Antiquated Software

[Fairlight]

A quick polite preface for John Esak:  You don't wanna start reading this
probably, because it'll prolly take 15min for JAWS to read.

There were just too many points to address.

Fair warning, proceed at your own risk.  :)


With neither thought nor caution, Dennis Malen blurted:
> We hire young people who came from other firms using windows. When they come 
> to us, the training is much easier and we always get comments on how simple 
> it is to use. Furthermore, the largest bank's in the country come on site to 
> audit us. They also reconfirm the ease of use. They come from banks that 
> would never use filePro because it is virtually unknown. A major bank can 

No, they would never use filePro because it's not Enterprise grade in terms
of sheer capacity, much less anything else--or at least hasn't been until
the forthcoming 5.6.  Sorry, but I did data entry for years as a temp, and
I'm telling you from experience that -just- the mutual funds department at
ONE bank's main office could not handle all their data entry needs, or the
Trust department at another I know of...could NOT handle 2gig limits.  That
bird simply won't fly.

Has nothing to do with the interface at all, there, I'm afraid.

> I interact with the outside work by sending and receiving data with no 
> problem, which involves a little filePro programming. I am able to access 
> off the shelf software for imaging etc. which interacts seamlessly with 
> filePro. I can also use any type of windows printer on my system. That only 
> involves the tweaking of UNIX (AIX in my case). I can run many different 
> reports at the same time.

Any decent system or networked system can run many different reports at the
same time.  That's what print spoolers are for.

> There are also outside vendors such as Sound Ideas. They have some great 
> utilities that work very well with filePro. Dave, the owner, was a filePro 
> user. Also many of the members of the list also sell products that interact 
> seamlessly with filePro. It is a matter of organizing the talent we have and 
> using it. For example, if we need a particular add on to filePro which 
> filePro does not have the resources to accomplish we should find out how 
> many of us would want it. We could put money in a kitty and designate 
> someone from the list with the talent to accomplish it to do it. They would 
> estimate what the development costs would be. They would own the rights to 
> the software. We would agree as to what they would sell it to us for after 
> it was completed. There is so much talent here it just needs proper 
> management.

Uhm, won't work. filePro itself has not responded to even close to the
needs of the requests of solid, diehard users over the years.  Whether
that's a resource shortfall or whatever, it doesn't happen.

What makes you think it works any better for 3rd party add-ons?  I can
assure you it doesn't.  I have a CGI solution that blows fpcgi out of the
water (I intentionally don't publish the comparison list -publicly- out of
politeness to fP-Tech, and because I don't like to play dirty, even if it
costs me), but after viewing the comparison charts -and- the lower pricetag
(first 38% less expensive, and then 51% less expensive), several people
-still- went with it strictly because of the company name being the same
for both products.  Hence I've raised the price to match theirs, since
competing on cost does me zero good.  Anyone that wants a solid solution
can make it worth my while more than they used to at this point, and I see
no reason to sell myself short when that method wasn't working.

But I've developed other things on spec for this community.  In general,
even when someone says they need it, sometimes even if it's directly
commissioned work, it turns out to be a no-go.  You may have a fully
functional piece of software that's -exactly- what was first elucidated to
you, make upgrades for weeks, and never sell a single copy.  Not for lack
of being good software, but because the marketplace is fickle.  First they
can afford something, then they can't.  First something will be done -next
week-, then it will be done -a year and a half or two- later when they
finally actually get to it For Real This Time[tm].  I can't tell you how
many 3-month to 3-year "we're about ready to do this project" lines I've
heard over the years.  If I had a nickel...

I've had a very BAD experience doing things on spec.  The people that
buy and use the software are really happy with it.  It's getting to the
point of making it worth the while to code, no matter how good an idea and
usefull it sounds at the time.

> Absolutely, filepro has its place in my business coupled with other programs 
> we need. The only point that was mentioned that has validity for me are size 
> the screens.

Someone talked about font sizing.  Technically, this can be done on xterms,
in emulators, and even on the linux console (which has 3 possible VGA font
sizes).  I wouldn't worry about size so much as raster area, which is
wholly addressable, and for which I've proposed numerous solutions covering
most (if not all) of the caveats to this over the years.  They Just Don't
Listen.  Or don't care.  Or whatever.  It never materialises, at any rate.

> We need to think about ways things CAN be accomplished not why it can't. It 
> is always easy to point fingers and degrade a well meaning attempt to solve 
> problems.

Yeah, well that's easy if you've to the resources to make it happen.  When
you have the choice of doing direct work for someone for $75/hr or coding
on something that may not even sell (and when it does, does so for
$0.69/hr), you'd have to be out of your gourd to go for a project on spec.

THIS is why it's a case of "show me the cash" before I'll code for anyone
in the community anymore, unless it's a feature request and they're already
a paying customer.

> ----- Original Message ----- 
> From: "Anthony Terrible" <tony at vegena.net>
> To: "File Pro List" <filepro-list at lists.celestial.com>; "Tips" 
> <tips at stanlyn.com>
> Sent: Monday, January 16, 2006 1:00 PM
> Subject: Re: Antiquated Software
> 
> 
[Attribution was missing.  --FLT]
> >I wasn't going to enter this flame although I have thought that Filepro is 
> >antiquated for some time now. So I am going to address some of the things 
> >that have been said and state my opinions. But before I begin let me state 
> >that  I have the utmost respect for all of you on this list and believe 
> >that all of you are better programmers than I.
> >
> > 1)      Data entry in text mode (mouse less) is faster than a GUI or using 
> > a mouse.
> >
> > While I believe this to be true you can design Windows data entry forms 
> > very easily that can be used without a mouse just like a text based 
> > screen. The tab key works very well and the tab stop can be set any way 
> > you want.

The problem is that if you give them (users) a squre inch, they'll take a
cubic mile.  If a mouse is present, the majority of 'lusers' (ie., regular,
unclued and barely adequately trained users) will -opt- for the mouse.  If
you -force- them into having keyboard-only entry, you -guarantee- optimal
performance, which translates into guaranteed ROI, especially when you have
15 temps sitting in an office at a cost of $27/hr each.  (I believe data
entry used to average me $7/hr between the highs and lows back in '89-'93,
and I was also told my rate was only 1/3 the cost of hiring a temp.)

If there's significant operator time involved, forcing that and leaving
-no- option is a Good Thing.  Sure, it -can- be done both ways, but nobody
will ever produce a time/motion/cost study that proves GUI users that
actually use the mouse are as efficient as pure-keyboard users, mostly
because it's not true.  Sure, if your mousers are at 75wpm and your
keyboarders are at 35wpm, it might average out or be better, but given
equal typing speed--no way in hell.

Bill C. menteioned the strengths of scripsit being control key
functionality and bindngs.  Should check out EMACS.  :)

> > 2)       Quoting Laura Brody, " I would bet cold-hard cash in a 
> > non-trivial amount
> > that a top-level data-entry gal who has been at it since Eisenhower was 
> > president would have a drop of productivity of around 50% or more if she 
> > had to navigate through the exact same screen with a mouse on some GUI 
> > database."
> >
> > While this is true, wake up Laura. The people entering data now are in 
> > their 20's and grew up with Windows and can navigate it quite easily, as 
> > well as that person using a text based system. Read  #1 above.

Their age and familiarity do not mitigate the actual time it takes to take
your hand OFF the keyboard, REACH OVER, use the mouse, and GO BACK to the
keyboard.  That is time.  Cumulatively, depending how much they "have" to
mouse, it could be as much as an hour of an 8hr workday when all is said
and done.  It wouldn't surprise me.  That'd be a 12% loss, financially.

And I guarantee you that I can hit tab 9 times before they've gotten the
mouse over the field--or possibly even located the cursor, given the
nature of some of the most obscenely blinding desktop wallpapers I've
actually seen used in real workplaces.  You practically have to have
mousetrails on at some places to see the thing move.  Even omitting that,
by the time they're just clicking in, I have the data keyed, period.
Anyone that does professional data entry from the old days would.  Read:
people who used CICS, DB4, Magic, or gee--filePro, for example.  I find
the multiscreen argument largely invalid.  I used CICS for years, and
concurrent multiscreen would have been irrelevant for hard data entry.
Whomever brought up the multiscreen argument is talking casual use where
you want the luxury of copy/paste.  -I'm- talking about HARDCORE data entry
here, where you're going through -hundreds- of pages of hardcopy an hour,
keying in from PAPER.  Multiscreen doesn't even enter into the equation.
Nor does it enter into a POS situation.

Not only can I hit that 9 keys faster, I can do it more accurately.  I know
-exactly- when I've overshot -if- I overshoot, and by how much, and can
reverse it probably before they're back to the keyboard.  Again, any -real-
data entry person can.

Again, use of the mouse in data entry environments is pretty much like
comparing consumer-grade printers with business-grade printers.  As has
been pointed out by many, there is a -large- difference.

I can only surmise that people advocating that the mouse even be there as
an -option- have never spent even a single week doing -hardcore- data entry
for a living, especially on work quota systems, nor have they paid their
employees under those conditions.  I'm talking about situations where if
you didn't hit 1000+ applications entered by the end of a day (I think this
particular one place was a mix of 5-7 possible 3-5 page applications that
needed keying), you weren't going to be gainfully employed the next day,
and oh by the way, mandatory overtime to add a few more hours of solid
keying some days.  I've -done- that kind of data entry for a living, on
and off for several years between word processing and spreadsheet jobs.  I
spent three months at one bank, six months at another, and had numerous
other 0.5-2 months stints at other places while I was a temp.  I've -lived-
that job multiple times, okay?  If you haven't, then listen to the voice of
hands-on experience, please.  I'm telling you, firsthand, it is -slower- by
a LARGE amount to use a mouse.

And it doesn't matter if you supply tab stops, if they -can- keyboard
navigate, whatever.  The fact remains that you're saying those Windows
users are now in their 20's and have grown up with the technology.  That's
absolutely right:  They've grown up on the technology that -allows- them to
be less productive, and they're -used- to being less productive by using a
mouse everywhere that's possible.  Probably 90% of the users out there
couldn't even tell you 5% of the Windows keyboard shortcuts.  Because,
"that's what mice are for."

When someone that's done data entry for 8-10 hours a day for at least six
months straight in a high-pressure environment tells me that using a mouse
is just as fast, I -might- believe them.  Maybe.  Anything else is sheer
speculation and wishful thinking.  And believe me, I know the quality of
temps and office workers out there.  Screw false modesty here, because this
is begs some serious questioning--I -was- one of the fastest temps at both
my agencies.  I've worked in rooms, side by side with other temps, for
months on end on the same projects.  Hell, we used to -race- for cigarettes
at once place.  I believe I ran into about one person over the years that
could actually beat me semi-consistently in terms of speed, and maybe three
others that could keep up with me once in a while.  That's over the course
of many assignments, and in varied environments.  That was also back when
I only went 85wpm instead of 114.  The acceptance speed for being a temp
was 35wpm.  I talked to my managers when I tested, and they were blown away
because the average speed they saw was 45-50wpm from the large majority
of temps.  Take that 45wpm and lop off 12%.  38wpm.  BARELY inside the
acceptance range.  Who wants to pay for that?

For that matter, take order-taking--be it in a restaurant or an order
fullfilment house.  I mean, I can type faster than most people can write
shorthand (what few can these days), and could when I was 40wpm slower
than I am now.  Now you've got a customer in front of you at a counter,
or on the phone, and you have a choice of having someone using memorized
keystrokes you've done a billion times or using a mouse to muddle their
way through while the customer gets impatient.  People -do- hang up, walk
away--I've seen it.  Hell, I've -done- it.  Make it too hard to spend
money with you, I'm gone to somewhere viable.  I have a lot of friends
that are that way too.  Ryan's relation of his POS experience was pretty
damned accurate.  And you -can't- point to employee training, as that's one
of the areas companies put in the LEAST amount of cash these days.  They
get so much churn at low paying jobs like keying that they barely bother
training them more than they have to--just toss them in.  Now after you've
done something like 200 times, you start burning into your head exactly
what you're hitting on a keyboard.  Actually, that's a physiological
reaction--your brain develops special neural pathways for specialised
behaviour that you'll need often and repeatedly.  That doesn't happen with
a mouse, as it's flicked off to a random location somewhere, you have to
get it somewhere, etc.  It's not 1,5,7,2,tab,tab,tab,key data off the top
of your head because you've done it a zillion times to the point you can
do it in your sleep, blindfolded, while smoking a cigarette, having a
drink, and listening to your iPod in one ear--it's suddenly having to find
where you are, get the cursor there, enter the fiel, etc.  That's very
unproductive.  

And if you -let- them, they will almost always do it.  I actually had a
place wanting replacement software for StarOffice recently because they
didn't want to train their operators on how to save files in the format MS
products at other customers' sites would accept.  I mean, if they don't
train them in something that -necessary-, you think they're going to teach
them to use the keyboard instead of the mouse?  -I- don't.

> > 3)      Quoting Bill Campbell, "I've never supported Windows, and 
> > seriously think that anybody who entrusts corporate assets and data to a 
> > Windows systems is guilty of malfeasance, and I refuse to be an accomplice 

Is there actually a feasance?  I mean, if there's a negation, there must be
a positive, right?  Merriam-Webster says no, though.  Cute.  :)

> > to that. Even if Windows were secure, and no data were ever lost to 
> > software crashes or random reboots, the employee time lost due to these 
> > crashes and reboots, when reliable alternatives are available, isn't 
> > something that should be tolerated by anybody who is responsible for an 
> > organization's bottom line."

Amen.  The problem, Bill, is that people aren't realising just -how- much
of the bottom line is getting blown.  They're underestimating the problem.
Worse than that, they're making excuses that perpetuate myths that counter
what aren't myths, but are actually common wisdom amongst those that have
done (and still do--I have friends that do data entry) this for a living.

I strongly suspect that the people speaking up on behalf of "going GUI"
have never actually worked "in the trenches" for more than an hour or two
as a make-do cover for someone that's late or out an afternoon ill.

> > 5)      For years I have listened to the list ask when is there going to 
> > be an ODBC driver for Filepro. The reason Filepro has never made an ODBC 
> > driver for Filepro is that they are very much aware that by doing this 
> > would be their death. If you could access the data stored in a Filepro 
> > file from any programming language why would you use Filepro? It would 
> > just be a matter of time before you totally get rid of Filepro. The only 
> > thing holding most people to Filepro is all the data they have in it's 
> > files that they can't get out without Filepro. Given the out, most would 
> > take the option. Filepro's file system is slow and clunky compared to 
> > modern data bases.

That's such a strawman argument.  It's not the first time I've seen it.
It's also counter to logic.  If you have a tool as easy to use as filePro,
and as easy to develop with, -why- would you want to switch?  The reason
you want ODBC server-side is so you can have filePro -at the CENTRE- of
your business, rather than on the periphery.  Keep all the data and the
main mechanisms in place, plus gain interoperability.  Don't
forget--migration costs a LOT of $$$.  It's not something you just jump at,
or I suspect there are more than a few people here that otherwise wouldn't
be.  The truth is, it'd make people more likely to KEEP their investment
right where it's been for 'x' years and save a large expenditure involved
with rewriting their business software from the ground up.

> > The world is changing. We need more data in many different formats then we 
> > ever did. People expect it, they demand it. We need pictures, email, 
> > internet access, the ability to open, view, modify  and run reports on 
> > several different files at the same time. This is just not possible with 
> > Filepro. It is however easily done with current day data bases running on 

Apparently whomever said this has never run fP integrated with a web
server, nor under X11 or Windows.  You can integrate it with a large
number of things.  The -relevant- question is not, "Can it be done?"  The
-relevant- question is, "How much are you willing to pay your developers
to re-develop the soft-code (processing, as opposed to hardcoded into fP
itself) workaround for something that could be 'more functional'?"

> > PC's with Windows. What you have now with Filepro is a small ugly screen, 
> > long learning curve for new employees, no real ability to view or save 
> > photo's, little to no ability to create text documents, lots of clumsy 
> > menus, the lack of ability to be in several files at the same time, 
> > inability to access the internet, reports that look like they came off a 
> > 1950's typewriter, the inability for management to create their own 
> > reports, inability to sort or group data without running a report, slow 
> > and clunky searches unless you use an index selection, no integration with 
> > current technology. And programming? There is more to life then IF-THEN. 

In order:

Agree on screen size.  Don't agree on ugly if someone with any aesthetic
sense designs it.  The tendency I've noticed is for developers to squeeze
as much onto a screen as possible to avoid multi-screen entry, STILL fail,
and end up with ugly and overcrowded screens rather than one extra screen
and the rest being easily readable.  Some just seem to have no aesthetic
design sense whatsoever. :)

You can view photos using 3rd-party apps.  You can scan in using same, and
bind the image pathnames to your data.  Embedding images (especially for
people forced to use TIFF--arguably THE most bloated image format around
when a compressed TIFF was still 57% smaller when gzipped) directly into
the data is just nuts.  There's only one database I know of that was
-designed- for that, and Bill Vermillion posted the info on that several
times over the years before he left this list.  The one that would do fuzzy
searches on "flower" and display a rose.  It was -designed- for it though.

Why can't you create text documents?  Raw file I/O plus EXPORT...what more
do you need?  

Menus can be avoided almost entirely, given good design, and implemented
minimalistically where needed.

What do you mean lack of ability to access several files at the same time.
LOOKUP helps you do just that.  I've seen some remarkable header/detail
code using associated fields and such.  I've seen full emulation of browse
lookups doing same, as well as data entry/modification using same.  If
you're talking about "views", that's another creature entirely.

Accessing the 'net.  When fP was demoed to us at my former place of
employment, someone pointed out that we were "abusing our spreadsheet"
because half the stuff in it really belonged in a database.  That line was
a source of humour for years, and I still chuckle about it.  The thing is,
there's a point where you do one thing and do it well, or you do many
things in a mediocre (at best) fashion.  Netscape went severely downhill
when they did Communicator after 3.2's pure browser, -specifically- because
they built so much in that it was this ungainly octopus.  However, there
are occasions where client ability might be useful in fP.  It's coming 
in 5.6, although I'd -never- use it for a server due to technical
limitations of both fP's programming paradigm, and lack of select().

Printing "pretty" reports is available through numerous 3rd party
solutions.

> > Using Filepro or thinking that it has anything to offer is like the person 
> > who refuses to give up on that 1960's automobile. Automobiles have come a 

Oh, come -ON-.  I've been a fairly large detractor over the years on
several fronts, but let's face it:  sweeping, blanket generalisations are
almost always wrong.  And there's no way in hell even -I- would suggest fP
has -nothing- to offer.  Actually, I think if they modularise and abstract
the storage layer, they'd be a great character-based RAD kit for MySQL, Pg,
or any other *SQL engine.

I'll tellya something...  If I had the choice between using an existing GUI
for *SQL and -writing- one in curses (ie., character based), I'd take the
latter every time.

Back to your statement...  That is so...utterly absurd.  Not just implying,
but actually stating that it has -nothing- to offer.  Even I can't buy that
one, sorry.

> > long way in 20 years, they are computer controlled in every aspect. They 
> > are faster then the cars of the 60's and 70's, get better gas mileage, 
> > don't pollute, safer, handle and drive better, last longer. So why would 
> > you want automobiles to go back to the way they were? Filepro was good 

Have you -read- the news lately?  Chevy and several others are going back
to designs made in the 60's and 70's as the basis for their new designs,
trying to bring back an era of appeal, focusing there rather than on the
efficiency, etc.  It's not like they'll downgrade, but people -do- look
back at what's been done right in the past and worked.

fP had a lot of potential.  If TPTB would unplug their ears and stop
chanting, "La la la, I can't hearrrrr youuuuuu!" it would have a lot more
future potential.

> > when it was made but it's time is done, has been. Now is the time to 
> > figure out how to migrate from Filepro. Total integration is the way. With 

I beg to differ.  "Total" integration brings forth any plethora of security
concerns that you really don't want to deal with for a database scenario,
and that I'm betting 90%+ of the people on this list aren't even qualified
to -handle- as a full-blown DBA.  Most of the people here seem to have far
less of a clue about database design than I not only do now, but did
several years back before learning *SQL.  The impression I get is that the
4GL Syndrome struck, and we have a bunch of people that can write
-applications- in filePro that may actually work, but they're not
-database- people qualified to design proper relational databases, along
with any system and security concerns attendant to the position.  Not
without a lot of reading, anyway.  All you have to do is look at some of
the questions we get here, and no offense to anyone, but it becomes obvious
that people even have trouble with the "dumbed-down" 4GL version of a
database, much less anything with (*gasp!*) internals to deal with at a
non-user-friendly level.

I'm all for giving people all the power, connectivity, and functionality
possible without artificial limits.  However, as the quote goes, "With
great power comes great responsibility."  Greater than 50% of this
community probably can't handle the modern tools without one hell of a
crash course, or without hiring someone to do it for them.

"Total integration is the way."  Gee, but that makes a dandy slogan!  You
should probably trademark that.  Only problem is that it's like a giant
buzzword--it gives you the feeling of going somewhere without giving you
the slightest idea of how you're going to make the trip.  It's phrases like
that which enable M$ to sell millions of copies of a word processor (and
numerous other products) that have actually pretty much hit their feature
threshholds and have nothing more to offer.  Sounds like empty hype to me.
Now if you cited specific examples of integration, I might consider it
differently and take it seriously.

> > Windows this can be done. Your clients never have to learn anything else 
> > but your program  No more switching to a picture viewer, an email program, 
> > a writing program, a web browser. All of this is done seamlessly and 
> > easily. Menus, right clicks, real help files make it easy to use and 
> > learn, tabs make it easy to enter data. Use the power of the PC. It will 
> > make your company better, your people more productive.

It depends what you mean by "better".  (Komptria!)  

There's "better" as in "slicker", and there's "better" as in "more
productive".  The two are not always in agreement, and in fact are (to
borrow Jay's word of the century) often orthogonal.  

It may be more seamless.  It may look slick.  But you pay a price for that
integration at -some- point along the way.  Could be a harder learning
curve, could be that you're stuck with far harder changes to make when you
modify something, could be that all the parts more or less work but none of
them are actually -fully- what they could be if you tacked pieces together
that, independently, are some of the best pieces of software in their
respective areas.  Do one thing, and do it well.  That's always been a good
maxim for software.  You don't drop everything but the kitchen sink into
something and pray for the best.  That's how you end up with a mess like
AIX, actually.  It was -supposed- to be the best of several worlds and
ended up being something that not many *nix admins actually like working
with.

I saw another subsequent post regarding fP's "direction", and damned if I'm
going to write another post, so I'll finish with that.  

"What direction?"

As long as I can recall (since 1993), fP has gone with whatever the pet
project du jeur was.  Anyone remember the fP Video dealie they had on
their site some years back?  Flop, from what I was told by some.  Now you
have Biometrics--not because it's actually all that useful, but because it
was supposed to be able to get a wider deployment for fP in general.  I'm
guessing price is an issue with that, not to mention that -everyone- I
talked to that saw the equipment said it looks like you can just rip the
scanner off the system--it's like a very expensive toy.  Then there's the
spell checker you'll see in 5.6, which apparently was intended to help sell
more upgrades for a very specific customer base.  It's always something
that sounds like a "sure lock".  And it usually sounds like exactly what
it is--a pet feature or side project that actually doesn't really enhance
the core nearly as much as needed.  Half the stuff the rumour mill has it
that's coming in 5.6 arguably should have been there for years (UNC's,
nested calls, 64bit I/O, more automatic indexes, etc.).

So...-what- direction?  They held up 1.0.14 because of multiuser
Biometrics, from what I heard somewhere at some point.  Did they even
-release- it yet?

For that matter, what communication?  Bud could just as easily step in
on this thread and address the entire lot of things with a Grand Unified
Vision worthy of Bill Gates' buzzword-filled and near-meaningless keynote
addresses that he's so fond of giving.  I wouldn't believe 90% of it, but
at least he'd be SAYING something. *listens to crickets chirping while we
wait for Bud to SAY SOMETHING*

The future direction would worry me if I were married to the product.
Actually, I said that over 3-5 years ago probably (maybe more), and it's
probably archived.  And yet, here they are, about to release 5.6.  Granted,
there's next to no progress along the lines many of us have advised
should be pursued.  But they're still there for now, and the existing
installations actually work, and it's still quite a usable product for what
it is.  Getting back to John's last post that I saw--that's exactly the
POINT:  FOR WHAT IT IS.  To be able to accept it and embrace it, you have
to be in a position where you can do just that--love (or at least tolerate)
it almost unconditionally.  If there's so much wrong that you need to
migrate because it's not for you, then that's your obvious move, and fine,
that's respectable.

But saying they're entirely irrelevant is just ignorant, I'm sorry.  Maybe
someone just phrased it poorly.  Happens to me lots of times.  But there's
no way, no matter how many things I think should be addressed with fP, that
I would even consider saying the product is -wholly- irrelevant to the
point that it has -nothing- to offer.  And if I ever went that far in the
past, I was -wrong-, and would readily admit it if called on it.  I used to
think more black and white than I do nowadays.  I still don't think I've
ever said anything -that- drastic though, from memory.  Just addressed lots
of things I think need attention.

I can't believe some of the uninformed rubbish that's backing some of the
opinions espoused in the last 2 days.  Everyone's entitled to theirs, but
at least try to have a -little- bit of a clue when you form an opinion on
one?  Some of these posts sound like me circa 1993 regarding SCO 3.2.4.x,
and that's just damned scary, actually, as there's no way in hell I should
have been spouting off about some things I just plain didn't know near
enough about as I should have.

mark->