I see several strong commercial releases that draw from the 'lower', smaller releases. In other words, a two-tiered approach. The commercial versions will appeal to the conservative businesses mindset, while the smaller distributions allow for innovation and growth. The conservative versions can introduce innovations and advancements after a Darwinian process determines the strengths and weaknesses of the innovations - which open source will then clean up as needed.
This is far better than Microsoft, which simply swallows the ideas of others and then regurgitates them in a closed (and usually badly reimplemented) form.
Unless I'm very mistaken, this may put a serious hurt on Cygwin and it's port of the Unix tools. I run Cygwin to get the bash shell and all the CLI utilities (find, grep, whatever). I'll need to pull this down When I Have Some Time (WIHST)(tm), so all of my speculation could be just so much crap. But yes, I see real use for this.
The basic issue is that the Napster system allows for an artist's work to be 'used' without any control system whatsoever. This is not listening on a record station (and them recording) or buying and then ripping for your personal pleasure (used as an example of 'fair use'). This is flat out intellectual property theft.
Having said that, what copyright.net is doing is nothing less than throwing gasoline on the raging fire. I don't trade on Napster based on my beliefs, and have stated those beliefs in the past. But that does not give me the right to use the kind of crude act that copyright.net uses to force my views down everybody elses throat.
Copyright.net is a far greater threat to the resolution of this entire issue than Napster ever was. They need to be taken down, preferably by legal means, and Tim Smith, copyright.net's chief executive, needs to have a new asshole ripped. His comment that he's not trying to antagonise anybody is pure bullshit. I've never seen a troll site before, but copyright.net sure fits my definition of one.
My father got me this book in middle school (for me, the mid-60's), and I used a number of the projects as starting points for my own hardware hacks. The most notable was the simple wind tunnel that used burning cones of incense to create streamers of smoke in the tunnel chamber. It was powered by a vacuum cleaner. I spent a fair amount of time making sure that air entering the chamber was even across the plenum. Another Scientific American experiment I started with was the construction of an electrostatic motor. I built a large one from plexiglass (12 inch diameter rotor, 18 inches long). I was into electrical and electronic hobbies, and this book was great just to read what others had done. My father never let me build the X-ray machine because he was afraid I'd irradiate myself and get cancer.
I'm happy somebody found it on the Web. But the book is far better.
I've read several comments where the esential point is "they can't do that". Unfortunately they can. Here's my story:
In February 2000 I went to work for Breakaway Solutions in Orlando, FL. (www.breakaway.com, BWAY). I left 87 days later in May, 2000. I got a good starting salary and stock options. I was told that the Orlando office was on track to become a solutions center, which is very high up in the pecking order. Along with that I was told my travel would be about 10%/15%, and that I had a pretty solid shot at becoming a team lead. All well and good. As I started to work on Breakway I worked on several side projects in defense, waiting for commercial work to start (I had done quite a bit of commercial development). Along the way I discovered that travel was 50% or higher, that Orlando would not become a solutions center because there was one in Boca (via the purchase of Eggrock Partners by Breakaway), and all those lovely options were rapidly turning into scrap paper. Other negative issues began to surface, and I eventually left Breakaway and went back to the company I worked at before Breakaway.
I was angry at how it had all turned out. On the Yahoo stock message board someone who identified themselves as another former employee made a comment about the company. In support of that comment I related my own experiences. I got a phone call and a letter from Breakaway legal telling me I was in violation of Breakaway's NDA clause of my seperation papers. I have since watched in sadistic satisfaction as Breakaway's stock tanked and they've gone through two rounds of layoffs (see www.fuckedcompany.com for more interesting details).
There is nothing more chilling when the legal cannons are shooting at you. At the very least I don't have the financial reserves to defend against, let along challenge this interpretation of their NDA clause. It's a hard choice sometimes when the choice is between being right and broke or knuckling under and staying solvent.
It's Not Here Yet - but Bell South is offering an interesting package deal in central Florida simetime in Q1 2001. The come-on is up to six months of free ISP service via standard dial-up, then connection to their nearly-finished DSL infrastructure. After that, you get an ISP connection, five mailboxes, standard DSL connectivity, and a free DSL modem for US$40/month. I don't know what the current exchange rate between US and Canada is, but it may be that US$40 is pretty close to C$50. And besides, if things get bad enough, I can always switch to Time Warner Cable Modems, since I get cable.
Well. Let's see. It's been a few hours since the last medication. Hmmm...
Well, I guess what I meant to say was that the salaries offered in various markets, such as Silicon Valley and Boston, look wonderful if you want to live in Orlando. But if you want to try to live on those salary offers in those specific locations, then you'll wind up living in the equivalent of a box.
Or to put it another way: I live in a nice house (appx 3500 sq ft) on 1/3 of an acre in the general vicinity of Universal Studios. I paid $88K for the original house, then took out a second mortguage for another $25K to add three new rooms to the back. The house is appraised now for about $160K. Do you think I could find an equivalent house for the same money in Silicon Valley or anywhere near Boston? Do you think I could find a house that big in Silicon Valley? No, I didn't think so.
So, it's fine where I'm at. I make a lot less, but I can afford a lot more. And I don't have to kill myself making a decent living.
I live and work in Orlando Florida (BlackRat Village) and it's fine right here, thank you. Every time I've chased a lead to Boston or the West Coast, and especially Silicon Valley, is that I'm somehow already rich enough to live there. I mean, how else can I understand how I get salary offers that would allow me to live like a king anywhere else except in Silicon Valley?
The other kicker is I did work for a dot-com based out of Boston. They have an Orlando office, and I went to work for them in February 2000. By May 2000 I had gone back to work for the company I was working at previously. While there I watched the stock rise then crash, I got to do lots and lots of travel (which was not supposed to happen), and in general I learned that the grass is not greener on the dot-com pastures. BTW, the former dot-com employer's stock hit a high of $86 dollars in March and is now trading around $3. Their earnings didn't meet expectations, and they had layoffs in September. My old employer gave me back my salary plus a reasonable raise (but not at the crazy heights the dot-com company gave me). The dot-com itch has definately died for me.
As already reported, it does not render sites too cleanly, such as slashdot. OK. Current HTML practices are attrocious. We know that. But what I have a problem with is the applications UI. It runs from little things like no busy indicator when it's loading an external page (indicator on the frame, or even a busy icon) to the orginization and rendering of controls. For a 4.0 release, this application puts forth a very poor quality image. Mozilla and Konqueror, just to name two, have never looked or worked this bad. I think it's time I pulled the source down and poked around a bit...
Oh. A word of warning. Trying to get the Windows binary from the HTTP site results in truncating the download to 4.096MB. Use the ftp site. And no, don't blame IE.
Absolutely. And Netscape helped make it that way, along with jwz's help. When Netscape set the precident, Microsoft simply followed (as they seem to do in so many different areas), producing a browser that has, over time and in so many ways, surpassed Netscape and others.
...it's all about market share and ease of use.
I would put ease-of-use over market share. Combine that with power and the ability to Do Something Useful, and you have that initial positive buzz that the marketeers can run with. After that, it's a self-supporting cycle as more users, attracted by the positive buzz themselves rave about the Product, inducing even more to come in and join the fun.
IE has a vast majority of the market share...
Earlier this year (2000) it had hit around 85% for IE, the rest for Netscape, Opera, etc.
My employers dictate how I write my web apps...
Your employers pick up their clues from the public. If the public is using IE by a wide margin, then that's what you have to cater to. Gone are the days when we had to essentially write multiple versions of the same site so that the widest margin of users could have the same experience. And good riddence.
Jamie (and others like Paul Allen) miss the point. What is tearing the high-tech centers apart, what they need more than anything, is affordable housing for the people who make up the bands that Jamie and Co. want to see. I'm glad that Jamie is working on a new loungue, but I have to wonder how long it will be before he raises prices in order to even break even in SF? When that happens, who but the rich will be able to see bands (or art works, etc).
Jamie and Co. need to look way beyond what he and his friends like and are into socially. If he can ever do that, and fix the important urban social problems, then he will have a more profound, longer lasting effect than anything he ever did at Netscape for the Internet.
Everytime I feel compelled to explain things this obvious, I worry that I've been trolled.
You're obviously not alone. I've never had a comment moderated around the block the way this one has. The comment was made tongue-in-cheek. I'm well aware of the gravity of the situation working for the CIA. When you go in there, you play by their rules, no exceptions. I just find the whole situation a little (black) humourous.
"The serious thing for us is people willfully misusing the computer system and trying to hide what they were trying to do," said one intelligence official. "If they were doing this with the KGB's computer system, we'd be giving them medals. Sadly, it was ours."
Now here's a perfect double standard. Fuck with the enemy's systems, and we'll give you a medal. Do the same with ours, and we'll shoot your ass. The funny part is that it was a chat room. Chat rooms are forums for essentially free speech. So the enemy probably would shoot you for attempting to practice your right to free speech. Thus, we have a situation here where they'd be damned by both sides.
This has all the smell of bad political infighting. As the Washington Post article points out, it seems "highly suspicious that all of those supervisors, not to mention the numerous component network administrators and security personnel, were unaware over a period of years of illicit computer usage by a group of 160 personnel". So something happens, and somebody who does know about this particular skeleton digs it up and uses it against "several officials, including members of the Senior Intelligence Service, a cadre of career officers at the upper reaches of the civil service system". They wind up with letters of repremand in their folders or worse, fired. In any event I strongly suspect there's a lot more going on that we don't know about - yet.
A version of BeOS from 2000 on a 1997 machine, compared with versions of Linux from 1997 on the 1997 machine and from 2000 on a 2000 machine? Or did i miss something in your argument?
I probably wasn't clear enough, my apologies. The best of the list to run on the Deskpro was Mandrake 7.1. It installed and ran out-of-the-box as well as any Windows release. I also installed BeOS 5 Pro next to it as well, just for the reference and the ability to do some testing. So the old Deskpro 2000 (from 1997) was running Windows 98SE, Mandrake 7.1, and BeOS 5. But this was after getting a 3dfx Voodoo 3000, a PCI modem that was not a Winmodem, and a 10/100 NIC that the TULIP drivers would recongnize on the Linux system.
For many years, Compaq has been one of the least compatable 'PC compatables' on the market.
Based on who's criteria? Windows? Every time I've installed or upgraded Windows on a Compaq machine, I've not had problems, even if it's from a 'pure' Microsoft distribution (i.e. not shipped with the system). If it's Linux, then yes, I can at least with my own experience say that Compaq has problems with some pieces of just about every distribution I've worked with. I find it ironic considering that Compaq has more than once advertised their Linux systems (Alphas) via click ads on Slashdot.
I regularly buy low cost motherboards from Taiwan with built-in everything and get Debian up and running in no time.
Two comments here.
The 'low cost motherboard' route is for the Linux enthusiast, which is fine. I've done the same. It's not for the "domain expert/casual computer user" who sees the computer as a means to an end, not an end unto itself. What's fine for you and me in this catagory is not even worth considering for a quite large (and important) group of people.
What version of Debian? If it's based on versions from 12 months ago, I went through two distributions, Debian and Corel, trying to install it on a Gateway PentiumIII 450 (E5500, I believe). Once again, the installed Windows NT 4 worked just fine. Once again, I had problems with the video card and the network card.
>> and when I've tried to test the 2.4.0-test10 kernel, I loose the network card
The word test isn't there because it looks cool, it means 'TEST', not 'ready to go out of the box'. Unless you want to TEST, don't use a test kernel.
"There are no known showstoppers, but I've asked all the major Linux houses to start deploying the current test kernels internally and start it through their test cycles," Torvalds said. "We've already found a few things that way, and hopefully, a month of this will shake out the worst.",
I felt that test10 was reasonably stable and close to being usable by mere mortals such as myself. That meant that stuff working reasonably well in the 2.2 kernel series would continue to work in the late 2.4 test series. I should have realized that it's probably another false hope, like the announcements in April and May. I stand corrected.
A school that just wants to plug in and go should buy their machines pre-configured and tested. If they choose a Linux based machine, they stand to save a lot of money.
Again, the comment concerning money saved is based on what studies, what public statistics to back this up? My personal experience is that after the initial cost of the software, the amount of time required to manage either system is the real clincher on cost, and it's generally a wash. If you hire good sysadmins for either system, both can be managed efficiently. But hire a mouthbreather in either position and you'll pay the price on higher maintainance costs due to induced downtimes. If it's the hardware, I've found that buying a reliable whitebox will cost you as much as an equivalent name-brand system. You get what you pay for, and if you trim the costs on the front end you'll usually get bit on back-end maintainance costs.
If the goal is to really teach kids about computers inside and out, they'll need one where looking inside is encouraged, not one that tries it's best to keep the hidden parts a secret. Imagine an auto mechanics course where the cars all have their hoods welded shut.
Over the past few decades, Microsoft and others have published rack after rack of information on operating system and application internals. Yes, there have been authors such as Andrew Schulman (Unauthorized Windows 95, etc) who've made a cottage industry out of documenting those dark corners that Microsoft 'forgot', but the system has been heavily documented based on customer feedback and need.
I mention Windows 95 because Microsoft needed to be dope slapped over the out-and-out lies concerning fundamental features of the OS. Andrew Schulman and others performed a great service in showing the emperor was a little bare in spots. But so should IBM on failing to take advantage of an incredible marketing opportunity in its inability to sell OS/2 over Win95. 1995 had to be the Year of the Software Clusterfuck.
In any event, the hood is not welded shut on that late model Windows system, any more than it's welded shut on anything else. In fact, based on some other comments concerning the teaching of C++ on Windows, the best environment is to get the (!free) Borland 5.5 command line compiler (and don't forget Turbo Debugger), and use that with Cygwin tools to teach a solid standard's compliant C++ course (don't use current gcc 2.59.2!). Then, if they want to teach widget programming on Windows, they can get an educational discount on the Borland Standard version and teach that portion of Windows. Want to teach Java? You've got everything from Sun's JDK and an editor (of which there are many) to an IDE such as Sun's Forte Community or Borland's JBuilder 4 Community. You can teach a tremendous amount concerning systems design, protocols, UI interaction, etc with Java. Oh. I forget. Java's not Politically Correct either. Oh, well.
I'm one of those 'Unix/Linux zealots', yet in spite of that, let me tell you a supporting story...
I have owned various computers since the Apple ][ (very late 70's vintage). I've been an x86 box user for some time now. Ever since I got ahold of Linux (early 94) I've managed to install it on every x86 box I've had since then. Compared to Microsoft, and especially Windows 3.x to Windows 95, Linux has been a joy when running on the same minimal hardware platform's I've either every run on, or owned. But a funny thing began to happen about four years ago, and reached it's peak two weeks ago.
Normally, I've built my x86 boxen out of pieces-parts picked up locally or mail-order. But I decided to get a Compaq Desqpro 2000 in 1997 (200MHz Pentium), just because it was getting to be a hassle with my real job to go hunting this stuff down. The deal I got on the box was pretty good. It came with Windows 95 installed, and all was good. As soon as I got it, I attempted to install Linux on it. I ordered a boxed set of RedHat to put on the system. I even purchased (and still keep up) an Infomagic subscription. I got the base Linux installed, but had problems with drivers for the network card, the video card (Cyrix on the motherboard), and the sound card (genuine Creative Awe32). Over the years, I've upgraded hardware (video and modem, new IDE drives) and kept the RedHat distribution up-to-date as well. And every time I've upgraded hardware I've run into compatibility issues. I've never had problems with basic installation on the Compaq, but let me try to run X or some other advanced application (like ppp (ha!)), and I've had hell to pay trying to get it to work reliably.
Two weeks ago I found a very good bargain on another Compaq; this one, a Presario 5000 with an Athlon 900MHz (Socket A), Hercules 3D Prophet II with Nvidia's GeForce2 MX chipset, 30Gig HD, 256MB ram, etc, etc, etc. Came pre-installd with Windows ME. Rather than hassle with re-partitioning the drive to accept Linux, I picked up a second drive (40GB, $159, !damn! this stuff is getting cheap!) and created a 10GB parition for Linux (along with a 10GB partition for Windows 2000 SP1). I will note that this box has USB ports, and I have a USB keyboard.This time, I attempted to install the following free operating systems:
RedHat 6.2
Slackware 7.1
Mandrake 7.1
SuSe 7.0 Professional
Storm Linux 2000 (Hail)
FreeBSD 4.1
OpenBSD 2.7
These were all boxed sets or from subscription services, not downloaded or borrowed (I try to support free software with cold hard cash as best I can). Of all the operating systems listed, only the first two booted all the way to installation, and only Slackware finished the installation and booted cleanly. Every OS listed after Slackware either locked up solid (such as Mandrake 7.1) or panicked (such as FreeBSD 4.1). Yet, Windows ME runs just fine, and Windows 2000, which was not part of the original package, loaded and found all hardware, and runs just fine next to Windows Me on a dual-boot system. The Hercules card has broken X on Linux, and I don't feel like hunting down free drivers or trying to install XFree86 4.x, since Slack now comes with it. I'll give credit to RedHat and Slackware both for working with the USB keyboard, but the drivers seem to have a problem with repeated keys (i.e. hitting the key 'd' pops up two d's, as in 'cdd'). Stability on both versions of Windows has been a joy (compared to Windows 95/98/SE). The installation of applications has been extremely easy. And the interesting thing is that Windows 2000 now sees fat32 volumes, meaning that I can reach everything on the WinMe volume under Win 2K. Getting the Linux totally up and running illustrates the irony of the current Linux situation: I have to boot into WinMe to use the modem to download patches, and when I've tried to test the 2.4.0-test10 kernel, I loose the network card (an Accton EN1207D Series PCI Fast Ethernet adapter, which the stock Slack kernel sees).
So what does this have to do with the current thread? Simply this: if the typical overworked underpaid teacher/professor can't get Linux installed and working out-of-the-box, then they're not going to bother. Today's new hardware is fast and cheap, and will continue this trend. The typical Linux whine that it's Compaq's fault for building 'non-standard' hardware won't fly. I know this, because along with the stack of operating systems I've collected I also have the BeOS 5 Pro that I picked up at BestBuy when it was on sale. Although I haven't installed it on the Athlon system, it has always worked just fine on the Desqpro 2000 that also gave Linux fits. I'll likely continue to use Linux as a server OS, and I'll more than likely go the route of running Linux in VmWare on Windows 2000 for some embedded work I have in mind, but from this point forward I'll not use it as my workstation OS until some major changes take place in the distributions. As for the 'free' apps, I run bash, emacs/xemacs, Python, Perl, gcc, and a whole raft of other GNUish tools via ActiveState, python.org, and Cygwin tools (for which I also purchased the 1.0 CD) just fine under Windows 2000, thank you very much. If today's kids somehow want that type of environment, they sure don't need Linux to get it.
Microsoft really has got nothing to worry about from Linux on the workstation with current hardware. Microsoft has made great strides overall with Windows 2000, and Microsoft can capitalize on this by offering a solid, substantial learning platform. And Microsoft knows what Unix and Apple have known before - if you capture the hearts and minds of the kids in school, there's a pretty good chance you'll keep them as they grow up into professionals. Linux has some serious challenges in front of it.
So mod this down if you like. But I hope somebody reads it, and thinks about it, before it happens.
I ordered Helix Code's Gnome on CD, and I was impressed with the broad support of many Linux distributions. (As an aside it's gotten to the point where I order rather than download, primarily because it makes for a high-quality convenient storage medium. Too many large downloads result in some minor corruption somewhere, causing installation problems later.) I was very impressed with 1.2 over 1.1 and earlier, specifically with its installation, usablity, and overall stability. This release makes up for the first Gnome release. I was not at all happy with that release, believing it had been pushed out the door far too soon.
In comparison to KDE 1.x and the soon-to-be released KDE 2, I believe Gnome has a much better look-and-feel. The look, in particular, is very professionally done, reminding me of Irix and the MacOS. However, I find that programatically I prefer the KDE APIs over the Gnome APIs. There is nothing wrong with QT, C++, or OO design and programming in general.
It is interesting to me that Motif/X/CDE gave us a butt-ugly UI and API. Now Gnome and KDE have split that 'paradigm', with Gnome giving us a very good UI and a merely ugly API, while KDE gives us a merely ugly UI and a very good API.
And finally, on the support of various Unixes, it would be nice if a Gnome package for OpenBSD were there along with the others. If Helix Code can support Solaris, then perhaps they could support OpenBSD and ship that as well.
Ok, there are gonna be alot of people bitching about how slow debian is on releases.
The complaint isn't a bitch, but the process is. You go on to talk about the long testing periods, but the best testing period is when it's released and out in the field. If I want to stay behind the curve with say, RedHat, I can go to RedHat's site and pick up patches against RedHat releases that are just as old as anything Debian puts out.
Your comment about long testing also runs counter to the claims of many in the Open Source community that when a problem comes up (such as a bug in the Linux kernel), it can be fixed rather quickly. Why should I sit behind the curve waiting for "stability" when I can get good stability, performance, and features on the curve with just about every other distribution that isn't Debian based? Damon Runyan's oft quoted "the race may not go to the strong and swift, but that's the way to bet your money" applies to Linux as to everything else in software; I'd rather bet my money on anything but Debian.
a former closed source company gets in trouble due to bad management/whatever, and decides to garner some attention/notoriety by dumping their former crown jewel application into the open source community.
Baan would have garnered a lot more attention and integrity if Baan had open sourced their work from a position of strength.
Insistence over control of the language!@?!?!? Didn't they invent the language? Does GM give its' cars away for free?
Wrong analogy. That's not the point I'm making. My point is how Sun, not once, but twice, went to two standards bodies (ISO and ECMA) to start the process of making Java a standard. In both cases Sun pulled out of the processes at the 11th hour. I don't have any complaints with Sun controlling Java as long as it is understood up front that this is their intention to do so, and they remain consistant about it. I liked the langauge even before Sun made the grand announcement about seeking standardization, which helped further Sun's cause. It is my feeling that Sun cynically played not only on anti-Microsoft sentiment to further their own position but on the strong pro feelings many have about open standards. It was a one-two combination that helped grease the skids of acceptance in a lot of quarters. Once Sun was assured of its position, however, it realized it had no need to complete either of the standardization processes; they had achieved what Microsoft has achieved, a defacto lockin in a lot of organizations. And many of us felt used because of it.
The ends never justify the means, not even when fighting against Microsoft. When you use Microsoft's tactics, you're no better than they are.
There isn't any real different. In all the intial press announcements and early books on Java, it was proclaimed that Java was a better C++ because it didn't allow pointers and had automatic garbage collection (memory management) than C/C++ (amoung other things). The problem with Java is Sun's desire to maintain control. What makes the control so blatant is that Sun went through two exercises where they were going to make it an open standard, only to kill that effort at the last minute. Rather than rehash the why and wherefore of both attempts, let me just say that it would ahve been better if Sun had never even made the effort in the first place. Sun has, IMHO (and many others) come across looking like M$ in the way it has handled not just Java but the community at large; cynical and manipulative for the ultimate benifit of Sun.
And I'm sorry you got modded down by someone who can't differentiate between a legitimate question and a real troll. I'd like to know who moderates comments so that we could moderate the moderators.
this is another fine example of what is wrong with Slashdot and "Open Source" in general. Openness is more than about software. It's also about open opinions and free speech, another subject that many slashdotters love to spill endless comments over. If criticism is construed as flamebaiting, then what we have is the Web's equivalent of the Animal Farm, and Napoleon is a penguin.
Slashdot Mirror
I see several strong commercial releases that draw from the 'lower', smaller releases. In other words, a two-tiered approach. The commercial versions will appeal to the conservative businesses mindset, while the smaller distributions allow for innovation and growth. The conservative versions can introduce innovations and advancements after a Darwinian process determines the strengths and weaknesses of the innovations - which open source will then clean up as needed.
This is far better than Microsoft, which simply swallows the ideas of others and then regurgitates them in a closed (and usually badly reimplemented) form.
Unless I'm very mistaken, this may put a serious hurt on Cygwin and it's port of the Unix tools. I run Cygwin to get the bash shell and all the CLI utilities (find, grep, whatever). I'll need to pull this down When I Have Some Time (WIHST)(tm), so all of my speculation could be just so much crap. But yes, I see real use for this.
The basic issue is that the Napster system allows for an artist's work to be 'used' without any control system whatsoever. This is not listening on a record station (and them recording) or buying and then ripping for your personal pleasure (used as an example of 'fair use'). This is flat out intellectual property theft.
Having said that, what copyright.net is doing is nothing less than throwing gasoline on the raging fire. I don't trade on Napster based on my beliefs, and have stated those beliefs in the past. But that does not give me the right to use the kind of crude act that copyright.net uses to force my views down everybody elses throat.
Copyright.net is a far greater threat to the resolution of this entire issue than Napster ever was. They need to be taken down, preferably by legal means, and Tim Smith, copyright.net's chief executive, needs to have a new asshole ripped. His comment that he's not trying to antagonise anybody is pure bullshit. I've never seen a troll site before, but copyright.net sure fits my definition of one.
The power of the press is very great, but not so great as the power of suppress.
Lord Northcliffe
And for those who believe in mirroring -
One has to multiple thoughts to the point where there aren't enough policemen to control them.
Stanislaw Lec
Is this another example of it doesn't matter (or even exist) unless it's on the web? Note at the bottom of the page where this was take from:
The preceding was taken in full from
Section IX. Optics, Heat, and Electronics;
Chapter 3. An Inexpensive X-ray Machine
The Scientific American Book of Projects for The Amateur Scientist
Library of Congress Card Catalog Number: 60-14286
© Copyright 1960 by C. L. Strong
My father got me this book in middle school (for me, the mid-60's), and I used a number of the projects as starting points for my own hardware hacks. The most notable was the simple wind tunnel that used burning cones of incense to create streamers of smoke in the tunnel chamber. It was powered by a vacuum cleaner. I spent a fair amount of time making sure that air entering the chamber was even across the plenum. Another Scientific American experiment I started with was the construction of an electrostatic motor. I built a large one from plexiglass (12 inch diameter rotor, 18 inches long). I was into electrical and electronic hobbies, and this book was great just to read what others had done. My father never let me build the X-ray machine because he was afraid I'd irradiate myself and get cancer.
I'm happy somebody found it on the Web. But the book is far better.
I've read several comments where the esential point is "they can't do that". Unfortunately they can. Here's my story:
In February 2000 I went to work for Breakaway Solutions in Orlando, FL. (www.breakaway.com, BWAY). I left 87 days later in May, 2000. I got a good starting salary and stock options. I was told that the Orlando office was on track to become a solutions center, which is very high up in the pecking order. Along with that I was told my travel would be about 10%/15%, and that I had a pretty solid shot at becoming a team lead. All well and good. As I started to work on Breakway I worked on several side projects in defense, waiting for commercial work to start (I had done quite a bit of commercial development). Along the way I discovered that travel was 50% or higher, that Orlando would not become a solutions center because there was one in Boca (via the purchase of Eggrock Partners by Breakaway), and all those lovely options were rapidly turning into scrap paper. Other negative issues began to surface, and I eventually left Breakaway and went back to the company I worked at before Breakaway.
I was angry at how it had all turned out. On the Yahoo stock message board someone who identified themselves as another former employee made a comment about the company. In support of that comment I related my own experiences. I got a phone call and a letter from Breakaway legal telling me I was in violation of Breakaway's NDA clause of my seperation papers. I have since watched in sadistic satisfaction as Breakaway's stock tanked and they've gone through two rounds of layoffs (see www.fuckedcompany.com for more interesting details).
There is nothing more chilling when the legal cannons are shooting at you. At the very least I don't have the financial reserves to defend against, let along challenge this interpretation of their NDA clause. It's a hard choice sometimes when the choice is between being right and broke or knuckling under and staying solvent.
It's Not Here Yet - but Bell South is offering an interesting package deal in central Florida simetime in Q1 2001. The come-on is up to six months of free ISP service via standard dial-up, then connection to their nearly-finished DSL infrastructure. After that, you get an ISP connection, five mailboxes, standard DSL connectivity, and a free DSL modem for US$40/month. I don't know what the current exchange rate between US and Canada is, but it may be that US$40 is pretty close to C$50. And besides, if things get bad enough, I can always switch to Time Warner Cable Modems, since I get cable.
Well. Let's see. It's been a few hours since the last medication. Hmmm...
Well, I guess what I meant to say was that the salaries offered in various markets, such as Silicon Valley and Boston, look wonderful if you want to live in Orlando. But if you want to try to live on those salary offers in those specific locations, then you'll wind up living in the equivalent of a box.
Or to put it another way: I live in a nice house (appx 3500 sq ft) on 1/3 of an acre in the general vicinity of Universal Studios. I paid $88K for the original house, then took out a second mortguage for another $25K to add three new rooms to the back. The house is appraised now for about $160K. Do you think I could find an equivalent house for the same money in Silicon Valley or anywhere near Boston? Do you think I could find a house that big in Silicon Valley? No, I didn't think so.
So, it's fine where I'm at. I make a lot less, but I can afford a lot more. And I don't have to kill myself making a decent living.
I live and work in Orlando Florida (BlackRat Village) and it's fine right here, thank you. Every time I've chased a lead to Boston or the West Coast, and especially Silicon Valley, is that I'm somehow already rich enough to live there. I mean, how else can I understand how I get salary offers that would allow me to live like a king anywhere else except in Silicon Valley?
The other kicker is I did work for a dot-com based out of Boston. They have an Orlando office, and I went to work for them in February 2000. By May 2000 I had gone back to work for the company I was working at previously. While there I watched the stock rise then crash, I got to do lots and lots of travel (which was not supposed to happen), and in general I learned that the grass is not greener on the dot-com pastures. BTW, the former dot-com employer's stock hit a high of $86 dollars in March and is now trading around $3. Their earnings didn't meet expectations, and they had layoffs in September. My old employer gave me back my salary plus a reasonable raise (but not at the crazy heights the dot-com company gave me). The dot-com itch has definately died for me.
As already reported, it does not render sites too cleanly, such as slashdot. OK. Current HTML practices are attrocious. We know that. But what I have a problem with is the applications UI. It runs from little things like no busy indicator when it's loading an external page (indicator on the frame, or even a busy icon) to the orginization and rendering of controls. For a 4.0 release, this application puts forth a very poor quality image. Mozilla and Konqueror, just to name two, have never looked or worked this bad. I think it's time I pulled the source down and poked around a bit...
Oh. A word of warning. Trying to get the Windows binary from the HTTP site results in truncating the download to 4.096MB. Use the ftp site. And no, don't blame IE.
The market has made the W3C irrelevant.
...it's all about market share and ease of use.
Absolutely. And Netscape helped make it that way, along with jwz's help. When Netscape set the precident, Microsoft simply followed (as they seem to do in so many different areas), producing a browser that has, over time and in so many ways, surpassed Netscape and others.
I would put ease-of-use over market share. Combine that with power and the ability to Do Something Useful, and you have that initial positive buzz that the marketeers can run with. After that, it's a self-supporting cycle as more users, attracted by the positive buzz themselves rave about the Product, inducing even more to come in and join the fun.
IE has a vast majority of the market share...
Earlier this year (2000) it had hit around 85% for IE, the rest for Netscape, Opera, etc.
My employers dictate how I write my web apps...
Your employers pick up their clues from the public. If the public is using IE by a wide margin, then that's what you have to cater to. Gone are the days when we had to essentially write multiple versions of the same site so that the widest margin of users could have the same experience. And good riddence.
Jamie (and others like Paul Allen) miss the point. What is tearing the high-tech centers apart, what they need more than anything, is affordable housing for the people who make up the bands that Jamie and Co. want to see. I'm glad that Jamie is working on a new loungue, but I have to wonder how long it will be before he raises prices in order to even break even in SF? When that happens, who but the rich will be able to see bands (or art works, etc).
Jamie and Co. need to look way beyond what he and his friends like and are into socially. If he can ever do that, and fix the important urban social problems, then he will have a more profound, longer lasting effect than anything he ever did at Netscape for the Internet.
You know, something that big spinning that fast gives a whole new meaning to the phrase "spin cycle".
Everytime I feel compelled to explain things this obvious, I worry that I've been trolled.
You're obviously not alone. I've never had a comment moderated around the block the way this one has. The comment was made tongue-in-cheek. I'm well aware of the gravity of the situation working for the CIA. When you go in there, you play by their rules, no exceptions. I just find the whole situation a little (black) humourous.
"The serious thing for us is people willfully misusing the computer system and trying to hide what they were trying to do," said one intelligence official. "If they were doing this with the KGB's computer system, we'd be giving them medals. Sadly, it was ours."
Now here's a perfect double standard. Fuck with the enemy's systems, and we'll give you a medal. Do the same with ours, and we'll shoot your ass. The funny part is that it was a chat room. Chat rooms are forums for essentially free speech. So the enemy probably would shoot you for attempting to practice your right to free speech. Thus, we have a situation here where they'd be damned by both sides.
This has all the smell of bad political infighting. As the Washington Post article points out, it seems "highly suspicious that all of those supervisors, not to mention the numerous component network administrators and security personnel, were unaware over a period of years of illicit computer usage by a group of 160 personnel". So something happens, and somebody who does know about this particular skeleton digs it up and uses it against "several officials, including members of the Senior Intelligence Service, a cadre of career officers at the upper reaches of the civil service system". They wind up with letters of repremand in their folders or worse, fired. In any event I strongly suspect there's a lot more going on that we don't know about - yet.
A version of BeOS from 2000 on a 1997 machine, compared with versions of Linux from 1997 on the 1997 machine and from 2000 on a 2000 machine? Or did i miss something in your argument?
I probably wasn't clear enough, my apologies. The best of the list to run on the Deskpro was Mandrake 7.1. It installed and ran out-of-the-box as well as any Windows release. I also installed BeOS 5 Pro next to it as well, just for the reference and the ability to do some testing. So the old Deskpro 2000 (from 1997) was running Windows 98SE, Mandrake 7.1, and BeOS 5. But this was after getting a 3dfx Voodoo 3000, a PCI modem that was not a Winmodem, and a 10/100 NIC that the TULIP drivers would recongnize on the Linux system.
Based on who's criteria? Windows? Every time I've installed or upgraded Windows on a Compaq machine, I've not had problems, even if it's from a 'pure' Microsoft distribution (i.e. not shipped with the system). If it's Linux, then yes, I can at least with my own experience say that Compaq has problems with some pieces of just about every distribution I've worked with. I find it ironic considering that Compaq has more than once advertised their Linux systems (Alphas) via click ads on Slashdot.
I regularly buy low cost motherboards from Taiwan with built-in everything and get Debian up and running in no time.
Two comments here.
>> and when I've tried to test the 2.4.0-test10 kernel, I loose the network card
The word test isn't there because it looks cool, it means 'TEST', not 'ready to go out of the box'. Unless you want to TEST, don't use a test kernel.
Well, excuse me. Based on an earlier article posted on slashdot (http://slash dot.org
"There are no known showstoppers, but I've asked all the major Linux houses to start deploying the current test kernels internally and start it through their test cycles," Torvalds said. "We've already found a few things that way, and hopefully, a month of this will shake out the worst.",
I felt that test10 was reasonably stable and close to being usable by mere mortals such as myself. That meant that stuff working reasonably well in the 2.2 kernel series would continue to work in the late 2.4 test series. I should have realized that it's probably another false hope, like the announcements in April and May. I stand corrected.
A school that just wants to plug in and go should buy their machines pre-configured and tested. If they choose a Linux based machine, they stand to save a lot of money.
Again, the comment concerning money saved is based on what studies, what public statistics to back this up? My personal experience is that after the initial cost of the software, the amount of time required to manage either system is the real clincher on cost, and it's generally a wash. If you hire good sysadmins for either system, both can be managed efficiently. But hire a mouthbreather in either position and you'll pay the price on higher maintainance costs due to induced downtimes. If it's the hardware, I've found that buying a reliable whitebox will cost you as much as an equivalent name-brand system. You get what you pay for, and if you trim the costs on the front end you'll usually get bit on back-end maintainance costs.
If the goal is to really teach kids about computers inside and out, they'll need one where looking inside is encouraged, not one that tries it's best to keep the hidden parts a secret. Imagine an auto mechanics course where the cars all have their hoods welded shut.
Over the past few decades, Microsoft and others have published rack after rack of information on operating system and application internals. Yes, there have been authors such as Andrew Schulman (Unauthorized Windows 95, etc) who've made a cottage industry out of documenting those dark corners that Microsoft 'forgot', but the system has been heavily documented based on customer feedback and need.
I mention Windows 95 because Microsoft needed to be dope slapped over the out-and-out lies concerning fundamental features of the OS. Andrew Schulman and others performed a great service in showing the emperor was a little bare in spots. But so should IBM on failing to take advantage of an incredible marketing opportunity in its inability to sell OS/2 over Win95. 1995 had to be the Year of the Software Clusterfuck.
In any event, the hood is not welded shut on that late model Windows system, any more than it's welded shut on anything else. In fact, based on some other comments concerning the teaching of C++ on Windows, the best environment is to get the (!free) Borland 5.5 command line compiler (and don't forget Turbo Debugger), and use that with Cygwin tools to teach a solid standard's compliant C++ course (don't use current gcc 2.59.2!). Then, if they want to teach widget programming on Windows, they can get an educational discount on the Borland Standard version and teach that portion of Windows. Want to teach Java? You've got everything from Sun's JDK and an editor (of which there are many) to an IDE such as Sun's Forte Community or Borland's JBuilder 4 Community. You can teach a tremendous amount concerning systems design, protocols, UI interaction, etc with Java. Oh. I forget. Java's not Politically Correct either. Oh, well.
I'm one of those 'Unix/Linux zealots', yet in spite of that, let me tell you a supporting story...
I have owned various computers since the Apple ][ (very late 70's vintage). I've been an x86 box user for some time now. Ever since I got ahold of Linux (early 94) I've managed to install it on every x86 box I've had since then. Compared to Microsoft, and especially Windows 3.x to Windows 95, Linux has been a joy when running on the same minimal hardware platform's I've either every run on, or owned. But a funny thing began to happen about four years ago, and reached it's peak two weeks ago.
Normally, I've built my x86 boxen out of pieces-parts picked up locally or mail-order. But I decided to get a Compaq Desqpro 2000 in 1997 (200MHz Pentium), just because it was getting to be a hassle with my real job to go hunting this stuff down. The deal I got on the box was pretty good. It came with Windows 95 installed, and all was good. As soon as I got it, I attempted to install Linux on it. I ordered a boxed set of RedHat to put on the system. I even purchased (and still keep up) an Infomagic subscription. I got the base Linux installed, but had problems with drivers for the network card, the video card (Cyrix on the motherboard), and the sound card (genuine Creative Awe32). Over the years, I've upgraded hardware (video and modem, new IDE drives) and kept the RedHat distribution up-to-date as well. And every time I've upgraded hardware I've run into compatibility issues. I've never had problems with basic installation on the Compaq, but let me try to run X or some other advanced application (like ppp (ha!)), and I've had hell to pay trying to get it to work reliably.
Two weeks ago I found a very good bargain on another Compaq; this one, a Presario 5000 with an Athlon 900MHz (Socket A), Hercules 3D Prophet II with Nvidia's GeForce2 MX chipset, 30Gig HD, 256MB ram, etc, etc, etc. Came pre-installd with Windows ME. Rather than hassle with re-partitioning the drive to accept Linux, I picked up a second drive (40GB, $159, !damn! this stuff is getting cheap!) and created a 10GB parition for Linux (along with a 10GB partition for Windows 2000 SP1). I will note that this box has USB ports, and I have a USB keyboard.This time, I attempted to install the following free operating systems:
RedHat 6.2
Slackware 7.1
Mandrake 7.1
SuSe 7.0 Professional
Storm Linux 2000 (Hail)
FreeBSD 4.1
OpenBSD 2.7
These were all boxed sets or from subscription services, not downloaded or borrowed (I try to support free software with cold hard cash as best I can). Of all the operating systems listed, only the first two booted all the way to installation, and only Slackware finished the installation and booted cleanly. Every OS listed after Slackware either locked up solid (such as Mandrake 7.1) or panicked (such as FreeBSD 4.1). Yet, Windows ME runs just fine, and Windows 2000, which was not part of the original package, loaded and found all hardware, and runs just fine next to Windows Me on a dual-boot system. The Hercules card has broken X on Linux, and I don't feel like hunting down free drivers or trying to install XFree86 4.x, since Slack now comes with it. I'll give credit to RedHat and Slackware both for working with the USB keyboard, but the drivers seem to have a problem with repeated keys (i.e. hitting the key 'd' pops up two d's, as in 'cdd'). Stability on both versions of Windows has been a joy (compared to Windows 95/98/SE). The installation of applications has been extremely easy. And the interesting thing is that Windows 2000 now sees fat32 volumes, meaning that I can reach everything on the WinMe volume under Win 2K. Getting the Linux totally up and running illustrates the irony of the current Linux situation: I have to boot into WinMe to use the modem to download patches, and when I've tried to test the 2.4.0-test10 kernel, I loose the network card (an Accton EN1207D Series PCI Fast Ethernet adapter, which the stock Slack kernel sees).
So what does this have to do with the current thread? Simply this: if the typical overworked underpaid teacher/professor can't get Linux installed and working out-of-the-box, then they're not going to bother. Today's new hardware is fast and cheap, and will continue this trend. The typical Linux whine that it's Compaq's fault for building 'non-standard' hardware won't fly. I know this, because along with the stack of operating systems I've collected I also have the BeOS 5 Pro that I picked up at BestBuy when it was on sale. Although I haven't installed it on the Athlon system, it has always worked just fine on the Desqpro 2000 that also gave Linux fits. I'll likely continue to use Linux as a server OS, and I'll more than likely go the route of running Linux in VmWare on Windows 2000 for some embedded work I have in mind, but from this point forward I'll not use it as my workstation OS until some major changes take place in the distributions. As for the 'free' apps, I run bash, emacs/xemacs, Python, Perl, gcc, and a whole raft of other GNUish tools via ActiveState, python.org, and Cygwin tools (for which I also purchased the 1.0 CD) just fine under Windows 2000, thank you very much. If today's kids somehow want that type of environment, they sure don't need Linux to get it.
Microsoft really has got nothing to worry about from Linux on the workstation with current hardware. Microsoft has made great strides overall with Windows 2000, and Microsoft can capitalize on this by offering a solid, substantial learning platform. And Microsoft knows what Unix and Apple have known before - if you capture the hearts and minds of the kids in school, there's a pretty good chance you'll keep them as they grow up into professionals. Linux has some serious challenges in front of it.
So mod this down if you like. But I hope somebody reads it, and thinks about it, before it happens.
I ordered Helix Code's Gnome on CD, and I was impressed with the broad support of many Linux distributions. (As an aside it's gotten to the point where I order rather than download, primarily because it makes for a high-quality convenient storage medium. Too many large downloads result in some minor corruption somewhere, causing installation problems later.) I was very impressed with 1.2 over 1.1 and earlier, specifically with its installation, usablity, and overall stability. This release makes up for the first Gnome release. I was not at all happy with that release, believing it had been pushed out the door far too soon.
In comparison to KDE 1.x and the soon-to-be released KDE 2, I believe Gnome has a much better look-and-feel. The look, in particular, is very professionally done, reminding me of Irix and the MacOS. However, I find that programatically I prefer the KDE APIs over the Gnome APIs. There is nothing wrong with QT, C++, or OO design and programming in general.
It is interesting to me that Motif/X/CDE gave us a butt-ugly UI and API. Now Gnome and KDE have split that 'paradigm', with Gnome giving us a very good UI and a merely ugly API, while KDE gives us a merely ugly UI and a very good API.
And finally, on the support of various Unixes, it would be nice if a Gnome package for OpenBSD were there along with the others. If Helix Code can support Solaris, then perhaps they could support OpenBSD and ship that as well.
Ok, there are gonna be alot of people bitching about how slow debian is on releases.
The complaint isn't a bitch, but the process is. You go on to talk about the long testing periods, but the best testing period is when it's released and out in the field. If I want to stay behind the curve with say, RedHat, I can go to RedHat's site and pick up patches against RedHat releases that are just as old as anything Debian puts out.
Your comment about long testing also runs counter to the claims of many in the Open Source community that when a problem comes up (such as a bug in the Linux kernel), it can be fixed rather quickly. Why should I sit behind the curve waiting for "stability" when I can get good stability, performance, and features on the curve with just about every other distribution that isn't Debian based? Damon Runyan's oft quoted "the race may not go to the strong and swift, but that's the way to bet your money" applies to Linux as to everything else in software; I'd rather bet my money on anything but Debian.
Probably the same inbred clueless moderator that marked it as over rated.
a former closed source company gets in trouble due to bad management/whatever, and decides to garner some attention/notoriety by dumping their former crown jewel application into the open source community.
Baan would have garnered a lot more attention and integrity if Baan had open sourced their work from a position of strength.
Insistence over control of the language!@?!?!? Didn't they invent the language? Does GM give its' cars away for free?
Wrong analogy. That's not the point I'm making. My point is how Sun, not once, but twice, went to two standards bodies (ISO and ECMA) to start the process of making Java a standard. In both cases Sun pulled out of the processes at the 11th hour. I don't have any complaints with Sun controlling Java as long as it is understood up front that this is their intention to do so, and they remain consistant about it . I liked the langauge even before Sun made the grand announcement about seeking standardization, which helped further Sun's cause. It is my feeling that Sun cynically played not only on anti-Microsoft sentiment to further their own position but on the strong pro feelings many have about open standards. It was a one-two combination that helped grease the skids of acceptance in a lot of quarters. Once Sun was assured of its position, however, it realized it had no need to complete either of the standardization processes; they had achieved what Microsoft has achieved, a defacto lockin in a lot of organizations. And many of us felt used because of it.
The ends never justify the means, not even when fighting against Microsoft. When you use Microsoft's tactics, you're no better than they are.
There isn't any real different. In all the intial press announcements and early books on Java, it was proclaimed that Java was a better C++ because it didn't allow pointers and had automatic garbage collection (memory management) than C/C++ (amoung other things). The problem with Java is Sun's desire to maintain control. What makes the control so blatant is that Sun went through two exercises where they were going to make it an open standard, only to kill that effort at the last minute. Rather than rehash the why and wherefore of both attempts, let me just say that it would ahve been better if Sun had never even made the effort in the first place. Sun has, IMHO (and many others) come across looking like M$ in the way it has handled not just Java but the community at large; cynical and manipulative for the ultimate benifit of Sun.
And I'm sorry you got modded down by someone who can't differentiate between a legitimate question and a real troll. I'd like to know who moderates comments so that we could moderate the moderators.
To the "moderator" who labeled this as flamebait;
this is another fine example of what is wrong with Slashdot and "Open Source" in general. Openness is more than about software. It's also about open opinions and free speech, another subject that many slashdotters love to spill endless comments over. If criticism is construed as flamebaiting, then what we have is the Web's equivalent of the Animal Farm, and Napoleon is a penguin.