Licensing

Ben makes some great points in post #13. They link well to this post, too. However, there are a couple problems.

First, and foremost, licensing is a method of ensuring that the consumer is more reliant on the licensor. Not merely by virtue of the license Ts & Cs but by virtue of the licensor being the "only" expert on their license. Also, the complexity of the agreements ensures that the licensee (that would be US) is pretty much liable for any violation the licensor can conceive. This second concept permits a long term potential for added revenue stream. (At least as long as the license or any derivitive license is valid.)

Second, Ben makes the point of posting licenses and applicability on web sites. This is great - in concept, but quite naturally, the legal kids have trashed the concept.

Simply put, unless you place a clear restriction in your actual license, the licensor can (and they do) change the license Ts & Cs on the web site at will.

The real key is having someone - not a lawyer - on staff who has been trained in strategic licensing; in how the products are used; and in how copyrighted products evolve across (and beyond) their life cycles. This isn't as difficult as it seems. However, this single individual can save the company huge amounts in both real money and diverted violations.

Any questions? Let me know & I'll do my best to provide a useful answer.

Of course. I still don't know exactly what reason Microsoft has for Client Access Licensing, I can't think of a valid argument, other than to generate income for Microsoft.

I think now Microsoft has decided to create a wider range of Windows versions that the Vista Business should come as standard with a Windows Server CAL included which covers any Windows Server versions.

Microsoft CAL's shouldn't be version specific either and the CAL should sit with the client allowing it to connect to any server.

First of all, why is it somehow wrong that Microsoft makes money? That's what every commercial enterprise tries to do. That's how all the people that work for Microsoft get paid. Unless one has somehow acquired independent wealth without doing anything to earn it (hit the MegaBucks grand prize or had the Prize Patrol stop by), that person did so by making money.

As may be, CALs are a perfectly reasonable way to make the cost match the product. Presumably, you do not find a problem with the 10 ounce package of some food costing less than the 24 ounce package. Why would you find fault with a company with 10 client computers paying less for server capability than a company with 100 client computers? That is what a CAL does - it related the utility to the cost.

All this notwithstanding, I heartily agree that Microsoft seems to positively delight in creating a licensing maze of impossible dimensions. Why, for example, are there so many variations of a Windows XP Pro license (retail, "over the counter" OEM, vendor-specific OEM, to name just three)? A XP Pro license should be just that: an XP Pro license. As long as I have it on just one computer at a time, there should be no problem.

I'm with you on every point. Particularly, I find it difficult to sympathize with a company that seems to be creating intentionally complex licensing schemes just so it can trick people into noncompliance even while they're doing their best to comply, and extract violation fees from them to enhance Microsoft's own revenue stream.

What other reason could Microsoft have for making things ever-more complex and difficult to maintain compliance if it can't solve the "piracy" problem for the previous, simpler licensing model? It's asinine and counter-productive, unless the intent is for people to violate the licensing terms whether they mean to or not.

. . . and yeah, government involvement is about the worst way to "fix" the situation.

Microsoft needs to simplify it's licensing and that's that. It is far too complex. I don't just mean for business either, how many home users do you imagine own legitimate copies of Windows and Office? Well here in the UK I don't know any home user with a legal copy of Office and those who upgrade their pre-installed Windows usually use a copy.

It's not that these people are intentionally avoiding paying for the licence, most people just very much assume that once someone has bought a single copy of Office or Windows that it gives them the right to install it anywhere (bit like a music CD or DVD, buy it and you can play it on any player, right?), OK the DVD may be a slight exception but thats a whole other argument for another time.

The average man on the street probably thinks terms like OEM and EULA are some kind of disease.

This is the question I recently had to get and answer to. Now you should know I got two different answers from Microsoft and a variety of answers from resellers.

I want to purchase Office on a volume licence agreement for 50 users, I want Office 2003 for now until I am ready to roll out Office 2007.

A. You can buy Microsoft Office 2003 now and upgrade to Office 2007 free when it is released.

B. You can buy Microsoft Office 2003 now with Software Assurance and upgrade to Office 2007 when it is released.

C. You can buy Office 2007 when it is released and downgrade to Office 2003 for now.

After spending hours writing emails and making phone calls I finally came to the conclusion that only B and C are true. There is a well hidden document on Microsofts website about downgrade rights (but read it very carefully).

My point is it should not be this complicated.

How many of you know what it is exactly Microsoft accepts as proof that you have purchased a peice of it's software?

I recently did a software licensing course with FAST (Federation Against Software Theft) in the UK and they quote MS as requiring CD keys, Paid Invoice, Delivery Notes, purchase order, Media (where possible), printed EULA, licence agreements and numbers, photo taken of the MS Authenticity case sticker and any manuals.

I have to admit to being fairly ignorant of novell Directory services. It probably would make a good approach, but its still a licensed product. As for kerberos, ldap, NFS, Im not that much of a linux ninja yet. But I see your point, there are options. I guess I should refine my statement to say that Open source directory services will not be competative until they reach the simplicity of AD. As for open source mail programs, configureing sendmail took me 4 days. I still dont have all the bugs out of it yet. I was able to put up an exchange server in 4 hours. Its been running for 3 years now. We did have to add some extra disk space, but thats it really for major maintainace. Once again, for me, Exchange just works, and keeps on working. My only complaint is the same one I have about the rest of MS products, Licensing. You buy the hardware, you buy the sergver software, you buy Exchange, you buy client workstations/OSs, and you still have to pay $ so clients can leagaly connect to the server. I know its not my money, but its the principal of the matter. For whqat we spend on licensing fees, we could have built a new server or two.

"[i]I beleive it's called ENTRAPMENT.[/i]"
Absolutely. I made roughly the same point, though less succinctly, in [url=https://www.techrepublic.com.com/5208-6230-0.html?forumID=102&threadID=205142&messageID=2132407][b]this response to alvinsilvain[/b][/url] above just moments before you posted this -- probably while you were typing it.

I agree with your statement %100. We run a mostly Linux shop. One thing that is funny regarding licensing is that we have to use a Windows server to run license tracking software from Quark for our Quark Xpress machines. And naturally, it doesn't work correctly..... Anyone have any open source programs that open Quark files for printing?

I thought you were a Debian guy.

I hate dealing with licensing, but I hate dealing with government regulations even more. Sure, let's give the same sorts of people that run the DMV control over how licensing is done. This is a bad idea.

You reckon maybe the government should slap a mandate down that a software company can only upgrade their product every 4 or 5 years? So much for progress, then. It's all well and good to say that you want the government to control this sort of thing, but you need to consider the process that would be involved before giving the government that mandate. I mean, really, what would you have the government do?

Sure, maybe IBM can negotiate better terms with M$. But some of us little guys can't afford office buildings full of lawyers.

My solution is to put these folks on notice that, they *DO* have competition, and many of us *ARE* willing to go the extra mile to make the alternatives work.

Take Microsoft Office as the perfect example:

It wasn't that long ago that the suite of office applications cost 6 or 8 HUNDREDS of dollars. SAY WHAT?!!

The only real competition -- Word Perfect -- was also "competitively priced", i.e, also in the stratosphere.

Along comes Star Office -- able to read MS files -- and it's FREE. (It's not free anymore, but it's still cheap.) Certainly not a perfect alternative -- definitely doesn't have half the features of MS Office -- rather kludgey to operate in many respects -- but you can't beat the price!

NOW -- you can get MS Office (for teachers and students ONLY, but the definition of "student" is very loose) for around $125 (less if you shop vigorously) -- and that even includes three licenses, so you can load up your computer and two kids' computers (as I have done).

Is Star Office responsible for this? Certainly not all by itself, but I think it helped.

I don't think MS's price would have become so reasonable without watching their market share begin to shrink, even if only a little bit.

That's the only option we "little guys" have. Take our business elsewhere when we can.

Any cooperative, collabrative effort to help nevigate the continually rough and turbulent waters of software licensing and compliance can only be a benefit.

Not to Pirate but to the rest of us, we need to stop the whining. We are professionals. We are the ones who are paid to find the solutions, save the company money, keep third-party audits at bay (BSA, SIIA, software manufacturers, etc.), educate people about intellectual property laws and regulations, etc.

It matters little that licensing is weighted in favor of the copyright holders. Alright...we all understand this so lets move on. It matters little that government is not very interested in helping. OK. This is part of why we have jobs.

But I also agree with Pirate. We can push back. I happen to work for a company that has a pretty good legal team and we have pushed back with some success. We are not victims of those who draft the LAs.

Let's find a way to row in the same direction and stop the grousing. I will come along side anybody who is willing to mount a cohesive, honest, professional effort to help all of us who do this kind of thing for a living or for those of us who educate others about this rather interesting and challenging field. I got game.

The only networks over which I've had responsibility that were noncompliant with software licenses were networks I inherited with their licensing issues already in place. It's pretty rare, in those cases, for the licensing noncompliance to be intentional: it's almost invariably a case of the licensing to be so complex and the time so short that people just decided to not worry about it "for now", and the time to start worrying never came.

Networks I've built from the ground up, myself, have been licensing compliant ever since I understood what that meant -- even though I personally feel that all this EULA and software copyright crap is a bunch of unethical nonsense designed to extort money from people who feel like they don't have any other options. I avoid license violations because it would be unethical of me to leave clients in a legally precarious position, and because in many cases the network I design doesn't really need licensing management anyway. That's one of the benefits of working primarily with open source technologies on a given project.

Slackware is pretty close to "bare metal" in terms of your options for easing system administration. Every Slackware network ends up being its own unique animal, as most of the automated system administration tasks need to be cobbled together by the sysadmin(s), since there are not the pleasantly slick tools you'd be used to from other distributions. Software management is a blast from the past with Slackware: yes, it has what they laughably call a "package manager" for Slackware, but it's pretty primitive and rudimentary beside tools like APT and YUM.

Gentoo has excellent software management tools, and great documentation on the Web. Unfortunately, it's also the "What's borken this week?" distribution. It is very much a bleeding-edge-oriented distribution, and ports tend to get included before adequate testing has been done, which means that updating software versions is really an adventure with Gentoo. This can be something of a challenge for production systems, where uptime and stable functioning are of primary importance.

NetBSD is kind of in a precarious state right now. The community is getting shaken up by internal turmoil, it has been losing popularity against FreeBSD and OpenBSD for years, and there are some serious questions about the effectiveness of the core team. The major claim to a benefit of using NetBSD is portability: there was a time when it was the most-ported general purpose modern OS in existence. Debian has surpassed it now, however, not only because Debian has been ported to a myriad of different platforms but also because NetBSD has been slacking on the platform porting. Its portability is now, mostly, a thing of myth and legend, from what I've seen.

OpenBSD is widely known as the most secure modern, general-purpose, easily available OS in existence. The OpenBSD project, in fact, was the source of the OpenSSH implementation of SSH, which is the most widely used secure/encrypted remote administration tool in the world. All major Linux distributions include OpenSSH by default, along with a number of other OpenBSD-originated tools. OpenBSD has had a grand total of one remotely exploitable vulnerability in default configuration, ever, in the history of the project -- and no documented security compromises in default configuration at all. It's about as stable as an OS can possibly be while still allowing you to do things like run a GUI web browser such as Firefox. On the other hand, it doesn't benchmark very well on system performance for a number of common tasks, in part because of all the internal security auditing it does. Additionally, in part because of the extremely close attention paid to security and stability, OpenBSD has very limited hardware support capabilities. Don't expect to play Unreal Tournament on OpenBSD any time soon.

While the size of package repositories is not necessarily of primary importance in an OS, there are other reasons to consider the availability of software in the software management system of choice for your OS. For instance, both Gentoo and FreeBSD use a "ports tree" to make commonly used software available to you in source code form with great ease. You can install the software from source in the ports tree, without having to use a package manager. You can also use a ports manager, such as emerge for Gentoo or portsnap/portupgrade for FreeBSD, to make the process of handling installation from source as easy as installing binary packages is in certain Linux distributions (such as using APT with Debian).

Because it's installing from source, however, you have greater control over the configuration of your software at compile time, you know exactly what you're getting whenever you install something, and stability of both the operating environment and the software management system is possible. The Gentoo project plays fast and loose with its ports, so this benefit is mitigated somewhat in Gentoo, but the FreeBSD project tends to lavish a fair bit more diligence on maintaining the stability of the software in its ports tree. In addition, because you're installing from source, there are fewer detriments to going "outside" the software management system of the OS to install something than there are with a system like APT or YUM.

There are benefits to using a system like FreeBSD's ports tree over just downloading tarballs from Sourceforge, as well. For one thing, it's a lot easier to search through available software to see what you want, since there's searching functionality built into the ports system. For another, the software is all vetted and extensively tested before inclusion in the official ports tree, and modified for maximum stability on FreeBSD. Software in the ports tree is by default configured to share libraries so that you don't need eight different versions of a given library to support eight different applications you've installed from tarballs -- or, worse yet, so you don't have two applications installed from tarballs that share a library, then one of them on a version upgrade changes its version requirements on that library dependency, leaving you with a system where one or the other application is broken unless you reinstall them both. You also get the benefit of increased security, because there's functionality in FreeBSD's software management system for checking for new security vulnerability patches and automatically auditing the entire ports tree, so you don't have to track vulnerability announcements for each and every separate program you've installed from a tarball you downloaded from Sourceforge.

Despite all this, you don't even have to install from source in FreeBSD if you don't want to. There are binary packages available for when you don't want to deal with compilation and are happy to go with all defaults. The packages, however, are basically exactly the same as ports, except that the process of compilation is already done for you -- so the whole system, between packages and ports, is unified. In fact, when you want to uninstall a port, you can use a package management tool to do so. Thus, once installed, packages and ports are interchangeable and indistinguishable, with the exception of the fact that if you installed from a port the system will remember to upgrade by compiling from the new version port rather than installing from the new version package. It's really a beauty to behold how well the system works as an integrated whole.

Oh, yeah, and audio configuration for FreeBSD is a lot easier in most cases than for Linux. That was a very pleasant surprise for me.