Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > Computer Information > 2008 HP Pavilion w/Vista SP2 - Facebook via Safari 5.1.7 All EFFED UP!!

Reply
Thread Tools

2008 HP Pavilion w/Vista SP2 - Facebook via Safari 5.1.7 All EFFED UP!!

 
 
thekmanrocks@gmail.com
Guest
Posts: n/a
 
      10-02-2013
2008 HP Pavilion w/Vista SP2 - Facebook via Safari 5.1.7 All EFFED UP!!


First of all - when I go to the login in screen, the headers describing the sections of FB look like "@$##@$#%$%$ - Share Photos with Your Friends". Gobledy gook.

Second off - I cannot upload pictures from my PC to the newsfeed or my own wall in Safari.

MS Internet Explorer is fine on both counts.
 
Reply With Quote
 
 
 
 
thekmanrocks@gmail.com
Guest
Posts: n/a
 
      10-14-2013
No one??

Come on Paul, you're the one here who can pilot a phone booth to Mars and back. Surely theres a reason for what I described.
 
Reply With Quote
 
 
 
 
Paul
Guest
Posts: n/a
 
      10-14-2013
http://www.velocityreviews.com/forums/(E-Mail Removed) wrote:
> No one??
>
> Come on Paul, you're the one here who can pilot a phone booth to Mars and back. Surely theres a reason for what I described.


What are the odds of me being able to reproduce that here ?
Pretty small.

Take a picture of Safari when it fouls up and post
it somewhere. Like on imageshack.us or Picasa or some
similar kind of image hosting site.

It could be a language preference that is screwed up
somewhere.

Note that, when web sites respond, they can geolocate.
So when a user in the UK sees a web site screw up on
them, I'm in Canada and the web site sends me entirely
different files. And that makes it pretty hard to
reproduce the problem. The odds are very poor that
I will see every detail the same.

Some problems are coded wrong, no matter where in the
world you are, or what browser you use, and that's about
the only kind of problem I could help with. Because
then I tend to be able to see it.

Paul
 
Reply With Quote
 
Paul
Guest
Posts: n/a
 
      10-14-2013
(E-Mail Removed) wrote:
> No one??
>
> Come on Paul, you're the one here who can pilot a phone booth to Mars and back. Surely theres a reason for what I described.


Problem is reproducible.
This is a good first step.

I'm going out now, so will look at this later.

http://imageshack.us/a/img69/3236/dlb.gif

Paul
 
Reply With Quote
 
thekmanrocks@gmail.com
Guest
Posts: n/a
 
      10-14-2013
On Monday, October 14, 2013 4:53:21 PM UTC-4, Paul wrote:
>
> > No one??

>
> >

>
> > Come on Paul, you're the one here who can pilot a phone booth to Mars and back. Surely theres a reason for what I described.

>
>
>
> Problem is reproducible.
>
> This is a good first step.
>
>
>
> I'm going out now, so will look at this later.
>
>
>
> http://imageshack.us/a/img69/3236/dlb.gif
>
>
>
> Paul

____________

Yes!! That is exactly what I see via Safari for Windows. Does not happen in MSIE.
 
Reply With Quote
 
Paul
Guest
Posts: n/a
 
      10-15-2013
(E-Mail Removed) wrote:
> On Monday, October 14, 2013 4:53:21 PM UTC-4, Paul wrote:
>>> No one??
>>> Come on Paul, you're the one here who can pilot a phone booth to Mars and back.
>>> Surely theres a reason for what I described.

>>
>>
>> Problem is reproducible.
>>
>> This is a good first step.
>>
>>
>>
>> I'm going out now, so will look at this later.
>>
>>
>>
>> http://imageshack.us/a/img69/3236/dlb.gif
>>
>>
>>
>> Paul

> ____________
>
> Yes!! That is exactly what I see via Safari for Windows. Does not happen in MSIE.


I'm not making much progress.

It's obviously a font problem (well, duh).

The CSS file (cascading style sheet) has no mention
of any outrageous fonts. Helvetica Neue is the font
that Facebook seeks to use, and that isn't available
on Vista (at least the Vista I've got here). The CSS
file gives a list of fonts used as fallbacks, including
vanilla Helvetica and Arial. None of those look goofy.
And the fallbacks are widely available and should have
worked.

I don't really have any good debug tools for this,
so I have to be content with observing side effects.

So I tried to record what fonts are actually being used.
I decided to use the Safari "print" function, to get some
information about the fonts. That didn't work out well at
all. After a couple hours of experiments, it turns out
that Safari just loves to dump bitmap prints with absolutely
no font info. ("Good" fonts use mathematical kinds of
glyphs, which scale well. Bitmap prints, when you blow
them up, have "jaggies" along the edge. That was my first
hint that the Safari printing was poor.)

Then I tried another tack. I decided to test other browser/OS
combinations (i.e. the ones that "work"). I logged the
fonts in those. And I found a reference to "Mangal" font,
which is a font that people in India might use when
browsing. There is no reason for that font to be
in the output.

And that's in a print and web view that are perfectly
normal looking. With a normal looking result, the printer
output says "Mangal font was used". Yet, I don't see any
Indian characters. In the broken case, the characters used
might indeed be Mangal (a font that Microsoft provides
apparently). Since they are punctuation in Mangal, it's hard
to find samples (font box style) which I can compare to
the bad Safari result.

So I haven't made much progress on the problem. I see
something goofy is going on, yet when I make a .webarchive
from the facebook.com main page, there is no mention of
"Mangal" at all in the ~700K text file that represents
the contents of the web page.

Whatever is happening, isn't a function of the style sheet.
A font substitution is being made somewhere. And the other
weird thing, is there might be some hidden Mangal letters
on the working web page, which are underneath somewhere and
not visible.

There probably will not be another release of Safari for
Windows, so if this is a problem with 5.1.7, you're stuck
with it.

Paul
 
Reply With Quote
 
Paul
Guest
Posts: n/a
 
      10-16-2013
Paul wrote:
> (E-Mail Removed) wrote:
>> On Monday, October 14, 2013 4:53:21 PM UTC-4, Paul wrote:
>>>> No one??
>>>> Come on Paul, you're the one here who can pilot a phone booth to
>>>> Mars and back. Surely theres a reason for what I described.
>>>
>>> Problem is reproducible.
>>>
>>> This is a good first step.
>>>
>>> http://imageshack.us/a/img69/3236/dlb.gif
>>>
>>> Paul

>> ____________
>>
>> Yes!! That is exactly what I see via Safari for Windows. Does not
>> happen in MSIE.


So I tried another experiment tonight.

I used Sysinternals Process Monitor (procmon) to trace all activity
for nine seconds it took for the facebook page to load and render.
In nine seconds it collected 93,000 events, which I sift through by
hand. A fair number of them, I can just hit the scroll bar and ignore.

What I was looking for, was font calls. This is the summary

tahomabd.ttf \
verdana.ttf \___ Plain Jane Microsoft fonts
arial.ttf /
clbcatq.dll (com+ catalog?)
mlang.dll (possibly in preparation for the next font)
mangal.ttf (Indian ?)
msgothic.ttc (Japanese ?)
simsun.ttc (Chinese)
atmfd.dll (OpenType Compact Font Format (CFF) driver, Adobe provided?)
micross.ttf (Microsoft Sans Serif) <--- source of garbage
autofill.js (from Safari resources folder)

Using the info, I found a Microsoft page with font samples of Sans Serif,
and the gibberish is Sans Serif. I suspect the atmfd.dll (possibly
written by Adobe), extracts what is necessary from micross.ttf.

If you look at the fonts with a font tool, there are many mappings
inside. The gibberish, if you wipe over the text in Safari, then
copy and paste into Notepad, the "real" letters show up. That tells
me the letters themselves are not screwy. When the font table gets
accessed, it's going to a page it shouldn't, like perhaps a bug
in the font handling.

Now, you might ask, why that set of fonts ?

Well, at the bottom of the Facebook page, it has links to click
in foreign languages. And the letters down there are printed
so foreigners will understand. That's why there is Japanese,
Chinese, and Indian text at the bottom. That's where the mangal
was coming from - it was for the Indian text at the bottom. So
that's no longer a mystery.

But that doesn't excuse the Sans Serif.

In the .webarchive file (which is how Safari saves a web page if
you ask it to), I collected these lines. I think these are from
a .css file (cascading style sheet). I don't know how to read one
of those, but on one site, I learned these are "font stacks".

font-family:'Helvetica Neue', Helvetica, Arial, 'lucida grande',
tahoma,verdana,arial,sans-serif;font-size:12px;
height:18px;line-height:18px
font-family:'Freight Sans Bold', 'lucida grande',tahoma,verdana,
arial,sans-serif;font-size:40px;font-weight:normal
font-family:'lucida grande',tahoma,verdana,arial,sans-serif;
font-size:11px
font-family:Helvetica, Arial, 'lucida grande',tahoma,verdana,
arial,sans-serif;font-size:12px;line-height:16px

What happens is, that's code for the web browser to use, and it
says "if you can't find the first font, try the next one". The
left-most font gives the nicest looking web page. It gets
uglier as you move to the right. And at the end of each font
stack, "sans-serif" is selected as the "I give up, just print
something" font. So that's where the Sans-Serif could be coming
from.

What I *suspect* is happening, is first Safari is asked to
open all those Indian/Japanese/Chinese fonts, and some of
those fonts contain huge tables. It causes something in the
Safari font handling code to barf. Safari attempts to
analyse its next font stack, perhaps looking for Helvetica Neue.
It fails to find any fonts (even though it loaded some of them
early in the trace). It resorts to Sans-Serif, as instructed.
It accesses a portion of table without a regular alphabet in it,
and prints that on the screen.

Now, if we compare Safari to Firefox, when opening wikipedia.org,
that site has similar decorations at the bottom of the page. And
uses a lot more fonts than Facebook does. And yet Safari manages
to do a good job on that page. It has about as many defects as
Firefox does, with some fonts missing because perhaps an
English version of Vista just doesn't have anything close.

I can't make any more progress, because I don't know how to
read a .css file.

Paul
 
Reply With Quote
 
thekmanrocks@gmail.com
Guest
Posts: n/a
 
      10-16-2013
On Wednesday, October 16, 2013 1:00:49 AM UTC-4, Paul wrote:
> Paul wrote:
>

..com wrote:
>
> >> On Monday, October 14, 2013 4:53:21 PM UTC-4, Paul wrote:

>
> >>>> No one??

>
> >>>> Come on Paul, you're the one here who can pilot a phone booth to

>
> >>>> Mars and back. Surely theres a reason for what I described.

>
> >>>

>
> >>> Problem is reproducible.

>
> >>>

>
> >>> This is a good first step.

>
> >>>

>
> >>> http://imageshack.us/a/img69/3236/dlb.gif

>
> >>>

>
> >>> Paul

>
> >> ____________

>
> >>

>
> >> Yes!! That is exactly what I see via Safari for Windows. Does not

>
> >> happen in MSIE.

>
>
>
> So I tried another experiment tonight.
>
>
>
> I used Sysinternals Process Monitor (procmon) to trace all activity
>
> for nine seconds it took for the facebook page to load and render.
>
> In nine seconds it collected 93,000 events, which I sift through by
>
> hand. A fair number of them, I can just hit the scroll bar and ignore.
>
>
>
> What I was looking for, was font calls. This is the summary
>
>
>
> tahomabd.ttf \
>
> verdana.ttf \___ Plain Jane Microsoft fonts
>
> arial.ttf /
>
> clbcatq.dll (com+ catalog?)
>
> mlang.dll (possibly in preparation for the next font)
>
> mangal.ttf (Indian ?)
>
> msgothic.ttc (Japanese ?)
>
> simsun.ttc (Chinese)
>
> atmfd.dll (OpenType Compact Font Format (CFF) driver, Adobe provided?)
>
> micross.ttf (Microsoft Sans Serif) <--- source of garbage
>
> autofill.js (from Safari resources folder)
>
>
>
> Using the info, I found a Microsoft page with font samples of Sans Serif,
>
> and the gibberish is Sans Serif. I suspect the atmfd.dll (possibly
>
> written by Adobe), extracts what is necessary from micross.ttf.
>
>
>
> If you look at the fonts with a font tool, there are many mappings
>
> inside. The gibberish, if you wipe over the text in Safari, then
>
> copy and paste into Notepad, the "real" letters show up. That tells
>
> me the letters themselves are not screwy. When the font table gets
>
> accessed, it's going to a page it shouldn't, like perhaps a bug
>
> in the font handling.
>
>
>
> Now, you might ask, why that set of fonts ?
>
>
>
> Well, at the bottom of the Facebook page, it has links to click
>
> in foreign languages. And the letters down there are printed
>
> so foreigners will understand. That's why there is Japanese,
>
> Chinese, and Indian text at the bottom. That's where the mangal
>
> was coming from - it was for the Indian text at the bottom. So
>
> that's no longer a mystery.
>
>
>
> But that doesn't excuse the Sans Serif.
>
>
>
> In the .webarchive file (which is how Safari saves a web page if
>
> you ask it to), I collected these lines. I think these are from
>
> a .css file (cascading style sheet). I don't know how to read one
>
> of those, but on one site, I learned these are "font stacks".
>
>
>
> font-family:'Helvetica Neue', Helvetica, Arial, 'lucida grande',
>
> tahoma,verdana,arial,sans-serif;font-size:12px;
>
> height:18px;line-height:18px
>
> font-family:'Freight Sans Bold', 'lucida grande',tahoma,verdana,
>
> arial,sans-serif;font-size:40px;font-weight:normal
>
> font-family:'lucida grande',tahoma,verdana,arial,sans-serif;
>
> font-size:11px
>
> font-family:Helvetica, Arial, 'lucida grande',tahoma,verdana,
>
> arial,sans-serif;font-size:12px;line-height:16px
>
>
>
> What happens is, that's code for the web browser to use, and it
>
> says "if you can't find the first font, try the next one". The
>
> left-most font gives the nicest looking web page. It gets
>
> uglier as you move to the right. And at the end of each font
>
> stack, "sans-serif" is selected as the "I give up, just print
>
> something" font. So that's where the Sans-Serif could be coming
>
> from.
>
>
>
> What I *suspect* is happening, is first Safari is asked to
>
> open all those Indian/Japanese/Chinese fonts, and some of
>
> those fonts contain huge tables. It causes something in the
>
> Safari font handling code to barf. Safari attempts to
>
> analyse its next font stack, perhaps looking for Helvetica Neue.
>
> It fails to find any fonts (even though it loaded some of them
>
> early in the trace). It resorts to Sans-Serif, as instructed.
>
> It accesses a portion of table without a regular alphabet in it,
>
> and prints that on the screen.
>
>
>
> Now, if we compare Safari to Firefox, when opening wikipedia.org,
>
> that site has similar decorations at the bottom of the page. And
>
> uses a lot more fonts than Facebook does. And yet Safari manages
>
> to do a good job on that page. It has about as many defects as
>
> Firefox does, with some fonts missing because perhaps an
>
> English version of Vista just doesn't have anything close.
>
>
>
> I can't make any more progress, because I don't know how to
>
> read a .css file.
>
>
>
> Paul

_____________

Face it Paul: the devil in the details is Facebook itself.

I truly believe there were few other events that might have changed world history for better than

1. Kennedy survived assassination/or was thwarted.

2. Facebook was never invented.

Seriously!




Just an update Paul on my other issue: Every day windows is attempting to run drivers install for my HP DeskJet printer.

I'm currently up to "HP 4100 Series Printer(Copy 4)"!!!

Oh well, at least I can print after each install. smdh..
 
Reply With Quote
 
Paul
Guest
Posts: n/a
 
      10-17-2013
(E-Mail Removed) wrote:

> Face it Paul: the devil in the details is Facebook itself.
>
> I truly believe there were few other events that might have changed world history for better than
>
> 1. Kennedy survived assassination/or was thwarted.
>
> 2. Facebook was never invented.
>
> Seriously!
>
>
>
>
> Just an update Paul on my other issue: Every day windows is attempting to run drivers install for my HP DeskJet printer.
>
> I'm currently up to "HP 4100 Series Printer(Copy 4)"!!!
>
> Oh well, at least I can print after each install. smdh..


You lurch along, getting whatever you can from the
computer before it tries its next mean trick

I think that is the reason to have multiple computers,
so they can't all pull the same mean trick at the
same time.

I think the Safari problem is a Safari bug. Web developers
have a large number of "quirks" they include in their
server code, to account for bugs like that. And while
it's a Safari bug, we know with the level of
support for Safari in Windows, only Facebook can fix it
now (by sending different web code).

Paul

 
Reply With Quote
 
thekmanrocks@gmail.com
Guest
Posts: n/a
 
      10-17-2013
Well, the printer issue has persisted since after replacing the CD/DVD rom/burner in my Pavilion.

Coincidence??

I checked my Device Manager, unfortunately printers do not appear there. Still, no black/yellow exclamation points(conflicts).
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
2008 HP Pavillian w/Vista SP2 - Facebook via Safari 5.1.7 All EFFED thekmanrocks@gmail.com Computer Information 0 10-02-2013 09:26 PM
Writing/Developing Facebook Applications in .NET using Facebook.NETSDK Dutt ASP .Net 1 02-08-2008 03:38 PM
Facebook app give feedback -Send birthday cards to buds http://apps.facebook.com/groupcards/start fringo Python 0 10-08-2007 08:04 PM
Facebook app give feedback -Send birthday cards to buds http://apps.facebook.com/groupcards/start fringo Ruby 0 10-08-2007 08:03 PM
It finally happened to me (effed up and got fullscreen) Michael Rogers DVD Video 17 11-05-2004 04:48 PM



Advertisments