Switching the programming staff from Solaris to Linux is no big deal. I'd be much more interested to hear what Oracle is doing with the PHBs, secretaries, marketers and other non-technical staff. I bet they're still on Windows.
Which may be the best route. I recall when I did some time on a Mainframe in the early 90's how ludicrous it seemed to have *everyone* using the same system to do their work: from the managers, engineers, developers, and clerical workers. All of these people had totally different jobs, but they all were forced to use the same setup to get their work done. The PC/LAN revolution was still gaining speed, and I recall thinking how much more efficient this would be: the engineers could upgrade systems rapidly for their uses, while the clerical staff could use more modest equipment that was geared for their jobs, and everyone would be happy now that they didn't have to use the same black Model T.
I felt this same derision when I was given a new box with Windows XP (I'm a developer). It seems like a return to those days where everyone is forced to use the same system. The file searching in XP is horrible for my uses, because it was altered to help newbies find their documents and digicam pics. The multitasking has degraded even more since Win2K, probably because it was optimized for home users who rarely run multiple heavy-lift applications. It feels like the mainframe days all over again: let's make the newbies and engineers all use the same system. What's old is new, I suppose.
It is my impression that Linux has momentum, but I think Sun is suffering more than Microsoft. To date, most of the major server migrations have been large companies switching from proprietary Unix systems to Linux.
However, it is effectively consolidating the Unix market into more or less a single front, which makes it a more formidable opponent to Windows in the long run.
My anecdotal observation shows a slow-simmering movement to open source in general by the "proles" of the IT industry: bread-and-butter IT departments for hospitals, industrial firms, etc, who don't really care about software religion, but just want to save money over the long haul. I knew when a friend of mine told me that the CIO of his rural hospital system was looking to migrate to OpenOffice/StarOffice to save costs, a slow movement based on raw economics was underway, techie religions be damned.
These types of migrations can stay under the radar for a long time before hitting a critical mass. Watching this unfold will keep things interesting, if nothing else.
One of the reasons I stopped using Mozilla was the bloat. I do not need one tool that does: web browsing, email, usenet, html editing and, now, ftp upload.
You can select browser only during the install, and this only loads the browser and html editor, and the email and other clients are not installed. I've run Moz like this for years, because I use other mail clients.
Au contraire! Microsoft has been shipping GPL software for a while. They call it "services for unix."
Thanks for the correction; in my head I was talking about MS not releasing core products like operating systems, office suites, etc under the GPL, but it didn't make it to the keyboard in my post...
Look, business is a dirty, bare knuckles kind of thing. You find the choicest customer, become his friend, and use that relationship to tar the competitor. With Linux, MS must discredit the very idea of working anybody but MS. True, a lot of customers think this way; but it is a result, not a strategy. MS wants to create this worldview, but it can't rely on it to be stable in and of itself
I wonder if this is a subtle change of policy for MS? By defining Linux as just another technology, that opens the door for MS using it, too. Not that Microsoft would ever release GPL'd software; but my prediction is that they will have a BSD-based Unix on the market around 2010. Apple did it, so they will too...:)
Sweet merciful crap! If it wasn't for these games, nobody would have wanted to grow up to be a game designer and create some of the games we see today. Some of those kids' parents need to just slap those brats across the face!
Funny, a friend of mine set up a MAME box for his seven-year-old son, and was laughing about how cool his son thought all the games were, and he would play Pacman for hours while Medal of Honor for PS2 sat gathering dust.
To a small extent, the success of games on Windows has put a lot of Windows PC's into the home, and by extension of familiarity, a lot of Windows PC's on managers desks and throughout companies.
From my experience, the machines at work dictate what people buy for home, not the other way around. In fact, a very large club the DOS/Windows market used against Macs and other systems were that they were "toy" systems focused mainly on gaming, whereas the PC was a "business" system that tacked on gaming as an afterthought.
Check out the AlphaSmart Dana, which is a compact keyboard/PDA based on PalmOS. Very rugged, long battery life, costs about $380 US new, can be found on eBay for less. I haven't personally used one, but they have gotten really good reviews for people who need to write on the road. The "used only once" eBay items are probably people who thought they were getting a really cheap laptop and were disappointed, but for your requirements a laptop would be a liability.
They were strong enough to carry the shuttles around on its back. According to the specs, the 747-400ER has a maximum takoff weight of 910,000 lbs. A fully-loaded 18-wheeler dirt truck averages around 80,000 lbs, to put that into perspective. I don't see how it gets off the ground. 11 trucks are heavy.
Considering that this experiment was done decades ago, you'd think you could go to Radio Shack and buy a junior double-slit experiment kit for $29. I believe the experiment, because a PhD physicist co-worker of mine vouched for it, but if not for him, I'd say the whole thing is a hoax. If it was done in the 50's with primitive technology, why isn't this experiment repeated more often? Other than this laser experiment, I've only seen pictures of a wooden box with a tube that looked like something built in the Victorian era, and charts that show what it's supposed to do. With so much pseudoscience and hogwash floating around, they can't expect us to just take their word for it. Show us!
BTW, I'm not asking you, I just needed to interject this comment amongst physicists, hoping that someone will say "moron, just go here and you can order a experiment kit for fifty bucks".
Thanks for this info. Where do you use this option?
It is a command-line option, so you can make a wrapper script and change your launchers/icons to call it instead of calling nautilus directly. If you named it (for instance) my.nautilus, it would be:
#!/bin/bash
nautilus --no-desktop $*
Save this in your path, and make it executable with "chmod 755 my.nautilus", then change your icons/launchers to launch my.nautilus. You can also just change every icon and reference to nautilus to "nautilus --no-desktop", but I have a habit of creating wrapper scripts like this and placing them in $HOME/bin, so that if I want later add other options I only have to change them in one place. Just make sure $HOME/bin is in your $PATH variable.
I can't find an official Honda page, but Temple of VTEC has an article where an Accord diesel achieved 76.6 MPG over a real-world 419-mile trip. In endurance track tests it achieved an average 130MPH over a 24-hour period.
This pretty much blows away the hybrids with a single-engine system and a nearly full-size car with no drawbacks. It's too bad that US diesel is too dirty for engines like this.
That's odd because the NT swapping strategy shouldn't do that (unless they've changed it for XP?).
I enjoyed NT4 as a dev box. With Win2K and WinXP, there are two basic features that have gradually deteriorated, however, and those are swapping and multitasking. Everything the posters have said is true: if you touch the filesystem in a big way (including the extremely borken file search utility in XP), it will swap out all your apps so that when you go to check your mail, etc, you get to watch your windows redraw pixel by pixel while the H/D goes nuts. How I wish the old Win9x MaxFileCacheSize setting was still there.
And without dual CPUs, WinXP is not a multitasking OS anymore. Try starting a compile/search/whatever and do something in the foreground... it's too frustrating, so it's best to just put the background task back in the foreground so it can get its priority boost and just go get some coffee. Yes, I've tinkered with the foreground/background switch; I don't think it's wired up anymore. I hate to see how boggy Longhorn will be; but hey, some newb will be able to share photos more easily once a year, so it's worth it to spend my 8 hour day in front of the PC frustrated.
But I'm not bitter.:)
For all the shit we give MS their kernel design is usually pretty damn good.
My guess is that the original authors are long gone or promoted, and the code is so taped-together that everyone is too scared to touch it anymore. Would you change a module of code if it could cause 20 million support calls when it crashes on a brandX mobo?
With all due respect, let's not give up the present to focus on an uncertain future. Microsoft has become legendary in their grand hand-waving about future products, and billg is one of the worst offenders, so betting our farm on what they say will happen two years from now is just not a solid product policy.
The fact is, Microsoft makes almost all their money on just two products: Windows (XP and 2003), and Microsoft Office. Two products that multiple sources have free equivalents. Even if XAML is the hottest thing going, it's still a development/deployment tool, and like VB.Net, ADO, or.Net, they do not really make Microsoft any money; they're just part of the Monopoly infrastructure.
How about we focus on pounding on the two products that matter; particularly Office, since we can run OpenOffice multi-platform. If you manage to get 15-20% market penetration, you can effectively influence the market, much like AMD has done with Intel (most people don't realize that AMD only has about 15% of the market). That is an achieveable goal by 2007, which is about how long it will take before large-scale deployment of Longhorn would begin, assuming it ships Summer 2006. If you have 15-20% of the Office market and the Linux market is trailing at 8-10% (but growing), it would create such a disruption that XAML would quickly become a back-burner issue.
Again, with all due respect, I would suggest focusing on those fights that are going on now and have tremendous potential, rather than allowing billg's intimidation to confuse our resolve. If you get 15-20% of the market that matters, history changes. Development tools are a sideline gig. Web client lock-in "standards" will quickly get scrapped by MSFT if their shareholders are letting management go because their two cash cows are losing ground.
Miguel, there are many articles and white papers that literally list out the deficiencies of desktop Linux and OOo vs. Windows and MSOffice, and you are in a position to influence developers and resources to fill those gaps, including improving solidarity between the Gnome and KDE efforts. I certainly am in no place to suggest to you how to proceed with your already stellar career, but take this as just as it is: a humble request to give it some thought.
Sorry, I should have been more expanatory, but I do have to get some work done during the day.:)
Eclipse is a bit different in its project-handling approach than many other IDEs like VS.Net, Borland Delphi, etc. All projects must be "imported" into the workspace before you can operate on them, and files within the project tree implicitly become part of the project. This is a relatively minor difference, but it can be strange at first when you're used to just opening a "project file" in other IDEs to begin working on them. The payback is that compilation is almost instantaneous and happens automatically when you save changes in the editor. I'm making a wild assumption that this feature is a benefit of having projects and their files imported or registered into the current workspace, since I haven't seen this feature in other IDEs.
Another big difference is that Eclipse, much like Mozilla, is a more of a reference design rather than a shrink-wrapped product, and as such is relatively sparse when you install it stock. You generally need to retrieve and install plugins to add functionality, but wow are there lots of good plugins out there. This is a Good Thing (tm) in my opinion, but it has caused no end of griping from our VS.Net developers who are using Eclipse. BTW, MyEclipse has pre-packaged plugins for Eclipse that are very affordable ($29/yr).
The philosophy of Eclipse, which is probably inherited from its IBM WebSphere roots, is that it seems to be designed for large or very large teams of developers in an enterprise situation, whereas most other PC IDEs seem to be geared more for smaller teams or lone wolf developers. That's not to say Eclipse isn't an excellent tool for small teams and individuals, any more than the other IDEs couldn't be used for computing in the large. I just get the sense when using Eclipse that there are large pools of power under the hood that I wouldn't be able to appreciate unless I had thirty other developers working on the workspace.
Rather, you have 12 year olds going on yahoo chat, trying to sneak some porn, look up the latest on Britney, or other such activities.
I agree with your concerns, but at the same time they are (in order with above) learning to read and type, solving the technical problems of getting around net nanny, and researching biographical information. Sure, it's pop culture, but the alternative activities for a 12yo would probably be TV or video games. And sooner or later they may get curious about all those shelves of books surrounding them.:)
I think the positives outweight the negatives, overral.
then could someone please explain what the Gates foundation actually does that could be construed as shady?
I think you'll find few complaints about the Gates Foundation, other than possibly charities who wish to receive funding who don't appreciate the stringent requirements to receive funding from the Foundation.
The shady sentiment mentioned in the article is probably confused with Microsoft Corp. "donating" software to schools out of goodwill or as a result of various antitrust trials. Donating $1 Billion of software is a misnomer when the cost of donation is a tiny fraction of the retail value of the items. Air would probably cost more to donate and deliver than a stack of license keys and CDs.
In the case of the Foundation, it is an independent charitable organization that is delivering hardware and software that the organization paid for. True, Bill G. probably got really deep discounts for Windows and Office, and likely discounts for the Dell PCs, but this is much closer to a true donation at retail value than MSFT donating pieces of paper (licenses) to schools which will have to upgrade later.
The Gates Foundation has had a tough time with legitimacy because it came about after Ted Turner basically called Gates out publicly for not donating any of his billions (Turner donated a third of his value, or $1Billion to the UN around 1998). A combination of pressure from Turner, Gates' father, and his wife reportedly caused the foundation to be formed. Gates initially ran the foundation much like Microsoft where he was heavily involved in the operations and ran the foundation in a fairly rigorous manner, so it was questioned in the mainstream press whether he was truly a philanthropist, or was this just another challenge/problem to solve for him. Time and money will eventually solve the image problem, and it already has improved in the last several years.
1) There seems to only be a single slide theme in there. OO needs to bundle more presentation templates in there. (Fortunately I have my own)
Agreed. However, what I usually suggest to people that need templates, fonts, etc, is to purchase StarOffice, which is Sun's "distribution" of OOo, with which they bundle extra fonts, lots of templates, a local database, etc. For about $75 at CompUSA for 5 personal installs, it's a good offer. Volume licensing is about $25-30 per seat, and schools get it for free. Plus, since it is 100% file compatible with OOo, it's relatively "future-proof": in case Sun drops the product, you'll still be able to work on your documents using OOo.
2) It's hard to browse between slides. With PowerPoint, all you have to do is hit page up or down to change slides in the editor. OO has these weird tab things you have to click on.
PgUp/Down should work fine. Also, if you look at the far right of the window, you'll see the small buttons for the different views: outline, slides, notes, fullscreen, etc.
KLyX is the KDE port of LyX, and would seem to be a good start. Instead of writing your own, why not join that team? If you're not a coder, then I'm sure the team would welcome usability suggestions to bring it up to speed. Writing your own app is very time-consuming, beyond the coding. It has to be polished, a web site has to be built and maintained, source patches accepted, documentation written, build and install scripts need to be written, etc. I've found that it's a lot better use of my time to join an existing project and make it better, rather than start from scratch.
As far as JBuilder being the best IDE for Java, that is what we used. It brought my 850MHz P3 w/ 768M memory to its knees being so slow and huge.
I haven't used older versions of JBuilder, so I can't speak to those, but I just installed the eval version of JBuilder X and it runs pretty well on my P4 1.8, a machine which can be bought from a namebrand vendor for about $500 with monitor. I'm sure it's slow on an 850 box, but most developers should have better machines than that by now.
"Not to mention the on-the-fly compiling."
-- This feature has been around for a long time (not just in Java or.NET stuff and before both). Besides, VS.NET 2001 did this so both have it.
That feature is not on by default; I just checked VS.Net 2003 to make sure I wasn't dreaming it. By on-the-fly I mean when I hit ctrl-s in eclipse to save after changing code, the project is automatically (and near-instantly) compiled and built, with compiler errors displayed and hightlighted with red squiggly lines. When I run it, there is no "build" step. VS.Net always "builds" and shows errors when I save code changes and run the project. How can I enable this feature in VS.Net? Or are we talking about two different things?
I'll be honest that I don't know what this "lightbulb" feature is.
Which is part of the problem in this business: few people will download and try new software, and even fewer will give fairly impartial people like myself the benefit of the doubt when I say there are genuinely good things about product X. I guess they assume that everyone has some sort of agenda or vendetta against company Y. Eclipse is a free download, but fairly sparse by default until you collect and install some plugins. The way it works with projects is a little wierd at first, but it's worth taking a look. The lightbulb (fix my code) feature will be included in VS.Net 2005, so you can get a leg up by seeing how it works in Eclipse.
This is all nice on Windows but have you seen it on Linux? It looks like shit.
I don't understand. I just copied an app we have running the jgoodies/jforms XP look and feel over to my RedHat box, and it looks identical to how it looks on my XP box. Ditto with the Liquid LnF.
I'm fairly new to Java/Swing, so I must have missed the warts of pre-1.4, but I'm surprised at how even Java developers dismiss Swing. I've been pleasantly surprised at how easy Swing development is. The widget set appears sparse at first, but especially the JTable control allows easier expansion than any other grid component I've worked with. It's just not as good for a ten-minute-get-it-working-quick session, but in real apps I tend to spend many hours altering the grid and overhauling the default behavior to suit the users, and usually get frustrated in the process. JTable has done a really good job of staying out of my way.
By having.NET on Linux I can write an app in Mono on Windows, then easily port it to run on Linux.
Disclaimer: I am not a Java fanboy, but if you really wanted to do this, why not use Java? Instead of waiting for Miquel to try and reimplement an unofficial port of a moving target (.Net), Java on Linux is officially supported by Sun. There are many IDEs available for Java. If you want GUIs, JBuilder is probably the best. For general coding, Eclipse is about the best IDE I've used, once you get used to its philosophy. After a couple of months with Eclipse, I had to go back to VS.Net 2003 for a couple of days, and I was shocked hollow it was and how dependent I had become on the "lightbulb" feature (fix my code) of Eclipse and the refactoring tools. Not to mention the on-the-fly compiling.
As an aside, VS.Net 2005 will have this lightbulb feature, and I predit the MS mainframers at our company will come running into my office to show this innovative "new" feature that Microsoft invented.
Anyway, the features you want are already available. Once you get the cheerleaders from both sides out of the room and get down to real work, Java is about the same as.Net as far as speed and GUI capabilities, and for real (not two-day petstore toys that the press loves) applications, they're about the same as far as productivity..Net has a little less cruft, but give it a few years and it will have similar cruft as Java. Check out javootoo for nice look and feels for Swing apps.
Imagine delivering an app via website that used native widgets and looked and felt like part of your OS, all while safely sandboxed.
This capability has been available with Java WebStart for a while now. Like many Sun and Apple products, they are consciously ignored until Microsoft "invents" them and the fanboys come running into my office to show this "new" technology on MSDN. Yawn. Trying to keep up with Mono is a Microsoft-sponsored hamster wheel, IMHO. If we really wanted.Net functionality on Linux, we would make peace with Sun and pull Java into the OSS world.
You can make really good Java Swing desktop or browser apps that look every bit as good or better than.Net apps. The Pluggable Look and Feels allow this; this site has a gallery of some. We've benchmarked real apps with Java and.Net, and the execution times are within 10% of each other. They both suck about the same.:)
Slashdot Mirror
Switching the programming staff from Solaris to Linux is no big deal. I'd be much more interested to hear what Oracle is doing with the PHBs, secretaries, marketers and other non-technical staff. I bet they're still on Windows.
Which may be the best route. I recall when I did some time on a Mainframe in the early 90's how ludicrous it seemed to have *everyone* using the same system to do their work: from the managers, engineers, developers, and clerical workers. All of these people had totally different jobs, but they all were forced to use the same setup to get their work done. The PC/LAN revolution was still gaining speed, and I recall thinking how much more efficient this would be: the engineers could upgrade systems rapidly for their uses, while the clerical staff could use more modest equipment that was geared for their jobs, and everyone would be happy now that they didn't have to use the same black Model T.
I felt this same derision when I was given a new box with Windows XP (I'm a developer). It seems like a return to those days where everyone is forced to use the same system. The file searching in XP is horrible for my uses, because it was altered to help newbies find their documents and digicam pics. The multitasking has degraded even more since Win2K, probably because it was optimized for home users who rarely run multiple heavy-lift applications. It feels like the mainframe days all over again: let's make the newbies and engineers all use the same system. What's old is new, I suppose.
It is my impression that Linux has momentum, but I think Sun is suffering more than Microsoft. To date, most of the major server migrations have been large companies switching from proprietary Unix systems to Linux.
However, it is effectively consolidating the Unix market into more or less a single front, which makes it a more formidable opponent to Windows in the long run.
My anecdotal observation shows a slow-simmering movement to open source in general by the "proles" of the IT industry: bread-and-butter IT departments for hospitals, industrial firms, etc, who don't really care about software religion, but just want to save money over the long haul. I knew when a friend of mine told me that the CIO of his rural hospital system was looking to migrate to OpenOffice/StarOffice to save costs, a slow movement based on raw economics was underway, techie religions be damned.
These types of migrations can stay under the radar for a long time before hitting a critical mass. Watching this unfold will keep things interesting, if nothing else.
One of the reasons I stopped using Mozilla was the bloat. I do not need one tool that does: web browsing, email, usenet, html editing and, now, ftp upload.
You can select browser only during the install, and this only loads the browser and html editor, and the email and other clients are not installed. I've run Moz like this for years, because I use other mail clients.
Au contraire! Microsoft has been shipping GPL software for a while. They call it "services for unix."
Thanks for the correction; in my head I was talking about MS not releasing core products like operating systems, office suites, etc under the GPL, but it didn't make it to the keyboard in my post...
Look, business is a dirty, bare knuckles kind of thing. You find the choicest customer, become his friend, and use that relationship to tar the competitor. With Linux, MS must discredit the very idea of working anybody but MS. True, a lot of customers think this way; but it is a result, not a strategy. MS wants to create this worldview, but it can't rely on it to be stable in and of itself
:)
I wonder if this is a subtle change of policy for MS? By defining Linux as just another technology, that opens the door for MS using it, too. Not that Microsoft would ever release GPL'd software; but my prediction is that they will have a BSD-based Unix on the market around 2010. Apple did it, so they will too...
Sweet merciful crap! If it wasn't for these games, nobody would have wanted to grow up to be a game designer and create some of the games we see today. Some of those kids' parents need to just slap those brats across the face!
Funny, a friend of mine set up a MAME box for his seven-year-old son, and was laughing about how cool his son thought all the games were, and he would play Pacman for hours while Medal of Honor for PS2 sat gathering dust.
To a small extent, the success of games on Windows has put a lot of Windows PC's into the home, and by extension of familiarity, a lot of Windows PC's on managers desks and throughout companies.
From my experience, the machines at work dictate what people buy for home, not the other way around. In fact, a very large club the DOS/Windows market used against Macs and other systems were that they were "toy" systems focused mainly on gaming, whereas the PC was a "business" system that tacked on gaming as an afterthought.
Check out the AlphaSmart Dana, which is a compact keyboard/PDA based on PalmOS. Very rugged, long battery life, costs about $380 US new, can be found on eBay for less. I haven't personally used one, but they have gotten really good reviews for people who need to write on the road. The "used only once" eBay items are probably people who thought they were getting a really cheap laptop and were disappointed, but for your requirements a laptop would be a liability.
But I thought 747's weren't particularly strong.
They were strong enough to carry the shuttles around on its back. According to the specs, the 747-400ER has a maximum takoff weight of 910,000 lbs. A fully-loaded 18-wheeler dirt truck averages around 80,000 lbs, to put that into perspective. I don't see how it gets off the ground. 11 trucks are heavy.
Yeah, here is a nice summary with pictures.
Considering that this experiment was done decades ago, you'd think you could go to Radio Shack and buy a junior double-slit experiment kit for $29. I believe the experiment, because a PhD physicist co-worker of mine vouched for it, but if not for him, I'd say the whole thing is a hoax. If it was done in the 50's with primitive technology, why isn't this experiment repeated more often? Other than this laser experiment, I've only seen pictures of a wooden box with a tube that looked like something built in the Victorian era, and charts that show what it's supposed to do. With so much pseudoscience and hogwash floating around, they can't expect us to just take their word for it. Show us!
BTW, I'm not asking you, I just needed to interject this comment amongst physicists, hoping that someone will say "moron, just go here and you can order a experiment kit for fifty bucks".
Thanks for this info. Where do you use this option?
It is a command-line option, so you can make a wrapper script and change your launchers/icons to call it instead of calling nautilus directly. If you named it (for instance) my.nautilus, it would be:
#!/bin/bash
nautilus --no-desktop $*
Save this in your path, and make it executable with "chmod 755 my.nautilus", then change your icons/launchers to launch my.nautilus. You can also just change every icon and reference to nautilus to "nautilus --no-desktop", but I have a habit of creating wrapper scripts like this and placing them in $HOME/bin, so that if I want later add other options I only have to change them in one place. Just make sure $HOME/bin is in your $PATH variable.
Preventing it from popping up in your startup session
Use the --no-desktop option and it will just do file management without the desktop "taking over" in XFCE.
I can't find an official Honda page, but Temple of VTEC has an article where an Accord diesel achieved 76.6 MPG over a real-world 419-mile trip. In endurance track tests it achieved an average 130MPH over a 24-hour period.
This pretty much blows away the hybrids with a single-engine system and a nearly full-size car with no drawbacks. It's too bad that US diesel is too dirty for engines like this.
That's odd because the NT swapping strategy shouldn't do that (unless they've changed it for XP?).
:)
I enjoyed NT4 as a dev box. With Win2K and WinXP, there are two basic features that have gradually deteriorated, however, and those are swapping and multitasking. Everything the posters have said is true: if you touch the filesystem in a big way (including the extremely borken file search utility in XP), it will swap out all your apps so that when you go to check your mail, etc, you get to watch your windows redraw pixel by pixel while the H/D goes nuts. How I wish the old Win9x MaxFileCacheSize setting was still there.
And without dual CPUs, WinXP is not a multitasking OS anymore. Try starting a compile/search/whatever and do something in the foreground... it's too frustrating, so it's best to just put the background task back in the foreground so it can get its priority boost and just go get some coffee. Yes, I've tinkered with the foreground/background switch; I don't think it's wired up anymore. I hate to see how boggy Longhorn will be; but hey, some newb will be able to share photos more easily once a year, so it's worth it to spend my 8 hour day in front of the PC frustrated.
But I'm not bitter.
For all the shit we give MS their kernel design is usually pretty damn good.
My guess is that the original authors are long gone or promoted, and the code is so taped-together that everyone is too scared to touch it anymore. Would you change a module of code if it could cause 20 million support calls when it crashes on a brandX mobo?
With all due respect, let's not give up the present to focus on an uncertain future. Microsoft has become legendary in their grand hand-waving about future products, and billg is one of the worst offenders, so betting our farm on what they say will happen two years from now is just not a solid product policy.
/deployment tool, and like VB.Net, ADO, or .Net, they do not really make Microsoft any money; they're just part of the Monopoly infrastructure.
The fact is, Microsoft makes almost all their money on just two products: Windows (XP and 2003), and Microsoft Office. Two products that multiple sources have free equivalents. Even if XAML is the hottest thing going, it's still a development
How about we focus on pounding on the two products that matter; particularly Office, since we can run OpenOffice multi-platform. If you manage to get 15-20% market penetration, you can effectively influence the market, much like AMD has done with Intel (most people don't realize that AMD only has about 15% of the market). That is an achieveable goal by 2007, which is about how long it will take before large-scale deployment of Longhorn would begin, assuming it ships Summer 2006. If you have 15-20% of the Office market and the Linux market is trailing at 8-10% (but growing), it would create such a disruption that XAML would quickly become a back-burner issue.
Again, with all due respect, I would suggest focusing on those fights that are going on now and have tremendous potential, rather than allowing billg's intimidation to confuse our resolve. If you get 15-20% of the market that matters, history changes. Development tools are a sideline gig. Web client lock-in "standards" will quickly get scrapped by MSFT if their shareholders are letting management go because their two cash cows are losing ground.
Miguel, there are many articles and white papers that literally list out the deficiencies of desktop Linux and OOo vs. Windows and MSOffice, and you are in a position to influence developers and resources to fill those gaps, including improving solidarity between the Gnome and KDE efforts. I certainly am in no place to suggest to you how to proceed with your already stellar career, but take this as just as it is: a humble request to give it some thought.
What is Eclipse's philosophy?
:)
Sorry, I should have been more expanatory, but I do have to get some work done during the day.
Eclipse is a bit different in its project-handling approach than many other IDEs like VS.Net, Borland Delphi, etc. All projects must be "imported" into the workspace before you can operate on them, and files within the project tree implicitly become part of the project. This is a relatively minor difference, but it can be strange at first when you're used to just opening a "project file" in other IDEs to begin working on them. The payback is that compilation is almost instantaneous and happens automatically when you save changes in the editor. I'm making a wild assumption that this feature is a benefit of having projects and their files imported or registered into the current workspace, since I haven't seen this feature in other IDEs.
Another big difference is that Eclipse, much like Mozilla, is a more of a reference design rather than a shrink-wrapped product, and as such is relatively sparse when you install it stock. You generally need to retrieve and install plugins to add functionality, but wow are there lots of good plugins out there. This is a Good Thing (tm) in my opinion, but it has caused no end of griping from our VS.Net developers who are using Eclipse. BTW, MyEclipse has pre-packaged plugins for Eclipse that are very affordable ($29/yr).
The philosophy of Eclipse, which is probably inherited from its IBM WebSphere roots, is that it seems to be designed for large or very large teams of developers in an enterprise situation, whereas most other PC IDEs seem to be geared more for smaller teams or lone wolf developers. That's not to say Eclipse isn't an excellent tool for small teams and individuals, any more than the other IDEs couldn't be used for computing in the large. I just get the sense when using Eclipse that there are large pools of power under the hood that I wouldn't be able to appreciate unless I had thirty other developers working on the workspace.
Rather, you have 12 year olds going on yahoo chat, trying to sneak some porn, look up the latest on Britney, or other such activities.
:)
I agree with your concerns, but at the same time they are (in order with above) learning to read and type, solving the technical problems of getting around net nanny, and researching biographical information. Sure, it's pop culture, but the alternative activities for a 12yo would probably be TV or video games. And sooner or later they may get curious about all those shelves of books surrounding them.
I think the positives outweight the negatives, overral.
then could someone please explain what the Gates foundation actually does that could be construed as shady?
I think you'll find few complaints about the Gates Foundation, other than possibly charities who wish to receive funding who don't appreciate the stringent requirements to receive funding from the Foundation.
The shady sentiment mentioned in the article is probably confused with Microsoft Corp. "donating" software to schools out of goodwill or as a result of various antitrust trials. Donating $1 Billion of software is a misnomer when the cost of donation is a tiny fraction of the retail value of the items. Air would probably cost more to donate and deliver than a stack of license keys and CDs.
In the case of the Foundation, it is an independent charitable organization that is delivering hardware and software that the organization paid for. True, Bill G. probably got really deep discounts for Windows and Office, and likely discounts for the Dell PCs, but this is much closer to a true donation at retail value than MSFT donating pieces of paper (licenses) to schools which will have to upgrade later.
The Gates Foundation has had a tough time with legitimacy because it came about after Ted Turner basically called Gates out publicly for not donating any of his billions (Turner donated a third of his value, or $1Billion to the UN around 1998). A combination of pressure from Turner, Gates' father, and his wife reportedly caused the foundation to be formed. Gates initially ran the foundation much like Microsoft where he was heavily involved in the operations and ran the foundation in a fairly rigorous manner, so it was questioned in the mainstream press whether he was truly a philanthropist, or was this just another challenge/problem to solve for him. Time and money will eventually solve the image problem, and it already has improved in the last several years.
1) There seems to only be a single slide theme in there. OO needs to bundle more presentation templates in there. (Fortunately I have my own)
Agreed. However, what I usually suggest to people that need templates, fonts, etc, is to purchase StarOffice, which is Sun's "distribution" of OOo, with which they bundle extra fonts, lots of templates, a local database, etc. For about $75 at CompUSA for 5 personal installs, it's a good offer. Volume licensing is about $25-30 per seat, and schools get it for free. Plus, since it is 100% file compatible with OOo, it's relatively "future-proof": in case Sun drops the product, you'll still be able to work on your documents using OOo.
2) It's hard to browse between slides. With PowerPoint, all you have to do is hit page up or down to change slides in the editor. OO has these weird tab things you have to click on.
PgUp/Down should work fine. Also, if you look at the far right of the window, you'll see the small buttons for the different views: outline, slides, notes, fullscreen, etc.
TeXmacs and LyX are close
KLyX is the KDE port of LyX, and would seem to be a good start. Instead of writing your own, why not join that team? If you're not a coder, then I'm sure the team would welcome usability suggestions to bring it up to speed. Writing your own app is very time-consuming, beyond the coding. It has to be polished, a web site has to be built and maintained, source patches accepted, documentation written, build and install scripts need to be written, etc. I've found that it's a lot better use of my time to join an existing project and make it better, rather than start from scratch.
As far as JBuilder being the best IDE for Java, that is what we used. It brought my 850MHz P3 w/ 768M memory to its knees being so slow and huge.
.NET stuff and before both). Besides, VS.NET 2001 did this so both have it.
I haven't used older versions of JBuilder, so I can't speak to those, but I just installed the eval version of JBuilder X and it runs pretty well on my P4 1.8, a machine which can be bought from a namebrand vendor for about $500 with monitor. I'm sure it's slow on an 850 box, but most developers should have better machines than that by now.
"Not to mention the on-the-fly compiling." -- This feature has been around for a long time (not just in Java or
That feature is not on by default; I just checked VS.Net 2003 to make sure I wasn't dreaming it. By on-the-fly I mean when I hit ctrl-s in eclipse to save after changing code, the project is automatically (and near-instantly) compiled and built, with compiler errors displayed and hightlighted with red squiggly lines. When I run it, there is no "build" step. VS.Net always "builds" and shows errors when I save code changes and run the project. How can I enable this feature in VS.Net? Or are we talking about two different things?
I'll be honest that I don't know what this "lightbulb" feature is.
Which is part of the problem in this business: few people will download and try new software, and even fewer will give fairly impartial people like myself the benefit of the doubt when I say there are genuinely good things about product X. I guess they assume that everyone has some sort of agenda or vendetta against company Y. Eclipse is a free download, but fairly sparse by default until you collect and install some plugins. The way it works with projects is a little wierd at first, but it's worth taking a look. The lightbulb (fix my code) feature will be included in VS.Net 2005, so you can get a leg up by seeing how it works in Eclipse.
This is all nice on Windows but have you seen it on Linux? It looks like shit.
I don't understand. I just copied an app we have running the jgoodies/jforms XP look and feel over to my RedHat box, and it looks identical to how it looks on my XP box. Ditto with the Liquid LnF.
I'm fairly new to Java/Swing, so I must have missed the warts of pre-1.4, but I'm surprised at how even Java developers dismiss Swing. I've been pleasantly surprised at how easy Swing development is. The widget set appears sparse at first, but especially the JTable control allows easier expansion than any other grid component I've worked with. It's just not as good for a ten-minute-get-it-working-quick session, but in real apps I tend to spend many hours altering the grid and overhauling the default behavior to suit the users, and usually get frustrated in the process. JTable has done a really good job of staying out of my way.
Combine Zero Install with Linux-level sandboxing and you have the same thing, but faster and language-neutral.
I hadn't heard of that. Thanks for the link.
By having .NET on Linux I can write an app in Mono on Windows, then easily port it to run on Linux.
.Net as far as speed and GUI capabilities, and for real (not two-day petstore toys that the press loves) applications, they're about the same as far as productivity. .Net has a little less cruft, but give it a few years and it will have similar cruft as Java. Check out javootoo for nice look and feels for Swing apps.
Disclaimer: I am not a Java fanboy, but if you really wanted to do this, why not use Java? Instead of waiting for Miquel to try and reimplement an unofficial port of a moving target (.Net), Java on Linux is officially supported by Sun. There are many IDEs available for Java. If you want GUIs, JBuilder is probably the best. For general coding, Eclipse is about the best IDE I've used, once you get used to its philosophy. After a couple of months with Eclipse, I had to go back to VS.Net 2003 for a couple of days, and I was shocked hollow it was and how dependent I had become on the "lightbulb" feature (fix my code) of Eclipse and the refactoring tools. Not to mention the on-the-fly compiling.
As an aside, VS.Net 2005 will have this lightbulb feature, and I predit the MS mainframers at our company will come running into my office to show this innovative "new" feature that Microsoft invented.
Anyway, the features you want are already available. Once you get the cheerleaders from both sides out of the room and get down to real work, Java is about the same as
Imagine delivering an app via website that used native widgets and looked and felt like part of your OS, all while safely sandboxed.
.Net functionality on Linux, we would make peace with Sun and pull Java into the OSS world.
.Net apps. The Pluggable Look and Feels allow this; this site has a gallery of some. We've benchmarked real apps with Java and .Net, and the execution times are within 10% of each other. They both suck about the same. :)
This capability has been available with Java WebStart for a while now. Like many Sun and Apple products, they are consciously ignored until Microsoft "invents" them and the fanboys come running into my office to show this "new" technology on MSDN. Yawn. Trying to keep up with Mono is a Microsoft-sponsored hamster wheel, IMHO. If we really wanted
You can make really good Java Swing desktop or browser apps that look every bit as good or better than