There already is a Gnus client for Slashdot. It allows you to use Gnus scorefiles and everything. Sometimes I wonder if Emacs hackers belong to a totally different species.
Now, now, I wouldn't go that far. While I don't particularly like to drive SUVs in the snow (for the reasons I stated above), I wouldn't go so far as to say that they have no place on the road.
Sometimes you actually do need a vehicle that will hold a large number of passengers over rough terrain. For example, I am a Scout Master, and I must admit that my buddy's SUV has come in handy a couple of times taking the scouts camping.
Personally, I prefer the lower cost and better handling and gas mileage that I get from my mini-van, but that's just me. After all, scouts can walk, and making them pack their gear in means less of a mess to clean up afterwards. I can generally get close enough to the campsite so that walking is feasible.
Of course, I wouldn't even pretend to dictate how others spend their money. If they want to pay for an SUV, well, that's fine with me. Just don't drive fast in a snowstorm.
Large four wheel drive vehicles are certainly helpful if you are plowing through snow drifts (you'll probably still get stuck, but you'll have fun doing it), but they are next to useless if the road is merely slick. If the road is merely slick you are far better off driving a smaller vehicle. This is because when the road is merely slick the problem isn't getting stuck, it's steering and stopping. Your SUV has a lot more mass and momentum than my Honda Civic, without much more contact surface area. I have lost track of the times where I have had some bozo in a SUV pass me in a snow storm only to see him in a ditch further up the road. And don't even get me started on Jeep Wrangler's (yes, I know they aren't an SUV) those short boxes are murder on slick roads.
Now, if by SUV you also include some of the 4 wheel drive suburu hatchbacks and station wagons then I agree. Those are quite good in the snow.
Well then, I agree with you:). Linux kernel developers should deport themselves with as much dignity and aplomb as possible. You are certainly correct that political opinions, even ones as relevant as this one, can certainly reflect badly on Linux in the eyes of some folks.
On the other hand, I can't help but think that Alan Cox deserves to be able to use the Linux kernel as a soapbox. After all, who knows if it would even exist without his tireless efforts. He almost certainly knew that some people wouldn't appreciate his statement, but apparently it meant enough to him to make it anyhow.
Linux Kernel hackers do what they do for the fun and the fame. An unfortunate side effect of this is that it creates a platform for opinionated, outspoken, "unprofessional" Unix hackers. Being able to speak out about issues like the DMCA and be heard is part of the reason that people hack Free Software. It's also a major difference between working on Linux and working on a commercial operating system. The guys hacking the Windows kernel get paid more than Alan, but hardly anyone knows who they are. Taking away Alan's right to use (and even abuse) his "fame" would make Alan both underpaid and under motivated.
Besides, he was right:). He could have been a little more professional, but apparently it is very important to him.
Yes, I realize that, but Hotmail is the primary draw for Passport. How many Passport users would there be without Hotmail? Probably not very many. Instead of millions of users Microsoft would be lucky to have tens of thousands of users.
Getting Ebay, Prudential, and 68 other companies to modify their websites to cater to such a small group of people would never fly. Microsoft needs to be able to wave the carrot of millions of potential customers, or it isn't worth the time.
Besides, even with millions of customers why should Ebay trust what Microsoft to store their important customer database if Microsoft can't even secure their own Passport enabled site. Ebay already has millions of their own customers. Not only that, but they know that AOL, the single largest ISP in the world, is working on an opposing standard. Microsoft has a Passport database filled with the information of millions of people who signed up for a free service, some of which no longer use the service. AOL, on the other hand, has a customer database that is just as big (if not bigger) of people who get a bill from AOL every month.
Microsoft knows that if their service is seen as insecure, then their customers will go elsewhere. Heck, unless Windows XP really takes off and grows the number of Passport users dramatically Microsoft's customers will probably go elsewhere anyhow (unless AOL really blows it).
Who would have thought I would be rooting for AOL.
Your example is flawed. Listing the names of private citizens who you think should be killed is not protected free speech in any civilized part of the world. However, if he were to put something in the changelog to the effect that abortion was bad (or good) without endangering people's lives or abridging their rights, then that would be different. If he were to want to include scriptures from the Koran in every comment, or wax on at length about the rain forest then that would be fine too.
I find it hard to believe that I have to explain this to "Reality Master 101" but people have differences of opinion. I personally disagree with the politics of quite a few well known hackers, but it doesn't keep me from using their software.
You are right, however, when you state that sometimes people let these unrelated political views cloud their judgement when it comes to technical matters. If Alan Cox worked for me, then I would probably ask him to refrain from similar politics in the future. He doesn't work for me, however, nor you. And truth be told he is a bit of an unprofessional freak. All you need to do is see a picture of him to know that:). Better yet, read his online journal some time.
The kernel development group is a meritocracy. If Alan were to start spending more time politicking than hacking, he would almost certainly get to the point where he had a hard time getting his patches included in the kernel source. That isn't likely to be the case any time soon.
If you are easily offended by such things, then there is almost certainly a Free Software group that would offend your sensibilities. One of the mailing lists I read is populated by developers that all have scriptures in their.sig, another has a heaping helping of homosexuals (nice alliteration, huh). Some hackers are gun nuts, some are ultra-liberal, some don't bathe regularly:). All of these people speak their mind, and all of them have opinions that are likely to converge from yours and mine. Some probably are idiots, but if they are they are software idiot savants.
Either way there is nothing we can do about it without limiting their freedom. They don't work for us, and don't have to abide by our rules to stay employed.
Fah, Kernel developers should do whatever the heck they want, and we should all be grateful. When you start paying Mr. Cox for his work then perhaps you might be able to gain some leverage over how he expresses his opinions. In the meantime, why should anyone listen to you?
Here's an idea. Why don't you code up something half as nifty as whatever Alan will code up over breakfast tomorrow, release it as Free Software for the world to see, and put in your Changelog that Alan is a wanker. Perhaps the people that use your software will care what you have to say.
As far as I am concerned Free Software is a perfectly good medium for espousing your political views. Especially since if you don't like Alan's views you can still use his excellent software without being subject to them.
Hotmail is also the source of all of the passport accounts. Microsoft knows that Windows XP is not going to generate enough Passport accounts to entice web sites to start including Passport hooks. Hotmail, on the other hand, is very popular, and already has millions of users. Besides, if Microsoft can't design a secure Passport site, what is the chance that the bozos at your bank are going to be able to design a secure Passport site?
In other words Hotmail is both the primary draw for Passport, and an important proof of concept. Unfortunately for Microsoft it is also a huge gaping pile of security holes.
The difference is, of course, that when you drink a Coke, it's gone. You can't drink any more Coke until you pay for it again. With software that isn't the case. My version of Word is good for as long as I want to keep using it. If Microsoft were to ever come out with a version of MS Office that was perfect their business would collapse because their customers would have no reason to ever upgrade again.
There is also the fact that software is much less expensive to produce and distribute than Coke. The costs associated with software are mostly R&D. Once the software has been created production and distribution is a piece of cake. In software it is the "recipe" that is worth something, with soft drinks it is the manufacturing capacity and distribution facilities. We'll save that part of the argument for another day and concentrate on the fact that software does not have to be a recurring cost.
You see, Microsoft isn't competing with Star Office, it is competing with old versions of it's own software. Most people are more than happy with Office 97, and even incompatible Office 95 is very prevalent. This is why Microsoft has changed their licensing agreements so that large companies will end up paying for new versions even if they don't have the time and manpower to actually roll them out. Clever huh?
Of course, because Star Office is functionally very similar to MS Office their strategy could very well backfire. If Microsoft is going to continually force their customers to migrate, then it makes sense for Microsoft's customers to at the very least take a look at how much it would cost to migrate to a competitive product. Especially if that competitive product is a Free Software product.
Microsoft's customers certainly aren't interested in Microsoft "features" that render their old documents obsolete, or that require hardware upgrades for their entire organization. So unless Microsoft makes changes that are actually useful making new incompatible formats is more likely to hurt them than help.
What's more, the PC manufacturers aren't particularly interested in selling systems where the hardware (the part they profit from) costs $400 a unit and has razor thin margins, while the software portion costs $500 most of which is pure Microsoft profit. This settlement would allow the PC OEMs to bundle Star Office without any possible repercussion, and it would allow the PC manufacturers to offer a much better value without at a lower cost, and with less of their profits going straight to Microsoft.
So it's not just the end users that would like to see the Free suites win. The fact of the matter is that everyone but Microsoft would be better off if Office Suites were a commodity product.
It will be interesting to see what happens to Microsoft. But even without the DOJ the market has been preparing a response for Microsoft for some time.
That's more that a little simplistic actually because it ignores the fact that most software development is currently proprietary software development. It shouldn't surprise anyone that three quarters of their revenue comes from commercial software developers, that's how business currently gets done. If the day and age ever comes where most people are using Free Software then there will be a great deal more customers for Sleepycat's service and support contracts.
There will always be businesses that are willing to pay for support, customization, fancy dan new extensions, and a whole host of other services. Right now Sleepycat sells software licenses to hardware OEMs, a rather limited field. If Free Software were to become the de-facto standard, then they would certainly see more people interested in their software, many of whom would be willing to pay for their particular expertise.
Besides, the alternatives are far worse. If the day comes that everyone switches to Free Software the commercial software companies will have no revenue. Free Software is going to be developed whether you like it or not. The smart thing to do is to make sure that some of it is yours.
Believe it or not there are those of us who actually prefer Unix desktops. By the time I get my Windows machines up to speed (cygwin, Emacs, Python, Perl, TeX, LaTeX, ghoscript, sgmltools, etc.) I generally find that it would have simply been easier to install Linux. Especially since getting all of these Unixy tools to work together under Windows is a real chore.
Besides, I can't even imagine trying to be productive without virtual screens. I also have a pile of ssh sessions going on, but each group is on a different virtual screen. That way I can easily tell which machine I am looking at.
Putty is nice, but if most of your actual work is done on Unix systems, there is simply so much more that you can do if you are likewise sitting down at a Unix system.
For a while there I was getting worried that I would have to switch just because Mozilla was taking such a long time to get to the point where it was useable. I was a little bit afraid that I would be stranded without a decent web browser. Nowadays, however, I no longer worry. I actually miss Mozilla when I sit down at my Windows machine. I really like having a modern browser that happens to have Emacs key bindings for keys like Ctrl-K, Ctrl-E, or Ctrl-A.
That is so spot on. I spent several years in the late 80s in Peru, and that was definitely the case there. They have plenty of natural resources, a very hard working population, but their system was so corrupt that it was impossible to run a business. Foreign companies especially had to be continuously on their guard for fear of being nationalized.
Why would any sane person invest there?
Especially when they could move a little further South and invest in Chile. I lived in Chile for several years in the early 90s and was surprised at the stark difference. Levels of corruption were much lower, and the people were far more educated (on average). Because of that business was booming.
Unfortunately, the people in power in Peru aren't interested in cleaning things up. After all, they have made millions extorting money from the Peruvian people. It's a sad fact that until the corruption of the Peruvian political system is cleaned up that no amount of money is likely to do the economy there any good.
First of all, this particular bug is not remotely exploitable. In other words the user would already need to be logged onto the machine with a valid account. Now, this does make exploits of Apache or bind or whatever remotely exploitable software you might be running more dangerous, but it certainly isn't nearly as deadly as Microsoft's latest exploits.
Secondly, most Linux installs use one of the "questionable binary kernel downloads" that you malign in your post. This is no different than the binary only kernel you get from Microsoft. Part of the fun of Linux is that you can compile most everything as a module and only load those modules you need. The stock Debian kernel that I am using supports an absolutely ridiculous array of hardware. Anyone that says that Linuxers have to compile their own kernels doesn't know what the heck they are talking about. That hasn't really been the case since before the 2.0 kernels years and years ago.
Heck, I regularly swap hard drives between machines from different manufacturers, Linux doesn't even bat an eye (try doing that with Windows).
Alternatively, Python should mature to the point that people can write a toolkit in Python itself, relying only on drawing primitives from the platform (the same could be said about Perl).
I don't think that this is a particularly good idea. Python is a scripting language, any GUI that relies too much on Python code is bound to be sluggish. That's the real beautty of wxPython (IMHO). It is a very nice wrapper around wxWindows, with a definite Python "feel." The fact that wxWindows is compiled means that WxPython GUIs are amazingly responsive. I ported some test apps from Java/Swing and was amazed at how much faster they ran and how much easier they were to write. WxWindows also uses the native toolkit, so your application appears to fit in with the rest of your environment.
Of course, I am more than a little biased, I personally think that a mixture of Python + C (or in this case C++) is about the perfect development blend. Prototyping is fast and easy, and Python is so easy to extend in C that making the application fast enough is very straightforward.
Anygui sounds like a nice idea, but any abstraction layer generic enough to include ncurses isn't likely to be capable of creating very complex GUIs.
Of course, if they prove me wrong, well I will be pleasantly surprised. I have wrong once or twice before:).
You need to get yourself a login so that someone can mod you up. I definitely agree with you when you say that Windows has gotten better. The home user is certainly getting a lot more bang for their buck with Windows XP than they did with Windows ME, and that's good. Stability is way up, functionality is way up, and price remains about the same.
That's all good.
I personally don't believe that Microsoft has done enough to convince consumers to throw out their old computer, and they certainly aren't going to plunk down $200 for the retail version of the OS. In fact, some of the new features to Windows XP are downright anti-consumer. If Microsoft keeps that up it is likely to be a big help for Linux. The days are over when Microsoft can pretend that Linux is completely useless as a desktop OS. If they squeeze their customers too tight, their customers will migrate away.
That's why Linux is good news for consumers whether or not you like Windows. Linux forces Microsoft to actually compete.
And while Microsoft has the lead in all three of your categories, apps, interface, and hardware support, they certainly don't have the commanding lead that they had 4 years ago, and Linux has the trump card of being a lot less expensive.
I recently built myself a new machine and I was absolutely surprised at how much hardware I could get for $450. Plunking down $200 for an operating system to run on that $50 machine is ridiculous. Especially since Windows, while it has gotten better since Windows 95, is almost completely useless with just an OS. With Linux, an extra $20 bucks gets you a couple of CDs with more useful software than you can shake a stick at, including development tools that would be downright expensive to replace with commercial tools.
And that's just for one computer. Imagine what Michael Dell must think every time he thinks about how much software costs him per machine. Even worse, he has to put up with Microsoft telling him what he can and can't do with that software. It is rapidly getting to the point where the cost of Microsoft software is a serious impediment to hardware sales.
Not that it matters with Windows XP. Microsoft isn't competing with Linux at this point. It is competing with Windows 98. If they can't convince consumers to buy new computers then Microsoft is just as cooked as if everyone had switched to Linux. Just what exactly is Microsoft going to have to give away to get folks to upgrade next time?
And as for.NET, the only really interesting part of that is Passport, and there is no way that is going to fly. First of all AOL is never going to support it (and, in fact, they will be actively opposing it with their own standard). A web site would have to be suicidal to require a technology that excluded AOL. Not to mention all of those folks who don't use Hotmail, and who aren't planning on upgrading to Windows XP.
.NET will be gangbusters on Microsoft sites, but those businesses with a lot of time and effort in their customer databases (like your bank) aren't likely to turn over their crown jewels. And Microsoft's competitors aren't likely to use it either, and nowadays it seems like Microsoft is competing against nearly everyone. This is especially true the closer Microsoft gets to being a "content provider."
Can you honestly show me a single proposed.NET service that you would pay money for?
This isn't closing a door for Linux. It has been possible to get a stable Windows OS for years, people just didn't care. Windows 9X was "good enough" and it was less expensive. Most folks simply weren't willing to pay an extra $100 for stability. Only geeks were willing to switch operating systems for stability. Normal people aren't going to switch until they can buy a computer preloaded with Linux that does everything they need in a computer and costs less than an identical computer running Windows.
For Joe Sixpack it's all about being "good enough" at the lowest price.
Which is why they certainly aren't going to rush out and buy a computer with Windows XP. Sure, some folks have been waiting around for the right time to upgrade, and so some computers will get bought. However, Microsoft will be lucky if PC sales don't continue their precipitous decline. Sales certainly aren't going back to their heyday levels anytime soon.
Which means that prices will continue to drop, and OEMs will continue to try and find some way to get rid of that most expensive of all components, Windows. At the very least they are going to consider bundling StarOffice for Windows. That would raise the total value of their system without raising their price. Once normal people start using StarOffice, and they will eventually, then Linux on the desktop gets that much closer. In the meantime, it's not like Linux is going away.
Microsoft could easily counteract this if they weren't too busy trying to cram stuff their customers don't want into these new versions. None of Microsoft's customers want things like digital rights management or the new activation "feature." Even worse, Microsoft is making it clear that third party developers are not welcome. Add personal firewalls, CD burning software , remote control software, etc. to the list of software products that now have no chance of competing against Microsoft. Perhaps your industry is next. Pundits are quick to point out that it is impossible to make money writing software for Linux, but it would appear that this is increasingly the case for software in general. Unless you are Microsoft, forget about writing desktop software. Third party developers have been a big part of Microsoft's success, but more and more of these folks are starting to take a look at what supporting the Penguin would entail.
In the end Microsoft can't compete against everyone. Right now Microsoft is the lesser of two evils for Windows developers, and they are a necessary evil for the OEMs, but Linux isn't in a hurry. Eventually the sheer value of what it brings to the table (for free) will be too much for these groups to overlook. In the meantime there are plenty of folks that are quite pleased with it. One thing is certain, no matter how cool Windows XP might be, it isn't going to cause the hordes of Linux developers to jump ship.
Key Largo is serving up 400 X-terminals with a commodity Intel-based Linux box. Now, if you had 450 users that spent all of their time compiling software, or rendering large documents with TeX, or something then I would agree, you might need more machine. However, if you are talking about 400 business desktops that spend most of their time in a word processor or reading email, then your clients probably will generate their highest load when their screen saver kicks in.
With the savings that you would realize from using thin clients and the consolidation of hardware you could easily afford to buy a machine specifically built to handle the peak period. This would give you the added bonus of acceptable performance the two weeks before finals, and fabulous performance the rest of the year.
Believe me, getting rid of PC hardware on every desk is a huge administration win. It makes upgrades a piece of cake, it allows for professionally administered desktops (who can say that with PCs), and it allows you to get more done with less staff. People have known this for years, but the problem has been that there hasn't been any software that actually runs on these beasts. Now, largely thanks to Linux, there is a fairly substantial set of software that can be used.
The fact of the matter is that most of the capacity of your average PC goes unused. And when you do need it for a big task it doesn't have the power that you could get from a big server. So even if your average case has a couple of users that are continually rebuilding the Linux kernel one firebreathing server and thin clients could work out in your favor (the folks running their compiles have *the* machine to run them on, and there still is enough horsepower to run everyone else's web browser).
Yes, but you are missing the most important part of the article. The UNIX solution was to put X-terminals on the users desks. Imagine a workplace where you have one machine to administer instead of a hodge podge of PCs all subtly different.
A good admin can make any box sing, but it takes a lot more manpower to keep a pile of desktop PCs running smoothly, even if you are skilled. Heck, just getting rid of those pesky hard drives is a big deal.
Thin clients, my friend, make all of the difference. Properly deployed thin clients make it possible to put a professionally sysadminned computer on everyone's desktop. That's a big deal.
Exactly, right now the "Enterprise" doesn't even hardly use Free Software, and there is still enough money floating around to pay folks like RedHat and Sendmail to keep writing the stuff. Just wait until it is the de-facto standard.
Not to mention the fact that 80% of software developers currently work outside of the software industry. Or, in other words, they work for some company that happens to need software written, but isn't interested in selling it. These folks happen to have a vested interest in quality low cost infrastructure software, and as they start to use Free Software more and more, they will realize that it often makes economic sense to "give back" parts of their code to the community.
Don't expect Free Software to go away anytime soon, it isn't going to, and don't expect it to become less popular, it's not going to happen.
Exactly. Even more telling is the fact that the two authors of this article happen to be living in the Czech Republic. What do you want to bet that they are willing to work for a heck of a lot less than our friends in the Silicon Valley?
My friends in Peru will probably work for even less, and they'll consider it a blessing.
The only way that Microsoft is going to maintain it's current position for 10 years is if they start drastically reducing prices, and start treating their customers like friends and not enemies. The problem with that, of course, is that if Microsoft doesn't keep up their revenues then Wall Street will punish them severely.
Good Luck, Microsoft. Maybe everyone in the world will sign up for the Premium.Net My Services:).
I honestly don't think so. AOL's working on a competing standard, as is Sun and a pile of banks. None of these standards are going to become so prevalent that sites will require them, at least not anytime soon.
The few surviving web business all are smart enough to realize that cutting off AOL customers is a long walk off a short pier. And there is no way that AOL is going to support Passport. Nor are people likely to rush out and replace their current PC with something running Windows XP (the prime driver for.NET). Lastly, it is not particularly likely that end users are going to pay a significant amount of money to have their computers automatically fill in web forms. Especially if this feature only works on a handful of sites.
After all, most folks already have a relationship with all of the sites that they currently visit. Does Microsoft honestly think that people are going to pay to have them manage this information? Even worse, does Microsoft actually think that organizations (like your bank) which have spent a great deal of time and effort on their customer databases are going to turn these crown jewels over to Microsoft? And for what?
Microsoft is desperate to create a bottleneck for the Internet so that they can create a tollbooth their (in much the same way that you pay your ISP for service). My guess is that they are going to have a hard time finding enough people, both users and developers, to make this a de-facto standard. Especially if one of the other authentication standards can be bundled with a service you already pay for (AOL would like your ISP to provide the service).
This doesn't even take into account the very real possibility that Microsoft will have a well publicised security breach, or that they won't be able to provide a reliable enough service for it to be useful. If either of those things happens, Passport will be finished for serious (read, your paying extra for it) use.
You have got to hand it to Microsoft, then have serious chutzpah, but they aren't likely to pull it off (IMHO).
Precisely. Microsoft wants getting viruses criminalized using the new set of "lock them up forever" terrorism rules. Then, once they have accomplished this, they will go after the security community and force them to stop pointing out security problems. At the very least they will force them to stop releasing example exploits.
Microsoft knows that these changes aren't likely to actually have much of an effect on the actual virus writers. But they know that it will have a profound effect on those people that are actively trying to secure our software.
You have got to hand it to Microsoft. I thought for sure that their recent spouting about the security community was nothing more than hot air. It would appear that the rules in the security industry are about to change.
Those are good points. I have worked tech support, and I agree that what you say is true. In fact, due to the flimsy nature of the rest of the article I must admit that I find this story hard to swallow. After all, I know for myself that Applixware's spell checker, while probably not the best example of a spell checker ever created, certainly worked well enough for anyone that writes for a living. Any user that wanted to switch from the company standard to something else with that sort of a flimsy excuse would get little sympathy and less support from me.
So no, I am not convinced, but even if that part of the story was the literal truth, it still wouldn't lead to the conclusion that the writer proffered. The capabilities of Applixware as an office suite, or the actions of one lowly tech working at RedHat, are completely irrelevant when trying to argue that Linux isn't ready for the desktop. Applixware isn't the best office suite available for Linux, and the actions of one Linux user, no matter how egregious, do not change the fact that Linux has plenty of useful tools.
In fact, for web publishing, it could easily be argued that it has better tools than the competition.
Which is why I think that the article in question was a very poor one indeed. Even if you accept that the author isn't inventing the story from scratch, it doesn't make his premise true (and I personally don't buy the story).
A lot of us would pay for that kind of serious support. This is partially why I think that Free Software is likely to do well in the Enterprise in the long term.
You see, currently there is little reason for a software manufacturer like Microsoft to provide exceptional support. After all, they know that no matter how bad their support is, you can't really get better support from someone else (only Microsoft has access to the source). In fact, the best you can do is sign up with a support vendor that is big enough so that when they complain on your behalf you have some chance of having Microsoft care. This is very expensive, and far from full proof.
The only strategy that is even somewhat effective is to threaten to switch to a competing product. However, this means that you have to absorb the cost of migrating your data, and you lose the sunk costs that you poured into acquiring the software in the first place. As you can guess, this is hardly a winning strategy, and is generally only useful as a last resort. Good competing products generally try and make this migration as painless as possible, but there are always problems. Still, if your vendor is really bad it may be your only option.
With Free Software, on the other hand, if you don't like your support vendor, you can hire a different one. Since everyone has the source, you can pick and choose the vendor that has the correct expertise at a price you can afford. More importantly, since these support vendors would know that you could switch vendors easily, and since they have the source, you could almost certainly persuade them to actually fix the software.
Slashdot Mirror
There already is a Gnus client for Slashdot. It allows you to use Gnus scorefiles and everything. Sometimes I wonder if Emacs hackers belong to a totally different species.
Now, now, I wouldn't go that far. While I don't particularly like to drive SUVs in the snow (for the reasons I stated above), I wouldn't go so far as to say that they have no place on the road.
Sometimes you actually do need a vehicle that will hold a large number of passengers over rough terrain. For example, I am a Scout Master, and I must admit that my buddy's SUV has come in handy a couple of times taking the scouts camping.
Personally, I prefer the lower cost and better handling and gas mileage that I get from my mini-van, but that's just me. After all, scouts can walk, and making them pack their gear in means less of a mess to clean up afterwards. I can generally get close enough to the campsite so that walking is feasible.
Of course, I wouldn't even pretend to dictate how others spend their money. If they want to pay for an SUV, well, that's fine with me. Just don't drive fast in a snowstorm.
Large four wheel drive vehicles are certainly helpful if you are plowing through snow drifts (you'll probably still get stuck, but you'll have fun doing it), but they are next to useless if the road is merely slick. If the road is merely slick you are far better off driving a smaller vehicle. This is because when the road is merely slick the problem isn't getting stuck, it's steering and stopping. Your SUV has a lot more mass and momentum than my Honda Civic, without much more contact surface area. I have lost track of the times where I have had some bozo in a SUV pass me in a snow storm only to see him in a ditch further up the road. And don't even get me started on Jeep Wrangler's (yes, I know they aren't an SUV) those short boxes are murder on slick roads.
Now, if by SUV you also include some of the 4 wheel drive suburu hatchbacks and station wagons then I agree. Those are quite good in the snow.
For those of you that don't know how MathWorld disappeared here's the whole story: http://www.mathworld.com/erics_commentary.html
Well then, I agree with you :). Linux kernel developers should deport themselves with as much dignity and aplomb as possible. You are certainly correct that political opinions, even ones as relevant as this one, can certainly reflect badly on Linux in the eyes of some folks.
On the other hand, I can't help but think that Alan Cox deserves to be able to use the Linux kernel as a soapbox. After all, who knows if it would even exist without his tireless efforts. He almost certainly knew that some people wouldn't appreciate his statement, but apparently it meant enough to him to make it anyhow.
Linux Kernel hackers do what they do for the fun and the fame. An unfortunate side effect of this is that it creates a platform for opinionated, outspoken, "unprofessional" Unix hackers. Being able to speak out about issues like the DMCA and be heard is part of the reason that people hack Free Software. It's also a major difference between working on Linux and working on a commercial operating system. The guys hacking the Windows kernel get paid more than Alan, but hardly anyone knows who they are. Taking away Alan's right to use (and even abuse) his "fame" would make Alan both underpaid and under motivated.
Besides, he was right :). He could have been a little more professional, but apparently it is very important to him.
Yes, I realize that, but Hotmail is the primary draw for Passport. How many Passport users would there be without Hotmail? Probably not very many. Instead of millions of users Microsoft would be lucky to have tens of thousands of users.
Getting Ebay, Prudential, and 68 other companies to modify their websites to cater to such a small group of people would never fly. Microsoft needs to be able to wave the carrot of millions of potential customers, or it isn't worth the time.
Besides, even with millions of customers why should Ebay trust what Microsoft to store their important customer database if Microsoft can't even secure their own Passport enabled site. Ebay already has millions of their own customers. Not only that, but they know that AOL, the single largest ISP in the world, is working on an opposing standard. Microsoft has a Passport database filled with the information of millions of people who signed up for a free service, some of which no longer use the service. AOL, on the other hand, has a customer database that is just as big (if not bigger) of people who get a bill from AOL every month.
Microsoft knows that if their service is seen as insecure, then their customers will go elsewhere. Heck, unless Windows XP really takes off and grows the number of Passport users dramatically Microsoft's customers will probably go elsewhere anyhow (unless AOL really blows it).
Who would have thought I would be rooting for AOL.
Your example is flawed. Listing the names of private citizens who you think should be killed is not protected free speech in any civilized part of the world. However, if he were to put something in the changelog to the effect that abortion was bad (or good) without endangering people's lives or abridging their rights, then that would be different. If he were to want to include scriptures from the Koran in every comment, or wax on at length about the rain forest then that would be fine too.
I find it hard to believe that I have to explain this to "Reality Master 101" but people have differences of opinion. I personally disagree with the politics of quite a few well known hackers, but it doesn't keep me from using their software.
You are right, however, when you state that sometimes people let these unrelated political views cloud their judgement when it comes to technical matters. If Alan Cox worked for me, then I would probably ask him to refrain from similar politics in the future. He doesn't work for me, however, nor you. And truth be told he is a bit of an unprofessional freak. All you need to do is see a picture of him to know that :). Better yet, read his online journal some time.
The kernel development group is a meritocracy. If Alan were to start spending more time politicking than hacking, he would almost certainly get to the point where he had a hard time getting his patches included in the kernel source. That isn't likely to be the case any time soon.
If you are easily offended by such things, then there is almost certainly a Free Software group that would offend your sensibilities. One of the mailing lists I read is populated by developers that all have scriptures in their .sig, another has a heaping helping of homosexuals (nice alliteration, huh). Some hackers are gun nuts, some are ultra-liberal, some don't bathe regularly :). All of these people speak their mind, and all of them have opinions that are likely to converge from yours and mine. Some probably are idiots, but if they are they are software idiot savants.
Either way there is nothing we can do about it without limiting their freedom. They don't work for us, and don't have to abide by our rules to stay employed.
Fah, Kernel developers should do whatever the heck they want, and we should all be grateful. When you start paying Mr. Cox for his work then perhaps you might be able to gain some leverage over how he expresses his opinions. In the meantime, why should anyone listen to you?
Here's an idea. Why don't you code up something half as nifty as whatever Alan will code up over breakfast tomorrow, release it as Free Software for the world to see, and put in your Changelog that Alan is a wanker. Perhaps the people that use your software will care what you have to say.
As far as I am concerned Free Software is a perfectly good medium for espousing your political views. Especially since if you don't like Alan's views you can still use his excellent software without being subject to them.
Hotmail is also the source of all of the passport accounts. Microsoft knows that Windows XP is not going to generate enough Passport accounts to entice web sites to start including Passport hooks. Hotmail, on the other hand, is very popular, and already has millions of users. Besides, if Microsoft can't design a secure Passport site, what is the chance that the bozos at your bank are going to be able to design a secure Passport site?
In other words Hotmail is both the primary draw for Passport, and an important proof of concept. Unfortunately for Microsoft it is also a huge gaping pile of security holes.
The difference is, of course, that when you drink a Coke, it's gone. You can't drink any more Coke until you pay for it again. With software that isn't the case. My version of Word is good for as long as I want to keep using it. If Microsoft were to ever come out with a version of MS Office that was perfect their business would collapse because their customers would have no reason to ever upgrade again.
There is also the fact that software is much less expensive to produce and distribute than Coke. The costs associated with software are mostly R&D. Once the software has been created production and distribution is a piece of cake. In software it is the "recipe" that is worth something, with soft drinks it is the manufacturing capacity and distribution facilities. We'll save that part of the argument for another day and concentrate on the fact that software does not have to be a recurring cost.
You see, Microsoft isn't competing with Star Office, it is competing with old versions of it's own software. Most people are more than happy with Office 97, and even incompatible Office 95 is very prevalent. This is why Microsoft has changed their licensing agreements so that large companies will end up paying for new versions even if they don't have the time and manpower to actually roll them out. Clever huh?
Of course, because Star Office is functionally very similar to MS Office their strategy could very well backfire. If Microsoft is going to continually force their customers to migrate, then it makes sense for Microsoft's customers to at the very least take a look at how much it would cost to migrate to a competitive product. Especially if that competitive product is a Free Software product.
Microsoft's customers certainly aren't interested in Microsoft "features" that render their old documents obsolete, or that require hardware upgrades for their entire organization. So unless Microsoft makes changes that are actually useful making new incompatible formats is more likely to hurt them than help.
What's more, the PC manufacturers aren't particularly interested in selling systems where the hardware (the part they profit from) costs $400 a unit and has razor thin margins, while the software portion costs $500 most of which is pure Microsoft profit. This settlement would allow the PC OEMs to bundle Star Office without any possible repercussion, and it would allow the PC manufacturers to offer a much better value without at a lower cost, and with less of their profits going straight to Microsoft.
So it's not just the end users that would like to see the Free suites win. The fact of the matter is that everyone but Microsoft would be better off if Office Suites were a commodity product.
It will be interesting to see what happens to Microsoft. But even without the DOJ the market has been preparing a response for Microsoft for some time.
That's more that a little simplistic actually because it ignores the fact that most software development is currently proprietary software development. It shouldn't surprise anyone that three quarters of their revenue comes from commercial software developers, that's how business currently gets done. If the day and age ever comes where most people are using Free Software then there will be a great deal more customers for Sleepycat's service and support contracts.
There will always be businesses that are willing to pay for support, customization, fancy dan new extensions, and a whole host of other services. Right now Sleepycat sells software licenses to hardware OEMs, a rather limited field. If Free Software were to become the de-facto standard, then they would certainly see more people interested in their software, many of whom would be willing to pay for their particular expertise.
Besides, the alternatives are far worse. If the day comes that everyone switches to Free Software the commercial software companies will have no revenue. Free Software is going to be developed whether you like it or not. The smart thing to do is to make sure that some of it is yours.
Believe it or not there are those of us who actually prefer Unix desktops. By the time I get my Windows machines up to speed (cygwin, Emacs, Python, Perl, TeX, LaTeX, ghoscript, sgmltools, etc.) I generally find that it would have simply been easier to install Linux. Especially since getting all of these Unixy tools to work together under Windows is a real chore.
Besides, I can't even imagine trying to be productive without virtual screens. I also have a pile of ssh sessions going on, but each group is on a different virtual screen. That way I can easily tell which machine I am looking at.
Putty is nice, but if most of your actual work is done on Unix systems, there is simply so much more that you can do if you are likewise sitting down at a Unix system.
For a while there I was getting worried that I would have to switch just because Mozilla was taking such a long time to get to the point where it was useable. I was a little bit afraid that I would be stranded without a decent web browser. Nowadays, however, I no longer worry. I actually miss Mozilla when I sit down at my Windows machine. I really like having a modern browser that happens to have Emacs key bindings for keys like Ctrl-K, Ctrl-E, or Ctrl-A.
That makes me very happy.
That is so spot on. I spent several years in the late 80s in Peru, and that was definitely the case there. They have plenty of natural resources, a very hard working population, but their system was so corrupt that it was impossible to run a business. Foreign companies especially had to be continuously on their guard for fear of being nationalized.
Why would any sane person invest there?
Especially when they could move a little further South and invest in Chile. I lived in Chile for several years in the early 90s and was surprised at the stark difference. Levels of corruption were much lower, and the people were far more educated (on average). Because of that business was booming.
Unfortunately, the people in power in Peru aren't interested in cleaning things up. After all, they have made millions extorting money from the Peruvian people. It's a sad fact that until the corruption of the Peruvian political system is cleaned up that no amount of money is likely to do the economy there any good.
First of all, this particular bug is not remotely exploitable. In other words the user would already need to be logged onto the machine with a valid account. Now, this does make exploits of Apache or bind or whatever remotely exploitable software you might be running more dangerous, but it certainly isn't nearly as deadly as Microsoft's latest exploits.
Secondly, most Linux installs use one of the "questionable binary kernel downloads" that you malign in your post. This is no different than the binary only kernel you get from Microsoft. Part of the fun of Linux is that you can compile most everything as a module and only load those modules you need. The stock Debian kernel that I am using supports an absolutely ridiculous array of hardware. Anyone that says that Linuxers have to compile their own kernels doesn't know what the heck they are talking about. That hasn't really been the case since before the 2.0 kernels years and years ago.
Heck, I regularly swap hard drives between machines from different manufacturers, Linux doesn't even bat an eye (try doing that with Windows).
I don't think that this is a particularly good idea. Python is a scripting language, any GUI that relies too much on Python code is bound to be sluggish. That's the real beautty of wxPython (IMHO). It is a very nice wrapper around wxWindows, with a definite Python "feel." The fact that wxWindows is compiled means that WxPython GUIs are amazingly responsive. I ported some test apps from Java/Swing and was amazed at how much faster they ran and how much easier they were to write. WxWindows also uses the native toolkit, so your application appears to fit in with the rest of your environment.
Of course, I am more than a little biased, I personally think that a mixture of Python + C (or in this case C++) is about the perfect development blend. Prototyping is fast and easy, and Python is so easy to extend in C that making the application fast enough is very straightforward.
Anygui sounds like a nice idea, but any abstraction layer generic enough to include ncurses isn't likely to be capable of creating very complex GUIs.
Of course, if they prove me wrong, well I will be pleasantly surprised. I have wrong once or twice before :).
You need to get yourself a login so that someone can mod you up. I definitely agree with you when you say that Windows has gotten better. The home user is certainly getting a lot more bang for their buck with Windows XP than they did with Windows ME, and that's good. Stability is way up, functionality is way up, and price remains about the same.
That's all good.
I personally don't believe that Microsoft has done enough to convince consumers to throw out their old computer, and they certainly aren't going to plunk down $200 for the retail version of the OS. In fact, some of the new features to Windows XP are downright anti-consumer. If Microsoft keeps that up it is likely to be a big help for Linux. The days are over when Microsoft can pretend that Linux is completely useless as a desktop OS. If they squeeze their customers too tight, their customers will migrate away.
That's why Linux is good news for consumers whether or not you like Windows. Linux forces Microsoft to actually compete.
And while Microsoft has the lead in all three of your categories, apps, interface, and hardware support, they certainly don't have the commanding lead that they had 4 years ago, and Linux has the trump card of being a lot less expensive.
I recently built myself a new machine and I was absolutely surprised at how much hardware I could get for $450. Plunking down $200 for an operating system to run on that $50 machine is ridiculous. Especially since Windows, while it has gotten better since Windows 95, is almost completely useless with just an OS. With Linux, an extra $20 bucks gets you a couple of CDs with more useful software than you can shake a stick at, including development tools that would be downright expensive to replace with commercial tools.
And that's just for one computer. Imagine what Michael Dell must think every time he thinks about how much software costs him per machine. Even worse, he has to put up with Microsoft telling him what he can and can't do with that software. It is rapidly getting to the point where the cost of Microsoft software is a serious impediment to hardware sales.
Not that it matters with Windows XP. Microsoft isn't competing with Linux at this point. It is competing with Windows 98. If they can't convince consumers to buy new computers then Microsoft is just as cooked as if everyone had switched to Linux. Just what exactly is Microsoft going to have to give away to get folks to upgrade next time?
And as for .NET, the only really interesting part of that is Passport, and there is no way that is going to fly. First of all AOL is never going to support it (and, in fact, they will be actively opposing it with their own standard). A web site would have to be suicidal to require a technology that excluded AOL. Not to mention all of those folks who don't use Hotmail, and who aren't planning on upgrading to Windows XP.
.NET will be gangbusters on Microsoft sites, but those businesses with a lot of time and effort in their customer databases (like your bank) aren't likely to turn over their crown jewels. And Microsoft's competitors aren't likely to use it either, and nowadays it seems like Microsoft is competing against nearly everyone. This is especially true the closer Microsoft gets to being a "content provider."
Can you honestly show me a single proposed .NET service that you would pay money for?
This isn't closing a door for Linux. It has been possible to get a stable Windows OS for years, people just didn't care. Windows 9X was "good enough" and it was less expensive. Most folks simply weren't willing to pay an extra $100 for stability. Only geeks were willing to switch operating systems for stability. Normal people aren't going to switch until they can buy a computer preloaded with Linux that does everything they need in a computer and costs less than an identical computer running Windows.
For Joe Sixpack it's all about being "good enough" at the lowest price.
Which is why they certainly aren't going to rush out and buy a computer with Windows XP. Sure, some folks have been waiting around for the right time to upgrade, and so some computers will get bought. However, Microsoft will be lucky if PC sales don't continue their precipitous decline. Sales certainly aren't going back to their heyday levels anytime soon.
Which means that prices will continue to drop, and OEMs will continue to try and find some way to get rid of that most expensive of all components, Windows. At the very least they are going to consider bundling StarOffice for Windows. That would raise the total value of their system without raising their price. Once normal people start using StarOffice, and they will eventually, then Linux on the desktop gets that much closer. In the meantime, it's not like Linux is going away.
Microsoft could easily counteract this if they weren't too busy trying to cram stuff their customers don't want into these new versions. None of Microsoft's customers want things like digital rights management or the new activation "feature." Even worse, Microsoft is making it clear that third party developers are not welcome. Add personal firewalls, CD burning software , remote control software, etc. to the list of software products that now have no chance of competing against Microsoft. Perhaps your industry is next. Pundits are quick to point out that it is impossible to make money writing software for Linux, but it would appear that this is increasingly the case for software in general. Unless you are Microsoft, forget about writing desktop software. Third party developers have been a big part of Microsoft's success, but more and more of these folks are starting to take a look at what supporting the Penguin would entail.
In the end Microsoft can't compete against everyone. Right now Microsoft is the lesser of two evils for Windows developers, and they are a necessary evil for the OEMs, but Linux isn't in a hurry. Eventually the sheer value of what it brings to the table (for free) will be too much for these groups to overlook. In the meantime there are plenty of folks that are quite pleased with it. One thing is certain, no matter how cool Windows XP might be, it isn't going to cause the hordes of Linux developers to jump ship.
Key Largo is serving up 400 X-terminals with a commodity Intel-based Linux box. Now, if you had 450 users that spent all of their time compiling software, or rendering large documents with TeX, or something then I would agree, you might need more machine. However, if you are talking about 400 business desktops that spend most of their time in a word processor or reading email, then your clients probably will generate their highest load when their screen saver kicks in.
With the savings that you would realize from using thin clients and the consolidation of hardware you could easily afford to buy a machine specifically built to handle the peak period. This would give you the added bonus of acceptable performance the two weeks before finals, and fabulous performance the rest of the year.
Believe me, getting rid of PC hardware on every desk is a huge administration win. It makes upgrades a piece of cake, it allows for professionally administered desktops (who can say that with PCs), and it allows you to get more done with less staff. People have known this for years, but the problem has been that there hasn't been any software that actually runs on these beasts. Now, largely thanks to Linux, there is a fairly substantial set of software that can be used.
The fact of the matter is that most of the capacity of your average PC goes unused. And when you do need it for a big task it doesn't have the power that you could get from a big server. So even if your average case has a couple of users that are continually rebuilding the Linux kernel one firebreathing server and thin clients could work out in your favor (the folks running their compiles have *the* machine to run them on, and there still is enough horsepower to run everyone else's web browser).
Yes, but you are missing the most important part of the article. The UNIX solution was to put X-terminals on the users desks. Imagine a workplace where you have one machine to administer instead of a hodge podge of PCs all subtly different.
A good admin can make any box sing, but it takes a lot more manpower to keep a pile of desktop PCs running smoothly, even if you are skilled. Heck, just getting rid of those pesky hard drives is a big deal.
Thin clients, my friend, make all of the difference. Properly deployed thin clients make it possible to put a professionally sysadminned computer on everyone's desktop. That's a big deal.
Exactly, right now the "Enterprise" doesn't even hardly use Free Software, and there is still enough money floating around to pay folks like RedHat and Sendmail to keep writing the stuff. Just wait until it is the de-facto standard.
Not to mention the fact that 80% of software developers currently work outside of the software industry. Or, in other words, they work for some company that happens to need software written, but isn't interested in selling it. These folks happen to have a vested interest in quality low cost infrastructure software, and as they start to use Free Software more and more, they will realize that it often makes economic sense to "give back" parts of their code to the community.
Don't expect Free Software to go away anytime soon, it isn't going to, and don't expect it to become less popular, it's not going to happen.
Exactly. Even more telling is the fact that the two authors of this article happen to be living in the Czech Republic. What do you want to bet that they are willing to work for a heck of a lot less than our friends in the Silicon Valley?
My friends in Peru will probably work for even less, and they'll consider it a blessing.
The only way that Microsoft is going to maintain it's current position for 10 years is if they start drastically reducing prices, and start treating their customers like friends and not enemies. The problem with that, of course, is that if Microsoft doesn't keep up their revenues then Wall Street will punish them severely.
Good Luck, Microsoft. Maybe everyone in the world will sign up for the Premium .Net My Services :).
I honestly don't think so. AOL's working on a competing standard, as is Sun and a pile of banks. None of these standards are going to become so prevalent that sites will require them, at least not anytime soon.
The few surviving web business all are smart enough to realize that cutting off AOL customers is a long walk off a short pier. And there is no way that AOL is going to support Passport. Nor are people likely to rush out and replace their current PC with something running Windows XP (the prime driver for .NET). Lastly, it is not particularly likely that end users are going to pay a significant amount of money to have their computers automatically fill in web forms. Especially if this feature only works on a handful of sites.
After all, most folks already have a relationship with all of the sites that they currently visit. Does Microsoft honestly think that people are going to pay to have them manage this information? Even worse, does Microsoft actually think that organizations (like your bank) which have spent a great deal of time and effort on their customer databases are going to turn these crown jewels over to Microsoft? And for what?
Microsoft is desperate to create a bottleneck for the Internet so that they can create a tollbooth their (in much the same way that you pay your ISP for service). My guess is that they are going to have a hard time finding enough people, both users and developers, to make this a de-facto standard. Especially if one of the other authentication standards can be bundled with a service you already pay for (AOL would like your ISP to provide the service).
This doesn't even take into account the very real possibility that Microsoft will have a well publicised security breach, or that they won't be able to provide a reliable enough service for it to be useful. If either of those things happens, Passport will be finished for serious (read, your paying extra for it) use.
You have got to hand it to Microsoft, then have serious chutzpah, but they aren't likely to pull it off (IMHO).
Precisely. Microsoft wants getting viruses criminalized using the new set of "lock them up forever" terrorism rules. Then, once they have accomplished this, they will go after the security community and force them to stop pointing out security problems. At the very least they will force them to stop releasing example exploits.
Microsoft knows that these changes aren't likely to actually have much of an effect on the actual virus writers. But they know that it will have a profound effect on those people that are actively trying to secure our software.
You have got to hand it to Microsoft. I thought for sure that their recent spouting about the security community was nothing more than hot air. It would appear that the rules in the security industry are about to change.
Those are good points. I have worked tech support, and I agree that what you say is true. In fact, due to the flimsy nature of the rest of the article I must admit that I find this story hard to swallow. After all, I know for myself that Applixware's spell checker, while probably not the best example of a spell checker ever created, certainly worked well enough for anyone that writes for a living. Any user that wanted to switch from the company standard to something else with that sort of a flimsy excuse would get little sympathy and less support from me.
So no, I am not convinced, but even if that part of the story was the literal truth, it still wouldn't lead to the conclusion that the writer proffered. The capabilities of Applixware as an office suite, or the actions of one lowly tech working at RedHat, are completely irrelevant when trying to argue that Linux isn't ready for the desktop. Applixware isn't the best office suite available for Linux, and the actions of one Linux user, no matter how egregious, do not change the fact that Linux has plenty of useful tools.
In fact, for web publishing, it could easily be argued that it has better tools than the competition.
Which is why I think that the article in question was a very poor one indeed. Even if you accept that the author isn't inventing the story from scratch, it doesn't make his premise true (and I personally don't buy the story).
A lot of us would pay for that kind of serious support. This is partially why I think that Free Software is likely to do well in the Enterprise in the long term.
You see, currently there is little reason for a software manufacturer like Microsoft to provide exceptional support. After all, they know that no matter how bad their support is, you can't really get better support from someone else (only Microsoft has access to the source). In fact, the best you can do is sign up with a support vendor that is big enough so that when they complain on your behalf you have some chance of having Microsoft care. This is very expensive, and far from full proof.
The only strategy that is even somewhat effective is to threaten to switch to a competing product. However, this means that you have to absorb the cost of migrating your data, and you lose the sunk costs that you poured into acquiring the software in the first place. As you can guess, this is hardly a winning strategy, and is generally only useful as a last resort. Good competing products generally try and make this migration as painless as possible, but there are always problems. Still, if your vendor is really bad it may be your only option.
With Free Software, on the other hand, if you don't like your support vendor, you can hire a different one. Since everyone has the source, you can pick and choose the vendor that has the correct expertise at a price you can afford. More importantly, since these support vendors would know that you could switch vendors easily, and since they have the source, you could almost certainly persuade them to actually fix the software.