WHY

[hrlngrv]

(e-mail address removed) wrote...

CA has a full VB compiler and IDE. They dropped it in like 1998 I believe.

If enough people from the Unix side would ask CA to opensource this; it
would happen; I am sure.

True VB or an approximate clone?

*x systems already have something called XBasic, but it's not much used.

 

[Harald Staff]

After I asked you pretty nicely to stop embarassing yourself, I've been a
week on the sunny island Tenerife, celebrating christmas with my family.
Returning home, I see that you still post the same arguments to the same
people. So I wasn't really asking for your opinion on speed and efficiency.

Best wishes Harald

 

[Frank Kabel]

You don't develop websites in VB.Net. you'll use plain HTML or Jave
servlets. Maybe even ASP.Net b ut not VB.Net. Personally i know many
developers for real business application s using a combination of C# and
VB.Net. Just depending what suits best. There's no signle development system
suits all application available. also you always have to take care of
existing code anyway

 

[Harlan Grove]

MDX doesn't have a lack of built in statistical tools

If you believe sum, average, count, min, max, variance, standard deviation
and correlation constitute an exhaustive set of statistical tools, then
maybe. In terms of built-in functions other than those provided by Excel
when Excel's available, there's no general linear modelling, not statistical
distributions, no more complicated statistics.

For ANOVA analysis and simple correlation analysis devoid of significance
testing, this may be sufficient. For real stats, it ain't.

I'm basing this on the online documentation, which may be incomplete. Can
you show how MDX handles linear regression with multiple independent
variables?

Re:

You basically bind a PivotTable to an olap datasource, and everything is
drag and drop-- with a big benefit over traditional pivotTables:

The pivot table being an array or a separate data type itself? I know Excel
functions handle arrays and worksheet ranges as arguments in most functions,
but I'm curious how you pass OLAP data to, e.g., Excel's SKEW or KURT
functions.

THE DIMENSIONS ARE HIERARCHIAL. YOU CAN DRILLDOWN!!!!

It's just absolutely wonderful how fast it is.. against Billions of records,
you can have a subsecond response time.

If MS OLAP works like Essbase, then in order to drill down, you have to be
in some 'cell' in the hypergrid. Being in that cell usually requires UI
actions. There are a lot of cycles that pass during human input, and often
more than enough cycles to fetch detail data underlying the value in the
active 'cell'. In other words, the apparent speed of drill down may be due
to all the extra time the computer has while humans move around the
interface which it can put to use caching information fetched via implicit
indexed queries.

There's no way this software is handling billions of records of even trivial
size in under a second. Just scanning through 1GB of RAM on a 2GHz machine
takes more than a second given limitation of RAM chips and OS overhead.
Multiply a billion records by the record length, and you have chunks of data
that won't fit into RAM even on ultra high end PCs much less the boxes on
most people's desks. Disk throughput is slower than RAM throughput, so data
in chunks of multiple gigabytes would need to come from disks, so
implausible you're processing billions of records in under a second on one
machine.

 

[Harlan Grove]

there are no intermediate results, you do everything in the database and
render it with a dumb report.

....

You just don't get it!

There are some computer applications that produce numbers but not reports,
unless you warp the defintion of 'report' to be any result that could be
displayed or printed. Under that sort of definition, every web page or Word
document would be a report.

There are some computer applications that require sorting as an intermediate
step between some calculations and other calculations. It'd be grossly
inefficient to deal with such sequential calculations by having to keep them
in DBMS tables rather than in simple and efficient arrays. Many
nonparametric statistics require order statistics, and the efficient way to
do that is to sort AS AN INTERMEDIATE RESULT.

I realize you just don't get the concept, so this is more wasted bandwidth.

 

[Harlan Grove]

almost anything that you want to do in Excel, you can do with MDX.. only
you don't need to copy and paste numbers into Excel LoL

....

So when a customer or intermediary sends you data in a dozen PDF files, how
do you get that data into your wonderful OLAP hypergrid? Pray? Telekenesis?

 

[Guest]

true VB.. I am pretty sure it was word for word compatable with VB6.

I used to work for CA.. I just can't remember the name of it... I'll keep on
looking

I think that all of us need to pressure CA into releasing the source of it--
they haven't done anything with this code for 3 or 4 years.. but I believe
it was basically VB6 compatable..

They just discontinued it since VS kindof cleaned up the market

 

[Guest]

whatever beancounter

ROTFLMAO - There comes a point when you really should realise you need to stop
digging that bloody great big hole you've dug for yourself. This is a game of
Chess and Harlan had you checkmated before you even started, even if you can't
see it.

--
Regards
Ken....................... Microsoft MVP - Excel
Sys Spec - Win XP Pro / XL 97/00/02/03

-------------------------------------------------------------------------- --
It's easier to beg forgiveness than ask permission
-------------------------------------------------------------------------- --




<snip>

 

[Guest]

more people _SHOULD_ have access.

Microsoft has 1/5 marketshare of RDBMS and they're obviously doing something
wrong. (I expected them to have double that market share by now)

They need to either DROP ACCESS or DO SOMETHING WITH IT.

When I say do something with it.. I mean:

a) include Access in the standard edition of Office-- all that other crap
like visio, infopath, and frontpage should be a part of the standard office
pro install
b) support their products and fix bugs in them
c) stop this mentality that *EVERYTHING MUST BE WEBBASED*-- it is a
distratction; please move on, Microsoft.
d) start competing more with IBM and Oracle-- push them out of business even
if it means you give away all of the SQL Server toolset.

That's all I'm saying.

I know that when I get to make decisions; that I will avoid Excel like the
plague. It isn't possible-- Excel doesn't have the best object model-- for
creating automated reports.

And I've seen company after company and manager after manager that wants
reports in Access.. and it's just like-- You have to watch out for what is
in their best interests.. and feeding someone data in Excel format is pretty
much a waste of time..

People should leave the number crunching to people with the right tools to
do it.

 

[Guest]

Access, SQL Server, OLAP, MDX and VB.Net.. oh gosh.. where do we start?

Access (ADP) connects to SQL Server. OLAP and MDX connect via Office Web
Components. If you don't have SQL Server; you can use MSDE... I think that
is an AWESOME strategy..

Access Data Projects are a familiar tool for Access people.. The main
difference between MDB and ADP is that ADP stores data in a SQL Server (or
MSDE) datastore, instead of in desktop (inefficient) files.

it is a revolution that happened 4 years ago.. and somehow beancounters were
left out.

It is all the same product. OLAP is magical.. if you haven't used it
(similiar to cognos, etc) then you are really short-changing yourself.

MDX is going to change the spreadsheet community.. and I hope for the best
of all of you-- that you spend some time and energy in broadening your
horizons..

MDX has what-- 80% of the functionality of Excel ... times the power of SQL
Server?

That is what all I am saying.

If you want to utilize OLAP long-term; you need to start doing your
'spreadsheet development' via Access.. hopefully ADP.. it looks just like
Excel for christ-sakes.. but it has real validation and real
programmability.

I'm NOT a VB.net guru, nor a proponent... Microsoft already has 3 flavors of
VB; why would they want another??

Microsoft needs to start offering a variety of tools-- like soemthing that
would help developers to translate between VB6 and VBscript.

They should have done that a long time ago.

When SQL2005 is more of a reality; I'm going to be pretty immersed in
VB.net-- for starters; I'll be able to start usign .net in DTS (I dont mean
bits and pieces-- i mean the whole thing).. as it is; it is too much of a
pain in the ass to have to maintain 3 different code bases.

 

[Guest]

I'll write webpages in whatever language I want.

I write them in VB (ASP with VBscript) because it is the easiest to
troubleshoot & debug.

any browser that doesn't support vbscript on the clientside doesn't fit on
my desktop.

the real question is why do firefox and opera not support client-side
vbscript?

it isnt' _THAT_ complicated of a language

 

[Guest]

and the fact that more poeple have access installed on their desktops-- this
means that they dont have to ask their boss for a copy of VS.. i mean-- you
can do a LOT of cool things with Access..

Access just has a lot of exciting programming features that Excel doesnt
have.

like the ability to email a report. this has been a part of access since
97; and just with Excel 2003; they decided to introduce some NEW imaging
solution.

it's just comical.

someone should have made an 'export to SNP' button in excel a long time
ago.. maybe i would take it seriously..

it just sux that you all email large spreadsheets around; and nobody can
really keep track of these spreadsheets since theyt live in 20 different
users' 'my documents' folders.

it's just comical.. that you guys-- you educated poeple-- think that it is
acceptable to use this POS obsolete langague called Excel.

 

[Guest]

I have more than 20.. and probably 10 carcasses.. that's probably the right
way to think of it.

 

[Guest]

Win2k baby

XP is for wimps

(e-mail address removed) wrote...

And they're all running all the time with every version of Windows from
3.1 through XP Professional, or are a few of them servers running
Windows Server 2003?

Let me echo your subject line: WHY?

 

[Guest]

I disagree that 'most websites run *nix'

I believe that apache does have quite a large marketshare (that doesnt mean
*nix); but most Fortune 500 companies rely on IIS, not apache.

and I have seen quite a few commercial apps written in VB; then again; I
have seen commercial apps ship with Access and the Access Runtime.. that is
a different argument tho.

I consult for a firm that has developed dozens of MS Access applications
that run against AS/400. I have built a couple of them myself.

These are legitimate apps-- they are easy to build, easy to troubleshoot.
They are fast; secure.. There isn't anything in this world that will make me
believe that an Access application with forms and reports isn't the right
way to do things.

I used to work for this ISP and we had this crm app written in vb6 and it
was commercial.

i totally agree that VB toolset was better than most others in the
mid-90s... and I'm pissed that Microsoft expects us all to move to C# just
because they *LIKE* it so much... (GAG)

Microsoft SQL Server DTS doesn't even support VB.net yet; there isnt' a
snowballs chance in hell that I'm going to focus on VB.net until SQL2005
comes out.. It's just not in my best interests to move code between:

VBA
VB6
VBS

(AND MICROSOFT DOESN'T HAVE ANY TOOLS FOR MEASURING VBS COMPLIANCE FOR
EXAMPLE)

and also re-write it to VB.net and/or C#-- there is like no benefit to it.

I'm not trying to talk sideways or anything.. I know VB.net well enough--
I've written a crazy spider / search engine/crawler in VB.net... I just
prefer to keep my coding in one language; and since I am a SQL Server
developer; and SQL Server doesn't really support .NET yet; I might as well
just bide my time..

I prefer Dreamweaver to anything else.. It is the best IDE for webdev, hands
down... I refuse to work where they won't let me use it... it has the
marketshare; it has the wizards.. it is a real IDE-- like Microsoft hasn't
come out with since VB6

whoever heard of an IDE where you can't edit your code when you debug?

That is like just ridiculous.. Like it was a step backwards. and that is
the reason that .net hasnt' gotten more traction-- it's like VB without the
VB debugging..

I just wish that MS would start innovating; start to really push the
envelope.

I wish that they would start recruiting and hiring bright people-- and they
would make these real people..

I don't think that i've ever met a person at MS that is qualified.. Maybe i
just haven't met the right people yet, i dont know.

They've got $70 billion in cash; and they can't afford to make stable and
secure programs.

I think that Microsoft needs to start firing people.. the people that made
the wrong strategic decisions soo many years ago.

Like naming it .net in the first place-- i mean.. couldn't we just PLZ call
things VS7 and VS8 and VS9?

And Office 2002 instead of Office XP.

And instead of re-naming products every 6 monhts; and adding products and
then discontinuing them-- it's just ridiculous to do things like that.

Mature companies shouldn't do things like that.

 

[Guest]

bottom of the ladder LoL

whatever bud.. i've got 20 machines at home; and i write code at home.. it's
a simple life.. i'm tired of being a consultant..

but i'm not going to accept that you get to be my manager just because you
know excel.

excel is the bottom of the ladder.

Access MDB isn't a one-size-fits-all solution.

But Access MDB scales a lot better than Excel does; and Access ADP is a
one-size fits all solution.

It scales to dozens of processors; with 255gb of ram..


--------------------------------------
- from the godfather http://www.fmsinc.com/tpapers/genaccess/DBOD.asp
Millions of databases are created in Excel spreadsheets each year, but only
a tiny percentage "graduate" to the next level: Access. Similarly, only a
tiny percentage of Access applications graduate to a more sophisticated
solution. In the interim, a huge number of database needs are solved
completely by Access. Access is simply the best at what it does.

An IT manager needs to understand and use Access strategically, and
anticipates that some Access applications migrate over time. This is not an
indictment on Access, but rather the natural process of database evolution
as business needs change. Sure, it would have been better to build that
Access application with a more sophisticated platform from the beginning,
but it was impossible to predict it would be that important when it was
first created. Similarly, is it possible to predict which 2% of databases
created this year need to migrate three years from now? Most will run
perfectly fine in Access forever or go extinct. Making a big investment
today makes no sense when a simpler, less risky Access solution is possible.
Let time determine which databases evolve and require additional investment
to take them to the next level. The key is to anticipate this.

Even when Access applications evolve to another platform, Access scales by
supporting the migration of Jet to SQL Server while preserving the
application development investment. The features developed for Access can be
rolled into the new platform guaranteeing the success of the new system (or
at least minimizing end-user objections). In that case, Access proved to be
a great prototype.

The savvy IT manager learns when Access is effective and when it's not. If
it can be done in Access, the ROI is superior to alternate technologies.
Taking advantage of the strengths of Access gives your organization a
significant competitive advantage both financially and in response to user,
market, and customer conditions.

 

[Guest]

in OLAP it would be this:

select {time.months} on columns,
{rate.children} on rows
from CUBENAME.

it would be easy to do in MDX; which means that you could do this in ONE
PLACE via MDX: or you could have this exact same logic in 100 different
places in an Excel spreadsheet.

What works the best when conditions change?

You can do all of this logic on the database side; and now that SQL2005 is
coming out-- you can even do your precious 'complex mathematical vba
functions' on the database side instead of the presentation tier.

it doesn't make sense to copy your business logic into 100 different places.

And your argument on templates was flawed.. if you dynamically build
something off of the source of the data; it's going to be more efficient to
change your logic.

I DONT WANT TO EVER CHANGE ANOTHER SPREADSHEET.

I DONT WANT TO STRESS OUT WHEN MY NUMBERS ARE WRONG.

I DONT WANT TO STRESS OUT WHEN YOUR NUMBERS ARE WRONG.

Aren't you just sick and tired of changing the same spreadsheet week in and
week out?

Don't you get tired when you document gets corrupted; or when it is too big
to email?

Don't you get pissed off when someone makes decisions 'off of the wrong
version of your spreadsheet'

The answer is to get rid of beancounters and spreadsheets and replace it
with databases and teams that decide business logic.

you can't afford to have numbers that you can't audit.

And I don't trust an auditor in the world that relies on Excel.

BECAUSE THEY DUPLICATE LOGIC IN 100 DIFFERENT PLACES; IT IS IMPOSSIBLE TO
HAVE PEACE OF MIND. IT IS IMPOSSIBLE TO _PROVE_ THAT YOUR NUMBERS WORK.

 

[Ken Wright]

Access MDB isn't a one-size-fits-all solution.

Correct

Access ADP is a one-size fits all solution.

Incorrect

There is NO SUCH THING as a one-size fits all solution, and your obsession with
Access has completely blinded you to the realities of life. It's a good
product, but it isn't the only thing in the world and never will be, and only a
fool would continue insisting it was.

but i'm not going to accept that you get to be my manager just because you
know excel.

Don't recall Harlan suggesting this at any time, so again smacks somewhat of
paranoia/obsession on your part.

It scales to dozens of processors; with 255gb of ram

Good, it can therefore do some of the things that you wouldn't use a spreadsheet
to do, and wouldn't want to use a spreadsheet to do. Reason being is because
most in here understand when another application is better suited to a
particular task, and are not obsessed with trying to fit their preferred
application where it doesn't belong.

--
Regards
Ken....................... Microsoft MVP - Excel
Sys Spec - Win XP Pro / XL 97/00/02/03

----------------------------------------------------------------------------
It's easier to beg forgiveness than ask permission
----------------------------------------------------------------------------

bottom of the ladder LoL

whatever bud.. i've got 20 machines at home; and i write code at home.. it's
a simple life.. i'm tired of being a consultant..

but i'm not going to accept that you get to be my manager just because you
know excel.

excel is the bottom of the ladder.

Access MDB isn't a one-size-fits-all solution.

But Access MDB scales a lot better than Excel does; and Access ADP is a
one-size fits all solution.

It scales to dozens of processors; with 255gb of ram..

<snip>

 

[Harlan Grove]

true VB.. I am pretty sure it was word for word compatable with VB6. ....
I think that all of us need to pressure CA into releasing the source of
it-- they haven't done anything with this code for 3 or 4 years.. but I
believe it was basically VB6 compatable..

....

Two things. First and picky, if CA did all the coding with nothing from MSFT
under license, how can you be sure it was 'true' VB? Second, if there was
any licensed code in the product, why do you believe CA could *legally* open
source it?

 

[Harlan Grove]

more people _SHOULD_ have access.

In the sense that there may be a few hundred high-end users who don't have
it who could put it to work, agreed. For the rest it's a cost/benefit
trade-off.

Microsoft has 1/5 marketshare of RDBMS and they're obviously doing something
wrong. (I expected them to have double that market share by now)

That or Microsoft (1) really can't compete when they face off against an
even bigger SOB (Oracle), (2) really can't compete on price performance
against *existing* mainframe databases, (3) can't compete against Linux and
BSD systems running other RDBMSs on web servers in terms of security (yeah,
I've heard that Windows Server 2003 will fix lots that's wrong with Windows
security, but I'm not holding my breath).

They need to either DROP ACCESS or DO SOMETHING WITH IT.

....

Of all the foolish things you've written to date, this is the most foolish.
Microsoft sells software to make people's lives wonderful?! No, Microsoft
sells software to make money, as much as they can for as little development
effort as possible. They are very, very good at doing so (though being a
monopoly in a country with a government that's the neutered and defanged lap
dog of big business doesn't exactly hurt). I think they have a MUCH BETTER
idea of how to make money than you do on their behalf, and if they believe
the expense of pushing Access isn't worth it, I'd be prepared to trust their
judgment. While I don't like a lot of what Microsoft does or sells, I give
them their due - they know how to wring money out of their customers.

d) start competing more with IBM and Oracle-- push them out of business even
if it means you give away all of the SQL Server toolset.

....

See previous comments about SOBs/Oracle.

With regard to IBM, I believe MSFT still licenses a lot of IBM patents. Hard
to drive a company out of business that can seriously screw up your own
products by revoking licenses. Also difficult to imagine how MSFT can do
anything against the mainframe market. Or the PPC market. Or any hardware
market. I believe mouses and the Xbox are the only nonsoftware products MSFT
has ever made money on.

People should leave the number crunching to people with the right tools to
do it.

Even more important, they should leave it to the people who know what number
crunching is and that it isn't database reporting.