End of "All-In-One" Microsoft OS

I have noted my displeasure with MS Windows, and its obvious drawbacks,
as have multitudes of others of users and analysts, and those drawbacks
have spawned a whole set of businesses designed JUST to let users
manage, fix and recover from the flaws of Windows, and the fixes for
the patches that don't work. MS has been made to look like a fool with
XP Pro's problems and failure of LongHorn/VISTA to reach market quickly
(and with the features they promised).

One user even implied or said I look at the world as if the "sky is
falling". If a MS Win XP Pro SP2 (SP2 is a long way into release)
operating system still gives problems and keeps people having to spend
time maintaining and recovering from various forms of crap and crashes,
it is not even a "Release Candidate".

Well, some research firms are writing about the "Vista will be the last
version of Windows" already. Techworld.com has run an article 8-25-06
by Matthew Broersma: Windows Vista the last of its kind:

I quote just 4 sentences:

Microsoft will be forced to migrate Windows to a modular architecture
tied together through hardware-supported virtualisation. "The current,
integrated architecture of Microsoft Windows is unsustainable - for
enterprises and for Microsoft," wrote Gartner analysts Brian Gammage,
Michael Silver and David Mitchell Smith.

The problem is that the operating system's increasing complexity is
making it ever more difficult for enterprises to implement migrations,
and impossible for Microsoft to release regular updates. This, in turn,
stands in the way of Microsoft's efforts to push companies to
subscription licensing.
[end quote]

Now, I can see further very good reasons to just keep using WinXP SP2
off the network and internet and just use what I have for another 3
years or so until everything shakes out. I really can't see bashing my
head on VISTA, only to have to migrate to another OS in a couple years,
and deal with the issues all over again.

I wonder how SolidWorks stability and upgrades will fare during the
coming couple transitions. I really really doubt SolidWorks will be
released to run on VISTA in 2007, so if VISTA creates to much
programming, I could easily see SolidWorks staying with Win XP Pro SP2
for years to come. (Better to work with the devil you know than the
devil that you don't AND that new devil costing you mucho millions in
programming costs.

OR, with virtualization around the corner, maybe the stability and
known quantity of UNIX is getting ready to be a viable solution for
complex end-user computing again.

I for one am going to be extremely cautious about accepting anything
Microsoft or SolidWorks has to say about new OSs and anything that
requires me to move to one. At the end, the last TWO WORDS of what I
quoted above, just made me CRINGE: SUBSCRIPTION LICENSING.

MICROSOFT HAS LOST MY TRUST: Microsoft tries to do everything, and in
the end has trouble just doing the OS right, and then they believe I
will bite for licensing? If I buy into licensing, then they are going
to have to guarantee 99.9% uptime or I get my money back plus
compensation for lost time beyond the .1%. I am sick of losing time to
fix their OS problems (frankly not on my machines anymore, as I have
stopped the problems: mostly other people's machines).

As much as I would like to see better features in SolidWorks, I might
do as a firm around the corner did, and lock into one version of
SolidWorks and just run that one version for 4-5 years.

You can't recover lost time, and you certainly do NOT get paid for it.

Bo
<http://www.techworld.com/news/index.cfm?RSS&NewsID=6718>

 

Windows is being forced to become more like *NIX everyday. The CLI
commands are growing so that repetive GUI based tasks are becoming
easier. There is much more scriptable access to the innards than I ever
knew existed in NT. Being a black box is becoming more of a detriment
to Windows now days. *NIX has been modular since day one. It can be run
with or without a GUI. There is a choice of GUIs. It will fit on a a
floppy, on an eprom or on a big hard drive in a supercomputer. Scaling
a system up and down can be done in a few minutes without a reboot. Try
that on WinServer 2003. You have to hang around till the network is not
being used to do the obligatory reboots. Just my opinion but Windows is
made up to a large part of bits of code bought from a bunch of sharp,
but proprietary developers while *NIX is made up of code that is to a
large extent open and constantly growing naturally and organically.

So if MSofts thinks *NIX is so good they have to copy it........

 

It may be that the ODF, Open Document Format may eventually be most
damaging to them, in that they lose their ability to lock in users to
MS Office.

Face it, most people with MSOffice do not use 90+% of the features in
the 'suite'.

ODF plus GPL'd open OS is a formidable challenge to MS long term.

Crappy undocumented, glitchy code doesn't survive in Open Software.

Bo

 

If Open Source was worth anything then

UGNX and Pro/E (at least for the time being) both support Linux.
If Dassault'd
http://www.hoise.com/primeur/06/articles/live/LV-PL-06-06-28.html
get smart instead of trying to bed down with Bill to get Catia V5
humping ... I'm going to keep hoping.

 

Are we talking about the operating system or applications here? I
thought it was the OS. And the LInux OS is stable. I have been running
it for years as a file server with nary a hiccup using SAMBA.

If we talk about applications, I have a shelf full of crappy,
undocumented Windows applications that I DID pay money for. There are
probably more of those than the Open Source variety. And they
frequently come from well known companies like Symantec and MicroSoft.
And unlike opensource, if they become obsolete or orphaned there is
absolutely nothing you can do about it whereas an opensource
application can be updated by anybody who cares enough about it to do
so.

When I use a Linux application I look for applications that are
supported and work. This is called shopping around. SAMBA, OpenVPN,
SmoothWall, mySQL, Apache, etc. are examples of Linux apps that you
may have used and not even known it.

And finally, proprietary apps, especially the kind used in an
enterprise generally come with support maintainance that must be paid
for and that are frequently worth little in day to day activities.

 

We are primarily talking primarily OS's.

NO DOUBT LINUX is mainstream for some things at this point. Part of
the reason has to be that if you really need a function, you can get it
implemented in house or with a consultant, and don't have to wait for
some megalithic corporation to do it on their time frame.

UNIX is also mainstream, and mostly open, so again, if you need a quick
code change for your use, it can be done.

Microsoft's Windows would have been great, but they failed to keep to
their promises to industry to provide an "Industrial Strength OS". A
friend who used to work as a CEO of a company working on applications
for the auto industry got badly burned when MS said they would deliver
and then didn't, and code crashed and MS wouldn't fix it. I think it
he knew MS wouldn't have delivered, he would not have sold his customer
on MS's OS.

Symantec sold me and my business copies of what was arguably the best
outliner on the Mac OS in the 90s, MORE. It could do outlines & slide
shows from your work, letters, etc., and, as they say, "Just Work".
Somewhere along the line, not only did they NEVER promote it, they then
stopped updating it.

Finally a company in Oregon recognized the need and started INSPIRATION
<www.inspiration.com> and made it for Windows & Mac, and it is a truly
wonderful basic outliner with ability to show them as Tree charts, and
then released a version used in grade schools as KIDSPIRATION. MS
Word's handling of Outline formatted text is about as clear to me as
Arabic, and I am not a newbie at using Windows apps. INSPIRATION is an
app you can use without reading the manual, and exports perfect
outlines to Word format, or PowerPoint in seconds.

Megalith Corporations have a function, but I am beginning to wonder
about their ability to survive in software.

CAD is such a specialty software with low user numbers compared to an
OS, that I can NOT see an app like SolidWorks being "free". The people
putting in programming hours have to get paid one way or another, and
CAD is literally only in its adolescent stage at best.

UNIX and Linux are arguably more mature, and the companies that make
money off those OSs do it with customization and training. That seems
like it is a supportable business model.

Windows? Yes in the short run, as end users are too afraid to leave
(like those in pre-WWII Germany). Long run, I do wonder about Windows,
but I will find a way to do it without hiccups as long as I use
SolidWorks. That may mean sticking with XP Pro SP2 and SWks 2006 or
2007 for the next 4-5 years, but that is just fine with me. I'm not
going to VISTA if it wastes lots of my time.

Bo

 

I noted " I'm not going to VISTA if it wastes lots of my time. "

If enough SolidWorks users say the same thing, SolidWorks is going to
have to support XP Pro way long into the future, and update code for
two versions of MS's OS, XP Pro & Vista.

If I read the news report right, Vista is going to be a moving target
for applications at least until somewhere after 2010.

Will SolidWorks support another OS? Not likely, but then their
programmers are already running Alphas, Betas of VISTA, and know of the
published roadmap for VISTA, and know of the prior MS plans and how
well MS did or DID NOT succeed in achieving results.

At this point I don't care. All I want is a bullet proof PC to run
SolidWorks. I have that now, though with restrictions I don't like (XP
+ SWks + MSO = one trick pony).

I don't care what the next OS I use is named, and I don't care whether
it runs on a free OS or one I pay for as that OS cost, even from MS, is
<5% the total cost of other pieces of software I buy.

Bo

 

....

Finally someone who can speak rationally about what is for others an
apparently emotional issue.

Well said. I fully agree.

 

I will agree that in the U.S. Linux is a desktop failure in terms of
market %. 3rd world markets are starting to go to it because of cost
pressures, and the improving GUIs are apparently acceptable to those
users (Ubuntu variant in India last week w/R. Stallman's help along the
line of Spain).

Since I use 2 desktop OSs day in and day out, I can truthfully state
that the Mac OS X is far and away my favorite, soley because:

I CAN ALWAYS COUNT ON OS X TO DELIVER W/O DOWNTIME.

There are people in this group that say I am <use one> a Mac zealot,
fanboy, machead. That is like someone saying SolidWorks users are
'SolidWorks Step and Fetchits'.

I use what works for me to get the most done in the least time, and
that means SolidWorks for 3D CAD, & PowerCADD for 2D CAD (Mac OSX).

Given the current trends where people are more and more recognizing
what OSs are more efficient, I predict:

1. Linux in servers is likely to continue to grow market share.
2. UNIX in various corporate, engineering, scientific and
mathematical uses will continue to grow, because it is so stable, well
known, and has such a large program & programmer base.
3. Mac OS X in desktop end user uses is likely to continue to grow
market share.
4. Windows OS in server, desktop, hand held PDAs, ebook, cell phone,
embedded devices, media center, et al, is simply unclear to me if they
can continue to grow doing it all.

When you have been King of the Mountain for a decade and every fawning
person around you keeps telling you, "We are the best" (or similar),
and you have the billions to "do everything", you start believing you
can "Do It All." Bill Gates is smart enough to be leaving near the
top. I get the feeling Ballmer will push and grind until he has a
heart attack on stage in mid leap. There is little precedence in
history of software to guide MS's CEO on how to "Do It All.", except to
note that IBM could not do so.

Bo

 

I think you are confusing success in numbers with functionality. There
is in fact a GUI for Linux that looks and feels just like Windows. The
average user probably wouldn't know the difference except that it would
be snappier than Windows on the MSoft OS, especially on slower
hardware.

MSoft spends millions on marketing, millions on getting developers to
port over applications and millions in getting Windows installed on new
computers. Linux doesn't.* So what Windows has over Linux is perception
and that is about it. If MSoft decided to become another RedHat you
would see Linux on every computer in the country in a year. In fact it
is scarry to think that MSoft could do what Mac did and put a *NIX OS
under their desktop.

MSoft as a desktop OS has serious technical drawbacks in
administration, performance and stability that while very real will not
be seen or understood by ordinary users.

Linux does well in many places other than servers. Lots of embedded
apps, routers, firewalls, NAS, etc. If you have a Linksys router
chances are you are running Linux.

*Maybe this Christmas we should see a Penguin inside commercial.

 

I don't know the GPL licensing issues, but suspect that Linux requires
whatever you put with it remain open also, but that is from my
non-legal type opinion.

Without MS being able to copyright and protect MS's Desktop GUI code in
number of ways, I don't see an M&Ms chance in hell that MS would use
Linux.

#1 reason is MS truly believes they know how to build a better system.

Ultimately, the marketplace buyers of products will decide who wins,
places and shows.

Bo

 

I'm not confusing anything.

The issue is TCO ( Total Cost of Ownership)
It's an absolute fact that the TCO for Linux system administration in an
enterprise
is substantially higher than Windows. The initial implementation is
cheaper but the
maintenance is a killer.

The fact is that MS Bashers (I consider myself one to a lesser extent)
who extol the virtues
of Linux, UNIX and Mac ignore the obvious in front of their own eyes.

Linux is a failure as per what it's own disciples have been predicting
for over 10 years.
It has not achieved any desktop traction and is destined to be a
miserable also ran on the desktop.
Linus T is also one of the main reason why GPL 2 simply won't be used
for it.
Linux isn't going anywhere on the desktop because you can't make a
profit by simply providing
a Linux distro. The closest to achieving anything toward this is
Linspire (formerly Lindows)

UNIX is NOT superior in terms of security or administration.
It certainly does have some advantages over Windows Networks especially
for large WANs.
However, UNIX is the OS that was routinely hacked at all levels in the past.
The DOD and Clifford Stoll can tell you something about it.
Steve Wozniak and his Apple buddies were Blue-Boxing International
phones calls for free.
These are only the most obvious intrusions.

All the Mac zealots heap praise on Apple and Macs in particular in spite
of the fact that
it is the epitome of a proprietary, CLOSED system and has been since day
one.
Apple Corp. treats 3rd party developers like the spawn of their
red-headed step children.

Boot Camp now provides dual-boot capabilities on Mac hardware and some
interesting
things have happened as a result. Benchmark tests show that WindowsXp
outperforms
OSX on the same hardware.

 

Linux isn't an operating system.

HP-UX is an operating system.
IRIX is an operating system.
Solaris is an operating system.
AIX is an operating system.
FreeBSD, OpenBSD, and NetBSD are operating systems.
Microsoft Windows are operating systems.

Linux is merely a kernel, and Linux desktop distributions are merely toys
for hobbyists and people who don't know any better.

I used Linux on my desktops and servers for quite some time. Now I know
better. My servers run FreeBSD and OpenBSD and my desktops run MS
Windows 2000 and XP. BSD systems are by far easier to install, administer
and maintain than Linux systems and Windows runs all the apps I need. No
real need for OS/X at this time.

Of course you can. See below.

BSD is growing just as much but without all the fanfare associated with
Linux, and fortunately BSD isn't hampered by the viral GPL which makes
it more attractive to developers.

NetBSD is more widely used than Linux and Windows in embedded
applications and suchlike, and it goes about doing it very quietly.

And you forgot something about OS/X. Under the hood it's Unix through and
through. You've got the Darwin Mach kernel which is based on 4.4BSD
Lite2, and a user-land based heavily on FreeBSD. OS/X is as stable and
robust as Unix because IT IS Unix.

OS/X is what Linux wants to be when it grows up, but Linux has no chance
of ever getting there.

 

Tim talks of TCO for an "enterprise", but my whole issue is software
used by an individual designer or small group, where by default, most
people have to do all their own software & OS maintenance.

I simply don't know what Linux might be, but I can tell you that
maintenance and training costs between Mac and Windows is not in
question, as even the Windows admin/tech trouble shooter guy next to me
admits he owes his ENTIRE income to the excessive maintenance required
of Windows systems.

If I lock down Win XP w/o network and Internet connections, I can get
it up to Mac OSX's level of up-time. But an OS like Win XP shouldn't
be restricted like that to keep it up and running.

Tim noted the closed vs open systems, but end users rarely care, and I
don't care what is open source and what is not. All they want is
things to be easy to use and stay up and running without all the fuss
and time.

I couldn't care which OS is "faster" as they are all fast enough for my
work.

It is VERY POSITIVE, that I can now run Win XP & Mac OSX & Linux on my
same Mac Book Pro, though I don't have an application which requires I
run Linux, so I don't.

For me to use Windows 100%, Microsoft is going to have to make a
massive improvement in ease of use and reliability on ALL levels.
Otherwise, if it continues with the inefficiencies, I'll stay with SWks
2006 or 2007 running on XP Pro on my Mac till hell freezes over (read
retirement).

Bo

 

I know something about TCO because we just added $10,000 to the cost of
setting up a server by going to Window 2003. Instead of Novell Linux we
had to go for Windows 2003, SQL Server 2000 and enough CALS to choke a
horse. SQL Server 2005 is the current version so we had to pay extra
for the obsolete but stable version. In addition to the cost of the
software (which was still higher than the extra hardware needed) we had
to upgrade the intended server with two RAID 10 arrays and a dual
Opteron setup to keep up with SQL Server. In addition to the $10,000
for the Windows "upgrade" we have to fork over $2,000 more for backup
software. And we now have to hire someone to run it because nobody
understands it, where the Linux system was easy to understand.

As far as the cost of system administration I have seen the White Paper
MicroSoft paid for. That paper only had a 35% cost increase for doing
common maintenance tasks. I find that hard to believe because I have
spent time trouble shooting NT, XP and Linux. I would far rather
trouble shoot Linux because it is so much easier to figure out what is
going wrong.

And when we installed the Server 2003 OS and SQL Server it wouldn't be
too much of a stretch to say that the time spent waiting for reboots on
the MSoft side was about equal the time to install the Linux system.

I know people with tight budgets for their enterprises and they run
Linux.

As far as security when we look at virus lists by OS, Microsoft is
invariably in the forefront by shear numbers. Has Linux been hacked?
Yes. Has Windows been hacked? Yes. But Windows is a virus magnet.

 

Last I knew it took about 1 sysadmin for each 50 MS Windows machines
and about 1 for each 150 Unix ones.
MS Windows has become much more complex since then
though.

What "maintenance" is at issue?

 

How do you arrive at that?
It has protections not even available on MS platforms last I knew.

Bad or poor sysadmins & users perhaps.

Have a link in mind?

 

Lots!

http://www.kernelthread.com/publications/security/uw.html

http://www.commercialventvac.com/~jeffs/WindowsVsLinuxSecurity.html

http://www.theregister.co.uk/2003/10/06/linux_vs_windows_viruses/

"To mess up a Linux box, you need to work at it; to mess up your Windows
box, you just need to work on it"

:-SecurityFocus columnist Scott Granneman.

Having worked on a dozen-odd flavours of Unix, even the worst-setup and
administered system is more stable than Windows.

Unix does not need to be re-installed every six months to bring it back
up to speed. The most important difference is, because Unix is 'hard',
i.e. has no pictures, users tend not to fiddle. After two days, a
newly-installed Windows system has had the background changed,new
default fonts, and 'a strange little thingie in the top of the screen
that won't go away'.

After twenty-five years in computer support, I would rather battle with
the obscure unix syntax, configuration files and oddly-named commands
than try to find one "Windows Protection Error".



Having lost most of my hair to computers, I now use them to run
engravers and routers - and we have never had to call in a support tech.

 

Total bunk.

Total bunk.

Learn how to properly administer MS Windows systems and you won't have
those problems.

HTH

--
Black Dragon

Camille's Axiom:
If you haven't asked yourself, "Why the hell did
I go to college anyway?", you must be teaching.

 

I will probably get a lot of flack for these comments but here goes:

If we give "constructive" criticism about some of our issues about SW,
there is a chance, (thought very slight) that SW may actually get
enough complaints that a problem will get fixed. And someone else may
actually have a work around etc. but, to moan and groan about MS in
this forum is probably a waste of time.

Then there is the question as if somone is having major problems with
SW on Windows the problem is probably the graphics driver the server,
(if there is one involved). My experience with Win2k is that it was
fairly stable and now that I have upgraded to XP, I can see that MS has
made a considerable effort to improveme the OS. There are plenty of
things about MS that I don't like either but the overall trend is that
every operating system has been a big improvement over the previous.
And the stability has improved.

Probably the biggest issue in the arguments that has been missed is the
cost. Here we are, (SW users- hopefully) and we are paying $4000+ for
the software and then we have to pay an average of $100 per month and
SW isn't always stable and there are tons of known bugs and has a major
upgrade every year, (where many of the known bugs don't seem to get
addressed) and now there is all this complaint about a piece of
software, (the OS) that is in the few hundreds of dollars range, is
fairly stable, etc. etc. and apparently MS does listen to it's client
base and has been making very good improvements. I am definatley not a
MS fan but the issues with SW compared to MS are so much more important
that this forum should probably focus on SW issues.

And if the arguments are taken to their logical conclusion, what is
being suggested is that SW should take part of the programming
resources and have them direct their efforts towards SW working on
other OS's. Personally, I would much rather have SW keep thier
programming resouces focused on resolving SW issues. At least until a
few more of the SW issues gets resolved.

Ed