Software licensing. As home users, it’s already an incomprehensible mess of legalese that nobody cares one bit about. However – we home users have it easy. The situation for business users and people managing IT departments is even worse (proprietary software, mostly, of course). Microsoft is a major culprit in this regard, and while the company acknowledges that the situation is messy, they claim they can’t really do anything about it.
Today, Microsoft held an event in London (the England one) where Steve Balmer addressed a crowd about all things Microsoft. He received a question from a member of the audience, who questioned him about Microsoft’s complicated licensing policies – to great applause from the crowd.
Ballmer was frank, and acknowledged that there are problems with Microsoft’s licensing policies. He said that Microsoft’s fine print certainly has its ‘gotchas’, but that his people should not harass companies about these gotchas. “I’m sure we have fine print we don’t need. We’re not saints,” he said.
However, it’s not something the company will solve any time soon. “I don’t anticipate a big round of simplifying our licensing,” Ballmer told the audience. The problem is that the fine print also gives a lot of customers to ability to get Microsoft’s software cheaper than without that fine print. “Customers always find an approach which pays us less money,” he claimed.
“Every time you simplify something, you lose something that people used to save money,” he added, “The goal is to simplify without a price increase. Our shareholders want simplicity without a price decrease.”
This does seem like a hole Microsoft kind of dug for itself. If the company had kept licensing simple and straightforward from the get-go, it wouldn’t be in the situation it is in now, where the company can’t simplify licensing because some of its customers depend on the licensing being complex.
They don’t want to do anything about it. They like the way it works, because it makes them a lot of money. People might be more money conscious if they understood everything Microsoft charges for. Plus, lumped together, the fees look like a lot compared to splitting them up.
I think that would bring some clarity to the licensing provided they don’t go about it the phone company way; “Your monthly plan will be 20$”. And, when you get teh bill; “Oh, plus 6$ connection fee, 5$ phone rental fee….”
But a selection shouldn’t be just done on the basis of costs – one should also ask oneself what open standards Microsoft support in full (not partial or bastardised implementations). The question one should also ask is whether one is willing to give up an ounce of freedom for convenience – for the sake of not having to read a book and learn a skill, is it worth the price in the long run? How many companies have gone Microsoft everything and are now regretting that decision because their whole business is absolutely so reliant on Microsoft that any move would be prohibitively expensive and time consuming?
It is unfortunate that IT employees in large companies are chosen by ‘recruitment agencies’ whose way of selection is jamming a CV through a parser looking for key words, and manager choosing from that list of ‘stellar’ performers by how ‘professional’ they look in their cheap polyester suite – ignoring the fact that the IT workers skill set will put their company on a path of higher IT costs. The clusterf–k of a system has less to do with Microsoft and the lack of ‘transparency’ and everything to do with poorly run companies who abdicate their responsibility through the use of recruitment companies and HR departments filled to the gills with clueless wonks with ‘HR Degree’s’ who couldn’t tell the difference between a micro-chip and a potato chip.
Edited 2009-10-05 19:22 UTC
That depends on a lot of things. For example, if it’s just me involved, and there’s a $30 proprietary app that’s a ton easier to use than a free (as in speech one), I’ll probably go for the former. But if the $30 app is $3,000, then I’d have to reconsider. Either way, the functionality of each app vs how much functionality I need will also factor into my decision. If paying $3,000 will help me get my job done in half the time as the alternative (even when I’m fully trained on the alternative), I might just decide that the cost is worth it.
Similarly, if I run a company that employes 500 people, how long is it going to take to get everybody up to speed with the new program(s)? Etc, etc.
I think that’s kind of a strawman, because if you invest heavily in any technology, whether open or closed, it’s probably going to be prohibitively expensive and time consuming to migrate away from. Just because you implement open standards/open solutions doesn’t mean it’s not going to cost you a sh*tload of time/money if you need to migrate to something else in the future. For example, I read somewhere on this site that the Konquerer developers wanted to migrate away from KHTML as the renderer and switch to Webkit, but the app is so dependent on KHTML, it wouldn’t really be practical to try and switch to WebKit. Hence, even though the solution they’re using is open, they’re still sort of stuck, unless they decide that the migration is worth the cost in time/money.
Sure, going with one vendor for your infrastructure has it’s drawbacks, but it also has its advantages. For example, if you have Linux servers with a LAMP stack (or perhaps, in the case of the company I work for, Oracle instead of MySQL), it’s easy for different vendors to point fingers at each other when something breaks, as I have witnessed happening on numerous occasions. However, if you’re using multiple solutions from the same vendor, you know who’s responsible.
I’m not saying that’s the way everybody should go, just that the discussion is not as cut and dry as you make it seem.
Edited 2009-10-05 23:51 UTC
Mate, I’ve seen people go out and purchase a copy of Exchange simply to use it as a mail server – I’m not joking. There are people who go out, purchase software that is unrequired but do so because it is a name they know – they don’t want an ’email client’, they want ‘outlook’. They don’t want a ‘web browser’, they want ‘Internet Explorer’. Too many people in IT have that mentality – and it is shocking to see the number who are supposed to be educated and qualified who turn out to be no smarter than the people they’re trying to serve.
If your company standardises on ODF with OpenOffice.org, for example, then if at a later date you wish to move to Lotus Symphony, it is just a matter of retraining the end users to use symphony. There is no long and laborious task of having to convert over large numbers of documents, correcting all the formatting errors that come as a result then converting all the templates across.
Again is a load of horsecrap. The example above – if one standardises on an open format, the only thing you have to do is retrain the staff and then deploy the software – there are no problems due to having to migrate large sways of data from one format to another and risk having large amounts of formatting being lost in the process.
Which has NOTHING to do with the discussion – stick to the topic.
And when you want to move in the case of Microsoft SQL – now what? you can just go and pick up all the stuff from SQL and start using it with some other package without major problems? where is this copy of Microsoft SQL for Linux or Solaris or FreeBSD? It isn’t just choice for the sake of choice, it is the ability for you to have control over your data so that you aren’t reliant on a single vendor anywhere in the stack.
You don’t simply have multiple vendors for the sake of having multiple vendors but don’t go out going 100% Microsoft, embracing every one of their proprietary formats under the sun – then turn around whining because when Microsoft puts the clamps on you, that you can’t take your data with you without having to jump through several fiery hoops. It has nothing to do with open source or multiple vendors – it has to with your data being saved in an open and portable format.
What the hell Webkit, which you mentioned before, has to do with the discussion God only knows.
How about before you reply, you actually read the whole post and learn the difference between:
Open Source
Open Standards
Then once you learn it, then actually READ the posts instead of scanning for keyboards, drawing a conclusion based on the appearance of those key words then firing off a response without putting your brain into gear.
Edited 2009-10-06 01:07 UTC
Obviously, you’ve missed my point. It’s not usually the data that causes problems; it is the code behind that data. Unless we are to assume that you work for a company that doesn’t actually do anything with the data it produces.
If my company wanted to switch from MS Office to Openoffice, for example, probably the biggest pain in the ass would be having to rewrite all those Excel macros and such. And even if you were using OpenOffice and ODF, when you wanted to switch from that to something else in 10-20 years, I don’t see how you’re going to be free from this hassle, unless you get lucky and whatever you switch to just happens to support Openoffice macros. Hell, I can hire interns to go through documents and make sure italics are in the right places, but what about that Excel spreadsheet that has a macro which hits an Oracle database and prints out graphs/charts based on the data that is returned?
If you’ve got a standards-compliant website and you use PHP on the back-end, if you wanted to change from PHP to another language, it’s still going to suck, even if you don’t have to change the formatting. Sure, it’ll be cheaper if you don’t have to reformat any of the data, but it’s not like, “Oh, we’re using open standards, so we can flip this switch and overnight we’ll be on a new platform/office suite/whatever.’
I guess my point is that once you get settled on a particular infrastructure, it’s difficult to move away from it, so there are risks involved no matter which way you go. When moving to something from something else, there’s almost always going to be some pain involved, whether you’re stuff is tied to a particular vendor or not. Rarely (if ever) will it be as cut and dry as you make it sound.
Edited 2009-10-06 03:20 UTC
Exactly, Licensing is fairly simple for home users and students… buy a PC with Windows and buy Student Office. They offer upgrades at a set cost per PC. Buy enough and you’re legal.
For companies that have more computers and more money the cost is “what the market will bear”. The game is that “full retail price” that’s god-awful expensive is the stick to the various complex “carrots” Microsoft offers such as multi-year schemes, upgrade insurance, and bulk licensing. Nobody pays the “shelf price” of MS software, but by keeping the “shelf price” of the products high it allows Microsoft to offer you an “offer you can’t refuse” or pay vastly more for software than your competition.
That is what makes investors happy, big profits…. of course explaining how they have 80%+ margins on Windows and Office divisions but take BILLIONS in losses on things like Xbox, Zune, Bing, etc. has got to make investors wonder why that money isn’t in THEIR pockets during a recession.
How can simplifying licenses increase costs? The BSD license is one of the simplest and requires no burden on development and almost zero burden on distribution. The GPL license is a bit more complicated, requiring no burden on development and only slight burden on distribution (if you make any changes). Borland’s license terms (the ones that made Turbo Pascal popular) were simple (unlimited unmodified runtime redistribution for library, barely more than standard copyright terms for development tools) more more complicated and had a burden on development but none on distribution.
In all these cases, having a license manager and having a lawyer counsel you on what you can and cannot do when using the software was overkill.
Each complication to the license adds yet another cost to good companies that try to play by the rules, and more costs to Microsoft since it has to find ways to enforce all those rules.
Lots of ways. For example there might be a license that says product X costs $1000 pr server, unless the server is a backup fail over server for a production server, at which point it will only cost you $300. This adds a layer of complexity about exactly what constitutes a fail over server. They could simply remove the clause and say $1000 pr server no matter what, and then the price would go up for a lot of users.
A lot of the complex clauses in MS licencing is about letting you use certain cheaper licenses in certain limited cases. Removing those clauses will lead to a lot of people who are using the cheap licenses in limited ways will now have to pay full price
That is a pile of crap!
You are saying that all servers should be charged for 1000. How about we do the following. We charge 650 for each, and we get the magic number of 1300.
My point is that Microsoft very well knows what they charge and when. They have the statistics to simplify the cost structure so that it would work.
BUT the problem is that Microsoft wants to make more money. Thus instead of charging 650 they charge 850 and say, “hey look you have a bargain because servers are cheaper.” But people look at the bottom line and say, “wait you are charging me more…”
This is the crux of the entire Microsoft price debate. Microsoft wants to charge to the waazoo but is unable to do so and hence by doing “special” clauses they can.
Look at their MSDN subscription pricing. It is absolutely insane, when originally it was about 1000 USD, now that same subscription costs about 3000.
In what way is it a pile of crap? An example was asked for and I provided one. Was my example somehow incorrect?
And there will still be someone who has one production server and two failover servers who now has to pay 1950 for what used to cost them 1600. I can’t see them being very happy.
They could simplify licensing by not having so many different version of the same thing. Let’s start with Windows desktop. How about one version. I will concede two, one for home and one for business.
Next we look at the server version. They don’t need an Enterprise and Standard version. One version will be fine. Datacenter version I will let slide as a datacenter provides a different function.
Office could be simplified into a standard and pro version.
As for SQL Server. One for business and one for data centers.
The best part is now Microsoft doesn’t need to support so many different versions. They save money.
Now we all know that you have full price users and those who qualify for discounts. Bam. Two licensing models. Next you throw in some Quantity discounts and your done. That’s pretty simple.
And costly for small people. All these changes would mean that cost of cheapest would go up since development costs of those extra features would be added on those. Reason why we have limited versions is because it makes stuff cheaper.
…but yet they keep taking mountains and mountains of cash per year, as extortion money.