Well anyway, today I *finally* got my Excel spreadsheet working again!

It stopped working when I upgraded from Office 97 to Office 2003. 
Specifically, my VBA macro was disabled. (Damn it, I spent *hours* 
writing that thing! I *want* it to work!!)

Apparently it's disabled as a security measure. You'd *think* there 
would be a button you could just press to say that *this* macro is OK to 
run. (That's what Office 97 had.) But no, apparently that would be too easy.

Google did at least manage to find me a relevant article without much ado.

http://msdn2.microsoft.com/en-us/library/aa141471(office.10).aspx

It turns out that the correct process for enabling a macro that you 
wrote to run on your own PC is as follows:

1. Find the product installation CD.

2. Install the certificate tools.

(By default they install VBA but not the tools apparently necessary to 
actually enable it to run. But then, have you *seen* VBA? I guess the 
assumption that you won't want to use it is fairly safe...)

3. Nagivate several system folders. (You know, the ones where Windows 
Explorer insists "This is dangerous. We have hidden it for your own 
safety. Do you *really* want to touch this stuff? We warned you!")

4. Run the certificate creation tool. This creates a self-signed 
certificate which you cannot export or back up. (Yay!)

5. Open the spreadsheet.

6. Open the VB Editor.

7. Find the right menu item.

8. Select the certificate you just created and add the signature.

9. Save the spreadsheet.

10. Close the spreadsheet.

11. Reopen the spreadsheet.

(The instructions fail to mention that the code being signed doesn't 
help you until the *next* time you open the thing...)

I wonder - how do you develop new code if it's always disabled until you 
sign it? (And - one hopes - every time you change it this invalidates 
the signature...)

All this fuss just so I can click a button to put today's date in a cell 
on the spreadsheet... sheesh!

Post a reply to this message

> All this fuss just so I can click a button to put today's date in a cell 
> on the spreadsheet... sheesh!

Or you could just have gone to Options -> Security and turned down Macro 
security to Medium, then at least you get the old style "do you want to run 
this or not" box for unsigned macros...

Post a reply to this message

And lo on Thu, 06 Dec 2007 10:41:58 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

> Well anyway, today I *finally* got my Excel spreadsheet working again!
>
> It stopped working when I upgraded from Office 97 to Office 2003.  
> Specifically, my VBA macro was disabled. (Damn it, I spent *hours*  
> writing that thing! I *want* it to work!!)
>
> Apparently it's disabled as a security measure. You'd *think* there  
> would be a button you could just press to say that *this* macro is OK to  
> run. (That's what Office 97 had.) But no, apparently that would be too  
> easy.

As Scott said you just turn down the security options from "I'm an idiot  
protect me" to "I'm a grown-up let me make my own decisions"

> Google did at least manage to find me a relevant article without much  
> ado.
>
> http://msdn2.microsoft.com/en-us/library/aa141471(office.10).aspx

or you could have checked your posts as I've told you how to do this twice  
now :-)

> It turns out that the correct process for enabling a macro that you  
> wrote to run on your own PC is as follows:
>
> 1. Find the product installation CD.
>
> 2. Install the certificate tools.
>
> 3. Nagivate several system folders. (You know, the ones where Windows  
> Explorer insists "This is dangerous. We have hidden it for your own  
> safety. Do you *really* want to touch this stuff? We warned you!")

or run a search for selfcert.exe

> 4. Run the certificate creation tool. This creates a self-signed  
> certificate which you cannot export or back up. (Yay!)

Um you mean except when you go to sign it, view details and pick "Copy to  
file"?

> 5. Open the spreadsheet.
>
> 6. Open the VB Editor.
>
> 7. Find the right menu item.

Tools|Digital Signature.. at least on 2k, hardly surprising.

> 8. Select the certificate you just created and add the signature.
>
> 9. Save the spreadsheet.
>
> 10. Close the spreadsheet.
>
> 11. Reopen the spreadsheet.
>
> (The instructions fail to mention that the code being signed doesn't  
> help you until the *next* time you open the thing...)

Well yeah by default the macros are disabled so you need to reopen it for  
the signature to be taken into account then it should say signed by blah  
trust this certificate and that's it.

> I wonder - how do you develop new code if it's always disabled until you  
> sign it? (And - one hopes - every time you change it this invalidates  
> the signature...)

It gets re-signed if the certificate that originally signed it is present,  
so someone else souldn't re-sign the code and distribute it unless they  
were using your computer/login, besides you should lock the project for  
viewing anyway.

> All this fuss just so I can click a button to put today's date in a cell  
> on the spreadsheet... sheesh!

But all that hassle was a one-off you've now got a certificate to sign  
anything VBA with.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Phil Cook wrote:

> As Scott said you just turn down the security options from "I'm an idiot 
> protect me" to "I'm a grown-up let me make my own decisions"

Well I pretty much assumed it would be an on/off settings, and even I am 
not too keen on turning the security off completely.

>> Google did at least manage to find me a relevant article without much 
>> ado.
>>
>> http://msdn2.microsoft.com/en-us/library/aa141471(office.10).aspx
> 
> or you could have checked your posts as I've told you how to do this 
> twice now :-)

I followed the instructions you gave me for Office 97 and they didn't 
work. (I forget where exactly it fails.) The instructions I found here 
actually worked.

>> 3. Nagivate several system folders. (You know, the ones where Windows 
>> Explorer insists "This is dangerous. We have hidden it for your own 
>> safety. Do you *really* want to touch this stuff? We warned you!")
> 
> or run a search for selfcert.exe

Actually I used a command prompt, but hey.

>> 4. Run the certificate creation tool. This creates a self-signed 
>> certificate which you cannot export or back up. (Yay!)
> 
> Um you mean except when you go to sign it, view details and pick "Copy 
> to file"?

I did follow the instructions for backing up the certificate, but 
apparently "the private key is marked as not exportable and cannot be 
saved". So you can save the certificate itself, just not the key. Very 
helpful.

>> 7. Find the right menu item.
> 
> Tools|Digital Signature.. at least on 2k, hardly surprising.

It's easy once you figure out it's under Tools, yes.

>> I wonder - how do you develop new code if it's always disabled until 
>> you sign it? (And - one hopes - every time you change it this 
>> invalidates the signature...)
> 
> It gets re-signed if the certificate that originally signed it is 
> present, so someone else souldn't re-sign the code and distribute it 
> unless they were using your computer/login, besides you should lock the 
> project for viewing anyway.

Interesting. So if you change something when you didn't mean to, it 
updates the signature? Oh dear...

>> All this fuss just so I can click a button to put today's date in a 
>> cell on the spreadsheet... sheesh!
> 
> But all that hassle was a one-off you've now got a certificate to sign 
> anything VBA with.

Until the next time I reinstall Windoze...

PS. Does OpenOffice provide a scripting language?

Post a reply to this message

Invisible a écrit :
> PS. Does OpenOffice provide a scripting language?

Several of them, it seems:
http://framework.openoffice.org/scripting/index.html

I never used that, though...

-- 
Vincent

Post a reply to this message

Vincent Le Chevalier wrote:
> Invisible a écrit :
>> PS. Does OpenOffice provide a scripting language?
> 
> Several of them, it seems:
> http://framework.openoffice.org/scripting/index.html
> 
> I never used that, though...

"OpenOffice Basic"? Oh, that's original...

Post a reply to this message

And lo on Thu, 06 Dec 2007 12:23:41 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

> Phil Cook wrote:
>
>> As Scott said you just turn down the security options from "I'm an  
>> idiot protect me" to "I'm a grown-up let me make my own decisions"
>
> Well I pretty much assumed it would be an on/off settings, and even I am  
> not too keen on turning the security off completely.

No amazingly they did add an adult option alongside the absolutes.

>>> Google did at least manage to find me a relevant article without much  
>>> ado.
>>>
>>> http://msdn2.microsoft.com/en-us/library/aa141471(office.10).aspx
>>  or you could have checked your posts as I've told you how to do this  
>> twice now :-)
>
> I followed the instructions you gave me for Office 97 and they didn't  
> work. (I forget where exactly it fails.) The instructions I found here  
> actually worked.

Hmm okay.

>>> 3. Nagivate several system folders. (You know, the ones where Windows  
>>> Explorer insists "This is dangerous. We have hidden it for your own  
>>> safety. Do you *really* want to touch this stuff? We warned you!")
>>  or run a search for selfcert.exe
>
> Actually I used a command prompt, but hey.
>
>>> 4. Run the certificate creation tool. This creates a self-signed  
>>> certificate which you cannot export or back up. (Yay!)
>>  Um you mean except when you go to sign it, view details and pick "Copy  
>> to file"?
>
> I did follow the instructions for backing up the certificate, but  
> apparently "the private key is marked as not exportable and cannot be  
> saved". So you can save the certificate itself, just not the key. Very  
> helpful.

Try makecert instead.

>>> 7. Find the right menu item.
>>  Tools|Digital Signature.. at least on 2k, hardly surprising.
>
> It's easy once you figure out it's under Tools, yes.

or read the article you linked to :-)

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

>> I followed the instructions you gave me for Office 97 and they didn't 
>> work. (I forget where exactly it fails.) The instructions I found here 
>> actually worked.
> 
> Hmm okay.

Well, to be honest it's getting quite hard to find information about 
Office 97. Hopefully Office 2003 will be an easier proposition...

>> I did follow the instructions for backing up the certificate, but 
>> apparently "the private key is marked as not exportable and cannot be 
>> saved". So you can save the certificate itself, just not the key. Very 
>> helpful.
> 
> Try makecert instead.

It's not there.

Post a reply to this message

And lo on Thu, 06 Dec 2007 12:57:53 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

>>> I followed the instructions you gave me for Office 97 and they didn't  
>>> work. (I forget where exactly it fails.) The instructions I found here  
>>> actually worked.
>>  Hmm okay.
>
> Well, to be honest it's getting quite hard to find information about  
> Office 97. Hopefully Office 2003 will be an easier proposition...

Well at least until Office 2010

>>> I did follow the instructions for backing up the certificate, but  
>>> apparently "the private key is marked as not exportable and cannot be  
>>> saved". So you can save the certificate itself, just not the key. Very  
>>> helpful.
>>  Try makecert instead.
>
> It's not there.

GIYF

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

On Thu, 06 Dec 2007 12:50:02 +0000, Invisible wrote:

> Vincent Le Chevalier wrote:
>> Invisible a écrit :
>>> PS. Does OpenOffice provide a scripting language?
>> 
>> Several of them, it seems:
>> http://framework.openoffice.org/scripting/index.html
>> 
>> I never used that, though...
> 
> "OpenOffice Basic"? Oh, that's original...

It's descriptive.  You'd be happier if it was called, I don't know, 
"OpenOffice Splash"?

Jim

Post a reply to this message

Invisible wrote:
> (That's what Office 97 had.)

Actually, Office 97 had a way for a macro to turn on the "don't disallow 
macros" flag. A true security cluster-fk. Like, huh? Disallow macros, 
unless the macro with the malware in it says to allow macros?

> (By default they install VBA but not the tools apparently necessary to 
> actually enable it to run. 

No. The tools to let it run, but not the tools to make new macros. Just 
like having by default a Java VM runtime installed without installing 
the Java compiler.

> I wonder - how do you develop new code if it's always disabled until you 
> sign it?

The same way you develop new code if it's always unrunnable until you 
compile it.

> (And - one hopes - every time you change it this invalidates 
> the signature...)

Yes. That's the point of it.

-- 
   Darren New / San Diego, CA, USA (PST)
     It's not feature creep if you put it
     at the end and adjust the release date.

Post a reply to this message

And MS could have made it easier by identifying the minimum subset of 
macro functions that are necessary for virus propagation, and 
eliminating enough of those functions to make virus propagation impossible.

Simply never writing macros to normal.dot would have stopped the 
propagation of many viruses, and depriving macros of the ability to 
disable menu commands would have helped, too.

Regards,
John

Post a reply to this message

John VanSickle <evi### [at] hotmailcom> wrote:
> Simply never writing macros to normal.dot would have stopped the 
> propagation of many viruses, and depriving macros of the ability to 
> disable menu commands would have helped, too.

  It's not the first time when MS's concept of fixing a security hole is
to either ignore it (by argumenting it's not a problem) or going completely
overboard, instead of actually fixing the problem itself.

  Somehow it gives the impression of a beginner and proud-of-itself
programmer who is given a bug report. He either is too proud to admit
the problem, or can't imagine a better solution to it than to disable
half of the functionality of the program. You know, like those cases you
can constantly read at the daily WTF.

-- 
                                                          - Warp

Post a reply to this message

I basically agree with everything you two have said.

How about you ask the user before letting a macro perform a potentially 
risky action? (Unless it's signed of course.)

OTOH, some idiots will click anything put in front of them, so... how 
about just turn off all potentially unsafe functionallity unless the 
macro is signed, and say "hey, get the macro author to sign this if you 
really want it to work"? (But provide no way to actually enable the 
macro just by clicking the window.)

The vast majority of macros are for auto-generating document content. If 
you turn off the ability to access other files / documents and disable 
changing the user's settings, it's pretty much impossible for a 
malicious macro to do anything except screw up the document it's already 
infected. Dude, how hard is that?

But hey, why do that when you can just completely disable all macro 
functionallity?

(Question: Has anybody ever actually *seen* a macro virus? I'm told they 
exist, but I've never ever come across one...)

Post a reply to this message

And lo on Fri, 07 Dec 2007 09:58:06 -0000, Invisible <voi### [at] devnull> did  
spake, saying:

> I basically agree with everything you two have said.
>
> How about you ask the user before letting a macro perform a potentially  
> risky action? (Unless it's signed of course.)

I believe that's called Vista ;-)

> OTOH, some idiots will click anything put in front of them, so... how  
> about just turn off all potentially unsafe functionallity unless the  
> macro is signed, and say "hey, get the macro author to sign this if you  
> really want it to work"? (But provide no way to actually enable the  
> macro just by clicking the window.)

Then you'd just get the 'unsafe' macros being signed, unless you want to  
force everyone to buy a certificate?

> The vast majority of macros are for auto-generating document content. If  
> you turn off the ability to access other files / documents and disable  
> changing the user's settings, it's pretty much impossible for a  
> malicious macro to do anything except screw up the document it's already  
> infected. Dude, how hard is that?

Except where you want a macro to be able to access other documents and  
files and change settings. For example IIRC in one version of Word to  
print out a document to a non-default printer via VBA requires you to  
change the default printer to the one you want to print to then change it  
back again.

> But hey, why do that when you can just completely disable all macro  
> functionallity?

Because it's easier

> (Question: Has anybody ever actually *seen* a macro virus? I'm told they  
> exist, but I've never ever come across one...)

In the early days when they were new, sure. Not so much nowadays.

-- 
Phil Cook

--
I once tried to be apathetic, but I just couldn't be bothered
http://flipc.blogspot.com

Post a reply to this message

Warp wrote:
> John VanSickle <evi### [at] hotmailcom> wrote:
>> Simply never writing macros to normal.dot would have stopped the 
>> propagation of many viruses, and depriving macros of the ability to 
>> disable menu commands would have helped, too.
> 
>   It's not the first time when MS's concept of fixing a security hole is
> to either ignore it (by argumenting it's not a problem) or going completely
> overboard, instead of actually fixing the problem itself.

Aside from the recurring buffer overruns bugs (it seems like one of 
those pop up every month[1]), every security hole appears to involve a 
feature that MS added for its own benefit, and not for the user's.

IE is a good example of this.  Frankly, everything that isn't directly 
related to displaying content formatted in HTML should be relegated to 
plug-ins that the user can shut off at will.  That includes automatic 
download, install, JavaCurse^H^H^H^H^HScript, and so on.

>   Somehow it gives the impression of a beginner and proud-of-itself
> programmer who is given a bug report. He either is too proud to admit
> the problem, or can't imagine a better solution to it than to disable
> half of the functionality of the program. You know, like those cases you
> can constantly read at the daily WTF.

Disabling half of the functionality in IE would be a pretty good idea.

Regards,
John

Post a reply to this message

John VanSickle wrote:

> Disabling half of the functionality in IE would be a pretty good idea.

Or perhaps... you know... providing a way to uninstall IE? ;-)

-- 
http://blog.orphi.me.uk/

Post a reply to this message

Orchid XP v7 <voi### [at] devnull> wrote:
> Or perhaps... you know... providing a way to uninstall IE? ;-)

  There is: Uninstall Windows.

-- 
                                                          - Warp

Post a reply to this message

"Orchid XP v7" <voi### [at] devnull> wrote in message
news:475f19d0$1@news.povray.org...
> John VanSickle wrote:
>
> > Disabling half of the functionality in IE would be a pretty good idea.
>
> Or perhaps... you know... providing a way to uninstall IE? ;-)

Add/remove programs -> Add/Remove Windows Components. Uncheck Internet
Explorer.
(At least on my windows 2000 box)

Post a reply to this message

Gail Shaw escribió:
> "Orchid XP v7" <voi### [at] devnull> wrote in message
> news:475f19d0$1@news.povray.org...
>> John VanSickle wrote:
>>
>>> Disabling half of the functionality in IE would be a pretty good idea.
>> Or perhaps... you know... providing a way to uninstall IE? ;-)
> 
> Add/remove programs -> Add/Remove Windows Components. Uncheck Internet
> Explorer.

Did you read the description of that "component" carefully? The only 
thing that will remove is the SHORTCUT.

Post a reply to this message

"Nicolas Alvarez" <nic### [at] gmailisthebestcom> wrote in message
news:475f676a$1@news.povray.org...
> Gail Shaw escribió:
> > "Orchid XP v7" <voi### [at] devnull> wrote in message
> > news:475f19d0$1@news.povray.org...
> >> John VanSickle wrote:
> >>
> >>> Disabling half of the functionality in IE would be a pretty good idea.
> >> Or perhaps... you know... providing a way to uninstall IE? ;-)
> >
> > Add/remove programs -> Add/Remove Windows Components. Uncheck Internet
> > Explorer.
>
> Did you read the description of that "component" carefully? The only
> thing that will remove is the SHORTCUT.

Ah, no. That would explain the 0k.

I know there is a way, did it some time back on a previous install. Can't
remember now. Left it on this one cause there's too many sites still that
won't work with firefox.

Post a reply to this message

Gail Shaw wrote:
> I know there is a way, did it some time back on a previous install. 

IE itself is a thin layer on top of a whole bunch of libraries that a 
whole lot of other things depend on. Like the help system, for example, 
as well as a bunch of download mechanisms. I don't think you ever 
actually managed to pull the tentacles of IE out of Windows. It would be 
like that scene from 20,000 Leagues with the giant squid, except it also 
has its tentacles plugged into Auntie Em, who yelps every time you tug.

-- 
   Darren New / San Diego, CA, USA (PST)
     It's not feature creep if you put it
     at the end and adjust the release date.

Post a reply to this message

> Or perhaps... you know... providing a way to uninstall IE? ;-)

As Warp said, you'll have to uninstall Windows to do that.  If somehow you 
actually managed to remove IE completely, then a whole load of other stuff 
would break.  WinPOV Help would be one thing, explorer would probably break 
too.

Post a reply to this message

Darren New <dne### [at] sanrrcom> wrote:
> IE itself is a thin layer on top of a whole bunch of libraries that a 
> whole lot of other things depend on.

  I think it started with Win98 and went only downhill from there.

  With Win98 MS had this braindead idea that they should make the desktop
and most applications to look&feel like webpages (or at least support making
them so). This, of course, is exactly the wrong direction to go.

  Webpages are very limited in functionality because HTML is very limited
in functionality. There are many, many things you can't implement in a
webpage because HTML was and still is purposefully simple. In a desktop GUI
trying to emulate a webpage is exactly the wrong direction to go, and it
should be the other way around: Webpages should aim to work like desktops,
not the other way around.

  What I really hate about Microsoft's braindead ideas is that most people
seem happy to blindly copy them. Windows 3 had several good design and
behavior decisions which were dropped from subsequent Windowses just for
the sake of change, just for the sake of making it look&feel *different*,
in many cases at the cost of reduced usability.

  Just as an example is the behavior of menus and sub-menus: In Win3 it was
standard practice that if you opened a menu and it had a sub-menu, you had
to click on the name of the sub-menu to open it, and then it stayed open
for as long as you clicked something else.
  Well, MS had this "great" idea to change this behavior from clicking to
hovering instead: Sub-menus open by hovering the sub-menu name and they
close when you unhover. Opening it with hovering is not completely bad
(although it becomes irritating when the sub-menu has something which takes
time to build, such as some of the sub-menus of the context menu of windows
explorer), but closing the sub-menu by unhovering is the most braindead
idea ever. You hover the name of the sub-menu, then start moving the mouse
to the right (or in some cases to the left...) to get to it, accidentally
unhover the name of the sub-menu by accident, and it closes. How this could
ever be a good idea is beyond by comprehension. You have to be really
proficient at aiming with the mouse in order to keep the sub-menu open. Not
all people are.

  Speaking of context menus, that's another step backwards MS took:
Previously you could open a context menu by right-clicking and then,
while keeping the mouse button down, select the item you wanted and then
release the mouse button and it would select it. This was a nice power-user
feature. For example selecting "Back" in a web browser was fast and easy
this way. Just press, move a bit down, release.
  But no, they had to change this. Now you can't do that. You have to press,
then release, then move down, then press, then release. Not a big deal, but
very irritating when you got used to the old behavior. The new behavior is
awkward.

  The biggest problem with this is that everyone is blindly copying these
braindead design decisions everywhere, often even in non-windows software.
For example KDE has the hover-to-open-and-close-submenus idea implemented,
which is irritating.

  Fortunately other systems have yet to copy the most braindead GUI ideas
from MS. For example menus which change contents every time you use them.
(I can't even begin to imagine how anyone could think that's a good idea.
When MS wants to make changes just for the sake of changes, just to make
a new Windows look different than the previous one, they will stop at
nothing.)

-- 
                                                          - Warp

Post a reply to this message

Warp wrote:
> Orchid XP v7 <voi### [at] devnull> wrote:
>> Or perhaps... you know... providing a way to uninstall IE? ;-)
> 
>   There is: Uninstall Windows.

Mmm. That's not an inherantly bad idea...

-- 
http://blog.orphi.me.uk/

Post a reply to this message

"Darren New" <dne### [at] sanrrcom> wrote in message
news:475f6e3b$1@news.povray.org...

> IE itself is a thin layer on top of a whole bunch of libraries that a
> whole lot of other things depend on.

Yeah, I know. I don't think the whole mess was missing. I know I've had
machines where internet explorer wasn't there.

Is all fine til some braindead installer decides to invoke IE directly,
rather than calling for the app that reads htm pages

Post a reply to this message

Warp wrote:
> the sake of change, just for the sake of making it look&feel *different*,
> in many cases at the cost of reduced usability.

Well, they *are* trying to make a profit. It's understandable, even if 
suboptimal. :-)

>   Well, MS had this "great" idea to change this behavior from clicking to
> hovering instead: 

Yah, I hate that too. Actually, I think there's a switch you can set to 
say don't do that, but I don't know if it disappears the menu if you 
move away after you click. I'll have to try that.

>   Speaking of context menus, that's another step backwards MS took:

Smalltalk did one better years earlier. Context menus not only open 
under the mouse, but position themself so that the previously chosen 
menu selection is under the cursor. If you're pasting the same thing 
into a dozen windows (because, say, you haven't yet automated 
broadcasting restarts to your server farm), you don't wind up with the 
mouse down in your lap before you're done.

>   Fortunately other systems have yet to copy the most braindead GUI ideas
> from MS. For example menus which change contents every time you use them.
> (I can't even begin to imagine how anyone could think that's a good idea.
> When MS wants to make changes just for the sake of changes, just to make
> a new Windows look different than the previous one, they will stop at
> nothing.)

Yah, well, that's for-fee software for ya. Gotta get people to buy the 
newer version, ya know.

-- 
   Darren New / San Diego, CA, USA (PST)
     It's not feature creep if you put it
     at the end and adjust the release date.

Post a reply to this message

scott wrote:
>> Or perhaps... you know... providing a way to uninstall IE? ;-)
> 
> As Warp said, you'll have to uninstall Windows to do that.  If somehow 
> you actually managed to remove IE completely, then a whole load of other 
> stuff would break.  WinPOV Help would be one thing, explorer would 
> probably break too.

What, you mean like if you turned RPC off? :-}

Post a reply to this message

Invisible escribió:
> scott wrote:
>>> Or perhaps... you know... providing a way to uninstall IE? ;-)
>>
>> As Warp said, you'll have to uninstall Windows to do that.  If somehow 
>> you actually managed to remove IE completely, then a whole load of 
>> other stuff would break.  WinPOV Help would be one thing, explorer 
>> would probably break too.
> 
> What, you mean like if you turned RPC off? :-}

Huh?

WinPOV Help is a .chm file. HTML Help. The viewer uses the IE rendering 
engine to show pages. Uninstall IE completely, including its rendering 
engine, and no more HTML help for you.

Post a reply to this message

>> What, you mean like if you turned RPC off? :-}
> 
> Huh?

Turning RPC off reputedly causes Windoze-based PCs to malfunction to the 
point of being unuseable. Great if you want to close this large security 
hole...

-- 
http://blog.orphi.me.uk/

Post a reply to this message

"Orchid XP v7" <voi### [at] devnull> wrote in message
news:47617baa$1@news.povray.org...
> >> What, you mean like if you turned RPC off? :-}
> >
> > Huh?
>
> Turning RPC off reputedly causes Windoze-based PCs to malfunction to the
> point of being unuseable. Great if you want to close this large security
> hole...
>

It's not that large a security hole. Unless you fiddle with the default
security, it's remote process communication on your own machine.

RPC is used by various processes of windows to communicate with each other

Post a reply to this message

Gail Shaw wrote:
> "Orchid XP v7" <voi### [at] devnull> wrote in message
> news:47617baa$1@news.povray.org...
>>>> What, you mean like if you turned RPC off? :-}
>>> Huh?
>> Turning RPC off reputedly causes Windoze-based PCs to malfunction to the
>> point of being unuseable. Great if you want to close this large security
>> hole...
>>
> 
> It's not that large a security hole. Unless you fiddle with the default
> security, it's remote process communication on your own machine.
> 
> RPC is used by various processes of windows to communicate with each other

Depends on your definition of "large".

As I understand it, anybody who can guess what your admin password is 
can have unlimited access to your PC via RPC. Nice...

-- 
http://blog.orphi.me.uk/

Post a reply to this message

"Orchid XP v7" <voi### [at] devnull> wrote in message
news:47617e7a$1@news.povray.org...
>
> As I understand it, anybody who can guess what your admin password is
> can have unlimited access to your PC via RPC. Nice...>

Anyone who's guessed your admin password owns your machine regardless.

del \\machine\c$\*.* /S

And that's not even considering them walking up to the machine (if its not a
server) and loging in. Or if it is a server, usign remote desktop/terminal
services

Strong admin password, perhaps?

Post a reply to this message

Gail Shaw wrote:
> "Orchid XP v7" <voi### [at] devnull> wrote in message
> news:47617e7a$1@news.povray.org...
>> As I understand it, anybody who can guess what your admin password is
>> can have unlimited access to your PC via RPC. Nice...>
> 
> Anyone who's guessed your admin password owns your machine regardless.
> 
> del \\machine\c$\*.* /S

Which, as I understand it, only works because RPC is enabled.

If you can't remotely talk to the machine, you can't do anything to it.

> And that's not even considering them walking up to the machine (if its not a
> server) and loging in. Or if it is a server, usign remote desktop/terminal
> services
> 
> Strong admin password, perhaps?

I was thinking more of the millions of people who have PCs in their 
homes, probably with a blank or default admin password, who have no idea 
that random people on the Internet can use that password to remotely do 
stuff to it.

-- 
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*

Post a reply to this message

Orchid XP v7 wrote:
> If you can't remotely talk to the machine, you can't do anything to it.

RPC includes things like programs talking to the graphics subsystem, 
login process talking to the protected storage process, and console 
programs talking to CRSS (console simulation subsystem).

Kind of like eliminating pipes and sockets on Unix.

-- 
   Darren New / San Diego, CA, USA (PST)
     It's not feature creep if you put it
     at the end and adjust the release date.

Post a reply to this message

Orchid XP v7 nous apporta ses lumieres en ce 2007/12/13 15:47:
> Gail Shaw wrote:
>> "Orchid XP v7" <voi### [at] devnull> wrote in message
>> news:47617e7a$1@news.povray.org...
>>> As I understand it, anybody who can guess what your admin password is
>>> can have unlimited access to your PC via RPC. Nice...>
>>
>> Anyone who's guessed your admin password owns your machine regardless.
>>
>> del \\machine\c$\*.* /S
> 
> Which, as I understand it, only works because RPC is enabled.
> 
> If you can't remotely talk to the machine, you can't do anything to it.
> 
>> And that's not even considering them walking up to the machine (if its 
>> not a
>> server) and loging in. Or if it is a server, usign remote 
>> desktop/terminal
>> services
>>
>> Strong admin password, perhaps?
> 
> I was thinking more of the millions of people who have PCs in their 
> homes, probably with a blank or default admin password, who have no idea 
> that random people on the Internet can use that password to remotely do 
> stuff to it.
> 
Turning RPC off mean that your programms can't accept any input from your 
keyboard and mouse. It mean that your programms can't even create ther main 
window, let alone display anything in them. It mean that you can't access the disks.
It mean that Windows itself can't accept keyboard and mouse inputs and can't 
access the display. It mean that the various parts of Windows can't communicate 
with each other. Most calls for DLL functions use RPC. Any call to code at an 
address that is more than 64K away may be a "remote" call.

A system that ask for a password, with a blank password, is more secure than the 
same system with a weak password. The reason is that when a password is asked, 
you don't just hit "enter", you try to find a password... and that cracker 
applications are not made to try the blank password.

-- 
Alain
-------------------------------------------------
Wiccan: An it harm none, let shit happen.

Post a reply to this message