On Fri, 06 Mar 2009 21:03:57 -0800, Darren New wrote:

> Jim Henderson wrote:
>> On Fri, 06 Mar 2009 16:12:42 -0800, Darren New wrote:
>> 
>>> Jim Henderson wrote:
>>>> Microsoft wrote Explorer.  Microsoft wrote UAC. I leave the rest of
>>>> my comment as an exercise for the reader. ;-)
>>> Sure. I was just pointing out that *you* could do better in *your*
>>> programs.
>> 
>> Just because others don't do a good job is no excuse for MS doing a
>> poor job.
> 
> 
> I quote myself: "Explorer is dumb here."
> 
> In what way is that saying MS didn't do a poor job?

It isn't.  But to say that the implementation of UAC is a good way to 
solve the problem is inaccurate; the team working on UAC and the team 
working on Explorer need to come to terms with the best way for the user.

I'm not saying that this will be *easy* to do - heck, I work in a 
software company that has problems like this itself - for example, 
SLES10SP2 breaks Open Enterprise Server (which runs on SLES10SP1).  It's 
a point of frustration for me when I see groups inside a company not 
talking to each other, and the user base accepting that "well, this is 
the way it is" instead of demanding something be done about it.

IOW, explorer being dumb here isn't the issue.  It's that UAC "fixes" it 
by also being dumb that is.  Two dumbs don't make a smart.

Jim

Post a reply to this message

Jim Henderson wrote:
> It isn't.  But to say that the implementation of UAC is a good way to 
> solve the problem is inaccurate; the team working on UAC and the team 
> working on Explorer need to come to terms with the best way for the user.

Certainly. That would be the "library" I was talking about.

> talking to each other, and the user base accepting that "well, this is 
> the way it is" instead of demanding something be done about it.

I think a lot of the broken security (w.r.t. the UAC) in Win7 is due to the 
user base demanding something be done about it, and Microsoft doing the 
wrong thing instead of fixing their applications the right way.

> IOW, explorer being dumb here isn't the issue.  It's that UAC "fixes" it 
> by also being dumb that is.  Two dumbs don't make a smart.

If you mean the UAC "fixes" in Win7, I'm in complete agreement. I think MS 
has a mess of code that doesn't run quite right with the UAC, just like 
everyone else (in the sense that it took years to get people to write 
software that runs unprivileged), and they're unwilling to spend the time to 
fix that code, so they make half-ass patches that break the security.

I think we're actually agreeing, and I'm just commenting "your code could 
probably be better than MS's if you took the time."

-- 
   Darren New, San Diego CA, USA (PST)
   My fortune cookie said, "You will soon be
   unable to read this, even at arm's length."

Post a reply to this message

On Sat, 07 Mar 2009 13:54:20 -0800, Darren New wrote:

> Jim Henderson wrote:
>> It isn't.  But to say that the implementation of UAC is a good way to
>> solve the problem is inaccurate; the team working on UAC and the team
>> working on Explorer need to come to terms with the best way for the
>> user.
> 
> Certainly. That would be the "library" I was talking about.

I see, yes.

>> talking to each other, and the user base accepting that "well, this is
>> the way it is" instead of demanding something be done about it.
> 
> I think a lot of the broken security (w.r.t. the UAC) in Win7 is due to
> the user base demanding something be done about it, and Microsoft doing
> the wrong thing instead of fixing their applications the right way.

Agreed. :-)

>> IOW, explorer being dumb here isn't the issue.  It's that UAC "fixes"
>> it by also being dumb that is.  Two dumbs don't make a smart.
> 
> If you mean the UAC "fixes" in Win7, I'm in complete agreement. I think
> MS has a mess of code that doesn't run quite right with the UAC, just
> like everyone else (in the sense that it took years to get people to
> write software that runs unprivileged), and they're unwilling to spend
> the time to fix that code, so they make half-ass patches that break the
> security.

In general, I think the idea of UAC as the "fix" is the problem.  
Granted, it's really a first attempt at fixing the problem, but still - 
MS is supposed to be hiring some of the smartest people out there.  I 
know some of them (not who work on the core OS, but on other parts of 
their product offerings) certainly are.

> I think we're actually agreeing, and I'm just commenting "your code
> could probably be better than MS's if you took the time."

We do have a habit of that, don't we? ;-)

Jim

Post a reply to this message

Jim Henderson wrote:
> In general, I think the idea of UAC as the "fix" is the problem.  

I dunno. From what I know of it, it seems like a reasonable fix. I see Mac 
fans saying they never have to type in their password, and Windows users 
saying they have to answer UAC prompts several dozen times a day, and I'm 
thinking "WTF?"  What in the world are Windows users doing that make them 
have to answer UAC prompts dozens of times a day, when I do it maybe once a 
week when making offsite backups?

> Granted, it's really a first attempt at fixing the problem, but still - 
> MS is supposed to be hiring some of the smartest people out there.  I 
> know some of them (not who work on the core OS, but on other parts of 
> their product offerings) certainly are.

Yeah.

Actually, I was thinking that UAC might not be quite enough. Now every 
program that needs to upgrade itself (like firefox, for example, or a virus 
scanner, or whatever) needs a privileged service running just so it can 
write over files without prompting for a password. I can't think how you'd 
fix that, except perhaps letting a program get overwritten by a new version 
that is signed by the same public key.

On the other hand, UNIX and OSX seem to get along OK without it. Linux needs 
the password to do updates, unless it's using a background process of 
course, so that's kind of the same except that almost all your updates come 
from the same place. I don't know what OSX does in this regard.

> We do have a habit of that, don't we? ;-)

Seems like it.  It seems I sometimes make comments that are only 
tangentially related to the stuff I'm quoting, which confuses people sometimes.

-- 
   Darren New, San Diego CA, USA (PST)
   My fortune cookie said, "You will soon be
   unable to read this, even at arm's length."

Post a reply to this message

On Sat, 07 Mar 2009 22:07:47 -0800, Darren New wrote:

> Jim Henderson wrote:
>> In general, I think the idea of UAC as the "fix" is the problem.
> 
> I dunno. From what I know of it, it seems like a reasonable fix. I see
> Mac fans saying they never have to type in their password, and Windows
> users saying they have to answer UAC prompts several dozen times a day,
> and I'm thinking "WTF?"  What in the world are Windows users doing that
> make them have to answer UAC prompts dozens of times a day, when I do it
> maybe once a week when making offsite backups?

Well, see, I never get them. ;-)  Come to think, I didn't get them while 
updating the Vista OS on this box, but that's because I turned them off.

> Actually, I was thinking that UAC might not be quite enough. Now every
> program that needs to upgrade itself (like firefox, for example, or a
> virus scanner, or whatever) needs a privileged service running just so
> it can write over files without prompting for a password. I can't think
> how you'd fix that, except perhaps letting a program get overwritten by
> a new version that is signed by the same public key.

That's kinda like the ActiveX control that gets installed in FF to allow 
software to be installed without the prompts.  Though it sounds like 
there's more to it than that.

> On the other hand, UNIX and OSX seem to get along OK without it. Linux
> needs the password to do updates, unless it's using a background process
> of course, so that's kind of the same except that almost all your
> updates come from the same place. I don't know what OSX does in this
> regard.

There's a reason I also don't do automatic updates on Linux as well.  I 
like to know what's being done to my system.

>> We do have a habit of that, don't we? ;-)
> 
> Seems like it.  It seems I sometimes make comments that are only
> tangentially related to the stuff I'm quoting, which confuses people
> sometimes.

I think we both do that, and that we both do it causes the confusion and 
violent agreements. ;-)

Jim

Post a reply to this message

Jim Henderson wrote:
> Well, see, I never get them. ;-)  Come to think, I didn't get them while 
> updating the Vista OS on this box, but that's because I turned them off.

I did too, while I was installing stuff. Then I turned it back on.

Come to think of it, I have to wonder how much of the griping about "too 
many UAC prompts" comes from vocal reporters whose job it is to do stuff 
like install and remove software and then write about it publicly.

The traditional "read email/surf web/write documents/balance checkbook" 
grandmothers that the Linux folks generally use as "typical home desktop 
user model" are unlikely to ever run across a UAC on Windows, except maybe 
during a software upgrade.

> That's kinda like the ActiveX control that gets installed in FF to allow 
> software to be installed without the prompts.  Though it sounds like 
> there's more to it than that.

I never heard of such an ActiveX, so I can't comment.

> There's a reason I also don't do automatic updates on Linux as well.  I 
> like to know what's being done to my system.

Me too. :-)  But you have to admit, there are groups of people who trust the 
Linux distributors (or Microsoft or Apple) more than they trust their own 
ability to make such decisions.

> I think we both do that, and that we both do it causes the confusion and 
> violent agreements. ;-)

Heh. I'll keep an eye out.

-- 
   Darren New, San Diego CA, USA (PST)
   My fortune cookie said, "You will soon be
   unable to read this, even at arm's length."

Post a reply to this message

On Sun, 08 Mar 2009 13:30:17 -0700, Darren New wrote:

> Jim Henderson wrote:
>> Well, see, I never get them. ;-)  Come to think, I didn't get them
>> while updating the Vista OS on this box, but that's because I turned
>> them off.
> 
> I did too, while I was installing stuff. Then I turned it back on.
> 
> Come to think of it, I have to wonder how much of the griping about "too
> many UAC prompts" comes from vocal reporters whose job it is to do stuff
> like install and remove software and then write about it publicly.

Maybe.  But they probably are supposed to install it like people who 
don't know computers use computers.

> The traditional "read email/surf web/write documents/balance checkbook"
> grandmothers that the Linux folks generally use as "typical home desktop
> user model" are unlikely to ever run across a UAC on Windows, except
> maybe during a software upgrade.

Possibly.

>> That's kinda like the ActiveX control that gets installed in FF to
>> allow software to be installed without the prompts.  Though it sounds
>> like there's more to it than that.
> 
> I never heard of such an ActiveX, so I can't comment.

I thought it was the one that started this thread (come to think, it was 
another) - a silently installed control in FF on Windows that comes with 
the .net framework update.

>> There's a reason I also don't do automatic updates on Linux as well.  I
>> like to know what's being done to my system.
> 
> Me too. :-)  But you have to admit, there are groups of people who trust
> the Linux distributors (or Microsoft or Apple) more than they trust
> their own ability to make such decisions.

Yes, but I think that comes down to poor education on computers in 
general. :-)

>> I think we both do that, and that we both do it causes the confusion
>> and violent agreements. ;-)
> 
> Heh. I'll keep an eye out.

Jim

Post a reply to this message

Jim Henderson wrote:
> Maybe.  But they probably are supposed to install it like people who 
> don't know computers use computers.

I think I did it again. :-) I'm not talking about first setting up the OS. 
I'm more talking about the people who complain they get dozens of UAC 
prompts a day.

>> I never heard of such an ActiveX, so I can't comment.
> 
> I thought it was the one that started this thread (come to think, it was 
> another) - a silently installed control in FF on Windows that comes with 
> the .net framework update.

Oh, I didn't realize that was ActiveX.  I wasn't aware you *could* "silently 
install" a plug-in in firefox. Every time I install one or it gets upgraded, 
Firefox is very in my face about it.

> Yes, but I think that comes down to poor education on computers in 
> general. :-)

Exactly. That's my point. If you're educated enough to know what you're 
doing, you're educated enough to adjust it to behave how you want it to.

-- 
   Darren New, San Diego CA, USA (PST)
   My fortune cookie said, "You will soon be
   unable to read this, even at arm's length."

Post a reply to this message

On Mon, 09 Mar 2009 10:14:49 -0700, Darren New wrote:

> Jim Henderson wrote:
>> Maybe.  But they probably are supposed to install it like people who
>> don't know computers use computers.
> 
> I think I did it again. :-) I'm not talking about first setting up the
> OS. I'm more talking about the people who complain they get dozens of
> UAC prompts a day.

We are talking about the same thing. :-)

>>> I never heard of such an ActiveX, so I can't comment.
>> 
>> I thought it was the one that started this thread (come to think, it
>> was another) - a silently installed control in FF on Windows that comes
>> with the .net framework update.
> 
> Oh, I didn't realize that was ActiveX.  I wasn't aware you *could*
> "silently install" a plug-in in firefox. Every time I install one or it
> gets upgraded, Firefox is very in my face about it.

My understanding from that thread was that it is ActiveX.  And I'm sure 
there's a way to install an extension for FF without invoking FF, that's 
probably what Microsoft has done.

>> Yes, but I think that comes down to poor education on computers in
>> general. :-)
> 
> Exactly. That's my point. If you're educated enough to know what you're
> doing, you're educated enough to adjust it to behave how you want it to.

Except that it should be designed properly to begin with. :-)

Jim

Post a reply to this message

Jim Henderson wrote:
> We are talking about the same thing. :-)

Could be.

> My understanding from that thread was that it is ActiveX.

I wouldn't be surprised. But being ActiveX doesn't preclude it being other 
things as well. Technically, Word is ActiveX as well. :-)

> And I'm sure 
> there's a way to install an extension for FF without invoking FF, that's 
> probably what Microsoft has done.

Yeah. And FF is designed to support that, so I'm at a loss why people thing 
something's done wrong. :-)

> Except that it should be designed properly to begin with. :-)

Hard to do, with conflicting requirements. Very difficult to design one car 
that supports both Indy racing and people who don't know how to drive.

-- 
   Darren New, San Diego CA, USA (PST)
   My fortune cookie said, "You will soon be
   unable to read this, even at arm's length."

Post a reply to this message