wrong.
Image processing has been a goal here for at least the last five years.
It'll probably be another five years before the IP system gets converted
to Image, and that's an optimistic projection.
Simply telling us that our systems are out of date is no solution. When
you process 13 tons of checks every day, it's a little difficult to yank
out one system and replace it with new one.
Your description of penalizing the customer because "the clearing house's
MICR machine can't read what is clearly printed on the paper" is a bit
convoluted. There are standards in place in the financial industry.
They've been here for decades. PC and LaserJet technology is almost an
overnight sensation in contrast with these standards. Most banks haven't
been able to adopt OCR/ICR technology for their main clearinghouse work.
I can't blame them, either.
Although Image processing is theoretically better, there's a large
infrastructure that has to be laid before it can be implemented.
Bandwidth and DASD space has to be expanded. Dump 3279 terminals need to
be replaced with high end PC's. System and application software has to be
evaluated, purchased, customized, tested, and re-tested. All of the
little ad-hoc reports need to be recreated to work with the new system.
The staff needs to be retrained. Corporate customers need to understand
their new requirements and also have to be trained.
There's more to Image processing than buying a Logitech handscanner and
some OCR software. Let's get real.
--Bill
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: Sal Cambareri 72137,3641 # 233802, 1 Reply
To: William Beem 72241,1111 Date: 01-Dec-92 11:41:21
I didn't mean to suggest anything as radical as "yanking out one system"
and replacing it with another. I just offered that it might be less
costly than manual processing to have a minimal OCR secondary system to
process the checks rejected by the primary MICR reader. i.e. A simple add
on.
Bill, I can certainly see that your clearing house faces some truly
formidable obstacles in getting from where you are to "image processing,"
and I both sympathize and agree that it is a real problem. I just
disagree with you about WHOSE problem it should be: You seem to think
that it's the customer's problem (reject fees) while I think that it
should be the stockholder's problem (competitive disadvantage.)
If the banking industry were truly competitive (capitalistic in the best
sense of the word) then the businesses that can adapt and use the most
efficent technology would thrive, and the others would not because
educated customers would always seek out the best deal for themselves. In
a heavily regulated (socialized) business such as banking the business can
mistreat ignorant consumers with cost-plus charges such as reject fees.
Here's a case in point: I have been using the UBCC MICR laser check
system for about 4 years for both business and personal checks. I've had
several bank accounts in two states (OR & NY), and I've never had a
rejected check. However if a bank ever charged me a reject fee, as an
educated consumer, I would IMMEDIATELY close that bank account and open
one in a different bank. (I have $0 invested in preprinted checks to
eat.) One person doing that has no effect, but in a perfect capitalistic
world that bank would soon be forced to change its policy. Ah Utopia!
>> get real << Actually I was thinking of slightly larger systems such as
the post office's or Federal Express' or even your local supermarket's
check out scanners. However even a hand scanner and a PC might be cheaper
than full manual processing of rejects.
Sal
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: les c. cseh 76424,2075 # 233799, 1 Reply
To: Sal Cambareri 72137,3641 Date: 01-Dec-92 10:51:01
Sal,
>>UBCC provides a font cartridge that causes the MICR to bond well with
special laser safety paper
There is nothing that in font design that can cause it to bond better or
worse; that is a function of the paper, the toner and the engine.
We have seen a dramatic difference between different papers and toners.
HP, Troy and Xerox have specifically formulated their toners to bond well
and stand up to numerous passes through a reader/sorter.
We have also seen how most 3rd-party toners build up a thick deposit on
the reader/sorter heads, which then ends up depositing on subsequent
documents.
>> I'm sure none of them spends an extra nickle for magnetic toner.
Being a MICR font and software designer, we have followed the rules and
made up large batches of checks for testing various reader/sorters. Stock HP
toner yields a signal strength around 25-30, while the bottom end called
for in the U.S. is 50, and in Canada 80. The readability drops off severely as
you drop below 50. Also remember that the performance of a virgin check
will be better than one that has been handled (folded, rubbed, etc.).
The fact that people have gotten away with using normal toner is a
function of the process. I write you a check, you deposit it, your bank tries
to process it, can't, so they apply a correction strip, send it back to my
bank for processing, and it works fine. At present, banks can only track very
high volumes of rejects, or those they find through random testing, or those
they find because people followed the rules and submitted checks for
examination prior to going live.
lcc.
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: Sal Cambareri 72137,3641 # 233806, * No Replies *
To: les c. cseh 76424,2075 Date: 01-Dec-92 12:24:19
Les,
I confess: I'm not an engineer so I don't really know WHY it works. I'm
just a simple application programmer, but I do know that the UBCC system
does work and that it is not a matter of "slipping between the cracks."
For example, among UBCC's clients is a large east coast accountancy that
over the last two years has been using the UBCC system to print payroll
checks for about 200 different companies a month. They use ordinary toner
because they can.
I know UBCC believes in it. They even offer a challenge. I'm not sure of
the exact wording, but it goes something like this: You and a UBCC rep
walk into any retail store selling HP IIIs. You (two) buy one, set it up
with the included toner, install UBBC's font cartridge, insert Duplex/UBCC
paper, and attach UBCC's laptop to the centronics port. You then print
some laser checks (with correct MICR information) and go cash them. The
fun happens at the end of the month when the bank statement comes. If
there are any correction slips UBCC has bought you a printer, but if there
aren't you've bought UBCC one. Ken Garen (70233,1662) tells me that
they've never had to buy a printer this way.
Sal
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: William Beem 72241,1111 # 233823, 1 Reply
To: Sal Cambareri 72137,3641 Date: 01-Dec-92 13:35:21
An add-on or secondary OCR system for rejects simply wouldn't work with
our current system. We're still talking about using sorters in check
clearing. The technology is clearly inferior to anything in today's
standards, but it's firmly entrenched.
Image processing is a drastic re-engineering project for the banking
community. It's one that has to happen, since, as you say, competition
and consumer choice will demand it.
As a small customer, you will probably never see a reject fee for a couple
of home-printed checks with a poor MICR line. Some operator might know
your name by heart, but your bank probably has a threshold for initiating
reject fees. After all, it wouldn't be economically feasible to process
the paperwork to charge you for the rejects if you didn't submit enough of
them. It's a very screwy system.
A larger customer, especially one that submits pre-encoded items, would
definitely be charged for rejects. Some examples would be department
stores or grocery stores.
--Bill
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: Sal Cambareri 72137,3641 # 233827, * No Replies *
To: William Beem 72241,1111 Date: 01-Dec-92 14:43:27
>> screwy system << That is surprising. I would never have imagined that
my bank(s) wouldn't charge me a fee because I'm so small. I always think
of banks as cutting deals for the big guys and getting us little guys
coming and going. I guess this time it's nice to be a gnat. <g>
Sal
----------------------------------------------------------------------------
Subj: MICRed laser checks Section: Views:Dvorak [E]
From: les c. cseh 76424,2075 # 234349, * No Replies *
To: Sal Cambareri 72137,3641 Date: 05-Dec-92 11:38:10
Sal,
It's an understandable conclusion ... I drew the same conclusion when we
first got into this. Keep two aspects in mind:
1. The normal toner does not bond well to check stock. I can show you the
thick accumulation built up on the reader/sorter heads from a batch of
laser
checks printed with normal toner. The problem is nasty, because the laser
check often won't fail, however the toner it leaves behind ends up
depositing on subsequent documents. How do you trace that?
2. The reader/sorters are designed to be extremely accurate when the
signal
strength is between 50 and 200 (80 and 200 in Canada). The further the
signal
strength strays, the more likely that some or all characters won't read,
or
will read incorrectly (be mistaken for another character).
The IBM 3890 reader/sorter is extremely forgiving, but there are several
other [considerably more finicky] types in use. We proceeded with our
initial products based on good success on 3890's. Some time later, we
were
horrified with the results on some of the other reader/sorters, and had to
make a significant investment improving the products.
In fact, because 'noise' tends to cause problems at the '50' level, there
are plans to retune the reader/sorters to 80 in the U.S. as well.
If you use a tuned font (HP, Troy, Xerox, CHEQsys) with good toner, then
checks will process properly the vast majority of the time. The further
you deviate from the rules, the greater the chance of reject (either
today,
or sometime down the road). If the checks are found to not comply with
the
published standards, then there is very reasonable justification for
service
charges.
As a consultant or vendor, the best and most responsible approach is to
recommend solutions that are designed to follow the rules, not ones that
'happen to work' at this point in time. Hope you'll join us ...
lcc.
----------------------------------------------------------------------------
Subj: MICRed laser checks Views:Dvorak [E]
To: les c. cseh 76424,2075
From: Sal Cambareri 72137.3641
Les,
You really do raise a valid question about what is the most responsible
approach. I certainly don't want to encourage anyone to do anything
that is damaging or anti-social. (eg Poisoning some one else's well by
telling them write MICR checks that might "clog the gears" of MICR
readers.) On the other hand, I don't want to be a Ludite trying to stop
the spread of new technology, in this case UBCC's ordinary toner system,
by denying it exists. A cynic might argue that I would be hastening the
future by giving clearing houses a real incentive to move from the
antiquated MICR readers to OCR readers. Further complicating the issue
is the fact that UBCC will continue to sell that technology regardless
of what I do. Should my users be denied what is available to others?
I think that, for now at least, the responsible approach would be for me
to let people know of the existence of the UBCC technology without
recommedation and to provide a copy of this complete thread with my
software. If neither you nor Bill (nor Ziffnet?) objects I think that's
what I will do. If there is more information you would like me to
include with the software please let me know.
Sal