08 Microsoft’s non-core
products often just don’t work; you must expect this and be willing to
quickly walk away from a test or tremendous amounts of time will be wasted.
Educating your clients of this is vital to a successful consulting business.
A recent example
involved MS Access and a SQL backend. I downloaded MS’s SQL Server 2005,
verified it worked by creating a new database and tables, and wasted hours
over 2 days attempting to connect to it from Access. The processes were
complex, badly documented and failed. I then downloaded an ODBC driver for
MYSQL; within minutes I was browsing it’s tables from Access.
Consider: these were
both new products from Microsoft , on the same PC; they were “designed” to
work together and didn’t. Consider these were my experiences years ago in a
How to explain this
? Look elsewhere for a moment; look at the recent fiasco of Vista; look too
at the profoundly bad engineering of the Office ribbon ( most importantly at
the needless lack of a switch to run the old menus the world knew by heart),
look at the destruction of VB 6 and the loss of countless yeas of
programming experience in it’s ruin. Look at the loop I fell into in Server
2008 as I attempted to turn on file sharing (this was evident of
amazingly poor programming and QC ). These kind of experiences are
utterly common when dealing with an MS product.
One might explain
MS’s continuation in the database server market as a strategic business
decision - but for the fact it never worked well for me and I hear endless
buzz in the industry of its mediocrity. How to make sense of this ?
Well, it’s clear
that the folks both writing the code and managing them get paid irrespective
of – of anything. Irrespective of SQL Server as a profit center.
Irrespective of it’s market penetration. Irrespective of feedback from me,
which they probably never get to read. If they needed it to work to get
paid, of course it would work as easily as MYSQL.
Worse though is that
they waste my time; they publish tools that don’t work and THEY DON’T CARE.
The sales of Windows and Office subsidize the wasting of my time. Whereas
in the late ‘80s and early ‘90s I supported MS, I would now benefit from
their demise; but for Access as a fine DB front-end I find nothing at MS
that cant be done better with Linux. From word processing to spreadsheets to
email to server processing...
Oddly someone is
not saying “We’re generating bad feelings doing this. If it does not work at
least warn people….”
Somewhere in the
byzantine world of MS is the absence of a core morality of business: “we are
here to be productive; to get rich by meeting market demand and in the
process help people” (see Adam Smith and Ayn Rand).
It could be that the
current crop of programmers and managers at MS are just not very good, just
unable to live up to those who handed them the code and the income stream -
but it often appears to me that no one working on the code or supervising
those that do care about what they are making; did anyone working on the
re-design of Access 2007 use it ? develop useful business tools with it ?
Do they even know people , like myself, who do ?
As for SQL Server,
you might say that they shouldn’t care of products that return no income.
They seem to take somewhat seriously the core products. I would guess that
hubris led to the mistakes of Vista , and one hears of sensitivity to this. Pride was common too at Ashton
Tate in ’88 and Novel in ’92 I would guess. Is this the reason no one
listened to ten of thousands of protests over the ruin of VB6 ? Is this the
reason they don't bother to supply useful help in Access ? How silly for a
company not to listen to it’s users.
As a consultant I
research hundreds of problems a year, thus I am in MS’s online “help”
hundreds of times a year; almost universally, literally 80% to 90% of the
time, it’s useless. It’s always at best dry and limited, but it’s often
plain wrong. It fails usually to admit there is any real problem. It seems
written by a corporate robot who was told to produce something –anything –
such that middle level management could tell upper : “yup, we have a help
system”. 3rd party help by comparison, when it can be found, is
always better. It’s written by people who sign their name and clearly care
to help their fellows ; more perhaps to the point it has no problem
admitting there is a problem.
We think this is an
odd way for a company to behave, because when we mistakenly produce
problematic code, our response is “ holly smokes ! fix it , and call the
customer and let them know ! “ We never get this sense from MS …
A funny example is
the time MS led me to waste in deciding if I should turn on per user or per
device licensing in Terminal Server 2008. After reading through pages of
horridly written MS “help” documentation I came across 3rd party
help that told me that (still ! I knew this of the 2003 version ) MS doesn’t
track the number of users logged in to TS ! How droll ! So I didn’t have to
worry or bother at all ! If MS cared about my experience with them or my
time they might have told me this themselves – but that would be admitting a
“problem” with their configuration of the 2008 OS; so they care as much
about the efficient use of my time as the town and state governments I deal
MS was once a
engineering firm, driven to make our use of computers more productive. It is
now a marketing agency with a programming shop in the back. Changes are
made not from logic, not to advance usefulness, but to force/justify new
sales. As an Access programmer with scores of apps in production I was
amazed at how Access 2007 was toyed with; hurt for no reason. Changes were
made for the sake of change (as with the ribbon generally). Clearly no one
who used or cared about the use of the product had any say over the
re-design. My fellows in the industry know this; to clients reading this
let me say: I get annoyed at it but I can make it work. ( and let me add
that the more I play with Linux the more I think you should consider it