Slashdot Mirror
Corporate America Wary of Subscription Software
medical_geek writes: "According to this
article
on cio.com,
MS's subscription service is failing in the business world. I guess
that personal users are not the only group that balks at paying a yearly
fee for software. My question is have you at your job bit the bullet
and signed up as an early adopter, or are you rolling the dice and seeing
if this experiment fails?" This article focuses only on Microsoft, but the same analysis probably explains why ASPs haven't taken off like they were supposed to, either.
It'll only be when the quality of the software is up to scratch that people will start thinking about its price. In the end, the total cost of ownership of software is much larger than the licence fee: putting in fixes after deployment is terribly expensive.
"Well, put a stake in my heart and drag me into sunlight."
Just a thought.. but Linux Just Makes Sense more and more now-a-days, even if in some cases, it is less capable than the Microsoft alternative.
eBay is one of the most successful ASPs on the Internet. Sellers seem to have no problem paying a per-auction fee to eBay for hosting the auction application. You can imagine an alternative where everone paid $10 for an eBay application that sat on their Windows desktop and did a P2P search of current auctions by communicting Gnutella-style with the other eBay applications. It would suck. The ASP version kicks its ass any day of the week.
Similarlly, I used to work for a company with an ASP remote access application. To circumvent firewalls that only allow outbound connections, the company routes all connections through their servers; there's no other way to do it if you want to support connections where both endpoints are firewalled. Hence, ASP. It's easy for me to justify paying a monthly fee to use this service because the application demands it. I have to use their servers. (The company includes free support and free upgrades with the subscription fee, too, which makes it rather more attractive than Microsoft's licensing scheme.)
As for ASP MS Office... At this point, my reaction is, "What's the point?" In the absence of ubiquitous thin-client computing, I can't see at all why I'd want to pay for a subscription. There's no value in an ASP model for lots of applications, include most of Microsoft's (with obvious exceptions like Hotmail).
ASPs didn't fail. They just succeeded where it was logical for them to succeed.
At my place of business, we develop VB applications (no flames please) that we use internally to control inventory, and to track units that we produce, as well as write test software to well, test our product before it ships. Besides running windows 98/NT/2000, we've decided to start converting our software over to Linux, writing in pure C, and using MySQL for databases. OK, I know this seems OT so far, but my point here is that we are doing all of this to avoid using XP and any other upcoming versions of MS windows. We have decided that we will do our best to make sure that 2000 is the last version of windows that we will ever buy, at least in our department. We've already been using Linux on servers within our department for about a year, using apache to run a simple intranet server, and have samba up doing file and print services. Since some of the brass have found out about our 'secret' of having high availability linux servers, they were intrigued and like the direction we've taken thus far. Hopefully we can avoid XP all together, and write software that can potentially be useful forever, by writing it for an OSS platform.
------
Random, useless fact: I type in startx entirely with my left hand.
...I'm using a company machine with WinNT 4.0 SP6a. With a "Windows(R) 2000 Professional 1-2 CPU" sticker on the side. We downgraded it, and for a very good reason.
It does everything that I need to do my job. It does nothing that I don't need. The issues are known. It doesn't require any more patching because it ain't broke, or it's broke in known and acceptable ways. It doesn't require our IT guys to have to ask what version of what OS I'm running, nor to hunt out the right ghost image for that combination of hardware and OS. It can be ghost installed or copied, which is vital for replicating software builds.
Windows 2000 would be a barely acceptable substitute. There are far too many unknowns with WinXP, plus it has that habit of knowing better than you what drivers you really want to use (I need to test beta drivers, for god's sake, give me an "I know what I'm doing" button!).
Windows.NET would be absolutely, utterly unworkable in a business environment, because neither I, nor our IT guys would know what exactly was on the machine, nor would it be possible to replicate that at a later date to reproduce a build exactly.
We cannot and will not upgrade to .NET. Ever. As application support for NT dies away at the same time as Linux support grows, it's looking like a better (corporate!) proposition every day, and not just in the server room.
If you were blocking sigs, you wouldn't have to read this.
Even Eric S. Raymond agrees with MS on this one :
"It is also worth noting that the manufacturing delusion encourages price structures that are pathologically out of line with the actual breakdown of development costs. If (as is generally accepted) over 75% of a typical software project's life-cycle costs will be in maintenance and debugging and extensions, then the common price policy of charging a high fixed purchase price and relatively low or zero support fees is bound to lead to results that serve all parties poorly.
Consumers lose because, even though software is a service industry, the incentives in the factory model all cut against a vendor's offering competent service. If the vendor's money comes from selling bits, most effort will go to making bits and shoving them out the door; the help desk, not a profit center, will become a dumping ground for the least effective and get only enough resources to avoid actively alienating a critical number of customers." - Eric S. Raymond
This is taken from The Magic Cauldron
I'm an IT Manager, and also opted not to take the Software Insurance offer. For one it meant that we'd have upgrade (or at least buy the licenses) all our MS products to latest version. All the NT + Exchange + SQL CALs, and then Windows + Office + Project... It adds up quickly.
... :-) But if you don't take software assurance then you can't just upgrade down the line. You have to pay the full purchase price. Which is all fine and dandy, except that they didn't write the product from scratch, they built it on top of older versions. And you already paid them for the intellectual property in the previous versions, so why do it again? Upgrade in the older, normal sense is much more fair, since you only pay them for the added value... when you need it.
But my main point that I wanted to make is that it's just plain unfair. Microsoft want you to pay them even when you *don't* want the latest and greatest. (And 95% can life without it... If some people think StarOffice 5.2 is good enough
Pardon the rant.
Because microsoft measures value based on increasing revenues each year, maximizing share holder value means shorten the product cycle from 2-4 years to 1 year. From a share holder perspective, it's great. A rapid product cycle is a good thing for share holders because it means you're getting more repeat business more frequently.
From a user perspective, a certain level of product stability is necessary to create a sense of value and reliability. If the product cycles at a faster rate the customer is comfortable with, the company begins to loose business. Think of a can opener. What if every can opener was only good for 10 uses and it would break. No one would buy can goods or can openers. Food manufacturers would use some other container, like a jar instead of cans. It doesn't matter if the can opener is only 2.00. No one wants to buy a new can opener every week.
Microsoft is not immuned to the same market principles. Making a product too good or really poor isn't good for the company, consumers or the economy. Back in the 80's Honda found a good way to make bearings in such a way that they would last 30-50 years. Well guess what. Honda stopped using them in cars because they were too good. Using those bearings in cars made them way too reliable and was hurting replacement parts sales. There has to be some middle ground where corporate and consumer needs are in balance.
Microsoft may or may not realize it before it is too late.
I'm not talking about servers, I'm talking about desktops. Desktops get moved around, recycled in different roles, and there's a lot of them, not all of which can run 2K let alone XP or .NET.
Managing 2 OS's, let alone 3 or 4 would increase the workload (and cost to us) of our IT guys. And rolling back an existing install is not the point. The point is that when a bit of my box fries (which it's done once), with half an hour I'm using another box from the same manufacturer (faster CPU, but same components) with a ghost of exactly the install of NT4 6a and apps that I was using ('cause I wasn't dumb enough to screw with my old box), and I and the IT guys know what's on it, and the box doesn't then try and second guess me and upgrade itself. Ever. If I can't get a recycled box with the same hardware, I can get a mostly equivelant (but known) ghost image on a newer box, or (worst case) we can start with a new box, rip whatever it comes pre-installed with, install it up to a known state that's not far off a ghost image in terms of repeatability.
Yes, we can do that with 2K. We can do it with XP - if we're careful. But with .NET and son of .NET, it's going to get harder and harder. I'm not saying we can't do it, I'm saying that it may become more trouble than it's worth, and it will push us into using alternatives.
I should make it clear that I develop and maintain telecomms software that's been in the field for years. When I have to fix a bug in software that was built 5 years ago, I need to replicate the original build environment exactly, no surprises, no enhancements. I previously worked in the nuclear industry, where this was an absolute requirement. We had 20 year old machines in storage that had to be maintained to replicate builds, and a firesafe packed full of complete OS and application tars to wipe these systems and put them in a known state.
Our problem is not one of administrating and updating a network, it's of stopping it being updated. That's a very real need for some developers.
If you were blocking sigs, you wouldn't have to read this.
We're jumping head-first into Microsoft products here, at a non-profit operation. Office XP alone will cost us several tens of thousands of dollars (US) per year. All this while we're being told there's no money. All this despite the fact that I had prepared an alternate path that relied on open source and free (as in beer) software which met all of our needs. Why? Because the executive committee, or at least the only one on it with a backbone, decided that anything that is downloaded is bad. Never mind the fact that I countered that by saying that all of the software required could be acquired without download. Unfortunately, there's still a mindset among the people who write the cheques that Microsoft is the best option. So they'll shell out $50,000 per for Office, while their underpaid staff continue to leave to work for better managed companies.
This is the second time since I started working here that management hasn't followed my recommendations. The first time was a disaster. Hmmm, let's peer into my crystal ball...
"Not to be off topic - but this is illegal. "
You can downgrade licenses purchased under a volume license agreement, but yep not OEM licenses.
"I seen a lot of posts that state NT4 is running fine for that web site (its much easier to secure - that's why people are still using it) "
Well that's certainly not true. Windows 2000 is much better as a web server and far easier to secure.
Furthermore, NT 4.0 goes on the unsupported list this year.