Yeah, that makes sense. However, it's not too difficult to set up an X terminal using the formual in the article. Heck, I wish that I would have thought about using Debian base system as the base for my X terminals. I rolled my own, and it was quite painful.
As for your problem trusting your firewall. If your attacker is sniffing packets on your local area network you have bigger problems than the fact that they might eavesdrop on your X session. If that's the case then one of your devices has already been compromised.
I did something similar with a 486/33 (20M ram) and a Cyrix P150+ (64M ram) without ssh, and it worked great. It wasn't slow at all. In fact, it was so much like actually sitting down at the Cyrix that it was spooky. My guess is that the latency added by compressing (and encrypting) your session killed your performance. Compression really only helps over low bandwidth links. If you aren't maxing out your pipe, you shouldn't be compressing your traffic.
I don't suppose you ran top on your 486 to see how badly your processor was swamped. My 486 would get to 100% processor utilization just running X. With the compression and encryption my guess is that you just asked to much of your client.
Exactly. X allows me to use my new AMD box out in my fort from my crusty old Cyrix P150+ in the bedroom. It honestly is better than having two computers for the price of one, because the X terminal is much easier to maintain than a second PC. And since the Cyrix machine has a pretty good video card in it I honestly can't hardly tell the difference between sitting down at my actual machine. I even have gotten sound to work (with a little help from Gnome and esd).
Yes, if you want to fork the code base you can. If you want your source code to end up in their CVS repository, on the other hand, you will assign your copyrights to Sun.
This is not necessarily a bad thing, however. The Free Software Foundation requires the same thing if you plan on submitting code to their projects. They do this to make sure that they can prosecute copyright infringers (only the copyright holder can prosecute), and to make sure that all of the source is legal to distribute.
In Sun's case it also means that they can re-license your code under non-free licenses. Not that this matters to you. You still have your copy of the source, and you can do what you want with it (including creating a fork). You can't sell closed source versions of the software, however, because Sun owns a large part of the copyrighted material.
In essence this sort of deal gives the originator of the code a slight advantage. Sun could, if they wanted, sell commercial versions of StarOffice with proprietary add-ons not found in OpenOffice. However, if they spent to much time on the proprietary add-ons then their userbase would probably fork their codebase, and Sun would lose the help of the Free Software community (and probably the bulk of their users).
Believe it or not their are business reasons for using the GPL.
I suppose they would do it for the same reason that Netscape rebrands Mozilla as Netscape. Marketing folks think that changing the name on the free product lends the non-free (yes, both Netscape and StarOffice will be released under commercial closed source licenses) product some dignity.
Personally I think that this particular trick has failed miserably. Mozilla is good enough to actually use as a browser, Netscape, on the other hand sucks. Right now StarOffice is certainly more stable and useable than OpenOffice, but once that is no longer the case I imagine that people will simply use OpenOffice (although your Sparcs will probably come with a copy of StarOffice).
All the better. Microsoft doesn't want a quality port of their CLI. They want something that barely works. Heck, they want something even more useless than the token "POSIX" Windows NT layer. Microsoft is just spreading more FUD about Linux and trying to stir things up in the Free Software community. They don't actually want FreeBSDers to be able to run.NET servers on their boxen, that would be competition.
Yes, that is true. StarOffice most certainly did come first. That wasn't at all clear from my post.
What I should have said was that StarOffice 6.0 will be the re-branded OpenOffice.
And Open Office really is getting better. For one thing it no longer relies on the goofy Xprint extensions to print (it can generate Postscript). I actually have been playing with the recent builds and they are getting to be pretty darn good (although soffice still takes forever to start).
I also agree with you about AbiWord and Gnumeric. The Gnome Office suite is getting to be pretty good. Either way it is nice to see an organization standardize on something besides MS Office. The fact that they are saving us taxpayers some money is also good news.
OpenOffice is the GPLed version, StarOffice is Sun's rebranded version of the same thing. Part of the fun with OpenOffice is that all hackers assign their copyrights to Sun. This allows them to relicense StarOffice as they please.
It's sort of like the difference between Netscape and Mozilla.
Just for the record, I personally think that this sort of licensing is good for all involved. Sun remains in control of their project, unless they are evil. In which case OpenOffice becomes the de-facto standard and StarOffice disappears. If Sun plays nice, everyone wins.
Whatever. Almost all Linux software nowadays comes as an RPM, and that RPM is almost always targetted at RedHat's latest distribution. Sometimes if a packager has really done his or her homework they will give instructions on how to force install on SuSE or Caldera. Now, I don't use RedHat anymore (I am a Debian user myself), but I am sure that PHP works fine (although they may still be using the GPLed version PHP3 instead of PHP4, I don't know).
The Python issue is similar. It's not RedHat's fault that Guido and Co. screwed up the licensing for Python 1.6 to 2.0. The fact that RedHat pays attention to these sorts of licensing problems is actually a good thing as it means that I don't have to hire a lawyer to do it for me. Python is a very popular "embedded" scripting language. The fact that several versions of Python were not GPL compatible was a big concern. RedHat did the only sane thing by not including it by default (you could still get Python2 rpms, I have seen them).
Debian did something very similar, and for very similar reasons. These organizations understand that legal issues like this are a big deal, even if they seem to be trivial.
The fact of the matter is that RedHat is on top because they concentrated on making Linux easy to use and then released their source. While SuSE and Caldera were busy trying to herd their users into proprietary Linux add-ons RedHat was becoming the de-facto standard by releasing software. RedHat's installers weren't as flashy, and their admin tools were second rate, but you could fix them, and you didn't have to worry about licensing fees or vendor lock. While Debian was impossible to install, and Slackware was aiming for the tarball elitists, RedHat was trying to make a Free operating system that normal people could use.
So now, like it or not, RedHat is the gold standard for Linux. Applications that don't run well on RedHat get fixed (perhaps by releasing RPMs of the necessary auxilarly packages, but they get fixed all the same). Commercial Applications that don't run well on Caldera (but do run well on RedHat) come with a free RedHat CD. It's as simple as that.
This is not about RedHat flag-waving. I don't even use their distribution, but I do recognize suicide when I see it. Caldera charging per seat licenses is suicide, plain and simple. They are going to drive their few paying customers to someone else (probably RedHat). Sure, Caldera has got some nifty features, but it has always had nicer features than RedHat, and it never has helped. Linux users would just as soon have a poor piece of free software that they can fix over a fancy piece of commercial software that they can't. You would think that Caldera would have learned this by now. They have watched their fancy commercial applications get upstaged by RedHat's wimpy free ones time and time again. And RedHat hasn't always been bigger than Caldera either. Back when Slackware was the distribution of choice Caldera's desktop was a hundred times better than RedHat (and they both used the precursor to RPM as their packaging manager, Love is always pointing out the fact that Caldera helped develop RPM). RedHat was free, Caldera was not. Caldera lost. And they will lose again with their newest hare brained scheme. They are practically driving the industry towards RedHat.
Yes, that's true, he probably will be recycled. But at least no one will take his word as an industry insider. When our PHB asks us about the Ransom Love column in ComputerWorld we'll be able to laugh and say:
"That's just some bozo that was CEO of the wacky Linux company that thought buying SCO was a good idea. He burned through millions of dollars in capital, and yet his distribution wasn't even as popular as a distribution created entirely by volunteers."
Quite frankly, even though Caldera is still kicking Ransom doesn't get much play as a Linux spokesperson. The only folks that ever pay any attention to him at all are the rabidly anti-Linux journalists, and they just trot him out to make Linux look bad. And now that RedHat actually has made a swipe at positive cash flow, he won't be able to say that making money with RedHat's business plan is impossible. If making money from Linux requires proprietary software and license fees then how come his company is still deep in the red while RedHat is in the black?
I don't have moderator points, and can't moderate in this thread anyhow, but this response deserves to be seen...
OK. Here's a comment from someone who helped test the first version of the Caldera distribution. Remember it? Caldera Network Desktop. It was (mostly) based on RedHat mothers-day (either 1.0 or 1.1 -- but RPP-based, no RPM systems yet!). For the record, I did buy CND 1.0 -- because it was the only way to get a copy of WP for Linux (6.0) at that time.
Here's what I got -- RH 1.x (re-branded to Caldera, with some updates, while RH was working on 2.x), and a bunch of proprietary software. Bearing in mind that this was a LONG time ago, these were useful:
Gallium font server (let you use TrueType fonts).
Looking Glass desktop (it was that, or fvwm -- mc was still an infant, and there wasn't a decent desktop project yet).
NetWare utilities (mars_nwe? HAH! ncpfs? not yet...) -- although they didn't support NW4 yet -- it was too new.
Oh, and having that license let me purchase WP for Linux 6.0. I wonder how many they actually sold...
The transition RH made with 2.0 to RPM shook up the Caldera distro quite a bit. (All recent bleating to the contrary, about helping RH develop RPM, they pretty much dug in their heels and refused to put it in for the preview -- it was the release that finally had it.)
It, however, convinced me not to use Caldera in the future. They stopped support on the version pretty soon after the next one (with a hefty upgrade price) came out, and the proprietary software they bundled had a tendency to have free (and better!) replacements within a year or two.
At the present, I work for a Linux company (we don't do distributions, thankfully). Our products have turned up missing (practically core!) libraries (libttf, libesd, among others), and the Caldera distros have consistently been far enough behind the curve that it's typically a headache to even install on their systems.
Not to mention that the RPM situation mentioned above is, well, problematic for the user. They can't update, and Caldera has (traditionally) been pretty bad about maintaining a decent contrib section.
The worst part is, the only real strength Caldera has is the add-ons. And those become outdated so fast that if you blink, you'll miss it.
Let 'em die piss in their own well by restricting their customers more. They haven't been a mainstream distro for years...
I personally am a Debian user, and I like Debian, but that doesn't make Debian a very useful distribution for newbies. Debian is the distribution you switch to after you have had to upgrade your RedHat machines manually for the fourth time.
Besides that, Debian is not any more "open" than RedHat. RedHat has consistently released the software they develop under the GPL, and they develop quite a bit of software. And RedHat's freely redistributable version of their software has been useable for a very long time (my first RedHat install was version 4.0). Even back in the early days when RedHat was bundling commercial software (like the Red Baron web browser or the commercial X Server) they still made a distribution that was perfectly useable without their proprietary bits. Caldera, on the other hand, wouldn't even install without their proprietary installation routine. They have also taken great pains to comply with published standards. The folks at Caldera have been belly-aching about RedHat and the LSB forever, and yet RedHat is more File System Hierarchy compliant than Caldera ever plans to be, and RedHat is more LSB compliant than anyone besides Debian. RedHat even took a stand on early version of KDE and the non-free versions of QT. This position allowed Mandrake to come into existence. Early versions of Mandrake were little more than RedHat + KDE.
In other words, RedHat has generally been pretty darn "open." Surpassed only in openness by Debian Linux.
However, while RedHat and Debian were both nearly equal in terms of "openness" RedHat was miles ahead in terms of ease of use. Back when RedHat was taking over the market Debian Linux was so hard to install that it was ridiculous. My first Debian install was version 1.3. I paid more for those CDs than I had ever paid for a RedHat distribution, and the dselect install was physically painful. RedHat also came with a neat set of administration tools. Debian had five versions of vi. Heck, until apt came along there really wasn't a reason to use Debian. It was simply too painful to install, with little benefit over it's RPM based brethren.
Debian has come a long way, but it still isn't nearly as easy to install as RedHat, and for newbies (the newbie market is an important market for Linux because there really aren't that many folks with experience) RedHat makes a lot of sense.
I really do believe that this is how RedHat got their marketshare. They had the best product at the lowest price, and they have followed up their early distributions with a fairly solid run of cutting edge Linux, nicely packaged. Debian might be slightly more "open," but RedHat is nearly as open, and still useable by mere mortals.
The Linux community doesn't even have a problem with making a profit licensing Linux. We simply aren't complete imbeciles. Honestly, why in the world would I purchase Caldera Linux with their ridiculous per seat license when I can purchase one copy of RedHat Linux (at a lower price) and install it on as many machines as I like.
Heck, RedHat is even more popular than Caldera, and it runs more software. Every piece of Linux software (free or proprietary) runs without major issues on RedHat. The same can not be said for Caldera.
And if my Linux servers are mission critical it still makes more sense to purchase RedHat for several reasons. First of all RedHat's per incident support means that I only pay for the support that I need and can afford. Second, since RedHat is more popular, it is easier to hire contractors that have a thorough knowledge of the RedHat distribution.
Caldera should be fighting RedHat by making their distribution more popular, not less. Raising prices is simply going to make them even more of a bit player.
Exactly, this isn't about GPL licensing issues. Caldera is perfectly within their rights to charge money for proprietary software that happens to run on Linux. However, they are competing in a marketplace with competitors that are willing to give an almost identical set of software away for free. Why would anyone in their right mind pay a per seat licensing fee for Caldera when they could get something as good (or perhaps better) without cost from one of Caldera's competitors?
Caldera can whine all they want about making a profit, but RedHat is turning a profit giving software away, and the Debian group doesn't need to turn a profit because they don't have employees. Both of these groups are more than happy to give their distribution away, and to be competitive Caldera will have to do likewise.
And the benefit of buying support from RedHat over paying Caldera's extortion (er licensing) fees is that you only pay for the level of service you need (and can afford). Plus, if RedHat has lousy service you can find another support contractor. All the software is Free Software, so there are plenty of people that can dig around in the source and fix things. RedHat knows this, and so they have plenty of incentive to do a good job.
With Caldera's new distribution important parts of the distribution are closed source commercial binaries (otherwise their license would be easy to circumvent). If Caldera's distribution doesn't work you are stuck with waiting for Caldera to fix it (even if you have the requisite skills yourself). Even worse, if their support is sub par there is nothing you can do about it. You are stuck. And since Caldera got their money from you up front, they have little or no incentive to have good service.
Bottom line is that this isn't going to work. Caldera Linux is simply going to become less popular than it already is and die. I personally can't wait. The sooner Caldera runs out of money and goes bankrupt the sooner Mr Love and his inane comments will disappear from the computer press.
Actually, Caldera spends quite a bit of time complaining about Linux compatibility. They realize that RedHat is the de-facto standard now, and that RedHat is what is used by most Linux developers (both in the free software and commmercial software world). This means that when a developer packages an RPM of their work they basically only test against RedHat, leaving the Caldera users to build the software themselves (assuming that it comes with source), or to install the correct libraries (if it is closed source). This is a pain. So when someone does pay for a Linux distribution, or signs up for service and support which distribution do they choose? Well duh, they choose RedHat. It's the most popular. It has the most developer support, and it runs all current Linux software "out of the box."
Caldera, in their infinite wisdom, has decided to combat this by charging a per seat licensing fee, which is basically guaranteed to make their distribution even more unpopular. I personally thought that Caldera saw the light some time ago when the GPLed their installation routine. Caldera has always had a fantastic distribution, but RedHat was more open (if not quite as good) and so RedHat got the market share (and the sales). It turns out that most Linuxers would rather have an inferior piece of software that they can fix over a fancy proprietary piece of software that they can't. Being open is the only way to survive in the Linux game. It is certainly possible to sell proprietary add-ons to Linux, but the base distribution has to be Free Software or your customers look elsewhere. There are very few Linuxers that are interested in building their software on anything but a truly open platform.
I would have guessed that Caldera had already learned that (the hard way), but apparently they need another smack or two with the clue stick.
Because, when all is said and done, the folks that are willing to pay for Linux software and support are much more likely to pay for the most popular distribution. Caldera is never going to be the most popular distribution if they keep these tactics up.
The Linux users that are testing their applications on a "free" version of RedHat are very likely to be the folks that sign up for service and support when their Linux application becomes mission critical. And you can bet they won't be signing up with Caldera.
Personally I didn't think that any of your answers were "fluffy," but I am glad that someone did:). Your answers to random posts in the forum have been nearly as enlightening as the original interview.
If you really want to make Microsoft squirm write the best piece of software ever, and then make sure that it is released under the GPL. Microsoft knows that their business model will never work with open protocols and GPLed source code. IBM, Sun, and some of the others (perhaps RedHat?) could get by selling consulting, services, and hardware, but Microsoft's entire business model is centered around creating de-facto closed standards by leveraging their desktop monopoly, and then using this standard as a toll bridge that the rest of the industry has to pay to cross to get to the end user.
Despite the fact that Dell, Compaq, and your local OEM down the street actually build the hardware and sell it, Microsoft controls what icons get deployed on the desktop, and what messages the machine displays while it boots. I imagine that it irks Dell to no end to realize that it actually competes with MSN (doesn't Dell have an ISP arm), and yet they have to include MSN icons by default on their boxen.
However, the more competition that Free Software (specifically GPLed Free Software) puts up the more leverage all of us have against Microsoft. Eventually either Microsoft will change its tune, or people will simply migrate away (or GNU/Linux will cease to be viable and Gates will control computing).
GPLed software already can not be used by Microsoft subversively. They can't close it up and charge for it, and they don't have a very good record as a service organization.
This is a fairly simple thing to fix as well. LDAP already would allow you to do this, and Mozilla (or whatever you chose as your personal security manager) could easily update a list of sites all at one time if an open protocol like LDAP was used. What's more, since you wouldn't really be interested in the passwords for the individual sites (after all, you would only need to remember the password that unlocked your security manager) the passwords that actually went over the Net could be ridiculously long and complicated. Mozilla could even automatically change all of your remote passwords every day when you first logged in. The user wouldn't even need to notice that this was happening.
And if this blob of information was stored in an encrypted format somewhere on the Internet you could access it from whatever device you happened to be using. More importantly, however, no one else would know what account information you were storing. Right now Microsoft is happily compiling a database of information about everyone that uses their passport services. They know who you are, they know where you live, they know what you buy over the Internet. They know which affiliated sites you visit and when. If Passport was at the center of authentication for the entire Internet then they would know nearly everything about everyone.
The same ease of use features could easily be created without anyone being in the middle (well, your credit card company would get involved when money changed hands). You would keep your own information privately, and could take steps to guarantee that it remained safe.
Ah, good point. Of course, that simply means that the information is stored on the file server (of my choice) out on the Internet in an encrypted fashion. The owner of the web site wouldn't need to know what was in this data packet, and wouldn't be able to read it, but I would be able to access it whenever I needed it. Barring that I could carry the information around on a smart card or a PDA. Most importantly, however, would be that I would choose who I shared that information with. My biggest problem with Microsoft's passport service is not that it's centralized, but rather that some other entity has access to my personal data. I would be perfectly happy with Microsoft's service if all they knew about me was that I was storing a blob of encrypted data on their network. As long as they don't have access to my information I don't care who stores it.
It is not like this would be any harder to set up. The reason that Microsoft didn't set passport up this way is that they aren't interested in providing a passport service (even if the users paid fees), they are interested in creating the largest database of marketing data ever compiled. Quite frankly, that scares me.
Well, yes and no. First of all the different PostgreSQL backends are already quite adept at reading the binary formats, and the changes (from what I understand) have generally not been that large. Binary formats certainly are more fragile than human editable text, but clearly it is possible to manage binary formats fairly safely, otherwise any database would be a poor place to put your information. Second of all, since the postmaster would be off the tool could safely do a
tar zcvf backup.tgz data/
Before running the format munger. The tool would have to guarantee you had enough space before hand, but that isn't a big deal (and if you failed you could simply exit with an error message).
Besides, buman readable is only useful to a point. For example, the backups on my PostgreSQL database are big enough that I run them through split so I don't end up with a single file greater than 2G. If something happens to one of my backups recovering that data is going to be a pain (though it would probably be possible).
When upgrading I generally shut down my PostgreSQL server for an entire day as it takes a long time to upgrade (during which time the database is unavailable). When I upgrade I generally dump out the database, load the database into a test machine running the new version (while making a tarball of my data directory just in case), I then verify that all is well, upgrade my production machine's version of PostgreSQL, and then load the database into my production machine. This takes a long time. Updating the binary files would almost certainly be faster, and with proper precautions it wouldn't be any more dangerous.
Not that I am complaining. PostgreSQL's structured text file dumps (like most everything in PostgreSQL) work quite well. But it's not very fast or convenient.
I can at least back up my hard drive. What happens to my passport credentials when Microsoft decides that I can't use them without paying for an "upgrade," and what happens when Microsoft gets hacked?
Heck, what happens when Microsoft "loses" my passport storage file. Microsoft doesn't guarantee that their site will be up, that they won't lose data, or that they won't be hacked. They don't guarantee that their prices won't go up, or that their service will even be available a year from now.
And if Passport becomes ubiquitous what is to stop Microsoft from using it to spam the lot of us? You think the EULA which specifically states that they can change ther terms of service is going to save us. To be honest, I don't even see Hailstorm as being all that useful. As a developer I personally am not really that interested in having my customers data on Microsoft's servers. I want to control my customers information myself. I want to have their email addresses, and mailing addresses in my database.
I imagine that Microsoft's competitors feel the same way, and now that Microsoft is a full-on media company they have a lot of competitors.
Two minor points. First of all the article (which is very good BTW) did a very scientific, highly detailed accounting of the software in a typical (RedHat) distribution. Over half of it (55%) was GPLed according to Mr. Wheeler. The article actually dealt with the GNU/Linux versus Linux debate and stated that while the Linux kernel + drivers is the largest single contribution, it is far overshadowed by the code that whose copyright belongs to the FSF (not too mention projects that are officially part of the GNU project but whose authors retain copyright).
Of course, I still usually call it Linux. But if I was talking to RMS, I would call it GNU/Linux, and I would prepend it with "Thank you sir." Without GNU software none of the free Unixes would even have a compiler, much less a useful set of tools.
Another small nitpick is that Perl is actually licensed under the GPL. If you don't believe me check it out. Perl is also dual-licensed under the Artistic license (which supposedly is supposedly not terrible well-written from a legal standpoint), but that doesn't mean that it isn't GPLed.
Python, on the other hand, is not GPLed. It also doesn't look like line noise, but that is another debate.
The overwhelming popularity of the GPL is why there is so much interest in making sure that open source licenses are GPL compatible. Even those Open Source developers who take issue with the GPL and the FSF have gone to great lengths to insure that their software is GPL compatible. This isn't because RMS has some sort of mystical mind ray that makes people submit to his wishes. Instead it is because GPLed code makes up the lion's share of Free Software. If the software isn't GPL compatible then it is cut off from being integrated with a very large pool of software.
Slashdot Mirror
Yeah, that makes sense. However, it's not too difficult to set up an X terminal using the formual in the article. Heck, I wish that I would have thought about using Debian base system as the base for my X terminals. I rolled my own, and it was quite painful.
As for your problem trusting your firewall. If your attacker is sniffing packets on your local area network you have bigger problems than the fact that they might eavesdrop on your X session. If that's the case then one of your devices has already been compromised.
Take care friend.
I did something similar with a 486/33 (20M ram) and a Cyrix P150+ (64M ram) without ssh, and it worked great. It wasn't slow at all. In fact, it was so much like actually sitting down at the Cyrix that it was spooky. My guess is that the latency added by compressing (and encrypting) your session killed your performance. Compression really only helps over low bandwidth links. If you aren't maxing out your pipe, you shouldn't be compressing your traffic.
I don't suppose you ran top on your 486 to see how badly your processor was swamped. My 486 would get to 100% processor utilization just running X. With the compression and encryption my guess is that you just asked to much of your client.
Exactly. X allows me to use my new AMD box out in my fort from my crusty old Cyrix P150+ in the bedroom. It honestly is better than having two computers for the price of one, because the X terminal is much easier to maintain than a second PC. And since the Cyrix machine has a pretty good video card in it I honestly can't hardly tell the difference between sitting down at my actual machine. I even have gotten sound to work (with a little help from Gnome and esd).
Yes, if you want to fork the code base you can. If you want your source code to end up in their CVS repository, on the other hand, you will assign your copyrights to Sun.
This is not necessarily a bad thing, however. The Free Software Foundation requires the same thing if you plan on submitting code to their projects. They do this to make sure that they can prosecute copyright infringers (only the copyright holder can prosecute), and to make sure that all of the source is legal to distribute.
In Sun's case it also means that they can re-license your code under non-free licenses. Not that this matters to you. You still have your copy of the source, and you can do what you want with it (including creating a fork). You can't sell closed source versions of the software, however, because Sun owns a large part of the copyrighted material.
In essence this sort of deal gives the originator of the code a slight advantage. Sun could, if they wanted, sell commercial versions of StarOffice with proprietary add-ons not found in OpenOffice. However, if they spent to much time on the proprietary add-ons then their userbase would probably fork their codebase, and Sun would lose the help of the Free Software community (and probably the bulk of their users).
Believe it or not their are business reasons for using the GPL.
I suppose they would do it for the same reason that Netscape rebrands Mozilla as Netscape. Marketing folks think that changing the name on the free product lends the non-free (yes, both Netscape and StarOffice will be released under commercial closed source licenses) product some dignity.
Personally I think that this particular trick has failed miserably. Mozilla is good enough to actually use as a browser, Netscape, on the other hand sucks. Right now StarOffice is certainly more stable and useable than OpenOffice, but once that is no longer the case I imagine that people will simply use OpenOffice (although your Sparcs will probably come with a copy of StarOffice).
All the better. Microsoft doesn't want a quality port of their CLI. They want something that barely works. Heck, they want something even more useless than the token "POSIX" Windows NT layer. Microsoft is just spreading more FUD about Linux and trying to stir things up in the Free Software community. They don't actually want FreeBSDers to be able to run .NET servers on their boxen, that would be competition.
Yes, that is true. StarOffice most certainly did come first. That wasn't at all clear from my post.
What I should have said was that StarOffice 6.0 will be the re-branded OpenOffice.
And Open Office really is getting better. For one thing it no longer relies on the goofy Xprint extensions to print (it can generate Postscript). I actually have been playing with the recent builds and they are getting to be pretty darn good (although soffice still takes forever to start).
I also agree with you about AbiWord and Gnumeric. The Gnome Office suite is getting to be pretty good. Either way it is nice to see an organization standardize on something besides MS Office. The fact that they are saving us taxpayers some money is also good news.
OpenOffice is the GPLed version, StarOffice is Sun's rebranded version of the same thing. Part of the fun with OpenOffice is that all hackers assign their copyrights to Sun. This allows them to relicense StarOffice as they please.
It's sort of like the difference between Netscape and Mozilla.
Just for the record, I personally think that this sort of licensing is good for all involved. Sun remains in control of their project, unless they are evil. In which case OpenOffice becomes the de-facto standard and StarOffice disappears. If Sun plays nice, everyone wins.
Whatever. Almost all Linux software nowadays comes as an RPM, and that RPM is almost always targetted at RedHat's latest distribution. Sometimes if a packager has really done his or her homework they will give instructions on how to force install on SuSE or Caldera. Now, I don't use RedHat anymore (I am a Debian user myself), but I am sure that PHP works fine (although they may still be using the GPLed version PHP3 instead of PHP4, I don't know).
The Python issue is similar. It's not RedHat's fault that Guido and Co. screwed up the licensing for Python 1.6 to 2.0. The fact that RedHat pays attention to these sorts of licensing problems is actually a good thing as it means that I don't have to hire a lawyer to do it for me. Python is a very popular "embedded" scripting language. The fact that several versions of Python were not GPL compatible was a big concern. RedHat did the only sane thing by not including it by default (you could still get Python2 rpms, I have seen them).
Debian did something very similar, and for very similar reasons. These organizations understand that legal issues like this are a big deal, even if they seem to be trivial.
The fact of the matter is that RedHat is on top because they concentrated on making Linux easy to use and then released their source. While SuSE and Caldera were busy trying to herd their users into proprietary Linux add-ons RedHat was becoming the de-facto standard by releasing software. RedHat's installers weren't as flashy, and their admin tools were second rate, but you could fix them, and you didn't have to worry about licensing fees or vendor lock. While Debian was impossible to install, and Slackware was aiming for the tarball elitists, RedHat was trying to make a Free operating system that normal people could use.
So now, like it or not, RedHat is the gold standard for Linux. Applications that don't run well on RedHat get fixed (perhaps by releasing RPMs of the necessary auxilarly packages, but they get fixed all the same). Commercial Applications that don't run well on Caldera (but do run well on RedHat) come with a free RedHat CD. It's as simple as that.
This is not about RedHat flag-waving. I don't even use their distribution, but I do recognize suicide when I see it. Caldera charging per seat licenses is suicide, plain and simple. They are going to drive their few paying customers to someone else (probably RedHat). Sure, Caldera has got some nifty features, but it has always had nicer features than RedHat, and it never has helped. Linux users would just as soon have a poor piece of free software that they can fix over a fancy piece of commercial software that they can't. You would think that Caldera would have learned this by now. They have watched their fancy commercial applications get upstaged by RedHat's wimpy free ones time and time again. And RedHat hasn't always been bigger than Caldera either. Back when Slackware was the distribution of choice Caldera's desktop was a hundred times better than RedHat (and they both used the precursor to RPM as their packaging manager, Love is always pointing out the fact that Caldera helped develop RPM). RedHat was free, Caldera was not. Caldera lost. And they will lose again with their newest hare brained scheme. They are practically driving the industry towards RedHat.
I remember a time when Jordan Hubbard didn't need an introduction here on slashdot. Oh well.
Yes, that's true, he probably will be recycled. But at least no one will take his word as an industry insider. When our PHB asks us about the Ransom Love column in ComputerWorld we'll be able to laugh and say:
Quite frankly, even though Caldera is still kicking Ransom doesn't get much play as a Linux spokesperson. The only folks that ever pay any attention to him at all are the rabidly anti-Linux journalists, and they just trot him out to make Linux look bad. And now that RedHat actually has made a swipe at positive cash flow, he won't be able to say that making money with RedHat's business plan is impossible. If making money from Linux requires proprietary software and license fees then how come his company is still deep in the red while RedHat is in the black?
Most PHBs understand those sorts of arguments.
I don't have moderator points, and can't moderate in this thread anyhow, but this response deserves to be seen...
I personally am a Debian user, and I like Debian, but that doesn't make Debian a very useful distribution for newbies. Debian is the distribution you switch to after you have had to upgrade your RedHat machines manually for the fourth time.
Besides that, Debian is not any more "open" than RedHat. RedHat has consistently released the software they develop under the GPL, and they develop quite a bit of software. And RedHat's freely redistributable version of their software has been useable for a very long time (my first RedHat install was version 4.0). Even back in the early days when RedHat was bundling commercial software (like the Red Baron web browser or the commercial X Server) they still made a distribution that was perfectly useable without their proprietary bits. Caldera, on the other hand, wouldn't even install without their proprietary installation routine. They have also taken great pains to comply with published standards. The folks at Caldera have been belly-aching about RedHat and the LSB forever, and yet RedHat is more File System Hierarchy compliant than Caldera ever plans to be, and RedHat is more LSB compliant than anyone besides Debian. RedHat even took a stand on early version of KDE and the non-free versions of QT. This position allowed Mandrake to come into existence. Early versions of Mandrake were little more than RedHat + KDE.
In other words, RedHat has generally been pretty darn "open." Surpassed only in openness by Debian Linux.
However, while RedHat and Debian were both nearly equal in terms of "openness" RedHat was miles ahead in terms of ease of use. Back when RedHat was taking over the market Debian Linux was so hard to install that it was ridiculous. My first Debian install was version 1.3. I paid more for those CDs than I had ever paid for a RedHat distribution, and the dselect install was physically painful. RedHat also came with a neat set of administration tools. Debian had five versions of vi. Heck, until apt came along there really wasn't a reason to use Debian. It was simply too painful to install, with little benefit over it's RPM based brethren.
Debian has come a long way, but it still isn't nearly as easy to install as RedHat, and for newbies (the newbie market is an important market for Linux because there really aren't that many folks with experience) RedHat makes a lot of sense.
I really do believe that this is how RedHat got their marketshare. They had the best product at the lowest price, and they have followed up their early distributions with a fairly solid run of cutting edge Linux, nicely packaged. Debian might be slightly more "open," but RedHat is nearly as open, and still useable by mere mortals.
The Linux community doesn't even have a problem with making a profit licensing Linux. We simply aren't complete imbeciles. Honestly, why in the world would I purchase Caldera Linux with their ridiculous per seat license when I can purchase one copy of RedHat Linux (at a lower price) and install it on as many machines as I like.
Heck, RedHat is even more popular than Caldera, and it runs more software. Every piece of Linux software (free or proprietary) runs without major issues on RedHat. The same can not be said for Caldera.
And if my Linux servers are mission critical it still makes more sense to purchase RedHat for several reasons. First of all RedHat's per incident support means that I only pay for the support that I need and can afford. Second, since RedHat is more popular, it is easier to hire contractors that have a thorough knowledge of the RedHat distribution.
Caldera should be fighting RedHat by making their distribution more popular, not less. Raising prices is simply going to make them even more of a bit player.
Exactly, this isn't about GPL licensing issues. Caldera is perfectly within their rights to charge money for proprietary software that happens to run on Linux. However, they are competing in a marketplace with competitors that are willing to give an almost identical set of software away for free. Why would anyone in their right mind pay a per seat licensing fee for Caldera when they could get something as good (or perhaps better) without cost from one of Caldera's competitors?
Caldera can whine all they want about making a profit, but RedHat is turning a profit giving software away, and the Debian group doesn't need to turn a profit because they don't have employees. Both of these groups are more than happy to give their distribution away, and to be competitive Caldera will have to do likewise.
And the benefit of buying support from RedHat over paying Caldera's extortion (er licensing) fees is that you only pay for the level of service you need (and can afford). Plus, if RedHat has lousy service you can find another support contractor. All the software is Free Software, so there are plenty of people that can dig around in the source and fix things. RedHat knows this, and so they have plenty of incentive to do a good job.
With Caldera's new distribution important parts of the distribution are closed source commercial binaries (otherwise their license would be easy to circumvent). If Caldera's distribution doesn't work you are stuck with waiting for Caldera to fix it (even if you have the requisite skills yourself). Even worse, if their support is sub par there is nothing you can do about it. You are stuck. And since Caldera got their money from you up front, they have little or no incentive to have good service.
Bottom line is that this isn't going to work. Caldera Linux is simply going to become less popular than it already is and die. I personally can't wait. The sooner Caldera runs out of money and goes bankrupt the sooner Mr Love and his inane comments will disappear from the computer press.
Actually, Caldera spends quite a bit of time complaining about Linux compatibility. They realize that RedHat is the de-facto standard now, and that RedHat is what is used by most Linux developers (both in the free software and commmercial software world). This means that when a developer packages an RPM of their work they basically only test against RedHat, leaving the Caldera users to build the software themselves (assuming that it comes with source), or to install the correct libraries (if it is closed source). This is a pain. So when someone does pay for a Linux distribution, or signs up for service and support which distribution do they choose? Well duh, they choose RedHat. It's the most popular. It has the most developer support, and it runs all current Linux software "out of the box."
Caldera, in their infinite wisdom, has decided to combat this by charging a per seat licensing fee, which is basically guaranteed to make their distribution even more unpopular. I personally thought that Caldera saw the light some time ago when the GPLed their installation routine. Caldera has always had a fantastic distribution, but RedHat was more open (if not quite as good) and so RedHat got the market share (and the sales). It turns out that most Linuxers would rather have an inferior piece of software that they can fix over a fancy proprietary piece of software that they can't. Being open is the only way to survive in the Linux game. It is certainly possible to sell proprietary add-ons to Linux, but the base distribution has to be Free Software or your customers look elsewhere. There are very few Linuxers that are interested in building their software on anything but a truly open platform.
I would have guessed that Caldera had already learned that (the hard way), but apparently they need another smack or two with the clue stick.
Because, when all is said and done, the folks that are willing to pay for Linux software and support are much more likely to pay for the most popular distribution. Caldera is never going to be the most popular distribution if they keep these tactics up.
The Linux users that are testing their applications on a "free" version of RedHat are very likely to be the folks that sign up for service and support when their Linux application becomes mission critical. And you can bet they won't be signing up with Caldera.
Personally I didn't think that any of your answers were "fluffy," but I am glad that someone did :). Your answers to random posts in the forum have been nearly as enlightening as the original interview.
Thanks!
If you really want to make Microsoft squirm write the best piece of software ever, and then make sure that it is released under the GPL. Microsoft knows that their business model will never work with open protocols and GPLed source code. IBM, Sun, and some of the others (perhaps RedHat?) could get by selling consulting, services, and hardware, but Microsoft's entire business model is centered around creating de-facto closed standards by leveraging their desktop monopoly, and then using this standard as a toll bridge that the rest of the industry has to pay to cross to get to the end user.
Despite the fact that Dell, Compaq, and your local OEM down the street actually build the hardware and sell it, Microsoft controls what icons get deployed on the desktop, and what messages the machine displays while it boots. I imagine that it irks Dell to no end to realize that it actually competes with MSN (doesn't Dell have an ISP arm), and yet they have to include MSN icons by default on their boxen.
However, the more competition that Free Software (specifically GPLed Free Software) puts up the more leverage all of us have against Microsoft. Eventually either Microsoft will change its tune, or people will simply migrate away (or GNU/Linux will cease to be viable and Gates will control computing).
GPLed software already can not be used by Microsoft subversively. They can't close it up and charge for it, and they don't have a very good record as a service organization.
This is a fairly simple thing to fix as well. LDAP already would allow you to do this, and Mozilla (or whatever you chose as your personal security manager) could easily update a list of sites all at one time if an open protocol like LDAP was used. What's more, since you wouldn't really be interested in the passwords for the individual sites (after all, you would only need to remember the password that unlocked your security manager) the passwords that actually went over the Net could be ridiculously long and complicated. Mozilla could even automatically change all of your remote passwords every day when you first logged in. The user wouldn't even need to notice that this was happening.
And if this blob of information was stored in an encrypted format somewhere on the Internet you could access it from whatever device you happened to be using. More importantly, however, no one else would know what account information you were storing. Right now Microsoft is happily compiling a database of information about everyone that uses their passport services. They know who you are, they know where you live, they know what you buy over the Internet. They know which affiliated sites you visit and when. If Passport was at the center of authentication for the entire Internet then they would know nearly everything about everyone.
The same ease of use features could easily be created without anyone being in the middle (well, your credit card company would get involved when money changed hands). You would keep your own information privately, and could take steps to guarantee that it remained safe.
Ah, good point. Of course, that simply means that the information is stored on the file server (of my choice) out on the Internet in an encrypted fashion. The owner of the web site wouldn't need to know what was in this data packet, and wouldn't be able to read it, but I would be able to access it whenever I needed it. Barring that I could carry the information around on a smart card or a PDA. Most importantly, however, would be that I would choose who I shared that information with. My biggest problem with Microsoft's passport service is not that it's centralized, but rather that some other entity has access to my personal data. I would be perfectly happy with Microsoft's service if all they knew about me was that I was storing a blob of encrypted data on their network. As long as they don't have access to my information I don't care who stores it.
It is not like this would be any harder to set up. The reason that Microsoft didn't set passport up this way is that they aren't interested in providing a passport service (even if the users paid fees), they are interested in creating the largest database of marketing data ever compiled. Quite frankly, that scares me.
Well, yes and no. First of all the different PostgreSQL backends are already quite adept at reading the binary formats, and the changes (from what I understand) have generally not been that large. Binary formats certainly are more fragile than human editable text, but clearly it is possible to manage binary formats fairly safely, otherwise any database would be a poor place to put your information. Second of all, since the postmaster would be off the tool could safely do a
tar zcvf backup.tgz data/
Before running the format munger. The tool would have to guarantee you had enough space before hand, but that isn't a big deal (and if you failed you could simply exit with an error message).
Besides, buman readable is only useful to a point. For example, the backups on my PostgreSQL database are big enough that I run them through split so I don't end up with a single file greater than 2G. If something happens to one of my backups recovering that data is going to be a pain (though it would probably be possible).
When upgrading I generally shut down my PostgreSQL server for an entire day as it takes a long time to upgrade (during which time the database is unavailable). When I upgrade I generally dump out the database, load the database into a test machine running the new version (while making a tarball of my data directory just in case), I then verify that all is well, upgrade my production machine's version of PostgreSQL, and then load the database into my production machine. This takes a long time. Updating the binary files would almost certainly be faster, and with proper precautions it wouldn't be any more dangerous.
Not that I am complaining. PostgreSQL's structured text file dumps (like most everything in PostgreSQL) work quite well. But it's not very fast or convenient.
Uh... percisely! :)
LOL. Time to sleep now.
I can at least back up my hard drive. What happens to my passport credentials when Microsoft decides that I can't use them without paying for an "upgrade," and what happens when Microsoft gets hacked?
Heck, what happens when Microsoft "loses" my passport storage file. Microsoft doesn't guarantee that their site will be up, that they won't lose data, or that they won't be hacked. They don't guarantee that their prices won't go up, or that their service will even be available a year from now.
And if Passport becomes ubiquitous what is to stop Microsoft from using it to spam the lot of us? You think the EULA which specifically states that they can change ther terms of service is going to save us. To be honest, I don't even see Hailstorm as being all that useful. As a developer I personally am not really that interested in having my customers data on Microsoft's servers. I want to control my customers information myself. I want to have their email addresses, and mailing addresses in my database.
I imagine that Microsoft's competitors feel the same way, and now that Microsoft is a full-on media company they have a lot of competitors.
Two minor points. First of all the article (which is very good BTW) did a very scientific, highly detailed accounting of the software in a typical (RedHat) distribution. Over half of it (55%) was GPLed according to Mr. Wheeler. The article actually dealt with the GNU/Linux versus Linux debate and stated that while the Linux kernel + drivers is the largest single contribution, it is far overshadowed by the code that whose copyright belongs to the FSF (not too mention projects that are officially part of the GNU project but whose authors retain copyright).
Of course, I still usually call it Linux. But if I was talking to RMS, I would call it GNU/Linux, and I would prepend it with "Thank you sir." Without GNU software none of the free Unixes would even have a compiler, much less a useful set of tools.
Another small nitpick is that Perl is actually licensed under the GPL. If you don't believe me check it out. Perl is also dual-licensed under the Artistic license (which supposedly is supposedly not terrible well-written from a legal standpoint), but that doesn't mean that it isn't GPLed.
Python, on the other hand, is not GPLed. It also doesn't look like line noise, but that is another debate.
The overwhelming popularity of the GPL is why there is so much interest in making sure that open source licenses are GPL compatible. Even those Open Source developers who take issue with the GPL and the FSF have gone to great lengths to insure that their software is GPL compatible. This isn't because RMS has some sort of mystical mind ray that makes people submit to his wishes. Instead it is because GPLed code makes up the lion's share of Free Software. If the software isn't GPL compatible then it is cut off from being integrated with a very large pool of software.