[OT] Microsoft "decline"
Chris Ahlstrom
Cheating Accusations Highlight Microsoft's Decline
By Katherine Noyes, PCWorld
So, what does it mean for users? Simply that the Microsoft name so longer
signifies anything about quality or innovation--quite the reverse, in
many cases. I'm not saying Google or any other company or product is
perfect, but Microsoft clearly does not have quality or its users' best
interests at heart.
The article overstates the case, but the patina is definitely wearing thin.
--
These goofs were made for stalking/
And that's just what they'll do/
JeffM
>:Katherine Noyes, PCWorld
>:[...]the Microsoft name
>:[no?] longer signifies anything about quality or innovation[...]
>
Dear Katherine,
It would be good if people making a claim for "Microsoft innovation"
would give a single example of something M$ created
--something M$ didn't simply copy from others or just buy up.
Once again for the clueless: "Microsoft innovation" is an oxymoron.
Ezekiel
"JeffM" <jef...@email.com> wrote in message
news:c52d6c01-57fd-4e59...@y3g2000vbh.googlegroups.com...
> Chris Ahlstrom wrote:
>>:Katherine Noyes, PCWorld
>>:[...]the Microsoft name
>>:[no?] longer signifies anything about quality or innovation[...]
>>
> Dear Katherine,
> It would be good if people making a claim for "Microsoft innovation"
> would give a single example of something M$ created
<quote>
* The Taskbar
Popularized with Windows 95, the taskbar was actually a feature of Windows
1.0. A crude representation of active tasks laid were represented in a
horizontal bar on the bottom of the screen. This was two years early than
the Acorn Author operating environment that many Anti-Microsoft people cite
as the first task bar. The taskbar was gone from Windows until Windows 95.
* Real Time Spell Checking
This was first featured in Microsoft Word 6.0, and has since been
incorporated into every word processing program on the planet.
* Rapid Application Development
RAD was a hugely popular industry buzzword during the 90's, and it was born
with Microsoft Visual Basic.
Introduced in 1991, Visual Basic was a innovative new way to rapidly develop
applications and it turned the programming industry on its head. The closest
competitor to Visual Basic (Borland Delphi) was not released until 1995.
* OLE and COM
OLE was originally designed by Microsoft in 1990 by the Microsoft Office
team out of necessity. It was used to allow real time embedding of Excel
documents into Word, and vice versa. This paved the way for what became
known as COM (Component Object Model) a technology that allowed any COM
compliant object to be used by any application. COM allowed reuse of objects
with no knowledge of their internal implementation because it provided
interfaces that are separate from implementation.
COM is heavily adopted and is used in nearly every windows application even
to this day and was actually imitated by the open source movement with
CORBA.
* Active Desktop
Introduced with IE 4.0 and Windows 98 and powered by COM/ActiveX technology,
Active Desktop was Microsoft's vision of integrating desktop applications
with web applications.
Litigation kept Microsoft from fully leveraging the concept, but the idea
lives on now in many open source projects :)
* XMLHTTPRequest and AJAX
Believe it or not, the radical technology that changed the web and the
technology that powers Gmail was stuck into IE 4.0 without any fanfare.
The Microsoft Outlook team needed a way to communicate behind the scenes on
their new Microsoft Web Mail client, so they built an ActiveX component for
IE 4.0. This little component remained under the radar for several years,
before the rest of the world discovered its power and AJAX was born.
I wonder if Google has ever said "thank you".
</quote>
> --something M$ didn't simply copy from others or just buy up.
Feel free to list all of the original "innovations" (such as "Search") that
Google developed or any other open-source company innovated.
> Once again for the clueless: "Microsoft innovation" is an oxymoron.
For the terminally clueless there isn't very much new/original "innovation"
happening in computer science these days. Most of what's being developed
today are refinements on ideas that have been around for a while. Most
everything is based on some form of "prior art." .NET for example was
influenced by Java and other prior languages. Going back much of the Java
syntax was influenced by C/C++ and the Java itself was based on the Oak
programming language. Same for HTML being a derivation of SGML which in turn
is a derivation of GML from the 1960s.
DFS
> Chris Ahlstrom wrote:
>> :Katherine Noyes, PCWorld
>> :[...]the Microsoft name
>> :[no?] longer signifies anything about quality or innovation[...]
>>
> Dear Katherine,
> It would be good if people making a claim for "Microsoft innovation"
> would give a single example of something M$ created
operating systems: MS-DOS, Windows 3x, 9x, NT/2000/2003, Vista/7
database: ODBC, MSDE and Jet db engines
office: first bundled/integrated office apps
programming/development: Visual Basic/VBA, ActiveX/COM, C#, Visual
Studio, PowerShell
multimedia: Media Center, Media Player, maybe the .wma/.wmv formats
Gaming: DirectX
Gaming consoles: XBox, XBox360
web: contributions to XML and SOAP
other: ClearType font technology
> --something M$ didn't simply copy from others or just buy up.
> Once again for the clueless: "Microsoft innovation" is an oxymoron.
Like "Linux works" or "smart JeffM"
JEDIDIAH
>
> "JeffM" <jef...@email.com> wrote in message
> news:c52d6c01-57fd-4e59...@y3g2000vbh.googlegroups.com...
>> Chris Ahlstrom wrote:
>>>:Katherine Noyes, PCWorld
>>>:[...]the Microsoft name
>>>:[no?] longer signifies anything about quality or innovation[...]
>>>
>> Dear Katherine,
>> It would be good if people making a claim for "Microsoft innovation"
>> would give a single example of something M$ created
>
><quote>
> * The Taskbar
> Popularized with Windows 95, the taskbar was actually a feature of Windows
> 1.0. A crude representation of active tasks laid were represented in a
> horizontal bar on the bottom of the screen. This was two years early than
Taskbar? That version of Windows didn't even have overlapping Windows.
[deletia]
--
Unfortunately, the universe will not conform itself to
your fantasies. You have to manage based on what really happens |||
rather than what you would like to happen. This is true of personal / | \
affairs, government and business.
Ezekiel
"JEDIDIAH" <je...@nomad.mishnet> wrote in message
news:slrnikoro...@nomad.mishnet...
> On 2011-02-04, Ezekiel <no_...@fake-zeke.com> wrote:
>>
>> "JeffM" <jef...@email.com> wrote in message
>> news:c52d6c01-57fd-4e59...@y3g2000vbh.googlegroups.com...
>>> Chris Ahlstrom wrote:
>>>>:Katherine Noyes, PCWorld
>>>>:[...]the Microsoft name
>>>>:[no?] longer signifies anything about quality or innovation[...]
>>>>
>>> Dear Katherine,
>>> It would be good if people making a claim for "Microsoft innovation"
>>> would give a single example of something M$ created
>>
>><quote>
>> * The Taskbar
>> Popularized with Windows 95, the taskbar was actually a feature of
>> Windows
>> 1.0. A crude representation of active tasks laid were represented in a
>> horizontal bar on the bottom of the screen. This was two years early than
>
> Taskbar? That version of Windows didn't even have overlapping Windows.
Maybe you should have bought the model of your computer that came with a
search engine.
<quote>
The Windows 7 taskbar is the latest in a series of evolutionary taskbar
designs that started with Windows 1.0. It represents a design and
architectural change with regard to user experience, and offers significant
improvements in user productivity. Figure 1 shows the Windows 1.0 taskbar.
(Note that the 1.0 taskbar bears more resemblance to the Windows 7 taskbar
than to the Windows Vista taskbar.)
Figure 1 Windows 1.0 taskbar
</quote>
Chris Ahlstrom
> "JeffM" <jef...@email.com> wrote in message
> news:c52d6c01-57fd-4e59...@y3g2000vbh.googlegroups.com...
>> Chris Ahlstrom wrote:
>>>:Katherine Noyes, PCWorld
>>>:[...]the Microsoft name
>>>:[no?] longer signifies anything about quality or innovation[...]
>>>
>> Dear Katherine,
>> It would be good if people making a claim for "Microsoft innovation"
>> would give a single example of something M$ created
>
> <quote>
> * The Taskbar
> * Real Time Spell Checking
> * Rapid Application Development
> * OLE and COM
> * Active Desktop
> * XMLHTTPRequest and AJAX
> I wonder if Google has ever said "thank you".
> </quote>
I'd quibble a bit about RAD, but generally a good answer.
>> --something M$ didn't simply copy from others or just buy up.
>
> Feel free to list all of the original "innovations" (such as "Search") that
> Google developed or any other open-source company innovated.
http://labs.google.com/papers/gfs.html
>> Once again for the clueless: "Microsoft innovation" is an oxymoron.
>
> For the terminally clueless there isn't very much new/original "innovation"
> happening in computer science these days. Most of what's being developed
> today are refinements on ideas that have been around for a while. Most
> everything is based on some form of "prior art." .NET for example was
> influenced by Java and other prior languages. Going back much of the Java
> syntax was influenced by C/C++ and the Java itself was based on the Oak
> programming language. Same for HTML being a derivation of SGML which in turn
> is a derivation of GML from the 1960s.
Personally, I would call "innovation" any process that finds good solutions
to old problems, or even solutions that are just much better than old
solutions. For example, the Boost libraries. Leveraging existing C++
features in innovative ways, some of which are going to be folded into the
next C++ standard.
The problem is that "innovation" has been co-opted by advertising interests
on one hand, and loud-mouthed bloggers and COLA posters on the other hand.
:-)
Time for me to innovate some beer.
--
This fortune intentionally says nothing.
Chris Ahlstrom
> "JEDIDIAH" <je...@nomad.mishnet> wrote in message
> news:slrnikoro...@nomad.mishnet...
>>>
>>><quote>
>>> * The Taskbar
>>> Popularized with Windows 95, the taskbar was actually a feature of
>>> Windows
>>> 1.0. A crude representation of active tasks laid were represented in a
>>> horizontal bar on the bottom of the screen. This was two years early than
>>
>> Taskbar? That version of Windows didn't even have overlapping Windows.
>
> Maybe you should have bought the model of your computer that came with a
> search engine.
>
> <quote>
> The Windows 7 taskbar is the latest in a series of evolutionary taskbar
> designs that started with Windows 1.0. It represents a design and
> architectural change with regard to user experience, and offers significant
> improvements in user productivity. Figure 1 shows the Windows 1.0 taskbar.
> (Note that the 1.0 taskbar bears more resemblance to the Windows 7 taskbar
> than to the Windows Vista taskbar.)
>
> Figure 1 Windows 1.0 taskbar
> </quote>
>
> http://msdn.microsoft.com/en-us/magazine/dd942846.aspx
Actually, with the inability to overlap windows in that crud, that "taskbar"
was a necessity.
Yeah, that's some innovation there.
Actually, TomB might agree, since he uses a window manager where the windows
do not overlap.
Anyway, chuckles aside, I remember using Microsoft's CodeView ages ago.
It forced tiled windows on the user. (This was a text-based implementation
of windows, by the way.)
The Borland debugger, on the other hand, not only supported overlapping
windows, but was prettier.
They both beat the hell out of DEBUG.COM.
All that's in the past, though. I do most of my debugging in a two-paneled
console window.
--
A cow is a completely automated milk-manufacturing machine. It is encased
in untanned leather and mounted on four vertical, movable supports, one at
each corner. The front end of the machine, or input, contains the cutting
and grinding mechanism, utilizing a unique feedback device. Here also are
the headlights, air inlet and exhaust, a bumper and a foghorn.
At the rear, the machine carries the milk-dispensing equipment as
well as a built-in flyswatter and insect repeller. The central portion
houses a hydro- chemical-conversion unit. Briefly, this consists of four
fermentation and storage tanks connected in series by an intricate network
of flexible plumbing. This assembly also contains the central heating plant
complete with automatic temperature controls, pumping station and main
ventilating system. The waste disposal apparatus is located to the rear of
this central section.
Cows are available fully-assembled in an assortment of sizes and
colors. Production output ranges from 2 to 20 tons of milk per year. In
brief, the main external visible features of the cow are: two lookers, two
hookers, four stander-uppers, four hanger-downers, and a swishy-wishy.
Ezekiel
"Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
news:iiicr0$fn8$2...@news.eternal-september.org...
It's was version 1.0 - you gotta start somewhere.
> Yeah, that's some innovation there.
>
> Actually, TomB might agree, since he uses a window manager where the
> windows
> do not overlap.
>
> Anyway, chuckles aside, I remember using Microsoft's CodeView ages ago.
I used CodeView way back when I worked for Fujitsu. It was about the time
that MFC 1.0 was introduced.
> It forced tiled windows on the user. (This was a text-based
> implementation
> of windows, by the way.)
I'm familiar with what it was. I never considered it as having windows -
they were more like panes to me.
> The Borland debugger, on the other hand, not only supported overlapping
> windows, but was prettier.
>
> They both beat the hell out of DEBUG.COM.
>
> All that's in the past, though. I do most of my debugging in a
> two-paneled
> console window.
On non Windows systems it's either gdb or dbx straight up (yuck) or
debugging via Emacs which at least shows you the source file as you step
through the code.
The one thing that continually bugs me (get it) about gdb/dbx is that it
forgets any and all breakpoints you've set every time you exit the debugger.
Visual Studio does it right by automatically saving all of your breakpoints
on a per project basis.
I know that you can type your breakpoints into a text file (and have gdb
initialize from that file but it's a clumsy solution. When debugging and
setting a bunch of breakpoints the last thing I think to do is to keep some
text file up to date. There's also the 'save breakpoints' command which I
always remember about 500 mSec after I exit the debugger.
Ezekiel
"Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
news:iiicbu$fn8$1...@news.eternal-september.org...
> Ezekiel wrote this copyrighted missive and expects royalties:
>
>> "JeffM" <jef...@email.com> wrote in message
>> news:c52d6c01-57fd-4e59...@y3g2000vbh.googlegroups.com...
>>> Chris Ahlstrom wrote:
>>>>:Katherine Noyes, PCWorld
>>>>:[...]the Microsoft name
>>>>:[no?] longer signifies anything about quality or innovation[...]
>>>>
>>> Dear Katherine,
>>> It would be good if people making a claim for "Microsoft innovation"
>>> would give a single example of something M$ created
>>
>> <quote>
>> * The Taskbar
>> * Real Time Spell Checking
>> * Rapid Application Development
>> * OLE and COM
>> * Active Desktop
>> * XMLHTTPRequest and AJAX
>> I wonder if Google has ever said "thank you".
>> </quote>
>
> I'd quibble a bit about RAD, but generally a good answer.
There were some things from that article that even I would quibble about.
Which is why I intentionally left those out and only pasted the ones that I
did.
Not that it's a big deal but the one thing that I can't find is who came up
with 'dynamic acceleration' (aka ballistic tracking) for mice. Every mouse
I've owned at one time all had linear tracking. The first mouse that had an
accelerated mode in the driver was my Microsoft "Dove Bar" mouse. But I
can't find who came up with the idea of accelerating mouse movements.
>>
>> For the terminally clueless there isn't very much new/original
>> "innovation"
>> happening in computer science these days. Most of what's being developed
>> today are refinements on ideas that have been around for a while. Most
>> everything is based on some form of "prior art." .NET for example was
>> influenced by Java and other prior languages. Going back much of the Java
>> syntax was influenced by C/C++ and the Java itself was based on the Oak
>> programming language. Same for HTML being a derivation of SGML which in
>> turn
>> is a derivation of GML from the 1960s.
>
> Personally, I would call "innovation" any process that finds good
> solutions
> to old problems, or even solutions that are just much better than old
> solutions. For example, the Boost libraries. Leveraging existing C++
> features in innovative ways, some of which are going to be folded into the
> next C++ standard.
Generally I do too because that's what we and most other software companies
do. That's why I qualified this in my first sentence as "new/original
innovation" because most of it is taking existing ideas and making them
better. It's not to say this is easy or isn't innovation - it's just not
creating new/original ideas.
Example: Our group is currently working on a way to improve table access
with multiple concurrent requests. Say you have 20 gigs of data that's
stored in a file or file(s). You need to process a query that requires a
full table scan so you have no choice but to read every record from that
file. About mid-way into that query a 2nd query arrives that also needs to
do a full table scan from the same file. At this point you basically have
disk thrashing because the large file size prevents disk caching from
helping you in either one of the two requests. Any disk data cached by the
1st reader will be flushed from the cache by the time the 2nd reader gets to
that file position.
This is a simplified example. Scale it up to files that are much larger than
20-gigs and several concurrent requests. We're already disk bound at this
point with just one request. Additional requests cause disk thrashing and
absolutely kill all of our I/O unless we employ request queuing and schedule
the I/O so that the disk reads are synchronized to the point where the OS
disk-caching can do it's thing.
> Time for me to innovate some beer.
I'm about to innovate some wine and watch MI-5.
Homer
> Ezekiel wrote this copyrighted missive and expects royalties:
>> "JeffM" <jef...@email.com> wrote in message
>> news:c52d6c01-57fd-4e59...@y3g2000vbh.googlegroups.com...
>>> Dear Katherine,
>>> It would be good if people making a claim for "Microsoft innovation"
>>> would give a single example of something M$ created
>>
>> <quote>
>> * The Taskbar
Introduced by Acorn Computers in 1987, eight years before the release of
Windows 95:
http://en.wikipedia.org/wiki/Icon_bar
Also, the Amiga had a similar "title bar" two years earlier, but lacked
a task manager:
http://en.wikipedia.org/wiki/Menu_bar#Commodore_Amiga
>> * Real Time Spell Checking
[quote]
In 1987 we released a spelling checker for the Sinclair QL by the name
of Spellbound, this checked your spelling in real time as you typed,
this is a feature that many thought that Microsoft invented in their
recent Microsoft Office software, we were doing this ten years earlier
than Microsoft.
[/quote]
http://www.sectorsoftware.demon.co.uk/history.htm
This was also a feature of Anor's Protext since its first release in
1985:
http://en.wikipedia.org/wiki/Protext_by_Arnor
>> * Rapid Application Development
A term first coined and defined by the independent British IT consultant
and author, James Martin, in 1991:
http://en.wikipedia.org/wiki/James_Martin_%28author%29#Rapid_Application_Development_.28RAD.29
However, the underlying paradigm was actually invented in the 70s by
Xerox PARC:
http://en.wikipedia.org/wiki/Smalltalk
And many development tools using RAD methodology were use years before
1991, dating back as far as early MUD creators. Here's a few early
examples:
http://en.wikipedia.org/wiki/SEUCK
http://en.wikipedia.org/wiki/Demomaker
http://en.wikipedia.org/wiki/3D_Construction_Kit
http://en.wikipedia.org/wiki/MUD#TinyMUD
http://en.wikipedia.org/wiki/Amiga_software#Application_Building_Tools
>> * OLE and COM
Based on Microsoft's Dynamic Data Exchange, first introduced in November
1987 (with the release of Windows 2.0), and preceded by Berkeley
sockets from 1983, four years earlier:
http://en.wikipedia.org/wiki/Berkeley_sockets
Also, Amiga ARexx, using its "ports" process communication system, was
created in early 1987.
http://en.wikipedia.org/wiki/ARexx
>> * Active Desktop
[quote]
The introduction of the Active Desktop marked Microsoft's attempt to
capitalize on the short-lived push technology trend led by PointCast.
Active Desktop placed a number of "channels" on the user's computer
desktop that provided continually-updated information, such as news
headlines and stock quotes, without requiring the user to open a Web
browser.
...
Active Desktop was largely considered to be a failure, with one of the
main problems being its high use of system resources and reduction in
system stability. Although little used, the availability of Active
Desktop was key to Microsoft's legal argument in the United States v.
Microsoft antitrust suit that Internet Explorer was a feature of Windows
rather than a separate product.
[/quote]
http://en.wikipedia.org/wiki/Active_Desktop#History
>> * XMLHTTPRequest and AJAX
Beaten by one year, by Java:
[quote]
Asynchronous loading of content first became practical when Java applets
were introduced in the first version of the Java language in 1995. These
allow compiled client-side code to load data asynchronously from the web
server after a web page is loaded. In 1996, Internet Explorer
introduced the iframe element to HTML, which also enabled asynchronous
loading.[4] In 1999, Microsoft created the XMLHTTP ActiveX control in
Internet Explorer 5, which was later adopted by Mozilla, Safari, Opera
and other browsers as the XMLHttpRequest JavaScript object.
[/quote]
http://en.wikipedia.org/wiki/Ajax_%28programming%29
And Microsoft did /not/ "invent" AJAX, which is actually a "broad group
of web technologies that can be used to implement a web application that
communicates with a server in the background", centring around
JavaScript - which was invented by Netscape in 1995. All Microsoft did
was create the XMLHttpRequest API, which was just a JavaScript method of
doing something that could already be done using Java.
>> Feel free to list all of the original "innovations" (such as
>> "Search") that Google developed or any other open-source company
>> innovated.
Well Mozilla (Netscape) and Sun accounts for much of the above, for
starters.
Face it, Microsoft invents /nothing/, it merely buys, plagiarises or
steals technology from others. Always has, always will.
>> For the terminally clueless there isn't very much new/original
>> "innovation" happening in computer science these days. Most of what's
>> being developed today are refinements on ideas that have been around
>> for a while. Most everything is based on some form of "prior art."
I agree, although the sum total of prior art attributable to Microsoft
is absolute zero.
> Personally, I would call "innovation" any process that finds good
> solutions to old problems, or even solutions that are just much better
> than old solutions.
Well that pretty much discounts everything Microsoft has ever done.
--
K. | Ancient Chinese Proverb:
http://slated.org | "The road to Hell is paved with
Fedora 8 (Werewolf) on sky | ignorant twits who know nothing
kernel 2.6.31.5, up 42 days | about GNU/Linux."
Chris Ahlstrom
> "Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
> news:iiicr0$fn8$2...@news.eternal-september.org...
>>>
>>> http://msdn.microsoft.com/en-us/magazine/dd942846.aspx
>>
>> Actually, with the inability to overlap windows in that crud, that
>> "taskbar"
>> was a necessity.
>
> It's was version 1.0 - you gotta start somewhere.
Then why was GEM on that Atari ST far better than that crap?
Did Microsoft have no shame, release such stuff. I mean, c'mon man!
Look at GEM 1.1
http://en.wikipedia.org/wiki/File:Gem_11_Desktop.png
> On non Windows systems it's either gdb or dbx straight up (yuck) or
> debugging via Emacs which at least shows you the source file as you step
> through the code.
You can also use
- cgdb (vi-like dual pane)
- ddd (primitive looking Motif style GUI, but fairly powerful, with a
graphical display feature that I don't believe you'll find anywhere
else, and it is a usable front end for a lot more than just gdb)
- Red Hats "insight" (http://sourceware.org/insight/)
- TotalView (commercial app)
- Eclipse (cdt plugin)
- Kdevelop
- Kdbg
- Qt Creator
- nemiver (GNOME)
- WinGDB (Visual Studio extension)
> The one thing that continually bugs me (get it) about gdb/dbx is that it
> forgets any and all breakpoints you've set every time you exit the debugger.
Well, then, don't exit the debugger. :-) gdb will automatically reload the
newly-built executable for you.
> Visual Studio does it right by automatically saving all of your breakpoints
> on a per project basis.
>
> I know that you can type your breakpoints into a text file (and have gdb
> initialize from that file but it's a clumsy solution. When debugging and
> setting a bunch of breakpoints the last thing I think to do is to keep some
> text file up to date. There's also the 'save breakpoints' command which I
> always remember about 500 mSec after I exit the debugger.
Use the Source, Zeke! Add that feature!
Naturally, the Python extensions to gdb make these problems trivial to
solve. In this post, I'll show you how to implement a new command to
save breakpoints to a file. This is a feature I've often wanted, and
which, strangely, has never been written.
It does now, though, as you note.
A useful command is "finish", to finish the current function and step out of
it. You can also "return" from the function, which cancels the execution of
the rest of it, but then steps out of the function, and returns an option
return value.
Also useful is "until", which save you from setting a breakpoint you
need only one time.
Instead of a "continue" of execution, you can tell gdb to "jump" (restart
execution) somewhere else.
Finally, there are a number of "reverse" commands for stepping, continuing,
and finishing backwards; breakpoints will work in reverse.
--
I'm having an emotional outburst!!
Chris Ahlstrom
> Verily I say unto thee, that Chris Ahlstrom spake thusly:
>
>> Personally, I would call "innovation" any process that finds good
>> solutions to old problems, or even solutions that are just much better
>> than old solutions.
>
> Well that pretty much discounts everything Microsoft has ever done.
I don't agree. However, I will say I've always found Microsoft's coding
styles and conventions to be a bit off-kilter; probably because I learned
programming well before Microsoft came to dominate the computing
environments I found myself in.
C#, in particular, is a mix of cool stuff and oddities. For example,
classes are "reference" types, while structures are "value" types, with
generally no visible clue to which kind an object is. For a C++ guy, used
to seeing the "&" that denotes a reference, this is a bit disconcerting.
Plus it leads to "boxing" and "unboxing" under the covers, which will
reduce the speed of your code unless you're careful.
All these languages seem to fix up some things that are "bad" about C++,
while sometimes losing some of the "good" things about C++. Lambda
functions have good notational support in C#, but you can use Boost Lambda
to do similar stuff in C++, without such nice notation.
--
The sweeter the apple, the blacker the core --
Scratch a lover and find a foe!
-- Dorothy Parker, "Ballad of a Great Weariness"
Chris Ahlstrom
> "Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
> news:iiicr0$fn8$2...@news.eternal-september.org...
>>>
>>> http://msdn.microsoft.com/en-us/magazine/dd942846.aspx
>>
>> Actually, with the inability to overlap windows in that crud, that
>> "taskbar"
>> was a necessity.
>
> It's was version 1.0 - you gotta start somewhere.
Then why was GEM on that Atari ST far better than that crap?
Did Microsoft have no shame, releasing such stuff. I mean, c'mon man!
Look at GEM 1.1
http://en.wikipedia.org/wiki/File:Gem_11_Desktop.png
> On non Windows systems it's either gdb or dbx straight up (yuck) or
> debugging via Emacs which at least shows you the source file as you step
> through the code.
You can also use
- cgdb (vi-like dual pane)
- ddd (primitive looking Motif style GUI, but fairly powerful, with a
graphical display feature that I don't believe you'll find anywhere
else, and it is a usable front end for a lot more than just gdb)
- Red Hats "insight" (http://sourceware.org/insight/)
- TotalView (commercial app)
- Eclipse (cdt plugin)
- Kdevelop
- Kdbg
- Qt Creator
- nemiver (GNOME)
- WinGDB (Visual Studio extension)
- NetBeans
> The one thing that continually bugs me (get it) about gdb/dbx is that it
> forgets any and all breakpoints you've set every time you exit the debugger.
Well, then, don't exit the debugger. :-) gdb will automatically reload the
newly-built executable for you.
> Visual Studio does it right by automatically saving all of your breakpoints
> on a per project basis.
>
> I know that you can type your breakpoints into a text file (and have gdb
> initialize from that file but it's a clumsy solution. When debugging and
> setting a bunch of breakpoints the last thing I think to do is to keep some
> text file up to date. There's also the 'save breakpoints' command which I
> always remember about 500 mSec after I exit the debugger.
Use the Source, Zeke! Add that feature!
Ezekiel
"Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
news:iijlfv$hd3$1...@news.eternal-september.org...
> Ezekiel wrote this copyrighted missive and expects royalties:
>
>> "Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
>> news:iiicr0$fn8$2...@news.eternal-september.org...
>>>>
>>>> http://msdn.microsoft.com/en-us/magazine/dd942846.aspx
>>>
>>> Actually, with the inability to overlap windows in that crud, that
>>> "taskbar"
>>> was a necessity.
>>
>> It's was version 1.0 - you gotta start somewhere.
>
> Then why was GEM on that Atari ST far better than that crap?
>
> Did Microsoft have no shame, releasing such stuff. I mean, c'mon man!
> Look at GEM 1.1
>
> http://en.wikipedia.org/wiki/File:Gem_11_Desktop.png
I didn't use Atari or GEM so this certainly is not from memory. But from
what I found GEM 1.1 isn't exactly an initial release. It seems to have been
based on GSX which was already around for a while.
>> On non Windows systems it's either gdb or dbx straight up (yuck) or
>> debugging via Emacs which at least shows you the source file as you step
>> through the code.
>
> You can also use
>
> - cgdb (vi-like dual pane)
> - ddd (primitive looking Motif style GUI, but fairly powerful, with a
> graphical display feature that I don't believe you'll find anywhere
> else, and it is a usable front end for a lot more than just gdb)
> - Red Hats "insight" (http://sourceware.org/insight/)
> - TotalView (commercial app)
> - Eclipse (cdt plugin)
> - Kdevelop
> - Kdbg
> - Qt Creator
> - nemiver (GNOME)
> - WinGDB (Visual Studio extension)
> - NetBeans
Unless all of these will also debug C/C++ code on AIX, HPUX, Solaris and zOS
then I don't plan on using any of them. I don't plan on using a different
debugger on every platform.
>> The one thing that continually bugs me (get it) about gdb/dbx is that it
>> forgets any and all breakpoints you've set every time you exit the
>> debugger.
>
> Well, then, don't exit the debugger. :-) gdb will automatically reload
> the
> newly-built executable for you.
It's not a matter of simply debugging one executable somewhere. It would be
nice if it worked like that but it doesn't. In most cases it's a daemon
process that forks off a script which then runs a series of apps. (These
scripts get generated dynamically.) Somewhere in that chain of apps that
gets run is the app that I need to attach to. The normal debugging
technique used (by everyone - not just me) is to add a dummy loop to where
you want to attach the debugger. Once the process starts spinning you attach
with the debugger, break out of the loop, and you're now in the code.
>> Visual Studio does it right by automatically saving all of your
>> breakpoints
>> on a per project basis.
>>
>> I know that you can type your breakpoints into a text file (and have gdb
>> initialize from that file but it's a clumsy solution. When debugging and
>> setting a bunch of breakpoints the last thing I think to do is to keep
>> some
>> text file up to date. There's also the 'save breakpoints' command which I
>> always remember about 500 mSec after I exit the debugger.
>
> Use the Source, Zeke! Add that feature!
No thanks. It's annoying but I'm not that intent on fixing it.
> http://tromey.com/blog/?p=501
>
> Naturally, the Python extensions to gdb make these problems trivial to
> solve. In this post, I'll show you how to implement a new command to
> save breakpoints to a file. This is a feature I've often wanted, and
> which, strangely, has never been written.
>
> It does now, though, as you note.
>
> A useful command is "finish", to finish the current function and step out
> of
> it. You can also "return" from the function, which cancels the execution
> of
> the rest of it, but then steps out of the function, and returns an option
> return value.
>
> Also useful is "until", which save you from setting a breakpoint you
> need only one time.
>
> Instead of a "continue" of execution, you can tell gdb to "jump" (restart
> execution) somewhere else.
>
> Finally, there are a number of "reverse" commands for stepping,
> continuing,
> and finishing backwards; breakpoints will work in reverse.
I use many of these (finish, return, etc) but haven't used the 'reverse'
commands. Debugging in the forward direction is enough fun for me!
Chris Ahlstrom
> "Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
> news:iijlfv$hd3$1...@news.eternal-september.org...
>>
>> You can also use
>>
>> - cgdb (vi-like dual pane)
>> - ddd (primitive looking Motif style GUI, but fairly powerful, with a
>> graphical display feature that I don't believe you'll find anywhere
>> else, and it is a usable front end for a lot more than just gdb)
>> - Red Hats "insight" (http://sourceware.org/insight/)
>> - TotalView (commercial app)
>> - Eclipse (cdt plugin)
>> - Kdevelop
>> - Kdbg
>> - Qt Creator
>> - nemiver (GNOME)
>> - WinGDB (Visual Studio extension)
>> - NetBeans
>
> Unless all of these will also debug C/C++ code on AIX, HPUX, Solaris and zOS
> then I don't plan on using any of them. I don't plan on using a different
> debugger on every platform.
Those are all front-ends to the same debugger, gdb.
>> Well, then, don't exit the debugger. :-) gdb will automatically reload
>> the newly-built executable for you.
>
> It's not a matter of simply debugging one executable somewhere. It would be
> nice if it worked like that but it doesn't. In most cases it's a daemon
> process that forks off a script which then runs a series of apps. (These
> scripts get generated dynamically.) Somewhere in that chain of apps that
> gets run is the app that I need to attach to. The normal debugging
> technique used (by everyone - not just me) is to add a dummy loop to where
> you want to attach the debugger. Once the process starts spinning you attach
> with the debugger, break out of the loop, and you're now in the code.
Okay, gotcha.
>> Use the Source, Zeke! Add that feature!
>
> No thanks. It's annoying but I'm not that intent on fixing it.
:-)
> I use many of these (finish, return, etc) but haven't used the 'reverse'
> commands. Debugging in the forward direction is enough fun for me!
:-)
--
There's no real need to do housework -- after four years it doesn't get
any worse.
amicus_curious
"JeffM" <jef...@email.com> wrote in message
news:c52d6c01-57fd-4e59...@y3g2000vbh.googlegroups.com...
I think you all continue to miss the essence of the situation. Microsoft is
incredibly successful due to an exquisite sense of timing for brining things
to the mass market. Babble all you want about "innovation", while confusing
it with "invention". The fact remains that Microsoft is the best that ever
was at selling prepackaged computer software to the world. You focus on
bits and bytes and only prove that you are a dweeb who cannot see the bigger
picture.
Tom Shelton
> Homer wrote this copyrighted missive and expects royalties:
>
>> Verily I say unto thee, that Chris Ahlstrom spake thusly:
>>
>>> Personally, I would call "innovation" any process that finds good
>>> solutions to old problems, or even solutions that are just much better
>>> than old solutions.
>>
>> Well that pretty much discounts everything Microsoft has ever done.
>
> I don't agree. However, I will say I've always found Microsoft's coding
> styles and conventions to be a bit off-kilter; probably because I learned
> programming well before Microsoft came to dominate the computing
> environments I found myself in.
>
> C#, in particular, is a mix of cool stuff and oddities. For example,
> classes are "reference" types, while structures are "value" types, with
> generally no visible clue to which kind an object is. For a C++ guy, used
> to seeing the "&" that denotes a reference, this is a bit disconcerting.
> Plus it leads to "boxing" and "unboxing" under the covers, which will
> reduce the speed of your code unless you're careful.
>
The Reference type vs Value type dichotomy can be a bit confusing at
first. It's partly the result of having a single rooted inheritance
and trying to solve the hole primitive vs class problem that you had in
Java... (int vs Integer for example).
Here's the deal - generally, anything you create should probably be a
class :) I know that's a huge over generalization.... But, I find it
useful. The main purpose for a structure in a .NET world - is
basically p/invoke or those occasional times you want soemthig to have
value type symantics - which really isn't all that often... Oh, and
you can avoid a lot of boxing/unboxing issues by using Generics.
It's not like there aren't confusing concepts in every language :) I
think C# (and .net in general) does a good job of keeping them to a
minimum.
> All these languages seem to fix up some things that are "bad" about C++,
> while sometimes losing some of the "good" things about C++. Lambda
> functions have good notational support in C#, but you can use Boost Lambda
> to do similar stuff in C++, without such nice notation.
Hmmm, your example doesn't seem to refelect your previous statement...
How did C#'s notational support take anything away from C++ (I love
lambda's by the way!).
What specifically do you think is a "good" thing you loose from C++?
--
Tom Shelton
JEDIDIAH
>
>
> "Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
> news:iijlfv$hd3$1...@news.eternal-september.org...
>> Ezekiel wrote this copyrighted missive and expects royalties:
>>
>>> "Chris Ahlstrom" <ahls...@xzoozy.com> wrote in message
>>> news:iiicr0$fn8$2...@news.eternal-september.org...
>>>>>
>>>>> http://msdn.microsoft.com/en-us/magazine/dd942846.aspx
>>>>
>>>> Actually, with the inability to overlap windows in that crud, that
>>>> "taskbar"
>>>> was a necessity.
>>>
>>> It's was version 1.0 - you gotta start somewhere.
>>
>> Then why was GEM on that Atari ST far better than that crap?
>>
>> Did Microsoft have no shame, releasing such stuff. I mean, c'mon man!
>> Look at GEM 1.1
>>
>> http://en.wikipedia.org/wiki/File:Gem_11_Desktop.png
>
> I didn't use Atari or GEM so this certainly is not from memory. But from
> what I found GEM 1.1 isn't exactly an initial release. It seems to have been
> based on GSX which was already around for a while.
GSX is just the graphics primitives. Although if it had already "been
around for awhile" even in 1984 or 1985 then that says quite a bit about
Digital Research.
[deletia]
--
On the subject of kilobyte being "redefined" to mean 1000 bytes...
When I was a wee lad, I was taught that SI units were |||
meant to be computationally convenient rather than just / | \
arbitrarily assigned.
Chris Ahlstrom
> on 2/5/2011, Chris Ahlstrom supposed :
>
>> All these languages seem to fix up some things that are "bad" about C++,
>> while sometimes losing some of the "good" things about C++. Lambda
>> functions have good notational support in C#, but you can use Boost Lambda
>> to do similar stuff in C++, without such nice notation.
>
> Hmmm, your example doesn't seem to refelect your previous statement...
> How did C#'s notational support take anything away from C++ (I love
> lambda's by the way!).
>
> What specifically do you think is a "good" thing you loose from C++?
Flexibility, I think.
And, ignoring language semantics, a committee of vendors and developers to
grow the language in a controlled and fairly standard fashion.
--
Economists state their GNP growth projections to the nearest tenth of a
percentage point to prove they have a sense of humor.
-- Edgar R. Fiedler
Tom Shelton
> Tom Shelton wrote this copyrighted missive and expects royalties:
>
>> on 2/5/2011, Chris Ahlstrom supposed :
>>
>>> All these languages seem to fix up some things that are "bad" about C++,
>>> while sometimes losing some of the "good" things about C++. Lambda
>>> functions have good notational support in C#, but you can use Boost Lambda
>>> to do similar stuff in C++, without such nice notation.
>>
>> Hmmm, your example doesn't seem to refelect your previous statement...
>> How did C#'s notational support take anything away from C++ (I love
>> lambda's by the way!).
>>
>> What specifically do you think is a "good" thing you loose from C++?
>
> Flexibility, I think.
>
I'm not sure I see that. Do you have a specific example?
> And, ignoring language semantics, a committee of vendors and developers to
> grow the language in a controlled and fairly standard fashion.
Hmmm... Ok. I'm not going to get in a debate over this one - C++ is
obviously worked on by a larger group. But, I will say, you realize
that C# is an ISO standard with a working group, composed of more than
just MS right?
--
Tom Shelton
Ezekiel
>
>"Chris Ahlstrom" wrote in message
>news:iik6gu$9nt$2...@news.eternal-september.org...
>
>Tom Shelton wrote this copyrighted missive and expects royalties:
>
>> on 2/5/2011, Chris Ahlstrom supposed :
>>
>>> All these languages seem to fix up some things that are "bad" about C++,
>>> while sometimes losing some of the "good" things about C++. Lambda
>>> functions have good notational support in C#, but you can use Boost
>>> Lambda
>>> to do similar stuff in C++, without such nice notation.
>>
>> Hmmm, your example doesn't seem to refelect your previous statement...
>> How did C#'s notational support take anything away from C++ (I love
>> lambda's by the way!).
>>
>> What specifically do you think is a "good" thing you loose from C++?
>
>Flexibility, I think.
And assembler is more flexible than C/C++. It's all a trade-off between how
much flexibility and fine control you want versus other features. For me and
a lot of application development C/C++ gives me a good balance of the two.
But other "code" is best done with shell-scripts or Python.
>And, ignoring language semantics, a committee of vendors and developers to
>grow the language in a controlled and fairly standard fashion.
It could also be argued that having large committees that move at a snails
pace isn't an ideal solution either. Even with formal committees things like
SQL are far from standardized. It also seems that changes/progress for
committee controlled things takes too long but vendor controlled things
change too fast.
Snit
Snit:
-----
... many in COLA complain about MS not innovating.
-----
JEDIDIAH:
-----
No one does that.
-----
Interesting how JEDIDIAH does not read Homer's posts.
--
[INSERT .SIG HERE]
Ezekiel
>
>"Chris Ahlstrom" wrote in message
>
>Ezekiel wrote this copyrighted missive and expects royalties:
>
>
>>
>> No thanks. It's annoying but I'm not that intent on fixing it.
>
>:-)
It wouldn't be a difficult change. I just don't feel like dealing with it.
If it were only my machine then that's one thing. But at work (even at home)
there's so many machines/platforms that it's easier just to live with it.
If I were going to change one thing (even easier for several reasons) I
would change the 'Places->Connect To Server' app on Ubuntu. It always
defaults to FTP which I use 1% of the time instead of Windows/Samba share
which I use almost all the time.
Chris Ahlstrom
> Chris Ahlstrom presented the following explanation :
>
>> Flexibility, I think.
>
> I'm not sure I see that. Do you have a specific example?
No, and in fact it is just my early opinion. It would take a long time to
develop a point-by-point comparison.
>> And, ignoring language semantics, a committee of vendors and developers to
>> grow the language in a controlled and fairly standard fashion.
>
> Hmmm... Ok. I'm not going to get in a debate over this one - C++ is
> obviously worked on by a larger group. But, I will say, you realize
> that C# is an ISO standard with a working group, composed of more than
> just MS right?
Sure.
But all I know is that, like Java, C# development is the pet of a single
company, and a powerful one, and that the pace of change of C# is akin to
that of, say, Perl and Python. The relative stability of C/C++ is definite
benefit of those languages.
Another benefit of C/C++ is that a lot of features are not part of the
language, but of libraries. That, of course, is my opinion, and many do not
agree.
Heh heh:
--
Don't worry about people stealing your ideas. If your ideas are any good,
you'll have to ram them down people's throats.
-- Howard Aiken
Chris Ahlstrom
>>"Chris Ahlstrom" wrote in message
>>news:iik6gu$9nt$2...@news.eternal-september.org...
>>
>>Flexibility, I think.
>
> And assembler is more flexible than C/C++. It's all a trade-off between how
> much flexibility and fine control you want versus other features. For me and
> a lot of application development C/C++ gives me a good balance of the two.
> But other "code" is best done with shell-scripts or Python.
I sort of believe that. And yet I feel most productive in C++, even where
its syntax is not as simple as, say Python.
>>And, ignoring language semantics, a committee of vendors and developers to
>>grow the language in a controlled and fairly standard fashion.
>
> It could also be argued that having large committees that move at a snails
> pace isn't an ideal solution either. Even with formal committees things like
> SQL are far from standardized. It also seems that changes/progress for
> committee controlled things takes too long but vendor controlled things
> change too fast.
I quite agree.
--
Women give to men the very gold of their lives. Possibly; but they
invariably want it back in such very small change.
-- Oscar Wilde
Tom Shelton
> Tom Shelton wrote this copyrighted missive and expects royalties:
>
>> Chris Ahlstrom presented the following explanation :
>>
>>> Flexibility, I think.
>>
>> I'm not sure I see that. Do you have a specific example?
>
> No, and in fact it is just my early opinion. It would take a long time to
> develop a point-by-point comparison.
>
Ok... I can live with that :) I won't say there aren't differences -
but, since moving to C# I can't say I've ever found anything I needed
to do that I couldn't :)
>>> And, ignoring language semantics, a committee of vendors and developers to
>>> grow the language in a controlled and fairly standard fashion.
>>
>> Hmmm... Ok. I'm not going to get in a debate over this one - C++ is
>> obviously worked on by a larger group. But, I will say, you realize
>> that C# is an ISO standard with a working group, composed of more than
>> just MS right?
>
> Sure.
>
> But all I know is that, like Java, C# development is the pet of a single
> company, and a powerful one, and that the pace of change of C# is akin to
> that of, say, Perl and Python. The relative stability of C/C++ is definite
> benefit of those languages.
>
C# has progress rather rappidly, I won't deny that. BUT, the good news
is that all versions are backward compatible. MS has been very careful
not to break old code by adding new language constructs.
> Another benefit of C/C++ is that a lot of features are not part of the
> language, but of libraries. That, of course, is my opinion, and many do not
> agree.
Well, I actually do agree with that - and this is actually mostly true
of C# as well (i say mostly, because lambda's and linq sort of blur the
line a bit).
--
Tom Shelton
Chris Ahlstrom
> Chris Ahlstrom explained on 2/5/2011 :
>> Tom Shelton wrote this copyrighted missive and expects royalties:
>>
>>> Chris Ahlstrom presented the following explanation :
>>>
>>>> Flexibility, I think.
>>>
>>> I'm not sure I see that. Do you have a specific example?
>>
>> No, and in fact it is just my early opinion. It would take a long time to
>> develop a point-by-point comparison.
>
> Ok... I can live with that :) I won't say there aren't differences -
> but, since moving to C# I can't say I've ever found anything I needed
> to do that I couldn't :)
I'm quite sure of that. C# and the .NET API seem extremely comprehensive.
Almost too many features!
>> But all I know is that, like Java, C# development is the pet of a single
>> company, and a powerful one, and that the pace of change of C# is akin to
>> that of, say, Perl and Python. The relative stability of C/C++ is definite
>> benefit of those languages.
>
> C# has progress rather rappidly, I won't deny that. BUT, the good news
> is that all versions are backward compatible. MS has been very careful
> not to break old code by adding new language constructs.
>
>> Another benefit of C/C++ is that a lot of features are not part of the
>> language, but of libraries. That, of course, is my opinion, and many do not
>> agree.
>
> Well, I actually do agree with that - and this is actually mostly true
> of C# as well (i say mostly, because lambda's and linq sort of blur the
> line a bit).
Yeah, one of the interesting things abou C# was that some of the language
features are actually converted to .NET calls under the covers.
The coolest thing I've read about C# so far is the "fluent syntax" for
database queries.
--
... there are about 5,000 people who are part of that committee. These guys
have a hard time sorting out what day to meet, and whether to eat croissants
or doughnuts for breakfast -- let alone how to define how all these complex
layers that are going to be agreed upon.
-- Craig Burton of Novell, Network World