I followed a few more of your links to the mailing lists, and I must say it seems to be pretty much unsubstantiated whining.
I'm sitting here in front of a Gnome 2 desktop running on SuSE 8.0. My bottom screen is taken up with a full-width panel, and I've removed all the other panels that were there by default. Took me a whole 15 minutes when I first switched over to Gnome from KDE, without actually reading any manuals.
The only legitimate gripe I found was the complaint about having the desktop be the desktop, instead of a Nautilus-based desktop. Personally I like the utility of a managed desktop, but I could see a few people wanting to just have the toolbar panels on their root window. Maybe they should be looking at less feature-rich environments than Gnome, rather than expecting the 90% population to adjust to their 10% wishlist.
If you were trying to be serious, all I can think to say is "Wah!".
No matter what changes are made to a system, there will always be those who object for various reasons. I followed the link to gconf and a couple others, and it seems to me most of the griping is about features that have a lot of utility.
Personally I'd like to see gconf use XML under the hood, but I haven't looked at the details of the implementation, and the whole intent of gconf is that I shouldn't need to look at the details!
You may know more about the history of Gnome than I, but I know I like what I'm using right now (Gnome 2), regardless of the history that got it here.
The cost of QT is per developer, so in order to have their customer's application developers use QT, they'd have to include a QT license with the distribution of Solaris development tools. Not cheap. Not cheap at all.
OpenWin was intended to run with DisplayPostscript, and did so very nicely. When the Unix standards wars and POSIX were ongoing, CDE was selected as the standard from various vendors contributions (components of HP's ToolTalk, Motif, etc.)
I've never run into anyone who thought CDE was better than OpenWin, but that's what was selected as the standard, and that's what Sun provided. If they hadn't, they would have been locked out of a lot of important markets.
It's not like there is a "constant change of GUIs" as you indicate. OpenWin was the Sun standard from about 1987 (not sure) until around 1990-1995, when CDE was spec'd. Now they're shifting to Gnome.
Note that all the way through, applications continued to run with the different desktop managers. Or were you under the impression that different versions of apps were running for different desktop managers?
What kind of licenses exist for Qt?
The Qt toolkit is available under two different licenses: The Professional and Enterprise Editions for commercial use on all platforms, and the Free Edition for developing free/open source software for the X11 platform.
For those thinking to develop with the free edition, then just buy a license when they're ready to deploy:
Can we use the Free Edition while developing our non-free application and then purchase commercial licenses when we start to sell it?
No. The Free Edition license applies to the development phase - anything developed without Professional or Enterprise Edition licenses must be released as free/open source software.
The price is very reasonable for the functionality, but I only have so much money to spend on tools, and I'm not willing to plunk down the coin now just in case I need to be able to use my code commercially (i.e. to support a client site.)
Last time I checked (almost a year ago), QT for Win32 was several thousand dollars, a far cry from 8 billable hours. Not that the price was/is unreasonable compared to similar products (such as the now-defunct Neuron Data Open Interface, which ran around $10K/developer.)
I agree. I was a die-hard KDE user from 1.0-3.0, periodically trying Gnome along the way, but always disappointed by the lack of stability. After trying Gnome again under SuSE 8.0, I finally found it stable enough to use daily (and do so.)
Some people argue about performance and resources, but they're both pretty bloated compared to simple window managers like CDE/Motif.
From a programming perspective, QT/KDE are nicer products, but Gnome is catching up rapidly on the GUI designer front (Glade.) I don't really use IDEs, so I can't comment on KDeveloper vs. Anjuta, but both look to be pretty full-featured on the surface.
Re:Over 10 years after DEC introduced Alpha ....
on
AMD's 64-bit Plot
·
· Score: 2
IIRC, PA-RISC was 40 or 48 bit addressing with a 32-bit instruction set at the time that Alpha came out. Alpha was out well before 64-bit MIPS processors (which were not DEC processors, BTW.) Alpha was also 64-bit before a refresh of the SPARC design brought it to 64 bit.
Sun has been babbling about the switch to Gnome from CDE for almost two years now. I use both KDE and Gnome, and both are far more a "desktop" than CDE ever was.
It also confirms my decision to use GTK for GUI development under Linux (I love QT's APIs and structure under KDE, but GTK lets me port to Win32 clients without cost.)
I gave it a good 5-6 episodes to turn into something worth watching, but had given up. It's a piece of drek, and good riddance. I can't believe they axed something entertaining like Dark Angel for a lame implementation of Rodenberry's "Wagon Train to the Stars" pitch line for Star Trek.
What really baffles me is how the show can possibly require the budget it has. The visual F/X are nothing special, rarely extending beyond CGI-only external shots. There has been nothing that would require extensive makeup, prosthesis, animatronics, or anything else to justify the cost of this series.
Then again, maybe they had to pay the "artists" who created the theme enough to live on for the rest of their lives -- that track is a career-killer if I've ever heard one!
Then maybe you just shouldn't be doing such "complex" programming. XSD should not take more than a couple days to understand, another few days to iron out the details, and by the end of 30-40 hours you should have produced a good sample of whatever you're modelling with XSD.
Too many "how to" documents and books are written for people who just don't "get it". While I sympathise that they need to learn, you have to grasp the concept of abstraction before some of these "languages" make sense. XSD, ERDs, SQL-DDL, etc. are just different ways of describing data structures and organization. Each is "complex" if you try to make it do more than it was intended for, but only takes a couple weeks to understand well enough to get by.
I tend to use mine for todo lists of offbeat items, like lists of books, CDs, DVDs, etc. that I want to pick up when convenient (i.e. I happen to be in a store, prices are good, and I have a few dollars to spare.)
It also lets me use the company categories on the address book to break down the list a bit. It's handy to be able to see the contact points for an active client site.
I tried using the todo lists to manage project/job tasks, but found it was too much hassle. I usually have some other todo facility for projects, whether it be a project management utility, a project status spreadsheet, or just a todo.html file in the project directory tree.
Other than that I installed some games I never play, an improved calculator that I rarely use, and just stopped trying to find "new uses" for it.
It's an address book and meeting reminder tool, with bonus reminder/todo lists. It's enough to be worth carrying around more often than a pager or a cell phone.
Some people like to be available for contact anytime anywhere (pager/cell-phone.) I prefer to maintain a personal life unless I'm providing formal offsite support to a client site. Leave a vmail or email message, and I'll look into it when I get back.
The technical concepts have been in use for decades, provided that you don't myopically focus on "console gaming" as the definition of platform.
Go wayyyyyy back to the days of the "dumb" terminal, the mid-80s. The "hot" video games such as Star Wars, Tempest, and their ilk had pathetic processors, but hardware vector graphics engines (i.e. line-drawing only) that would have required CPUs 10-20 times as powerful to render the same data to a raster display. "Intelligent" display terminals used for CAD/CAM often had at least as much raw CPU power as the monstrous VAX midframes they talked to.
Over the decades, the balance of power has constantly shifted between the processing center (CPU or data center) and the presentation/client (smart TTYs, X-terminals, fat clients, thin clients, browsers, browsers with plugins, java download clients, etc.)
You'd be amazed how much of modern graphics and CPU technology was actually theorized as far back as the 60s or 70s, but just could not be implemented with the technology of the day. Some of the algorithms discussed in my 400-series graphics class in the late 80's were only "theoretical" algorithms because even a Cray would have taken a couple days to calculate a single frame using them. Now we take it for granted and argue about which is "better", myopically ignoring all which came before us.
Case in point: Darwin Peachey was one of the graphics grads/profs at the University of Saskatchewan when I graduated. He left that year to work for a little startup you might have heard of: Pixar. Most of what he'd studied was research, and he saw Pixar as a chance to work with a bleeding-edge company that wanted to make the theory real. To the public their work is "new", but to their staff it's something they've been working on for 15-20 years, and far from new!
But does it squeal like an outraged pig after falling down the stairs like ED-209? Is it smart enough to crush it's remote after you've armed it? And most importantly, does it run Linux? (After all, if it's another WinCrap partnership you just know it's just going to stare at you with lifeless blue eyes blinking "BSOD" when you need it most!)
I've often had to resort to registry shredding to get things moved around. Just open up regedit and do a global search for subdirectory of the app you're trying to move. Similarly, you can search/substitute X: values with Y: values if you're shifting partitions around.
Obviously this isn't something most users know how to do (or should do!), but it can be done.
My bigger beef with the registry is that apps scatter pieces of their initialization all over the place. It makes it impossible to backup and restore individual apps by saving their registry tree, because you usually find that there are critical pieces of information whose names bear no resemblance to the application.
The 450/500 GBP prices (roughly 704/780 USD) are just for the mobo with CPU, not a complete system. Assume another $20 for shipping (which would be cheap!), and you're looking at $800USD just for the mobo. You still need to add memory, a case, video card, HDD, CD[-RW]/DVD[-+RW], keyboard, mouse, and monitor.
Lets assume for sake of argument you're going cheap, cheap, cheap, so:
40 - memory
50 - case
25 - video card
50 - 40G HDD
25 - CD-R
15 - keyboard
15 - mouse
100 - 15" monitor
You're now running $320 in basic components, bringing the price up to $1045-1120 (700/800MHz variants.) Or you can get an eMac for $1100 (700MHz) that upgrades you to a CD-RW with a better video card, modem, and FireWire port that is pre-installed with a currently shipping copy of OSX. Alternatively, $1300 gets you an iBook with a 12.1" screen (slightly smaller 30GB HDD.)
Having decided to buy the AmigaOne mobo anyhow, you now have the option of running PPC Linux or waiting for the new OS. Either way, you miss out on the commercial product support for Linux (DB/2, Oracle, Sybase, et. al. are x86 binaries, not PPC.) Assuming pure open source is just fine by you, you've still got a box that is woefully underpowered to a similarly priced/configured AMD system (and maybe even Intel P4.)
Much as I loved my Amiga 1000, I just can't see any reason I'd want one of these new "Amiga" systems. Most of the reasons I loved my A1000 just aren't valid anymore -- everyone has hardware accelerated video and audio now, video capture and processing cards are common, and I'd rather be coding *nix than a system with no mind/market share.
Font anti-aliasing is completely useless when the monitor itself is blurry. Even so, I'd much rather use extreme resolutions with large fonts, as that has the same end result as anti-aliasing. A "good" monitor effectively blends adjacent pixels when fed resolutions higher than the physical display can support, which is exactly what anti-aliasing is trying to do via software. For example, most 17" monitors have a maximum physical display of 1280x1024. Feed them a 1600x1280 signal, and you get hardware "aliasing".
The typical office environment with modular office cube is just about the worst ergonomic environment there is. The cube itself has several problems:
Placing the monitor in the corner means the desk wings are constantly in the way of elbows.
Desk height is non-adjustable, so it's either too high or too low for the majority of people.
Keyboard trays (mounted under the desk) take up enough space that anyone over 5' 10" is going to get whacked in the knees if they try to use it.
These units were designed before mice were common, and have no support for them on the keyboard trays.
The overhead shelves/bins prevent raising the monitor to an ergonomic height.
Add to that issues of non-adjustable chairs, cheap flat keyboards (vs. ergonomic designs), poor mouse designs, and you've got a guarantee of neck, shoulder, and back problems before you even turn on the monitor.
Turning to the monitors, it's usually not a simple issue of LCD vs. CRT, but of monitor quality. Most cube-lands are full of poor quality monitors that came with pre-negotiated system bundles, often meaning they were outdated before the supply contract was first signed.
As a consultant, I have spent at least half my career stuck in front of blurry, non-adjustable, cheap monitors that I wouldn't even consider for a kids computer, much less someone who has to spend hours a day on it. While far from the worst I've dealt with, my current client environment consists of 5+ year old 17" monitors with poor color, poor contrast, glare problems, low refresh rates, and focus problems (the monitor I'm saddled with can't handle more than 1280x1024 at 60Hz NI. While the resolution is theoretically good enough, refresh rates under 75Hz cause serious eye strain.)
TFT LCD displays would help the eye strain, but so would high quality CRT monitors (e.g. Sony G420 19", Hitachi CM715 19", et. al.) The problem is that companies are not going to pay for replacement monitors because they can get entire system bundles with cheap monitors for very little more.
MKS Toolkit does a terrific job of integrating a Unix ksh-based CLI with WinNT/2K/XP, and couples closely with IBM DB/2 so that you can run DB/2 UDB scripts as you can under AIX. It is expensive, but worth it if you can afford it.
Qt is a very, very nice cross-platform development library. Another option you could look at used to be Neuron Data's Elements Environment, but they renamed the company (www.blazesoft.com) and I don't know if they still sell EE as a seperate product (Advisor is a repackaging of their rule-base software, which was built on EE.)
There are also the ever-present Rogue Wave class libraries, but I don't think they'll address your GUI requirements. However, if you use it to split out the core application functionality from the presentation (GUI), it might be helpful.
A solid set of macros with compiler/platform detection directives can help a great deal for porting code, though many people prefer to use sed or perl scripts instead (ala config.) Macros have the advantage of dealing with portability more consistently, and localize the changes for platforms to the headers and migration binding code (usually done as a base library.)
Using cross-platform libraries such as xml.apache.org libraries, IBM's ICU (Unicode support), et. al. can also make your code much more portable without requiring extra work after the initial coding.
There are also various open source projects that provide portable thread libraries, portable GUI toolkits, etc.
... ClearType...requires a set pixel layout, which traditional CRTs don't have.
Actually, I prefer ClearType on my Sony monitor, but not on my "regular" monitor at work. Trinitron tubes use square/rectangular phosphor cells instead of dots. Seems to me that trinitron tubes might be close enough to a "set pixel layout" to have the same effect.
Anyone else out there who can compare tube/gun types?
Slashdot Mirror
I followed a few more of your links to the mailing lists, and I must say it seems to be pretty much unsubstantiated whining.
I'm sitting here in front of a Gnome 2 desktop running on SuSE 8.0. My bottom screen is taken up with a full-width panel, and I've removed all the other panels that were there by default. Took me a whole 15 minutes when I first switched over to Gnome from KDE, without actually reading any manuals.
The only legitimate gripe I found was the complaint about having the desktop be the desktop, instead of a Nautilus-based desktop. Personally I like the utility of a managed desktop, but I could see a few people wanting to just have the toolbar panels on their root window. Maybe they should be looking at less feature-rich environments than Gnome, rather than expecting the 90% population to adjust to their 10% wishlist.
If you were trying to be serious, all I can think to say is "Wah!".
No matter what changes are made to a system, there will always be those who object for various reasons. I followed the link to gconf and a couple others, and it seems to me most of the griping is about features that have a lot of utility.
Personally I'd like to see gconf use XML under the hood, but I haven't looked at the details of the implementation, and the whole intent of gconf is that I shouldn't need to look at the details!
You may know more about the history of Gnome than I, but I know I like what I'm using right now (Gnome 2), regardless of the history that got it here.
The free version of QT for Win32 is outdated.
If you really edited this mess that many times, surely you could have figured out how to capitalize the text.
I wonder how many people realize your post is a sarcastic parody? (Surely it can't be intended as serious!)
The cost of QT is per developer, so in order to have their customer's application developers use QT, they'd have to include a QT license with the distribution of Solaris development tools. Not cheap. Not cheap at all.
OpenWin was intended to run with DisplayPostscript, and did so very nicely. When the Unix standards wars and POSIX were ongoing, CDE was selected as the standard from various vendors contributions (components of HP's ToolTalk, Motif, etc.)
I've never run into anyone who thought CDE was better than OpenWin, but that's what was selected as the standard, and that's what Sun provided. If they hadn't, they would have been locked out of a lot of important markets.
It's not like there is a "constant change of GUIs" as you indicate. OpenWin was the Sun standard from about 1987 (not sure) until around 1990-1995, when CDE was spec'd. Now they're shifting to Gnome.
Note that all the way through, applications continued to run with the different desktop managers. Or were you under the impression that different versions of apps were running for different desktop managers?
From the Trolltech FAQ:
For those thinking to develop with the free edition, then just buy a license when they're ready to deploy:
The minimal price for a single platform commercial license is $1240USD. See Trolltech - Pricing Desktop.
The price is very reasonable for the functionality, but I only have so much money to spend on tools, and I'm not willing to plunk down the coin now just in case I need to be able to use my code commercially (i.e. to support a client site.)
One refresh it was, at which point I commented. After refreshing again, the previous post showed up.
*sigh* Still waiting for the fluke... *LOL*
Last time I checked (almost a year ago), QT for Win32 was several thousand dollars, a far cry from 8 billable hours. Not that the price was/is unreasonable compared to similar products (such as the now-defunct Neuron Data Open Interface, which ran around $10K/developer.)
It's only taken five years for me to fluke a legitimate "first post". *g*
I agree. I was a die-hard KDE user from 1.0-3.0, periodically trying Gnome along the way, but always disappointed by the lack of stability. After trying Gnome again under SuSE 8.0, I finally found it stable enough to use daily (and do so.)
Some people argue about performance and resources, but they're both pretty bloated compared to simple window managers like CDE/Motif.
From a programming perspective, QT/KDE are nicer products, but Gnome is catching up rapidly on the GUI designer front (Glade.) I don't really use IDEs, so I can't comment on KDeveloper vs. Anjuta, but both look to be pretty full-featured on the surface.
IIRC, PA-RISC was 40 or 48 bit addressing with a 32-bit instruction set at the time that Alpha came out. Alpha was out well before 64-bit MIPS processors (which were not DEC processors, BTW.) Alpha was also 64-bit before a refresh of the SPARC design brought it to 64 bit.
Sun has been babbling about the switch to Gnome from CDE for almost two years now. I use both KDE and Gnome, and both are far more a "desktop" than CDE ever was.
It also confirms my decision to use GTK for GUI development under Linux (I love QT's APIs and structure under KDE, but GTK lets me port to Win32 clients without cost.)
I gave it a good 5-6 episodes to turn into something worth watching, but had given up. It's a piece of drek, and good riddance. I can't believe they axed something entertaining like Dark Angel for a lame implementation of Rodenberry's "Wagon Train to the Stars" pitch line for Star Trek.
What really baffles me is how the show can possibly require the budget it has. The visual F/X are nothing special, rarely extending beyond CGI-only external shots. There has been nothing that would require extensive makeup, prosthesis, animatronics, or anything else to justify the cost of this series.
Then again, maybe they had to pay the "artists" who created the theme enough to live on for the rest of their lives -- that track is a career-killer if I've ever heard one!
Then maybe you just shouldn't be doing such "complex" programming. XSD should not take more than a couple days to understand, another few days to iron out the details, and by the end of 30-40 hours you should have produced a good sample of whatever you're modelling with XSD.
Too many "how to" documents and books are written for people who just don't "get it". While I sympathise that they need to learn, you have to grasp the concept of abstraction before some of these "languages" make sense. XSD, ERDs, SQL-DDL, etc. are just different ways of describing data structures and organization. Each is "complex" if you try to make it do more than it was intended for, but only takes a couple weeks to understand well enough to get by.
I tend to use mine for todo lists of offbeat items, like lists of books, CDs, DVDs, etc. that I want to pick up when convenient (i.e. I happen to be in a store, prices are good, and I have a few dollars to spare.)
It also lets me use the company categories on the address book to break down the list a bit. It's handy to be able to see the contact points for an active client site.
I tried using the todo lists to manage project/job tasks, but found it was too much hassle. I usually have some other todo facility for projects, whether it be a project management utility, a project status spreadsheet, or just a todo.html file in the project directory tree.
Other than that I installed some games I never play, an improved calculator that I rarely use, and just stopped trying to find "new uses" for it.
It's an address book and meeting reminder tool, with bonus reminder/todo lists. It's enough to be worth carrying around more often than a pager or a cell phone.
Some people like to be available for contact anytime anywhere (pager/cell-phone.) I prefer to maintain a personal life unless I'm providing formal offsite support to a client site. Leave a vmail or email message, and I'll look into it when I get back.
The technical concepts have been in use for decades, provided that you don't myopically focus on "console gaming" as the definition of platform.
Go wayyyyyy back to the days of the "dumb" terminal, the mid-80s. The "hot" video games such as Star Wars, Tempest, and their ilk had pathetic processors, but hardware vector graphics engines (i.e. line-drawing only) that would have required CPUs 10-20 times as powerful to render the same data to a raster display. "Intelligent" display terminals used for CAD/CAM often had at least as much raw CPU power as the monstrous VAX midframes they talked to.
Over the decades, the balance of power has constantly shifted between the processing center (CPU or data center) and the presentation/client (smart TTYs, X-terminals, fat clients, thin clients, browsers, browsers with plugins, java download clients, etc.)
You'd be amazed how much of modern graphics and CPU technology was actually theorized as far back as the 60s or 70s, but just could not be implemented with the technology of the day. Some of the algorithms discussed in my 400-series graphics class in the late 80's were only "theoretical" algorithms because even a Cray would have taken a couple days to calculate a single frame using them. Now we take it for granted and argue about which is "better", myopically ignoring all which came before us.
Case in point: Darwin Peachey was one of the graphics grads/profs at the University of Saskatchewan when I graduated. He left that year to work for a little startup you might have heard of: Pixar. Most of what he'd studied was research, and he saw Pixar as a chance to work with a bleeding-edge company that wanted to make the theory real. To the public their work is "new", but to their staff it's something they've been working on for 15-20 years, and far from new!
But does it squeal like an outraged pig after falling down the stairs like ED-209? Is it smart enough to crush it's remote after you've armed it? And most importantly, does it run Linux? (After all, if it's another WinCrap partnership you just know it's just going to stare at you with lifeless blue eyes blinking "BSOD" when you need it most!)
I've often had to resort to registry shredding to get things moved around. Just open up regedit and do a global search for subdirectory of the app you're trying to move. Similarly, you can search/substitute X: values with Y: values if you're shifting partitions around.
Obviously this isn't something most users know how to do (or should do!), but it can be done.
My bigger beef with the registry is that apps scatter pieces of their initialization all over the place. It makes it impossible to backup and restore individual apps by saving their registry tree, because you usually find that there are critical pieces of information whose names bear no resemblance to the application.
I have to wonder if the 3D-Realms team didn't use Hurd as their development/scheduling model...
The 450/500 GBP prices (roughly 704/780 USD) are just for the mobo with CPU, not a complete system. Assume another $20 for shipping (which would be cheap!), and you're looking at $800USD just for the mobo. You still need to add memory, a case, video card, HDD, CD[-RW]/DVD[-+RW], keyboard, mouse, and monitor.
Lets assume for sake of argument you're going cheap, cheap, cheap, so:
You're now running $320 in basic components, bringing the price up to $1045-1120 (700/800MHz variants.) Or you can get an eMac for $1100 (700MHz) that upgrades you to a CD-RW with a better video card, modem, and FireWire port that is pre-installed with a currently shipping copy of OSX. Alternatively, $1300 gets you an iBook with a 12.1" screen (slightly smaller 30GB HDD.)
Having decided to buy the AmigaOne mobo anyhow, you now have the option of running PPC Linux or waiting for the new OS. Either way, you miss out on the commercial product support for Linux (DB/2, Oracle, Sybase, et. al. are x86 binaries, not PPC.) Assuming pure open source is just fine by you, you've still got a box that is woefully underpowered to a similarly priced/configured AMD system (and maybe even Intel P4.)
Much as I loved my Amiga 1000, I just can't see any reason I'd want one of these new "Amiga" systems. Most of the reasons I loved my A1000 just aren't valid anymore -- everyone has hardware accelerated video and audio now, video capture and processing cards are common, and I'd rather be coding *nix than a system with no mind/market share.
Font anti-aliasing is completely useless when the monitor itself is blurry. Even so, I'd much rather use extreme resolutions with large fonts, as that has the same end result as anti-aliasing. A "good" monitor effectively blends adjacent pixels when fed resolutions higher than the physical display can support, which is exactly what anti-aliasing is trying to do via software. For example, most 17" monitors have a maximum physical display of 1280x1024. Feed them a 1600x1280 signal, and you get hardware "aliasing".
The typical office environment with modular office cube is just about the worst ergonomic environment there is. The cube itself has several problems:
Add to that issues of non-adjustable chairs, cheap flat keyboards (vs. ergonomic designs), poor mouse designs, and you've got a guarantee of neck, shoulder, and back problems before you even turn on the monitor.
Turning to the monitors, it's usually not a simple issue of LCD vs. CRT, but of monitor quality. Most cube-lands are full of poor quality monitors that came with pre-negotiated system bundles, often meaning they were outdated before the supply contract was first signed.
As a consultant, I have spent at least half my career stuck in front of blurry, non-adjustable, cheap monitors that I wouldn't even consider for a kids computer, much less someone who has to spend hours a day on it. While far from the worst I've dealt with, my current client environment consists of 5+ year old 17" monitors with poor color, poor contrast, glare problems, low refresh rates, and focus problems (the monitor I'm saddled with can't handle more than 1280x1024 at 60Hz NI. While the resolution is theoretically good enough, refresh rates under 75Hz cause serious eye strain.)
TFT LCD displays would help the eye strain, but so would high quality CRT monitors (e.g. Sony G420 19", Hitachi CM715 19", et. al.) The problem is that companies are not going to pay for replacement monitors because they can get entire system bundles with cheap monitors for very little more.
MKS Toolkit does a terrific job of integrating a Unix ksh-based CLI with WinNT/2K/XP, and couples closely with IBM DB/2 so that you can run DB/2 UDB scripts as you can under AIX. It is expensive, but worth it if you can afford it.
Qt is a very, very nice cross-platform development library. Another option you could look at used to be Neuron Data's Elements Environment, but they renamed the company (www.blazesoft.com) and I don't know if they still sell EE as a seperate product (Advisor is a repackaging of their rule-base software, which was built on EE.)
There are also the ever-present Rogue Wave class libraries, but I don't think they'll address your GUI requirements. However, if you use it to split out the core application functionality from the presentation (GUI), it might be helpful.
A solid set of macros with compiler/platform detection directives can help a great deal for porting code, though many people prefer to use sed or perl scripts instead (ala config.) Macros have the advantage of dealing with portability more consistently, and localize the changes for platforms to the headers and migration binding code (usually done as a base library.)
Using cross-platform libraries such as xml.apache.org libraries, IBM's ICU (Unicode support), et. al. can also make your code much more portable without requiring extra work after the initial coding.
There are also various open source projects that provide portable thread libraries, portable GUI toolkits, etc.
Actually, I prefer ClearType on my Sony monitor, but not on my "regular" monitor at work. Trinitron tubes use square/rectangular phosphor cells instead of dots. Seems to me that trinitron tubes might be close enough to a "set pixel layout" to have the same effect.
Anyone else out there who can compare tube/gun types?