Re: B-Tree Index Usage

Concealing my true feelings, I will just ask:

Is it possible to discuss issues without being badgered, please? I get the
following from being told...oh...more than once: RD doesn't read this list.
Your support provider must apply pressure. You must file the paperwork or
don't expect results. Don't complain if you haven't tested the beta. Don't
complain if the bug's been there for more than x days.

I'd just like to read about actual issues. That's what this forum should be
for.

Thank you,

Chandru Murthi
"Tony Gravagno" <g6q3x9lu53001@xxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:gm20e25iav4qrckqd2l4sffvfvs1lesvs0@xxxxxxxxxx
(Sigh, this started out to be about 10 lines...)

Peter:

If I've misunderstood you, I'm sorry. If some documented function
works one place and not another, or it's not documented properly, tell
your vendor(s) - we can't do anything about that here.

RD Support won't come to you. No one from RD is going to post a note
in a public forum thanking you for your report. They won't promise to
have things working differently in a new emergency cut of the
software. What result are you expecting from any of us outside of
confirmation that you've identified something that (perhaps) should be
working differently? Yes, indexes should index everything. Can we go
now?
You keep evoking ***'s name and citing how things were so many years
ago. *** made a lot of bad business decisions and so did his
successors. I think it's time to get over that and deal with the
companies that provide your software now.

I've told you how RD works. I understand you don't like what you're
reading, but that is how they work nonetheless and arguing with me
won't change anything. If you DO work with the system you will get
results. If you do not, you will find things don't change. The fact
that you continue to approach things in the same way and nothing
changes should be proof of that.

TData is your direct Support provider. If you don't or can't
communicate with RD, take it up with TData. If you're reluctant to do
any of this, then I get the impression you just want a sounding board
here and not solutions. Great, I hope you feel better when you're
done, but other than agreeing with you there's probably nothing that
anyone in this forum can say that will make you happy about this
specific topic.

More below.


"Excalibur" wrote:
As I said in an earlier post I asked Pick Systems about this when it first
came out. Funnily enough no one took any notice and some of us had a life
to get on with. Things have changed dramatically since then in the
computing world. How the heck would I know what RD are doing as they
don't
even answer their own support board and claim to look at this one but not
respond.
Did you ever succeed in compelling *** to do anything.

You've been told here at least once...
The RD forum has the terms in their registration...
Public forums are not official channels for support.


If it's so important, why didn't anyone compel RD to make this change
10 years ago or anytime in the time period since? I'm trying hard to
answer all of your questions with complete candor on a topic that I
don't care the least about - I keep asking this single question and I
don't get any response from anyone, now or then.

If someone has asked RD about it, what's the action item number? What
was the date and resolution? Are you aware that RD is now purging old
action items that no one asks about any more? The reasoning is that
if no one asks for something in about 5 years that it's not worth
keeping in the database anymore, especially since releases change and
new features may satisfy the need or eliminate the problem which
prompted the original request.

Please read the above again, I don't think you're taking it in. "I
asked Pick Systems about this when it first came out" is not the same
as filing an action item and riding their asses about it until a
change is made. (When you say "I asked Pick Systems" are you still
talking about your conversation with *** 15 years ago???)

The issues you're citing were obviously as important to you then as
they are now - not important enough to file a real report so someone
there can duplicate the problems, prioritize, change, and test.


Maybe you've put your finger on yet another very pertinent point.
You're talking about driving the truck, hitting the switch, etc..
Most D3 users are quite content to run in R83 compatibility mode. The
fact that you're even looking for a switch to hit puts you into a very
small group, and that's a significant factor in the RD decision making
process for enhancements and fixes.

How did you decide this. To the best of my knowledge the closest anyone
ever got to a discussion with RD was when they came out to announce the
new
venture and that was crucified within a few months.

At Pick Systems I held the position of Support technician (liaison to
Engineering), Quality Assurance Manager, and Corporate Technical
Account Manager for Sales and Marketing. At Raining Data I held the
position of DBMS Product Manager. Before, during, and after my time
there I have seen the way many end-users and developers manage their
environments and write their code. From what I've seen, most
applications running on D3 are largely unchanged from their initial
development which occurred any time from the 80's and on into the mid
90's. Based on the comments in this forum and elsewhere it's apparent
that many people don't use many of the "newer" features in D3
(anything from 1998 forward?), and that they don't make use of the
enhancement and resolution reports that come with each new release.

I have had many discussions with RD personnel over the years about the
concept that they waste a lot of effort putting in neat features that
Marketing and Sales don't know about and/or don't advertise. The
decision to not enhance some aspects of the software has always been
influenced by the fact that many sites simply don't use the features
that have already been added.

This directly relates to my comment below about chicken and egg: If RD
doesn't tell people about new features, people don't use them.
Similarly, if some feature has bugs, people have a tendency to just
not use it, rather than reporting the problem. The path of least
resistence is the one that is taken most often. This gives RD the
idea that some feature is useless and should be ignored or obsoleted,
obviously because no one is using it. I know that makes no sense but
that's what's been happening for years. In my opinion the core
problems like with Marketing not telling people what's in there, Sales
not showing the client base the value of their continued licensing
fees, and unfortunately I've heard for years that people really don't
like dealing with RD Support.

All of this leads to diminished product usage and very little quality
feedback getting to the people who can make a difference. The
behaviour of the market only reinforces the policies over there. We
had an RDTA VAR committee that had no official agenda, no list of
critical bugs, no demands for new features or product changes. When
the market says there is no problem, what is a vendor to do but assume
and act upon the belief that there is no problem?

And where does _that_ lead? Migration. The D3 sites that actually
cared about "features" and pro-active vendors have moved on to vendors
who showed more concern - primarily U2 and jBASE. The people who
migrated aren't complaining much anymore. The people who are still
using D3 don't really ask a lot out of the software or the company
that provides it. Their software is business-functional, they don't
sell a lot of new systems, and the current user base doesn't ask for a
lot more than what's already in there. They pay their support fees,
they are entitled to make a phone call and get re-activated if they
really need it, they get new cuts now and then to fix whatever was
broke by the last release.

So... Getting back to my original assertion, that leaves you in a
small group of people who reads documentation and expects full
functionality out of your product. You're the type that goes right
from the ref manual to the keyboard to check things out and if it
doesn't work as advertised you get ticked, especially when you've been
there before. You're not wrong to do so. But - you need to focus
your energies in the right direction. Present your concerns to your
vendors in a business-approproate manner. If your vendors do not
address issues satisfactorily, I recommend lowering your standards
(not my choice) or finding a vendor that you can work with. I caution
you though - very few vendors will accept a bug report from a public
forum. Such things are done through individual and personal
initiative, not officially.


This is the chicken and egg scenario that I used to argue with Mark
about. (I hope you don't disagree because I am and always have argued
for the very point that you and Tom are trying to make here.) If RD
doesn't support a feature, rather than requesting it people dump the
platform because the feature isn't supported. There are then fewer
people who might have wanted the feature or related features. So
fewer people ask for related features, which also fall by the wayside
as far as PS/RD are concerned because apparently no one wants them.
Their attitude is, if we don't hear about it, it's not a problem.
Your attitude is, if it's not in there, we're leaving.

I object intensely to this statement. Unlike you I have not switched to
JBASE.

You don't understand my situation here.

First, when I say I used to argue with Mark about stuff I'm talking
about when he was Manager of Continuing Engineering and I was DBMS
Product Manager. (The discussions weren't adversarial, they were
entirely philosophical, and neither of us lacked the influence to make
any changes anyway.) The point is we were intimately involved with
how reports came in and how they got processed, we weren't just casual
observers. All I'm doing above is telling you how people behave, I
don't know what it is that you object intensely to. That's just the
way it is.

Second, I use D3 intensely every day because I like it, always have,
though there are some components I have chosen to not use simply
because they don't work for reasons described above. My lonely voice
usually isn't enough to compel RD to change their software - though I
appreciate the many times when it has. Many of my clients are D3
sites and I want their software to work as error-free as possible.
When I find a problem on my system or my clients', I report it to RD
and we get "some kind" of resolution.

Third, I've formed a business relationship with jBASE because I have
had too many business issues with RD. I've taken my own advice, but
that doesn't stop me from liking D3, working with it, even
recommending continued usage for many of my clients.

And please also note that I haven't just jumped ship to jBASE. I did
do some testing on the D3 v7.5 Beta. I tried to make it better
software through my contributions (though I admit I wasn't able to
participate as much any of us hoped). If I had known about your
indexing issues a few months ago you can be sure I and others would
have been jumping on RD with test cases and all sorts of rhetoric to
try to get changes prior to production. That's why I'm a little
ticked that all of this comes up only AFTER that effort and after the
release of software that has been in development for so long.


The whole point of this thread is my attempt to get RD to sort some
of the issues so that I do not get shot down in flames by "relational
experts" when trying to sell. So that I can extend my package without
re-inventing the wheel.

Then you need to file action items with Support and if they aren't
prioritized to your satisfaction then you can take up the issue with
Marketing. That's the process.


Hiding the capability of the OSFI is stupid.

They don't hide it, they just don't expand on it. I don't care where
that line is drawn.

I look at other products to see if we are missing the market even though
we
have the best database engine. I have raised these issues many times over
the years at forums which unfortunately died out during the '90s so we had
less opportunity. On several occasions I have had people come up to me
and
say thank you for asking , we all think it but are scared for our jobs.
In
particular I asked several questions of *** Pick on the crap rap trip and
got the lamest answers ever. In fact on that occasion I was again
approached with the comment, good questions lousy answers.

Talking to *** during a marketing show probably wasn't the most
effective way to report a bug. ;) ***'s been gone for 14 years now,
it's time to bring up the topic again with someone else.


T

Peter McMurray
That just
encourages RD to support less, to believe D3 users want less, to
believe D3 sites will leave no matter what they put in, etc. Without
communication there is a snowball effect that I've been warning both
sides about for at least 8 years. I wish people spent as much time
sending emails to their software vendors as they do in public forums,
we'd see a lot fewer complaints in here.

T




.