[Info-vax] HP wins Oracle Itanium case

[Paul Sture]

On Thu, 23 Aug 2012 15:38:51 -0600, Howard S Shubs wrote:

> In article <k1651d$h31$1 at dont-email.me>,
>  Stephen Hoffman <seaohveh at hoffmanlabs.invalid> wrote:
> 
>> I was not referring to Windows, there.
> 
> I try to ignore the fact that they put a registry on VMS.  I hope it's
> not related to the Windows one.

IIRC it arrived as a result of the <cough> Affinity Program <apologies 
for using bad language>, and was there for things like COM, SOAP etc.

As to the Windows Registry itself, the idea of putting stuff in a central 
repository rather than having it scattered all over the disk was a good 
one - think SYSGEN versus all the .INI files that were in Windows 3.1.

However, the implementation left much to be desired.  Don't take my word 
for it though, here's the opinion of someone who reverse engineered the 
Windows Registry (Feb-2010):

"Why the Windows Registry sucks … technically"

http://rwmj.wordpress.com/2010/02/18/why-the-windows-registry-sucks-
technically/

Here are the points he addresses:

1. It’s a half-arsed implementation of a filesystem
2. Hello Microsoft programmers, a memory dump is not a file format
3. The implementation of reading/writing the Registry in Windows NT is
   poor
4 Types are not well specified
5 Interchange formats are not well specified
6 The Registry arrangement is a mess
7 The Registry is a filesystem
8 Security, ha ha, let’s pretend
9 The Registry is obsolete, sorta

My observation with the Registry and Windows Server 2008?  Well yes, 
there are tons of potentially useful features in the Windows Server 
products, but when I saw as part of a patch over 100K Registry entries 
being updated I was wondering if the Registry hadn't just got a bit out 
of control...

-- 
Paul Sture