Searching \ for '[OT] Firefox security gripe' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=firefox+security
Search entire site for: 'Firefox security gripe'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Firefox security gripe'
2007\07\24@110648 by Mike Hord

picon face
For the first time ever I have a gripe about Firefox.

I'm working on my company Wiki trying to create a
page which contains synopses of our formalized
problem solving procedure documents (so that we
can actually access the solutions in the future), and
Firefox just will not allow the website to pull up the
documents located on our fileserver.

IE does it just fine (via a file://server/dir/filename.doc
type structure) but Firefox freaks out and dumps a
security error into the error log.  Doesn't TELL you
it's doing that, it just does nothing visible and dumps
the error into the error log.

Furthermore, there's no security setting in Firefox
that would let me say "I trust this website, let it open
the file".  Nope, just can't do it.

Anyone have a workaround, besides "Use IE"?

Mike H.

2007\07\24@114143 by Michael Rigby-Jones

picon face


{Quote hidden}

If possible map a drive letter to the server, it works ok then.

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2007\07\24@115019 by Paul Hutchinson

picon face
> -----Original Message-----
> From: piclist-bouncesspamKILLspammit.edu On Behalf Of Mike Hord
> Sent: Tuesday, July 24, 2007 11:07 AM
>
> For the first time ever I have a gripe about Firefox.
>
> I'm working on my company Wiki trying to create a
> page which contains synopses of our formalized
> problem solving procedure documents (so that we
> can actually access the solutions in the future), and
> Firefox just will not allow the website to pull up the
> documents located on our fileserver.
>
> IE does it just fine (via a file://server/dir/filename.doc

Try using three extra slashes like this:
file://///server/dir/filename.doc

The extra slashes override the local file access security feature of
Firefox.

If you map a drive letter for the server share then only one extra slash is
required.
file:///X:/dir/filename.doc

Paul


{Quote hidden}

2007\07\24@134822 by Mike Hord

picon face
> If possible map a drive letter to the server, it works ok then.
>
> Regards
>
> Mike

Not for me it doesn't.

Mike H.

2007\07\24@134902 by Mike Hord

picon face
> Try using three extra slashes like this:
> file://///server/dir/filename.doc
>
> The extra slashes override the local file access security feature of
> Firefox.
>
> If you map a drive letter for the server share then only one extra slash is
> required.
> file:///X:/dir/filename.doc

No, and no.

I'm beginning to wonder if this is specific to my installation.  Did our IT
guys do something here?

Mike H.

2007\07\24@165730 by Paul Hutchinson

picon face
> -----Original Message-----
> From: .....piclist-bouncesKILLspamspam.....mit.edu On Behalf Of Mike Hord
> Sent: Tuesday, July 24, 2007 1:49 PM
>
> > Try using three extra slashes like this:
<snip>
> > required.
> > file:///X:/dir/filename.doc
>
> No, and no.
>
> I'm beginning to wonder if this is specific to my installation.
> Did our IT guys do something here?
>
> Mike H.

Could be a network specific problem but I do have a couple other ideas that
may help.

Try Firefox safe mode and also try with a new Firefox profile. These
attempts will eliminate Firefox bad extension and corrupt configuration
problems.

Try opening a file with a type other than Word document. If other types work
then the file types association in Firefox may be broken. You can fix the
association in the "Options - Content" settings page.

Paul

2007\07\24@213602 by Gerhard Fiedler

picon face
Mike Hord wrote:

>> Try using three extra slashes like this:
>> file://///server/dir/filename.doc
>>
>> The extra slashes override the local file access security feature of
>> Firefox.
>>
>> If you map a drive letter for the server share then only one extra slash is
>> required.
>> file:///X:/dir/filename.doc
>
> No, and no.

FWIW, both methods work here (simple Win2k workgroup shares) and always
have worked.

Gerhard

2007\07\25@021543 by Peter P.

picon face
Mike Hord <mike.hord <at> gmail.com> writes:
> Firefox just will not allow the website to pull up the
> documents located on our fileserver.

Maybe I did not understand the question exactly, but I think that you should
know that pulling local documents from the local filename space (file:///) from
*any* other domain (i.e. ) is a security violation and FF and other
properly built browsers will block it and allow no recourse, and that's the way
it should be.

You can think of namespaces. The namespace http://somedomain (even if localhost)
is distinct from the domain file:/// and it is a really bad idea to allow a page
or a script from http://somedomain load anything from file:///, because file:///
is really sensitive (even if trusted and/or precisely because it is trusted and
precious) . You can either load your documents into the web server's namespace
(so they each have a namespace location), or use a wiki that generates
static pages and then the wiki will be in the namespace of the files since it
will now have *only* file:/// addresses. So basically you can go from file:///
to but not the other way around, because it would be like leaving the
front door open.

It is bad enough as it is now with some local domains being 'untrusted' and thus
loadable by external pages with impunity. This week there was a security update
in FF because of that and I caught at least one webpage trying to access my
local web services through this mechanism. If you think that this is not
serious, answer yourself the question: 'would I like someone to browse my
internal network and access my router's setup pages, my network printer's web
access page, and my local network file storage ?'. I think that the answer to
that is NO.

hope this helps,
Peter P.


2007\07\25@062154 by Gerhard Fiedler

picon face
Mike Hord wrote:

> For the first time ever I have a gripe about Firefox.
>
> I'm working on my company Wiki trying to create a page which contains
> synopses of our formalized problem solving procedure documents (so that
> we can actually access the solutions in the future), and Firefox just
> will not allow the website to pull up the documents located on our
> fileserver.
>
> IE does it just fine (via a file://server/dir/filename.doc type
> structure) but Firefox freaks out and dumps a security error into the
> error log.  Doesn't TELL you it's doing that, it just does nothing
> visible and dumps the error into the error log.
>
> Furthermore, there's no security setting in Firefox that would let me
> say "I trust this website, let it open the file".  Nope, just can't do
> it.
>
> Anyone have a workaround, besides "Use IE"?

Come to think of it, considering what Peter says, the file: protocol is
actually the workaround :)

The real solution would probably be to set up a web server on the file
server and serve the files in question through http:. Then the link would
be http://server/virtualdir/filename.doc. Setting up a web server is easy
enough on both Win2k and Linux.

Gerhard

2007\07\25@062505 by Gerhard Fiedler

picon face
Peter P. wrote:

{Quote hidden}

How would they be able to browse my network? The way the file: protocol
seems to work is that I can "save" or "open" (which is a form of "save")
such links. I don't think they get loaded into the page context.

Gerhard

2007\07\25@070251 by Russell McMahon

face
flavicon
face
>> It is bad enough as it is now with some local domains being
>> 'untrusted'
>> and thus loadable by external pages with impunity. This week there
>> was a
>> security update in FF because of that and I caught at least one
>> webpage
>> trying to access my local web services through this mechanism. If
>> you
>> think that this is not serious, answer yourself the question:
>> 'would I
>> like someone to browse my internal network and access my router's
>> setup
>> pages, my network printer's web access page, and my local network
>> file
>> storage ?'. I think that the answer to that is NO.

I usually decline to do computer support even for friends but I looked
at a PC problem in very special circumstances. It was running Windows
XP home and advised that it was unable to run Windows update until it
had had three URLs added to the trusted internet category so that "it
could access updates in all situations". I'd never seen anything ever
ask for that before. Two of the urls appeared to be valid microsoft
pages but the third used the word "update" in the url but went to an
uncertain location.

I accessed the url specified from another PC and it transferred me to
a valid microsoft update site. Suspicions nevertheless highly aroused
and as the user concerned insists on downloading every helper, addon,
search bar etc they can find (despite large written warnings stuck on
the front of the PC) * I deleted the whole browser and linked kaboodle
and did a clean install.

I've added some reasonably rigorous blocking programs and we'll see
what happens "next time".




           Russell


* Yes, I too would decline to support such an irresponsible user in
almost any circumstance. Alas, this is the exceptional circumstance
that one cannot easily avoid.


2007\07\25@091643 by Mike Hord

picon face
> Mike Hord <mike.hord <at> gmail.com> writes:
> > Firefox just will not allow the website to pull up the
> > documents located on our fileserver.
>
> Maybe I did not understand the question exactly, but I think that you should
> know that pulling local documents from the local filename space (file:///) from
> *any* other domain (i.e. ) is a security violation and FF and other
> properly built browsers will block it and allow no recourse, and that's the way
> it should be.

In general, yes, it should be a security violation.  In this specific
case, I think
there should be an easy way to say "Trust this domain" and allow it to load
from the local filespace- the Wiki website is on our intranet, as are the files,
but the website is accessed via http: and the files are via file:.

Of course, IE has NO problem with this and will happily load files from the
local filespace upon request from the website.

> It is bad enough as it is now with some local domains being 'untrusted' and thus
> loadable by external pages with impunity. This week there was a security update
> in FF because of that and I caught at least one webpage trying to access my
> local web services through this mechanism. If you think that this is not
> serious, answer yourself the question: 'would I like someone to browse my
> internal network and access my router's setup pages, my network printer's web
> access page, and my local network file storage ?'. I think that the answer to
> that is NO.

I think more serious is the fact that allowing a remote website to access the
local filespace could allow a website that you visit to activate trojans or DOS
bots which were piggyback installed with questionable software.

Mike H.

> hope this helps,
> Peter P.

2007\07\25@091951 by Mike Hord

picon face
> The real solution would probably be to set up a web server on the file
> server and serve the files in question through http:. Then the link would
> be http://server/virtualdir/filename.doc. Setting up a web server is easy
> enough on both Win2k and Linux.

If this were on my network at home I'd agree. However, the scope of this
project does not warrant that outlay of resources, since I'd have to pull
in at least a couple of other people to make that happen.

I think what's going to end up happening is the modification of this
project somewhat- be it the suggestion that this page be loaded in IE
instead of Firefox or a note to copy and paste the URL.

Mike H.

2007\07\25@115621 by Peter P.

picon face
> I think what's going to end up happening is the modification of this
> project somewhat- be it the suggestion that this page be loaded in IE
> instead of Firefox or a note to copy and paste the URL.

I believe that making static pages with the wiki is the easiest way to get
things done without running a web server. Simply look at the wiki pages to see
how to get rid of the full urls and make sure none says . Once inside a
realm (or namespace), you need only the leaf name in any href="" argument. e.g.
file:///some/where/really/complicated.doc should become
href="/some/where/really/complicated.doc"

Once all the links are modified (and don't forget to remove the base url="" from
pages, if any) they are all in the same realm and the browser should stop
complaining.

Publishing static wiki pages is very easy once you get the scripts right. Please
take a look here for a Wiki smorgasbrod. You probably want to choose one that
does not use a database or php. That makes static pages really easy:

 http://www.wikimatrix.org/

Maybe this will give you some ideas on how it's done without a server:

 http://en.wikipedia.org/wiki/TiddlyWiki

hope this helps,
Peter P.


2007\07\25@191050 by Gerhard Fiedler

picon face
Mike Hord wrote:

>> The real solution would probably be to set up a web server on the file
>> server and serve the files in question through http:. Then the link would
>> be http://server/virtualdir/filename.doc. Setting up a web server is easy
>> enough on both Win2k and Linux.
>
> If this were on my network at home I'd agree. However, the scope of this
> project does not warrant that outlay of resources, since I'd have to pull
> in at least a couple of other people to make that happen.

Another (maybe) option: Since you have a company wiki, your company
probably already has a intranet web server. The files you want to link to
are possibly on shares to which the web server has (read-only) access, or
at least can reasonably easily made to have. The only thing missing then is
a virtual directory on the intranet web server that points to the file
share. With that, you can use http: to link to the files.

Gerhard

2007\07\25@192429 by Xiaofan Chen

face picon face
On 7/25/07, Gerhard Fiedler <EraseMElistsspam_OUTspamTakeThisOuTconnectionbrazil.com> wrote:
> >
> > Anyone have a workaround, besides "Use IE"?
>
> Come to think of it, considering what Peter says, the file: protocol is
> actually the workaround :)
>
> The real solution would probably be to set up a web server on the file
> server and serve the files in question through http:. Then the link would
> be http://server/virtualdir/filename.doc. Setting up a web server is easy
> enough on both Win2k and Linux.
>

I hate this thingy of serving files through http expecially under Windows
with long file names. Very ugly.

The real solution seems to use IE;-)

<rant> Personally I really do not like Firefox. It is the most often crashed
program under my Ubuntu Linux installation (the most used as well
unfortunately). The KDE browser is not good enough unfortunately. </rant>

2007\07\26@034443 by Nate Duehr

face
flavicon
face

On Jul 25, 2007, at 5:24 PM, Xiaofan Chen wrote:

{Quote hidden}

My rant notwithstanding, they make a nice "open this page in IE" add-
on for Firefox.  It works wonders for such problems, and you can give  
people a URL to install it themselves, go home and have a beer far  
faster that way.

--
Nate Duehr
@spam@nateKILLspamspamnatetech.com



2007\07\26@083339 by Gerhard Fiedler

picon face
Xiaofan Chen wrote:

> I hate this thingy of serving files through http expecially under Windows
> with long file names. Very ugly.
>
> The real solution seems to use IE;-)

What's different between serving files that have long names with embedded
spaces from your file server using the http: and the file: protocol? They
all end up as long names with ugly characters embedded anyway, no?

Gerhard

2007\07\26@084724 by Peter Bindels

picon face
On 26/07/07, Xiaofan Chen <KILLspamxiaofancKILLspamspamgmail.com> wrote:
> I hate this thingy of serving files through http expecially under Windows
> with long file names. Very ugly.

file:// is a hack. It should never have worked and can only work
properly on the Windows systems that propagated the hack. Same goes
for IE - it's quite a hack to support it in the first place, making it
incompatible from the rest of the world (who don't have a file://E:/
or such) and completely defeating the purpose of serving through HTTP
- not requiring anything else, standards etc.

> The real solution seems to use IE;-)

IE isn't the solution. IE is the problem.

> <rant> Personally I really do not like Firefox. It is the most often crashed
> program under my Ubuntu Linux installation (the most used as well
> unfortunately). The KDE browser is not good enough unfortunately. </rant>

I like Firefox a lot more than IE since it's more stable, more
responsive, more standard-conformant (although the webpages usually
aren't), crossplatform and more usable in general. Did you try to use
IE with a file://C:/ link on your Linux machine? Wanna bet it doesn't
work - and that the HTTP solution does?

Regards,
Peter

2007\07\26@090613 by Xiaofan Chen

face picon face
On 7/26/07, Peter Bindels <RemoveMEdascandyTakeThisOuTspamgmail.com> wrote:
> On 26/07/07, Xiaofan Chen <spamBeGonexiaofancspamBeGonespamgmail.com> wrote:
> > I hate this thingy of serving files through http expecially under Windows
> > with long file names. Very ugly.

I will still say this.

> file:// is a hack. It should never have worked and can only work
> properly on the Windows systems that propagated the hack. Same goes
> for IE - it's quite a hack to support it in the first place, making it
> incompatible from the rest of the world (who don't have a file://E:/
> or such) and completely defeating the purpose of serving through HTTP
> - not requiring anything else, standards etc.
>
> > The real solution seems to use IE;-)
Sorry I seem to make a mistake here.

> IE isn't the solution. IE is the problem.
I see. Now I kind of agree with you.

> > <rant> Personally I really do not like Firefox. It is the most often crashed
> > program under my Ubuntu Linux installation (the most used as well
> > unfortunately). The KDE browser is not good enough unfortunately. </rant>

I will still stick to my opinion unless the next version of Firefox gets better.
Firefox 2.x (quite bad) and 1.5.x are both not very good under Ubuntu Linux.

> I like Firefox a lot more than IE since it's more stable, more
> responsive, more standard-conformant (although the webpages usually
> aren't), crossplatform and more usable in general. Did you try to use
> IE with a file://C:/ link on your Linux machine? Wanna bet it doesn't
> work - and that the HTTP solution does?
>

I do not use IE on my Linux machine. There is no native IE for linux.
If yes I will try it. I have used Linux at home for quite a while and
I think Firefox is getting worse, kind of bloated now.

To say Firefox is more responsive under Windows is not true based
on my experience. It stars slower. It load web pages not faster than IE.


Regards,
Xiaofan

2007\07\26@090902 by Xiaofan Chen

face picon face
On 7/26/07, Gerhard Fiedler <TakeThisOuTlistsEraseMEspamspam_OUTconnectionbrazil.com> wrote:
> Xiaofan Chen wrote:
>
> > I hate this thingy of serving files through http expecially under Windows
> > with long file names. Very ugly.

Still true.

> > The real solution seems to use IE;-)
>
> What's different between serving files that have long names with embedded
> spaces from your file server using the http: and the file: protocol? They
> all end up as long names with ugly characters embedded anyway, no?

Yeah I see now. I made a mistake here.

Anyway, I'd like to use explorer to browse to the shared folder there and
not through a browser. The whole file:// idea is not good.

Regards,
Xiaofan

2007\07\26@173249 by Dario Greggio

face picon face
Peter Bindels wrote:
>>The real solution seems to use IE;-)
>
>
> IE isn't the solution. IE is the problem.

Terminator 3: skynet ;-)

2007\07\26@174752 by Peter P.

picon face

> > I like Firefox a lot more than IE since it's more stable, more
> > responsive, more standard-conformant (although the webpages usually

For speed please try out Opera. http://www.opera.com *really* fast. It is also
available for embedded devices (phones etc) and for *nix, mac, and windows, and
it can even render wml (a bonus for people who write pages for phone surfing).

Peter P.

2007\07\26@194234 by Xiaofan Chen

face picon face
On 7/26/07, Peter P. <RemoveMEplpeter2006spamTakeThisOuTyahoo.com> wrote:
>
> > > I like Firefox a lot more than IE since it's more stable, more
> > > responsive, more standard-conformant (although the webpages usually
>
> For speed please try out Opera. http://www.opera.com *really* fast. It is also
> available for embedded devices (phones etc) and for *nix, mac, and windows, and
> it can even render wml (a bonus for people who write pages for phone surfing).
>

For most of the people Opera is not an option. It does not properly
render many web pages. IE/Firefox are better in this aspect. This
might not be the fault for Opera but it is the reality.

And I still do not agree it is faster than IE under Windows. IE has
an edge under Windows in terms of speed and responsiveness.

2007\07\26@225820 by Jake Anderson

flavicon
face
Xiaofan Chen wrote:
> On 7/26/07, Peter P. <plpeter2006EraseMEspam.....yahoo.com> wrote:
>  
>>>> I like Firefox a lot more than IE since it's more stable, more
>>>> responsive, more standard-conformant (although the webpages usually
>>>>        
>> For speed please try out Opera. http://www.opera.com *really* fast. It is also
>> available for embedded devices (phones etc) and for *nix, mac, and windows, and
>> it can even render wml (a bonus for people who write pages for phone surfing).
>>
>>    
>
> For most of the people Opera is not an option. It does not properly
> render many web pages. IE/Firefox are better in this aspect. This
> might not be the fault for Opera but it is the reality.
>
> And I still do not agree it is faster than IE under Windows. IE has
> an edge under Windows in terms of speed and responsiveness.
>  
tried fasterfox addon?

2007\07\27@032047 by Peter Bindels

picon face
On 27/07/07, Xiaofan Chen <EraseMExiaofancspamgmail.com> wrote:
> And I still do not agree it is faster than IE under Windows. IE has
> an edge under Windows in terms of speed and responsiveness.

It has so much of an "edge" that my music freezes for a split second
when I use it. It appears to completely hog the CPU that split second
for all of its things that basic things like any other program don't
get the chance to fill their buffers at all. Given the fact that
Firefox doesn't do that, I think it's not quite a fair edge, nor one
that I would want. Not to mention that if you have used a file dialog
during your Windows session (at all!), most of it is preloaded already
(in particular the caching is fully preloaded). They also most likely
won't be unloaded since just about any program uses the "common
dialogs" that now drag in IE.

Regards,
Peter

2007\07\27@065552 by Xiaofan Chen

face picon face
On 7/27/07, Peter Bindels <RemoveMEdascandyEraseMEspamEraseMEgmail.com> wrote:
> On 27/07/07, Xiaofan Chen <RemoveMExiaofancspam_OUTspamKILLspamgmail.com> wrote:
> > And I still do not agree it is faster than IE under Windows. IE has
> > an edge under Windows in terms of speed and responsiveness.
>
> It has so much of an "edge" that my music freezes for a split second
> when I use it.

I am not talking about all the bad things about IE. I am just tired of
people alway claiming that Firefox is faster/stable and that is not
true from my personal experiences. Under Linux I still use firefox...

Xiaofan

2007\07\27@070216 by Xiaofan Chen

face picon face
On 7/26/07, Jake Anderson <RemoveMEjakeTakeThisOuTspamspamvapourforge.com> wrote:
> > And I still do not agree it is faster than IE under Windows. IE has
> > an edge under Windows in terms of speed and responsiveness.
> >
> tried fasterfox addon?

Not yet. Actually the speed is not a probelm for me. I was just
pointing out my observation since so many people always
relates Firefox to stability and speed and I do not.

And the biggest problem for me is the stability problem under
Ubuntu Linux. Speed is not a problem.

Xiaofan

2007\07\27@121152 by Neil Cherry

picon face
Xiaofan Chen wrote:
> On 7/26/07, Jake Anderson <EraseMEjakespamspamspamBeGonevapourforge.com> wrote:
>>> And I still do not agree it is faster than IE under Windows. IE has
>>> an edge under Windows in terms of speed and responsiveness.
>>>
>> tried fasterfox addon?
>
> Not yet. Actually the speed is not a probelm for me. I was just
> pointing out my observation since so many people always
> relates Firefox to stability and speed and I do not.
>
> And the biggest problem for me is the stability problem under
> Ubuntu Linux. Speed is not a problem.
>
> Xiaofan

While I'm not arguing with your experience I find it odd. Firefox
1.5 & 2.0 have been very stable for me on my Linux boxes. I have
a Ubuntu laptop that ran 1.5 but now runs 2.0 (I prefer 1.5 for
a lot of little reasons). On my Fedora box it's still 1.5. I tend
to crash other applications on another OS (it ain't VMS :-).

Which version of Firefox on Ubuntu and what add-on (if any) are
you running? Are you hitting a lot of Java or Javascript pages?
I normally have about 15 or more tabs open at any one time on
my Fedora system. On my Ubuntu system I keep it to down to 8 or
9. Both systems get hit pretty heavy if I use Google maps or
some other site the uses Google maps (I create a lot of bike
routes). Also I tend to hit Accuweather.com to check out the
local radar. That uses Java and I've found that I need to
stop the animated radar otherwise it just keeps running even
after I close the tab. Otherwise I tend to leave the Fedora/
Firefox running for weeks at a time.

--
Linux Home Automation         Neil Cherry       RemoveMEncherryKILLspamspamlinuxha.com
http://www.linuxha.com/                         Main site
http://linuxha.blogspot.com/                    My HA Blog
Author of:            Linux Smart Homes For Dummies

2007\07\27@130326 by Xiaofan Chen

face picon face
On 7/27/07, Neil Cherry <ncherrySTOPspamspamspam_OUTcomcast.net> wrote:
> While I'm not arguing with your experience I find it odd. Firefox
> 1.5 & 2.0 have been very stable for me on my Linux boxes. I have
> a Ubuntu laptop that ran 1.5 but now runs 2.0 (I prefer 1.5 for
> a lot of little reasons). On my Fedora box it's still 1.5. I tend
> to crash other applications on another OS (it ain't VMS :-).

I have both Ubuntu 7.04 and 6.06LTS installations with Firefox
2.0 and 1.5 respectively. 1.5 is better. I've tried the bundled
version and the standalone version. It is not that bad but it
is the single-most often crashed program other than
my alpha/beta testing of some Linux software packages.

And IE crahes under Windows. No doubt about it. For me
the chance of crash is the same for IE under Windows
and Firefox under Linux. So I think they are the same
in term of stability for me. I should say I take good care
of my Windows installation and I might be a better Windows
user than a Linux user even though I prefer to use
Linux at home since I use Windows at work. Still
I will say I am quite good at using Linux (no
programming) and I can find solutions to
my Linux problem through Google most of the time.

> Which version of Firefox on Ubuntu and what add-on
> (if any) are you running?

Google search plug-in and flash plug-in.

> Are you hitting a lot of Java or Javascript pages?
No Java. Javascript, yes.

Some of the crashes are related to Chinese pages. Some
of the crashes might be related Ubuntu version of Firefox.
Some of the crashes might be related to Firefox 2.0. I do
not know.

Do not get me wrong, I like Ubuntu. It is my preferred Linux
distribution. And Linux is the prefer OS I will use at home.

Regards,
Xiaofan

2007\07\27@130832 by Herbert Graf

flavicon
face
On Fri, 2007-07-27 at 06:55 -0400, Xiaofan Chen wrote:
> On 7/27/07, Peter Bindels <spamBeGonedascandySTOPspamspamEraseMEgmail.com> wrote:
> > On 27/07/07, Xiaofan Chen <KILLspamxiaofancspamBeGonespamgmail.com> wrote:
> > > And I still do not agree it is faster than IE under Windows. IE has
> > > an edge under Windows in terms of speed and responsiveness.
> >
> > It has so much of an "edge" that my music freezes for a split second
> > when I use it.
>
> I am not talking about all the bad things about IE. I am just tired of
> people alway claiming that Firefox is faster/stable and that is not
> true from my personal experiences.

Faster? I'd say 50/50. There are pages that render much faster under
firefox then IE, yet there are other situations where IE is much faster.
On the whole, IE and firefox are pretty close in speed, on average.

As for stability, without question Firefox wins. While firefox certainly
crashes, and possibly more often the IE, when it DOES crash it has NEVER
taken the system with it. And when it's restarted, it very helpfully
allows you to automatically reopen every page you had open at the time
of the crash as if the crash never happened.

IE OTOH may crash less, but in almost every case a crashing IE takes a
good chunk of the system with it. Often Explorer goes with IE, meaning
all the task bar stuff disappears when Explorer restarts. That's on a
good day, I've seen cases where IE crashing cases Explorer to crash and
NOT respawn, leaving you with a windows system that is basically useless
unless you reboot.


> Under Linux I still use firefox...

I agree, for Linux firefox is by far the best choice. Oddly firefox
under Linux is far more stable then firefox under windows, to the point
where I don't even recall the last time firefox has crashed under linux
for me.

TTYL

2007\07\27@134656 by Neil Cherry

picon face
Xiaofan Chen wrote:
{Quote hidden}

I suspect this from your postings, you seem to be a
DIY type of person.

>> Which version of Firefox on Ubuntu and what add-on
>> (if any) are you running?
>
> Google search plug-in and flash plug-in.

I'm not using the Google search plug-in. I usually
just type in 'g what ever I want' and use that as
my search. I sometimes use simple Google 'regex'
to limit my searches. I don't know if the search
plugin helps or hinders the problems. Have you
tried running without it?

>> Are you hitting a lot of Java or Javascript pages?
> No Java. Javascript, yes.
>
> Some of the crashes are related to Chinese pages. Some
> of the crashes might be related Ubuntu version of Firefox.
> Some of the crashes might be related to Firefox 2.0. I do
> not know.

I think that the use of the Chinese character set is the
key difference in our two setups. I don't visit many Chinese
sites. I don't know why it makes a difference but I get it
has something to do with the handling of (mine (ASCII, utf-8,
ISO8859-1?) is probably single byte character while I'll bet
the Chinese characters are 2 byte (could be more).

> Do not get me wrong, I like Ubuntu. It is my preferred Linux
> distribution. And Linux is the prefer OS I will use at home.

I work with lots of OSs. I've never really grown attached to
Windows but I'm a long time Unix user so I think in terms of
Unix. This makes dealing with Windows difficult at times.
My preferred OS is obvious. :-) My preferred distribution is
a toss between Ubuntu for the Desktop (on my laptop) and
Fedora for my server/dev machine. Yes I still have Windows
as I need that for any school work I need to do.

--
Linux Home Automation         Neil Cherry       @spam@ncherry@spam@spamspam_OUTlinuxha.com
http://www.linuxha.com/                         Main site
http://linuxha.blogspot.com/                    My HA Blog
Author of:            Linux Smart Homes For Dummies

2007\07\27@153305 by James Newtons Massmind

face picon face
> > tried fasterfox addon?

Fasterfox is a royal pain in the butt for webmasters. It automatically
follows all the links on a web page, requesting each of them no matter if
the user ever intended to click them or not. It does so at high speed,
burning up bandwidth that will probably never be viewed. (Note to the
clueless: Webmasters PAY for bandwidth)

It is also a robot, so it makes the web server think it is being indexed
either by a search engine or by a spammer trying to harvest email addresses
because it follows links that would normally not be visible to a user.

Let me give you an example of why this is a bad thing. Lets say you have an
account on Yahoo or whatever that greets you with a "Hello username: If you
are not username, click here to log out" Fasterfox has to try to figure out
that it should never follow that link, because to do so would automatically
log you out without even letting you know that you have done so. If
Fasterfox can't figure that out, then the web site has to try to identify
that the request is coming from fasterfox and ignore it. How either of those
things are suppose to work reliably is beyond me.

Website: "Click here to delete all files" Fasterfox: Pant, pant, duh, I'd
better prefetch that page. Website:"Really delete all pictures?" fasterfox:
Duh, I'd better get that one too, master might want it later... Website:"Ok,
deleting all file" user: "Hello? Tech support? YOUR WEB SITE JUST LOST ALL
MY FILES!!!!"

I strongly recommend against the use of that program. It is a bomb, waiting
to go off.

---
James.


2007\07\27@164243 by Xiaofan Chen

face picon face
On 7/27/07, Herbert Graf <spamBeGonemailinglist3spamKILLspamfarcite.net> wrote:
{Quote hidden}

Ah thanks. This is exactly my experience!

Xiaofan

2007\07\27@164549 by Xiaofan Chen

face picon face
On 7/27/07, Xiaofan Chen <.....xiaofancspam_OUTspamgmail.com> wrote:
{Quote hidden}

Oops, hit the send button too fast.

Thanks a lot for the fair observation and I agree with you that when IE crashes
things is not as good. What I mentioned "stability" was more based on the
chance of crashes. I think they are on par.

Xiaofan

2007\07\28@034627 by Jake Anderson

flavicon
face
James Newtons Massmind wrote:
>>> tried fasterfox addon?
>>>      
>
> Fasterfox is a royal pain in the butt for webmasters. It automatically
> follows all the links on a web page, requesting each of them no matter if
> the user ever intended to click them or not. It does so at high speed,
> burning up bandwidth that will probably never be viewed. (Note to the
> clueless: Webmasters PAY for bandwidth)
That mode is an optional extra by default its turned off. That sort of
thing has been tried a few times before and has never really caught on
for the reasons you give.
The default mode for it basically just changes a few settings within
firefox to turn things like http-pipelining on, They are off by default
because some webservers don't support it (i've never come across one),
If you are on a high latency connection to a server who likes putting
their buttons as pictures it can make the difference between a 10 second
page load time and a 40 second page load time.

You can do it all through about:config but this way is easier ;-> If
somebody knows of something that does the same tweaks without all that
other crap i'd like to hear it.


'[OT] Firefox security gripe'
2007\08\01@081302 by Peter Bindels
picon face
On 27/07/07, Herbert Graf <TakeThisOuTmailinglist3KILLspamspamspamfarcite.net> wrote:
> IE OTOH may crash less, but in almost every case a crashing IE takes a
> good chunk of the system with it. Often Explorer goes with IE, meaning
> all the task bar stuff disappears when Explorer restarts. That's on a
> good day, I've seen cases where IE crashing cases Explorer to crash and
> NOT respawn, leaving you with a windows system that is basically useless
> unless you reboot.

Ctrl-shift-esc, file->run, explorer.exe. Happened so often I got
annoyed and figured out how to fix it without rebooting.

2007\08\01@102319 by Herbert Graf

flavicon
face
On Wed, 2007-08-01 at 14:13 +0200, Peter Bindels wrote:
> On 27/07/07, Herbert Graf <.....mailinglist3spamRemoveMEfarcite.net> wrote:
> > IE OTOH may crash less, but in almost every case a crashing IE takes a
> > good chunk of the system with it. Often Explorer goes with IE, meaning
> > all the task bar stuff disappears when Explorer restarts. That's on a
> > good day, I've seen cases where IE crashing cases Explorer to crash and
> > NOT respawn, leaving you with a windows system that is basically useless
> > unless you reboot.
>
> Ctrl-shift-esc, file->run, explorer.exe. Happened so often I got
> annoyed and figured out how to fix it without rebooting.

Yes, I have used that, but invariably in cases where explorer doesn't
restart automatically, the system is so hosed that manually trying to
restart explorer this way also doesn't work.

Fortunately, I made the switch to Linux a few years ago, and I haven't
looked back. TTYL

2007\08\01@115141 by Xiaofan Chen

face picon face
On 8/1/07, Herbert Graf <RemoveMEmailinglist3spamspamBeGonefarcite.net> wrote:
> On Wed, 2007-08-01 at 14:13 +0200, Peter Bindels wrote:
> > Ctrl-shift-esc, file->run, explorer.exe. Happened so often I got
> > annoyed and figured out how to fix it without rebooting.
>
> Yes, I have used that, but invariably in cases where explorer doesn't
> restart automatically, the system is so hosed that manually trying to
> restart explorer this way also doesn't work.

Restart is kind of universal cure for many Windows problem.

> Fortunately, I made the switch to Linux a few years ago, and I haven't
> looked back. TTYL
>
That is not quite possible for work. At home I use Linux most of the
time now. I will go back to be with my Linux computer soon. ;-)
At work we still use Windows and I am in US now for a business
trip.

Xiaofan

2007\08\01@133010 by Herbert Graf

flavicon
face
On Wed, 2007-08-01 at 11:51 -0400, Xiaofan Chen wrote:
> On 8/1/07, Herbert Graf <spamBeGonemailinglist3@spam@spamspam_OUTfarcite.net> wrote:
> > On Wed, 2007-08-01 at 14:13 +0200, Peter Bindels wrote:
> > > Ctrl-shift-esc, file->run, explorer.exe. Happened so often I got
> > > annoyed and figured out how to fix it without rebooting.
> >
> > Yes, I have used that, but invariably in cases where explorer doesn't
> > restart automatically, the system is so hosed that manually trying to
> > restart explorer this way also doesn't work.
>
> Restart is kind of universal cure for many Windows problem.
>
> > Fortunately, I made the switch to Linux a few years ago, and I haven't
> > looked back. TTYL
> >
> That is not quite possible for work. At home I use Linux most of the

Are you sure? I'm running Linux at work as well. Many of the tools I use
are run on remote servers. For things that must be run on Windows I use
Vmware. TTYL

2007\08\01@144832 by Xiaofan Chen

face picon face
On 8/1/07, Herbert Graf <TakeThisOuTmailinglist3spamspamfarcite.net> wrote:
> On Wed, 2007-08-01 at 11:51 -0400, Xiaofan Chen wrote:
> > That is not quite possible for work. At home I use Linux most of the
>
> Are you sure? I'm running Linux at work as well. Many of the tools I use
> are run on remote servers. For things that must be run on Windows I use
> Vmware. TTYL

I am pretty sure about this. How many of us use Linux at work?
Actually in most of the companies the IT standardize the client
software at Windows. On the server side, that is another world,
Windows/Sun/IBM Mainframe/Linux/etc all have its places

Most of the client or client tools we use at work run on Windows. The
customers use Windows as well. This is the reality no matter what
people say.

Regards,
Xiaofan

2007\08\01@164537 by Herbert Graf

flavicon
face
On Wed, 2007-08-01 at 14:48 -0400, Xiaofan Chen wrote:
> On 8/1/07, Herbert Graf <mailinglist3EraseMEspamfarcite.net> wrote:
> > On Wed, 2007-08-01 at 11:51 -0400, Xiaofan Chen wrote:
> > > That is not quite possible for work. At home I use Linux most of the
> >
> > Are you sure? I'm running Linux at work as well. Many of the tools I use
> > are run on remote servers. For things that must be run on Windows I use
> > Vmware. TTYL
>
> I am pretty sure about this. How many of us use Linux at work?
> Actually in most of the companies the IT standardize the client
> software at Windows.

True, which is why I use Vmware with a Windows image. Whenever I REALLY
need to do something in windows I start up the image, do my thing and
shut it down again.

Obviously you have to guage whether the effort is enough for you. I
switched my home machines to Linux a while ago, and just like sitting in
first class on a plane, once you experience something better, you can't
imagine going back to economy. It took a while but eventually I switch
my work machine to Linux, and I haven't looked back.

> On the server side, that is another world,
> Windows/Sun/IBM Mainframe/Linux/etc all have its places
>
> Most of the client or client tools we use at work run on Windows. The
> customers use Windows as well. This is the reality no matter what
> people say.

Again, it's up to how much of a pain it is for you. Many of the windows
apps I have run directly under Linux using Wine. For those that don't I
use my Vmware image. Thanks, TTYL

2007\08\01@171445 by Xiaofan Chen

face picon face
On 8/1/07, Herbert Graf <RemoveMEmailinglist3EraseMEspamspam_OUTfarcite.net> wrote:
> On Wed, 2007-08-01 at 14:48 -0400, Xiaofan Chen wrote:
> > Most of the client or client tools we use at work run on Windows. The
> > customers use Windows as well. This is the reality no matter what
> > people say.
>
> Again, it's up to how much of a pain it is for you. Many of the windows
> apps I have run directly under Linux using Wine. For those that don't I
> use my Vmware image. Thanks, TTYL
>

Seriously, your IT department allows you to do that?

Somehow I have a bad feeling using Wine even though I start
my first GNUPIC post with a howto on MPLAB with Wine. It
extends to Vmware even though it might be much better than Wine.
I prefer to use dual-boot. I prefer to use real Linux or real Windows.

This is just my opinion though.

And I know some purist Linux user will not want to use
Vmware or Wine, especially for those who hates Windows
for religious reasons. I do not agree with them on the
attitude with Windows. But somehow I kind of agree with
their attitude toward Wine/Vmware.

Xiaofan

2007\08\01@175053 by Dave King

flavicon
face
Any idea if Wine or one of the commercial versions could handle something
like solidworks or acad? (maybe the odd first person shooter ;-])

Dave

{Original Message removed}

2007\08\01@182919 by Shawn Tan

flavicon
face
On Wednesday 01 August 2007 22:51:00 Dave King wrote:
> Any idea if Wine or one of the commercial versions could handle something
> like solidworks or acad? (maybe the odd first person shooter ;-])

But you don't need Wine for FPS. [-:
http://zerowing.idsoftware.com/linux/

Solidworks/AutoCAD
http://appdb.winehq.org/appview.php?iAppId=318
http://appdb.winehq.org/appview.php?iAppId=86

--
with metta,
Shawn Tan

Aeste Works (M) Sdn Bhd - Engineering Elegance
http://www.aeste.net

2007\08\01@183443 by Shawn Tan

flavicon
face
On Wednesday 01 August 2007 22:14:44 Xiaofan Chen wrote:
> Somehow I have a bad feeling using Wine even though I start
> my first GNUPIC post with a howto on MPLAB with Wine. It
> extends to Vmware even though it might be much better than Wine.
> I prefer to use dual-boot. I prefer to use real Linux or real Windows.

Hmm. I would think that virtualisation is a better way of running Windows apps
than WINE. At least, it's running real Windows code instead of a Win32
compatibility layer. I use virtualisation from time to time for software
work.

One good thing with virtualisation is having virtual "hardware" to play with.
Althought I've never tried it, I could imagine it being useful for debugging
purposes.

--
with metta,
Shawn Tan

Aeste Works (M) Sdn Bhd - Engineering Elegance
http://www.aeste.net

2007\08\02@080021 by Gerhard Fiedler

picon face
Shawn Tan wrote:

> One good thing with virtualisation is having virtual "hardware" to play with.
> Althought I've never tried it, I could imagine it being useful for debugging
> purposes.

It is. It doesn't hose your working system if something goes wrong. And it
allows one to have a setup with two servers and a client right in your
notebook at the client's site.

I also don't share Xiaofan's reservation about virtual machines. I think
they are great. (You need the memory, though -- that's the part they don't
virtualize :)

Gerhard

2007\08\02@081330 by Peter Bindels

picon face
On 02/08/07, Gerhard Fiedler <@spam@listsRemoveMEspamEraseMEconnectionbrazil.com> wrote:
> It is. It doesn't hose your working system if something goes wrong. And it
> allows one to have a setup with two servers and a client right in your
> notebook at the client's site.
>
> I also don't share Xiaofan's reservation about virtual machines. I think
> they are great. (You need the memory, though -- that's the part they don't
> virtualize :)

There are virtual machines that let you debug an instruction at a time
with full virtualization - pretty much like looking in the processor
with a microscope and slowing the clock to about a tenth of a Hertz
without hardware failing on you - and with reliable timers every few
thousand or million cycles. I love it, they let me log (not print out
- usually a few hundred thousand lines) a trace of the entire program
to see what it exactly does.

For normal uses, it's the best you can get - virtual hardware doesn't
suffer the problems real hardware has. It's never shorted out, it
doesn't fail due to solder masks, it's never out of stock, it always
responds perfectly and within time bounds, unless you tell it not
to... It's great for testing.

2007\08\02@095147 by Xiaofan Chen

face picon face
On 8/2/07, Gerhard Fiedler <EraseMElistsspam@spam@connectionbrazil.com> wrote:
> Shawn Tan wrote:
>
> > One good thing with virtualisation is having virtual "hardware" to play with.
> > Althought I've never tried it, I could imagine it being useful for debugging
> > purposes.
>
> It is. It doesn't hose your working system if something goes wrong. And it
> allows one to have a setup with two servers and a client right in your
> notebook at the client's site.
>
> I also don't share Xiaofan's reservation about virtual machines. I think
> they are great. (You need the memory, though -- that's the part they don't
> virtualize :)

Yes it is good for testing purpose.

However the situation here is different. In this case, they are use it
for real work. They like to use Mac OS X or Linux for its so-called
"superiority against Windows". Then in reality they have to use
MPLAB with ICD2 to debug theire PIC code. So they use VMware
and Windows. To me it is kind of irony. This is actually a good
solution in the aspect of technology. If they do not really claim
thjey can get rid of Windows and use only Linux at work, this will not
be a problem. Windows under VMware is still Windows.

On the other hand, I'd really like to see Linux getting better
and better and can be really used at work. I'd like to have
choices. That is why I spent quite some of my spare time
helping to test piklab and pk2. In this aspect, I think Wine
or Vmware actually hinders the development of alternative
applications. The best is of course for the vendor to
support Linux directly (MPLAB/ICD2/etc). In reality, this is
not possible in the forseeable future.

I am really sorry and I know this is a bit controversial. I need
some good rest...

Regards,
Xiaofan

2007\08\02@095345 by Xiaofan Chen

face picon face
On 8/2/07, Peter Bindels <@spam@dascandyspam_OUTspam.....gmail.com> wrote:
> For normal uses, it's the best you can get - virtual hardware doesn't
> suffer the problems real hardware has. It's never shorted out, it
> doesn't fail due to solder masks, it's never out of stock, it always
> responds perfectly and within time bounds, unless you tell it not
> to... It's great for testing.

It is great for testing. For real work, I am not so sure. It is slower,
it does not support all the hardwares. For example, you can get
ICD2 to work but not RealICE or ICE2000. For real work, stick
with the original OS...

Xiaofan

2007\08\02@102632 by Shawn Tan

flavicon
face
On Thursday 02 August 2007 14:51:23 Xiaofan Chen wrote:
> However the situation here is different. In this case, they are use it
> for real work. They like to use Mac OS X or Linux for its so-called
> "superiority against Windows". Then in reality they have to use
> MPLAB with ICD2 to debug theire PIC code. So they use VMware
> and Windows. To me it is kind of irony. This is actually a good
> solution in the aspect of technology. If they do not really claim
> thjey can get rid of Windows and use only Linux at work, this will not
> be a problem. Windows under VMware is still Windows.

Ah, you are pointing out the "religious" issues. Not technical issues. I'm
actually one of those people who try my best to use Open Source software
wherever possible.

> On the other hand, I'd really like to see Linux getting better
> and better and can be really used at work. I'd like to have
> choices. That is why I spent quite some of my spare time
> helping to test piklab and pk2. In this aspect, I think Wine
> or Vmware actually hinders the development of alternative
> applications. The best is of course for the vendor to
> support Linux directly (MPLAB/ICD2/etc). In reality, this is
> not possible in the forseeable future.

That's true. If Windows was to disappear tomorrow, you can be sure that the
hardware people will sit down, and take some time to build all the necessary
plugins/support software. It wouldn't take us very long to do it, but we just
don't have the reason to.

Actually, this brings up another consideration. We hardware people need good
software tools. However, we're not often willing to spend the time to write
these tools. We tend to use whatever is off-the-shelf. To people like us,
software isn't the ends, it's just a means to get there. That's probably why
pure software people can get more religious than the hardware engineers.

Cheers.

--
with metta,
Shawn Tan

Aeste Works (M) Sdn Bhd - Engineering Elegance
http://www.aeste.net

2007\08\02@115241 by Gerhard Fiedler

picon face
Xiaofan Chen wrote:

>> I also don't share Xiaofan's reservation about virtual machines. I think
>> they are great. (You need the memory, though -- that's the part they don't
>> virtualize :)
>
> Yes it is good for testing purpose.
>
> However the situation here is different. In this case, they are use it
> for real work.

I don't know what you mean by that difference between testing and real
work. For some, testing /is/ real work. And when done well, it's no less
demanding than a production run (if that is what you mean); often more so.

> They like to use Mac OS X or Linux for its so-called "superiority against
> Windows". Then in reality they have to use MPLAB with ICD2 to debug
> theire PIC code. So they use VMware and Windows. To me it is kind of
> irony. This is actually a good solution in the aspect of technology.

I think in this paragraph you mixed two things. I agree with the last
phrase; and that's about virtualization technology. The rest is about why
some people use it and how they talk about it... I don't think this has
much to do with how well it works or what it can be used for.

I really think that having the possibility of two OSes running at the same
time on one computer is really great; they may be different (as in Linux
and Windows) or the same (as in a client-server setup). I don't see
anything wrong with either. In fact, if I'll move to Linux some day, that's
probably how I'll do it. It's just too big a disruption to do it all at
once. Dual-boot is a pain; you don't have access to the applications on the
other system. I don't see any advantage in dual-boot vs virtual machines;
they are pretty stable now, and their major growth is in the server world,
where stability is more important than in the desktop world.

So you can start with a Linux installation in a virtual machine to become
familiar with it and start to think about how to set it all up. You can
play with it when you have the time, and don't need to reboot for that (and
don't need to reboot to switch back). Once familiar enough, you switch to
Linux as a host and leave your previous Windows system running in a virtual
machine. This provides (almost) full access to everything you had before.
With time get rid of anything in the Windows installation that has been
moved to the Linux system. Some things maybe remain, maybe not. Much less
disruptive than any other way.

Unless you /want/ to become religious about this :) I don't see a reason to
talk about virtualization technology in the terms of some OS fanatics.
Virtualization allows us to take advantage of whatever we like. Need
Windows for some things but like the Apple notebooks? Get one and run
Windows under VMware. Like to work on Windows but are developing web apps
that run on Linux hosts? Run Linux under VMware on your Windows system. And
so on.

> In this aspect, I think Wine or Vmware actually hinders the development
> of alternative applications.

I don't think so. You maybe can have that impression if you only look at a
certain application, but once you factor in the boost that Wine and VMware
gave to Linux as a whole (in that they allow people to switch to Linux but
still run their few remaining Windows-only applications without
dual-booting), things may look different. Difficult to judge, at least for
me, where the balance is.

> The best is of course for the vendor to support Linux directly
> (MPLAB/ICD2/etc). In reality, this is not possible in the forseeable
> future.

Right. And that's why virtualization is a practical solution that's
independent of any vendor support for this or that OS.

Gerhard

2007\08\02@115424 by Xiaofan Chen

face picon face
On 8/2/07, Shawn Tan <spamBeGoneshawn.tanEraseMEspamaeste.net> wrote:
>
> Actually, this brings up another consideration. We hardware people need good
> software tools. However, we're not often willing to spend the time to write
> these tools. We tend to use whatever is off-the-shelf. To people like us,
> software isn't the ends, it's just a means to get there. That's probably why
> pure software people can get more religious than the hardware engineers.
>

Thanks a lot for the insight. I was wondering about this question
from time to time why highly intelligent people can get so religious
in the open source community.

I highly respect Linus though. He seems to be much a more
practical guy. Apparently he is a software guy but he is quite
different from the others in the same community.

Ok I'd better shut up. Apparently I am not in a good mode
these two days. I should be happy now since I will go back
home tomorrow. It is so good to be home...


Xiaofan

2007\08\02@115951 by Xiaofan Chen

face picon face
On 8/2/07, Xiaofan Chen <xiaofancspamBeGonespamgmail.com> wrote:
> Thanks a lot for the insight. I was wondering about this question
> from time to time why highly intelligent people can get so religious
> in the open source community.
>
> I highly respect Linus though. He seems to be much a more
> practical guy. Apparently he is a software guy but he is quite
> different from the others in the same community.
>
> Ok I'd better shut up. Apparently I am not in a good mode
> these two days. I should be happy now since I will go back
> home tomorrow. It is so good to be home...
>

I realized that I may offended some people in the list by my
comments. I apologize. My comments are my thoughts and
they are not targeted to some paticular list members. Actually
I highly respect the other people's views in general.


Regards,
Xiaofan

2007\08\02@134217 by Herbert Graf

flavicon
face
On Wed, 2007-08-01 at 17:14 -0400, Xiaofan Chen wrote:
> Seriously, your IT department allows you to do that?

What my IT department doesn't know...

> Somehow I have a bad feeling using Wine even though I start
> my first GNUPIC post with a howto on MPLAB with Wine. It
> extends to Vmware even though it might be much better than Wine.
> I prefer to use dual-boot. I prefer to use real Linux or real Windows.

Vmware is FAR FAR FAR different from Wine, you can't compare the two.

Vmware basically creates a virtual PC, you can load whatever OS you want
on this virtual PC. In fact, once you configure the virtual machine, you
have to "install" windows EXACTLY like you would a normal machine. The
Virtual machine even BOOTS exactly like a normal machine, complete with
a system CMOS you can tweak.

As for your "real windows" statement, a virtual machine under vmware IS
running a "real windows", windows has no clue it's running on a virtual
machine, it just sees a PC, everything else is the same. You can run
sound, video, access USB devices, etc.

Once you've tried using a virtual machine the reasons for dual booting
become very short in number IMHO.

TTYL

2007\08\02@134418 by Herbert Graf

flavicon
face
On Thu, 2007-08-02 at 09:53 -0400, Xiaofan Chen wrote:
> On 8/2/07, Peter Bindels <RemoveMEdascandy@spam@spamspamBeGonegmail.com> wrote:
> > For normal uses, it's the best you can get - virtual hardware doesn't
> > suffer the problems real hardware has. It's never shorted out, it
> > doesn't fail due to solder masks, it's never out of stock, it always
> > responds perfectly and within time bounds, unless you tell it not
> > to... It's great for testing.
>
> It is great for testing. For real work, I am not so sure. It is slower,
> it does not support all the hardwares. For example, you can get
> ICD2 to work but not RealICE or ICE2000. For real work, stick
> with the original OS...

But it IS the original OS, it's the machine that's virtualized. You
install the OS exactly like you would on a "real" machine.

And for the record, I do "real work" on my virtual machines, how do you
define "real work"?

TTYL

2007\08\02@134645 by Herbert Graf

flavicon
face
On Thu, 2007-08-02 at 09:51 -0400, Xiaofan Chen wrote:
{Quote hidden}

Right, but you get all the benefits of Linux being the OS underneath.
Windows under vmware is actually more stable then windows running native
on the machine. Plus, you only run the apps you NEED to run under
windows, so when windows does crash, you only loose the work that was
running under windows, the Linux machine runs as if nothing went wrong
and you can continue your Linux work as the virtual machine reboots.

Yes, you're still running windows, but you're relying on it alot less,
and you'll pretty much never have to sit at your computer waiting for a
reboot anymore.

2007\08\02@155658 by Xiaofan Chen

face picon face
On 8/2/07, Herbert Graf <.....mailinglist3RemoveMEspamfarcite.net> wrote:
> > Windows under VMware is still Windows.
>
> Right, but you get all the benefits of Linux being the OS underneath.
> Windows under vmware is actually more stable then windows running native
> on the machine. Plus, you only run the apps you NEED to run under
> windows, so when windows does crash, you only loose the work that was
> running under windows, the Linux machine runs as if nothing went wrong
> and you can continue your Linux work as the virtual machine reboots.
>
> Yes, you're still running windows, but you're relying on it alot less,
> and you'll pretty much never have to sit at your computer waiting for a
> reboot anymore.
>

Thanks for not offended by my comments. Now I feel better. Of course
this is also due to the fact the tough meeting was over and I will
go back to Singapore tomorrow morning. ;-)

Now forget about the religious things and go back to the technical
benefits of visualizations.

Firstly, I am not a software guy and I do not know
much about programming and struggling with firmware programming.
(eg: http://forum.microchip.com/tm.aspx?m=270049)

Still I "waste" a lot of time reading IT related things, especially
last time. I can see there are many benefits of visualization.
It is the buzz word now after all. Vmware is going to public soon.

Still I think two workstations running Windows and Linux superlatively
is better than a single workstation running Vmware to help with the
shortcomings of Windows or Linux.

I am not so sure about the comment that Windows under
Vmware is more stable than Windows. The only thing is that
you perhaps have kind of standard hardware and driver.

There are also limitations of Vmware. Hardware driver is a
big thing. I know that ICD2 is working. But it is said that
RealICE does not.

Then  when you are doing work, you anyway need to
save things. If your vmware windows sessions failed, it
is the same as when your run it under Windows. You
still lose the unsaved changes.

I am not so sure if I understand the sentence that "you only
run the apps you NEED to run under windows". You run
only the apps you NEED to run under Windows as long
as you are a knowledgeable and responsible user.


Regards,
Xiaofan

2007\08\02@184031 by Nate Duehr

face
flavicon
face
Jake Anderson wrote:

> tried fasterfox addon?

Used it for a while, but some websites are now redirecting folks who are
opening multiple sessions (fasterfox has a "fingerprint" of sorts, when
you're looking at the server logs) to a page that says, "You're abusing
our webserver by opening multiple connections."  And refusing to display
the site.

In other words, some of the more aggressive settings in FasterFox bug
webserver sysadmins, and they're starting to retaliate... they
shouldn't, they just just build their servers appropriately, but...
well, we'll see how it plays out.

Many server admins are too clueless to do the above, so there's a
built-in barrier to the playground...

Nate

2007\08\02@192917 by James Newtons Massmind

face picon face
> In other words, some of the more aggressive settings in
> FasterFox bug webserver sysadmins, and they're starting to
> retaliate... they shouldn't, they just just build their
> servers appropriately, but...

Errr... How is a web server built "appropriately" to manage bandwidth abuse?

And when did the firefox developers and or users start paying my bandwidth
bill?

---
James.


2007\08\02@211353 by Nate Duehr

face
flavicon
face
James Newtons Massmind wrote:
>> In other words, some of the more aggressive settings in
>> FasterFox bug webserver sysadmins, and they're starting to
>> retaliate... they shouldn't, they just just build their
>> servers appropriately, but...
>
> Errr... How is a web server built "appropriately" to manage bandwidth abuse?
>
> And when did the firefox developers and or users start paying my bandwidth
> bill?

In order to answer that question, you'd have to define "bandwidth abuse"
and that's a tall order, since it varies by your site's traffic and
overall ability to handle HTTP traffic overall.

Just as a small point, the "Firefox developers" didn't write FasterFox,
someone else did... a user of Firefox, so there must have been an itch
they wanted to scratch.

I personally think that the HTTP pipelining feature and some of the less
aggressive FasterFox "enhancements" are fine, but as others have
mentioned, pre-caching is retarded... and never used it.

Wish I could remember which site actually threw me over to a "you're
running something like FasterFox and we don't like it" page... it was a
lot bigger than PicList, and could/should have easily handled the little
extra bandwidth from a small percentage of users... since not even a
large percentage of Firefox users ever use FasterFox (from what I've
seen in various server logs on systems I take care of).

I think the crux/driving motivation behind a lot of people adopting
things like FasterFox really stems from sites that have rediculous
amounts of advertising links on them, and heavy CSS/frames/whatever
design choices that cause pages to seem like they're rendered at a
snail's pace, even when bandwidth between the webserver and the browser
are fine... of course, that's not always even hosted on the same server
(usually not) so... people's perceptions vs. what's really happening are
definitely at play here.

In other words... anyone can do whatever they want with their servers...
but if they find that 80% of their users are hitting them with Firefox
and all the aggressive FasterFox stuff turned on... if they want/need
the traffic, they'd better just set up the server farm to handle it.
Whining about it with a redirect page is only a temporary and false way
to deal with it.

I also agree that some of the things FasterFox does goes too far, and
it's pretty clearly labeled inside the extension that those items really
aren't a good idea... but people are going to mess with them anyway.  It
has different "levels" of insanity, but people will be people and try
all of them, anyway.

Even more interesting is, none of the changes FasterFox makes are
TECHNICALLY wrong (other than pre-caching), they're just "different than
the norm".  So if a web host wants to block "non-normal" types of
traffic, that's up to them...

Nate

2007\08\02@214036 by Jake Anderson

flavicon
face
James Newtons Massmind wrote:
>> In other words, some of the more aggressive settings in
>> FasterFox bug webserver sysadmins, and they're starting to
>> retaliate... they shouldn't, they just just build their
>> servers appropriately, but...
>>    
>
> Errr... How is a web server built "appropriately" to manage bandwidth abuse?
>  
If pre-cache is disabled then fasterfox should not increase bandwidth.
At all.
The highest level before precache sets several options that violate RFC
specs. IE number of simultaneous connections to a http V 1 host and the
like.
It would increase the "instant" load on the server which some people
might not like. But really if a P2 300 can flood a 100mbit link then
they must have pretty serious server issuses to be worried about that.
> And when did the firefox developers and or users start paying my bandwidth
> bill?
>
> ---
> James.
>  
>
>  

2007\08\02@221150 by Gerhard Fiedler

picon face
Xiaofan Chen wrote:

> Still I think two workstations running Windows and Linux superlatively is
> better than a single workstation running Vmware to help with the
> shortcomings of Windows or Linux.

Not necessarily. If you have good screens connected to one machine, you may
not have the money to get a second set. Or you may have to take your
notebook along with your development environment. In both cases having both
systems running on one machine may be an advantage. It all depends on the
specific situation.

> Then  when you are doing work, you anyway need to save things. If your
> vmware windows sessions failed, it is the same as when your run it under
> Windows. You still lose the unsaved changes.

Of course, but if it was due to a Windows crash, that's the same whether
you have a second physical machine or run it in a VMware box.

> I am not so sure if I understand the sentence that "you only run the apps
> you NEED to run under windows". You run only the apps you NEED to run
> under Windows as long as you are a knowledgeable and responsible user.

I think Herbert meant that if you have a Linux system and run Windows
inside a virtual machine on it, that's usually because of a few
applications that are Windows-only -- and all the other apps you need you
run on the same machine under Linux. You can do that (even on a notebook
that you can take with you) because of virtualization. Otherwise you'd have
to use two machines (sometimes not practical) or dual-boot (which would
require you to run several applications like email on both systems -- which
is what Herbert seems to have thought of).

Gerhard

2007\08\02@223931 by Herbert Graf

flavicon
face
On Thu, 2007-08-02 at 15:56 -0400, Xiaofan Chen wrote:
> Still I think two workstations running Windows and Linux superlatively
> is better than a single workstation running Vmware to help with the
> shortcomings of Windows or Linux.

Define "better"? Two workstations take up twice the space, twice the
power, and you have twice the chance of a hardware problem.

One workstation has a higher impact of hardware failure, but takes up
half the space and power. It does require more memory and CPU power then
each separate single workstation, but doesn't require the doubling that
two workstations have.

> I am not so sure about the comment that Windows under
> Vmware is more stable than Windows. The only thing is that
> you perhaps have kind of standard hardware and driver.

I can't explain it, perhaps it's the fact that windows "sees" less
complications that makes it more stable. All I know is in every case of
my experience the windows virtual machines have been more stable then
their native counterparts (might I add, identical hardware in both
cases).

> There are also limitations of Vmware. Hardware driver is a
> big thing. I know that ICD2 is working. But it is said that
> RealICE does not.

If it's USB it generally works. Older versions of vmware did have some
problems, the latest has improved so much that I haven't seen any
problems. The ICD2 runs perfectly, as do numerous other USB devices I've
tried (programmers/debuggers for other platforms mostly).

> Then  when you are doing work, you anyway need to
> save things. If your vmware windows sessions failed, it
> is the same as when your run it under Windows. You
> still lose the unsaved changes.

I don't understand what you are saying here.

> I am not so sure if I understand the sentence that "you only
> run the apps you NEED to run under windows". You run
> only the apps you NEED to run under Windows as long
> as you are a knowledgeable and responsible user.

I'm not saying virtualization is for everyone, most people seem happy
with the hoops that windows has had them jump through for years, so be
it. Me, I've experienced the "other side", and I won't be going back any
time soon.

As I said before, I do still run windows, even in non virtual form. My
laptop still runs windows. The reason being the apps I use on the road
won't work under wine (I've tried), I haven't liked the competing apps,
and the machine doesn't have enough memory or HD space to run linux
+vmware comfortably, so for now, it stays windows. Hopefully that will
soon change, it's still by far the least stable machine in the house.

I've recently been working on upgrading one of my servers, and as a
result had to reboot the old machine. Before reboot I had a look at how
long it had been up, I don't remember the exact count, but it was in the
300 day range. Fedora Core 5. The machine just works. That's how an OS
should be IMHO. TTYL

2007\08\03@004123 by jamodio

flavicon

After many years of countless unkown error messages and having to reboot
the machine every time William the Gates sneezes I decided for the dual
machine approach.

I have a dual core machine running linux in a corner of my office so it
doesn't
take any desk space and it has been running for ages and I use vnc on the
windoze.

I like vmware but when you are running any flavor of windoze is like running
a controlled version of malware with a gui interface that requires tons of
resources just to do the most simple things.

Just my humble opinion.

Cheers.

--
View this message in context: www.nabble.com/Firefox-security-gripe-tf4136687.html#a11977381
Sent from the PIC - [OT] mailing list archive at Nabble.com.

2007\08\04@220006 by Xiaofan Chen

face picon face
On 8/3/07, Herbert Graf <.....mailinglist3STOPspamspam@spam@farcite.net> wrote:
> > There are also limitations of Vmware. Hardware driver is a
> > big thing. I know that ICD2 is working. But it is said that
> > RealICE does not.
>
> If it's USB it generally works. Older versions of vmware did have some
> problems, the latest has improved so much that I haven't seen any
> problems. The ICD2 runs perfectly, as do numerous other USB devices
> I've tried (programmers/debuggers for other platforms mostly).

RealICE is USB. There are many USB reported not working. Anyway,
this might not be the fault of Vmwaree. Many USB device have flacky
firmware which prevent them working even natively under Linux or
Windows. Many USB drivers are also not quite good. USB seems to
be very difficult in terms of firmware and driver even though the
hardware seems to be simple.

> As I said before, I do still run windows, even in non virtual form. My
> laptop still runs windows. The reason being the apps I use on the road
> won't work under wine (I've tried), I haven't liked the competing apps,
> and the machine doesn't have enough memory or HD space to run linux
> +vmware comfortably, so for now, it stays windows. Hopefully that will
> soon change, it's still by far the least stable machine in the house.
>

Good to know this.

Xiaofan

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...