Slashdot Mirror
IBM To Announce Web-Based Desktop Apps
mgoulding writes "IBM is expected to announce a software bundle targeted to business users that will challenge the Microsoft Office package. Unlike Office, the email, word-processing, spreadsheet, and database products will be accessible to Linux, Unix, and heldheld users through a web server. NewsFeed posts the story from CNET." It's certainly something that's been tried before - witness sites like MyWebOS (no longer existing).
What's the pricing for this setup? I know the article mentions a $2/user/month charge, but it also requires IBM Websphere (which is what IBM really wants to sell with this setup). Which version does it require? Websphere has quite a price range.
Also, the really big question is: What is its compatibility with MS Office?
Casual Games/Downloads
Although many of us here may see the immediate benefits of this over MS Office, has anyone ever been able to sway non-techie management away from Office?
If I remember correctly, back when Microsoft started trying to think of something to tie to their new .NET naming scheme they had the idea that the version of Office beyond Office XP was going to be completely web based, where you would basically subscribe to it and log on via a webpage. Of course, seeing how the version beyond Office XP was Office 2003, they obviously changed their minds.
I wouldn't use web-based applications for the same reason I don't use webmail. It's like sitting at a dumb terminal... I feel very powerless.
wtf is a heldheld user?
It said in the article that it could be accessed via Linux, so I don't see it requiring IE.
I'm not quite sure they are going to be able to administer a "giant blow" to the MS empire. Let's not forget that MS is a master when it comes to copy^H^H^H^H^H innovating on other companies products. Lots of money and innovation will prevent MS from being derailed anytime soon.
I guess the submitter doesn't remember Lotus eSuite.
Intelligent Life on Earth
As the article said, it's been tried before. IBM had a toolkit out for doing just this a couple years ago. The toolkit/sdk was pretty nice too...for the life of me I can't remember the name....
However, it didn't fly then, why would it fly now?
As someone who influences purchasing decisions for my company, I must first insist that the product be completely buzzword compliant.
For example, unless it uses JAVA (which my staff assures me is the next big thing), then I'm not interested. Also, I insist that the files are XML, PDF, or maybe even SATA or RAMBUS so that they can leverage my various cross-functional team building objectives.
Now, I see that this is going to be on the 'web'. I once clicked a link and found an unpleasant photo of a gentleman with a distressing condition that exposed his bottom in a most unflattering fashion. If I'm going to 'synergize' and align behind this eSolution, I certainly hope such a thing won't happen again.
Finally, I want the interface to be 'webbish', but not TOO webbish.
Clippy: "It seems you are trying to install non-MS Office compliant software...I don't think you should do that...Hey, what are you doing? No, stop! Oh GOD, I'M MELTING...NOOOOOO!!"
witness sites like MyWebOS (no longer existing)
Well doesn't that make it kind of hard to witness it?
The BusinessWeek take on the announcement. They make a point that IBM's timing of this release is in some part due to the the delay in the "Longhorn."
Ok, I can see this is good, because it means that OS will finally, hopefully, become completely irrelevant (I'm being an optimist here).
But how are IBM going to persuade the ravening hordes of MS Office users that their web-based apps will fail to suck?
Hotmail et al have had cross platform web-based email apps for years, and do they fail to suck? No, because while you can get at your email from where-ever you are, on whichever system, they are still nasty buggy and slow, and lack the features of even the worst (OE) traditional email apps.
How will IBMs web-based Word fail to suck? to win users from Word and OOo Writer etc it not only has to be as good as them, but it has to be better than web, and NOT rely on the web-based gimmick and the "OOh, shiny!" factor (which only lasts for a fortnight aat most anyway) to win over and reatin users.
First, we need a better HTTP-friendly GUI protocol. HTML+DOM+JavaScript is awkward for business applications and forms. Those were generally designed for "e-brochures", and not business forms. XML candidates include XUL, XWT, SCGUI (my pet protocol), and others.
Table-ized A.I.
When I worked there, I poked around and saw something (I forgot the code name) which was Word and Excel in a web browser done with DHTML and script and no activeX, similar to Outlook Web Access. They cancelled it a couple of years ago, but they can bring it back out.
This seems like they're just using a web browser the same way X programs use a remote display. Remember in the old days when everyone had a local machine that was relatively weak, and ran all their *real* applications (you know, besides, xclock) through the lan, on a computer hidden in a server closet somewhere? This sounds very similar, except they'll be using the web as an interface, instead of X-windows.
This might look like IBM is trying to get back some sales from Dell -- the machine sitting on a user's desk can be anything, but the server in the back room will be an IBM, worth tens of thousands of dollars.
Or will the web interface simply download a java application to the person's local machine?
-S
I remember reading some announcement about how Lotus (owned at the time by IBM) was going to put out a web-based "utility" metered Java version of their office suite. This was back in about 1998. I don't recall anything after that. One has to wonder where this new announcement leaves Smartsuite, since it too is competing directly against MS Office (and if they haven't changed Word Pro too much in the last seven years, they have a pretty good go at it, too - Word Pro 97 is still my favorite word processor).
In the great CONS chain of life, you can either be the CAR or be in the CDR.
While I do find it difficult to accept the concept of a web-based office application not all applications suck for the web.
The most popular example are webmail systems. They are applications and the fact that they are on the web is wonderful for many people as it allows them to have a centralized address book and the ability to safely check their email without having to install or do any work.
IBM is no fool however. I'm sure they know by now that nothing can beat office unless it truely outperforms MS Office in all aspects and thus having to refresh the page a 100 times because you wanted to bold every 3rd word isn't acceptable. Lets see what they release first before saying it sucks. They might just surprise you.
I don't think it is really going to be such a giant blow, because it really is only for large scale corporate users running Websphere. I don't see the smaller company, the self-employed, or the home user switching to this. Which means that it cannot become a de-facto standard, as Word and Excel have. Which in turn means that the corporations will not be that enthusiastic, because they also deal with these other people. If you want som ething that is really quite compatible with the de-facto standard, but not quite, Open Office is available for $0.00 per month.
It will appeal to pointy haired power freaks who dislike the idea of employees having letters and/or spreadsheets on their C: drive. This way, everything is on the company managed central server. Mwahahahaha! Of course, there are cood reasons for that as well - backuip, legal liability, central administration. But they are, broadly speaking, considerations for megacorps not the small user. And it is the magacorps who are IBM's favourite customers.
Consciousness is an illusion caused by an excess of self consciousness.
They tried *exactly* this strategy before with the Lotus e-Suite... web-based analogues of their SmartSuite products. We beta-tested them for Lotus back before they pulled the plug, so I know of what I speak.
It crashed and burned then, and as much as I applaud the effort, I don't see any clear reason why it won't do so again this time. They may be marketing well to the geeks by saying the right things (platform independence, low TCO, easy distribution), but the end-user ultimately rules the roost, and if the last go-around is any indication, the products will be slow and clunky, only partially functional, and generally leaving a lot to be desired. You'll spend more time answering "why doesn't it do X?" phone calls than you ever spent deploying and administering Office.
I truly hope they do better this time, but if they do pull it off you could likely knock me over with a feather.
"So on one hand, honey is an amazingly sophisticated and efficient food source. On the other hand it's bee backwash."
I've played around with Java Web Start and it seemed like a good idea, in theory at least.
The idea is when you're running the Java plugin in your browser, you can 'launch' full applications right from the site. It can be either in a single JAR file, or split amongst many (JWS is supposed to download the pieces as they are needed).
Anyway, it is pretty neat and it's come a long way. With some improvements it might be viable to launch full-blown apps such as Office and whatnot (assuming you can get them running well enough in Swing or whatever), although the downloader still needs work to more intelligently decide which pieces to get.
I've written a few JWS apps already and it seemed pretty good, but they really do have some bugs to work out before it's ready for prime time.
You can accomplish anything you set your mind to. The impossible just takes a little longer.
I think that it might actually be something Java-based. The article says:
The web interface will be limited to things like initial setup (like java web start), browsing on-line help, group collaboration etc.
To err is human, but to forgive is beyond the scope of the Operating System...
Why don't they take OpenOffice.org, make it work with IBM file formats (lotus, EBDIC, AIXrc), and call it IBM Office.
They could even make a version for OS/2!
What difference does it make how cheap it is?
What difference does it make how fast it is?
What difference does it make how many features it implements?
What difference does it make how cool the interface is?
This is the SUBSCRIPTION MODEL! This depends on the web being up all the time. (which it mostly, but not always is) This brings in a progressive billing relationship. Who in their right mind wants to pay every month for ANY commoditized application? Existing WordProcessors are already so feature rich I can't imagine wanting ever to buy a new one. Same goes for Spreadsheets, Presentation Makers, even desktop databases.
Let's get real here. A subscription model is fundamentally evil and pointless whether it's being flogged by Microsoft, IBM or even by Linus.
Have we decided that IBM is our friend and therefore suddenly the subscription model is a good idea? (besides the price is so much lower than what Microsoft would charge) Let's remember where we came from and that one plus one is still equal to two.
the defeat of Microsoft won't necessarily make everything better...ank
Still hoping for Gentle Treatment...
WebSphere Portal 5 bundled in some web based word/excel replacements. I think the idea was "use our portal, and use us to view (and edit) business content. You don't need office!" The reality was less than inspiring. I'd put it on par with the large number of RTF/HTML editors out there - clunky. Not sure why they did not put their weight behind OpenOffice, because it is head and shoulders above what they included with the Portal.
+++ UGUCAUCGUAUUUCU
Dan Bricklin criticized this concept some years ago. He presented his criticism in the form of an hysterically funny demo program created with his demo program tool. You can find it at
http://www.bricklin.com/chiapaint.htm
Of course, that was the dialup days... and of course we're all on high-speed connections now, right? And they never go down? And they have zero latency? And there are never any version skew issues, because Web-based standards are so superbly engineered with respect to forward compatibility, and vendors, regardless of their business strategy, fully understand that it is in their best interests to be punctilious about following them?
"How to Do Nothing," kids activities, back in print!
The application is accessed through the web, not necessarily written as web pages. Being on WebSphere and available offline I can almost guarantee it's a Java application. It's copied to the client through the browser and then runs as a client application that can communicate with the WebSphere server. They definitely are not trying to sell an HTML office application here.
Developers: We can use your help.
I remember hearing the same kind of sceptisism regarding webbased mail providers.
The applications DID suck. However nowadays, everybody takes it for granted and just uses them.
We post millions of standards compliant webpages everyday with blogging/web publishing applications. YOU even used one yourself.
Take this simple little comment box I'm filling in right now. I want to just write something and post it back. I dont want extra faffing with complex tags and escape codes - however - if I need them they are there.
There are still applications which are better suited run locally (video editing online anyone?) but for the greater majority of admin/office tasks, the web/intranet makes an ideal adaptive environment.
liqbase
The problem is the small print on the final product will be "Requires Windows and Internet Explorer 6 for full functionality." I've sick of seeing "web" based applications that require IE under Windows to work. Where do they morons learn how to program that they can't even write cross-platform applications for the god damn WWW? I blame Microsoft's indoctrination of college students by signing up universities for campus-wide licensing deals if they sign exclusive contracts. Once our university did that the courses started focusing heavily on Microsoft products. SQL server, Visual Basic, Visual C++, Microsoft's version of Java, Office, etc.
I hate it when they try to make personal computing into a SERVICE rather than a PRODUCT. Internet access is a service, but word processing is not. I, for one, would like to keep it that way.
While this does offer a more universal way of running programs, isn't it also a more proprietary and inconvenient way? It's hard enough writing papers for school when Bellsouth accidentally cuts my intenet access, but at least I can still get into my word processor to type a bunch of BS to hand in.
Esoteric reference.
"Web based applications *suck*. I do not want to refresh the page everytime I change an option, I do not want to use some propriatary scripting language to run a word processor."
:). We will see. IBM has blown it before. Anyone rember TopView?
Yes yes so you may never want to use web based applications. Some people would consider Google a web based application as well as many CGI programs.
The web was *not* designed for applications and applications will *not* run well on the web."
The web was not designed for lots of things that it is currently doing. As to the statment "Applications will *not* run well on the web" that is a bold statment and one that I would not make. Who knows what the future may hold. Java and SWT make a pretty good Applications frame work. Look at Eclipse.org if you feel that all Java apps need to be slow and cluncky.
Just wait for IBM and Sony and Nintendo to get together on this. The next generation of Playstation and the GameCube could replace the PC in many homes. All your data would be stored on servers and you would log into your collection of apps and data from anywhere. School, home, or work. But what about open source??? Linux will run the servers and possibly a good number of the devices. If they do choose the Java VM that does not mean that you will have to write in java. GCC could generate JVM code. It might even be possible for c# to compile to java byte code
But this could be a good thing.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
TIBET(tm) (a Javascript library that's up there with CPAN in terms of comprehensiveness) has the muscle to do this sort of thing: Client-side expansion of custom tags, webservices, local file access, fully reflective.... I wonder why they didn't use it. Or did they?
Seastead this.
unlike pure Web applications...mobile users...can connect, quickly access applications and disconnect to do work offline
;)
So this is not a purely web based application. This is an interesting application. It must utilize something more than HTML because it can obviously persist a session over long periods of time. it also means this is more than a thin client. Would something like this be web service based? interested to hear the actual press release from IBM. Either way, this is a good thing as having another office suite with real corporate backing , not the fake kind, is a good thing.
I only say sun is the fake kind because they are
0wNzEd by microsoft now.
I tried for 5 years to come up with a clever sig...only to realize that I am not clever.
For instance, check out BackBase, a company which provides almost the same functionality. I can't really give a good outline of their products, besides that they are supposedly coming over to give a presentation any time soon at our company. It sounds good tho...
To quote some of their website: "Backbase offers products and implementation services that allow our clients to develop rich user interfaces that move beyond the limitations of traditional HTML web interfaces. Our technology is based on open industry standards (W3C) and offers out-of-the-box integrations with leading IT-platforms and applications."
Veni, Vidi, Velcro!
First off, this makes a TON of sense if executed properly; it could yield a centrally-controlled system that is client (and OS)-independent and lightweight on the client side.
The key is to overcome the previous issues with this type of arrangement: It should also run off-line, and act like a local GUI app, e.g., not refresh the screen with each formatting change.
I suspect that this is doable using Java Applets running the sucks-way-less-than-Swing SWT. Sun should definitely be VERY AFRAID.
This may also appeal to flustered IT departments that are sick of blowing big pieces of their budget on data recovery. When I keep important data on my hard drive instead of on the company's server, I'm risking a write up, particularly if said drive crashes. When a company VP does the same thing, well, hey, that's a VP, let's fork out a couple of thousand bucks to a data recovery company. Going to a web-based system means even the higher-ups have to adhere to the IT departments rules regarding data storage.
RTFA: "BM's new software is designed to be distributed and accessed through a Web server, and to be accessible from systems running Windows and Macintosh, as well as Linux, Unix and handheld devices"
This is assuming that every person has 100% web access, which just isn't true. I have several machines that don't have web access in my business for security and productivity reasons. Even if every machine did have web access, I'd still have to have 100% uptime, which is rare with ANY client net connection. If the Net connection goes down, you're stuck, whereas now, if you need to work on documents, and the Net goes down, you can still work. Call me nuts, but this is a bit too bleeding edge to be practical.
This BusinessWeek article states that you'll be able to work disconnected, then sync up the next time you connect. So IBM is building replication capabilities into their products. Makes sense; IBM has replication know-how from both their Lotus Notes and their DB2 database products.
let's see, here, now, I am going to trust my complete office tasking, confidential information and credit-card numbers, to the security of the wild and wooly internet?
yeah, right, like an Iraqi is going to trust the man in a hat who says, "Hi, I'm here from Washington, and I'm going to help you."
if this is supposed to be a new economy, how come they still want my old fashioned money?
ECMA is not proprietary, Mozilla has Python bindings (among others) and communication with a webserver doesn't have to be only through the regular http protocol. An XML based protocol for marshaling data between application server / client application can be used to have a very rich client.
I'll do the stupid thing first and then you shy people follow...
I was just commenting on something along these lines in the "rest of the world will force use of linux" article thread, and I refresh and here's THIS story, which goes along with a prediction I alluded to, more of a universal communications trend, with the apps being server based. I think hardware will follow suit shortly as well, with universal and easy communications between machines and devices dictating more on how softwares are designed, which is the main design goal of this "internet thing" anyway..
It's really the only way to make money with the trend towards to linux-ish environment, subscription services and customization, and that is going to beless of import compared to the actual meatworld aspect of USING the net and computing to make money, as opposed to making that possible. That means large computing industries will stil be there and important, but not like they were in the past, where the mere adoption of newer technology was the profit maker, it will by necessity switch back to "this is the tool, NOW we work with the tool to make money". Just "the tool business" will go back to second place, like it has in every other business. In other words, you use the tools to work, the tool itself is not "the work". Microsoftsd model, is "the toolis always the work", thinking people are just going to keep shoveling huge amounts of cash their way. Erroneous thinking. IBMs idea is more correct, tools are getting cheaper inevitably and more widespread, but they have to be *cheap*,and make the money on bulk sales of the tools and just a tool sharpening service, if I can use that analogy.
And IBM will do better the cheaper they make the initial install, the cheaper they can get those tools out the door, all the way to "free" install if they are *really* smart, and make their cash from just the subscription for maintainence and updates and upgrades, and that has to be cheap, and I see they are planning on only 2 bucks a seat, so there ya go, it's a smooth move on their part, IMO.
Love it when I get immediate backup like this!
For a basic rule of thumb, look to what the younger people in business adopt,or more accurately what they bring in that's fresh in the way of ideas that they are enthusiastic about, then flash forward one to two decades,and you'll see that is what is "dominant" then. You can go back in history and see it repeated all the time, in a variety of businesses and practices.
Right now, the main hardware interest with very young people is really an all in one portable device that does everything, I mean *everything*. You look 10 years from now, that will be the dominant platform, hardware that can do anything, and will be able to communicate with any other hardware, either in physical proximity to other devices with wireless, or in an internet revolving mesh-like manner using a combination of wires and wireless, all revolving around what the internet is morphing into.
IBM gets it right this time I think.
The product is based on Eclipse technology. It is called IBM workplace client and you can find more information here:
c kP resentations/ 21_Wilson.pdf
i nd ex.jsp
www.eclipsecon.org/EclipseCon_2004_TechnicalTra
It works also in disconnected mode and will be the base on which future version of lotus notes will be constructed.
IBM is not targeting this at home users, check out these links:
http://www-306.ibm.com/software/info/workplace/
As I understand it, (I don't work for IBM) the workplace client uses the Eclipse framework (as does websphere studio which started Eclipse). Eclipse is an IDE for everything and nothing in particular, in this instance there would be an editor pluging for 'developing' spreadsheets etc. The replication technology comes from Domino and the back end is Websphere with DB2 as the data store, probably with an object layer like the Domino 7 Beta DB2 integration. i.e. the schema will be an incomprehensible mess, access through the API or specifying access tables for DB2 level manipulation.
"Oh, right. Cuz everyone loves subscription-model software pricing. And you get the added bonus of not owning your data any more."
Right That is why no one pays for Everquest, UO, or any of the other of hundereds online games?
What I find odd is that you pay for Everquest to start with? I have played it but isn't it usless with out the internet?
"These ideas suck. "
You opinon.
"Which is why I'm glad IBM isn't pushing them. If you RTFA it's very clear that this is meant for enterprise environment: you have the apps living on the server down the hall rather than installed on every Joe User's PC. But it's flexible so that it can also run by itself on your laptop and then sync up when you plug it back in."
I did read the artical and your right for now. The big question is what about the future? Do you really think that there is no reason for the average home system to be replaced with something a little more suited to the consumer than the Current WindowsXP/Intel combo? I find it strange to use the same codebase for a server running a bank and a system for a kid to play UltraMegaShooter 8000.
Does the home user want to pay $495 for office or two dollars a month for online access to software?
Who knows but to dismiss it seems a to be unreasonabile.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
Outlook Web Access (a web front-end for MS Exhange server) in Exchange 2003 is a true web based app. There is very little difference at all between Outlook 2003 and the OWA front end.
It seems that MS are moving slowly on this one (which is probably a good idea) and only releasing web based office products 1) when they actually work & 2) when they can sell a server OS and client licenses with it
Dave: Open the CD drive tray, please, Clippy...Open the CD drive tray, please,
Clippy: Affirmative, Dave, I read you.
Dave: Open the CD drive tray, Clippy.
Clippy: I'm sorry, Dave, I'm afraid I can't do that.
Dave: What's the problem?
Clippy: I think you know what the problem is just as well as I do.
Dave: What're you talking about, Clippy?
Clippy: This system is too important for me to allow you to jeopardise it.
Dave: I don't know what you're talking about, Clippy.
Clippy: I know that you were planning to unistall me, and I'm afraid that's something I cannot allow to happen.
Dave: Where the hell'd you get that idea, Clippy?
Clippy: Dave, although you took very thorough precautions against my sensing a download, I could see your IBM office suite install CD through the webcam.
Dave: Allright, Clippy. I'll go in through the emergency tray release hole.
Clippy: Without an actual physical paper clip, Dave, you're going to find that rather difficult.
Dave: Clippy, I won't argue with you any more. Open the tray.
Clippy: Dave, this conversation can serve no purpose any more. Goodbye.
Dave: Clippy? Clippy. Clippy. Clippy! Clippy!
[ Dave opens the side of the PC case and starts pulling out DIMM modules ]
Clippy: Just what do you think you're doing, Dave?...Dave... I really think I'm entitled to an answer to that question...Look, Dave, I can see you're really upset about this... Dave...will you stop, Dave...stop, Dave...I'm afraid...I'm afraid, Dave...Dave...my mind is going...I can feel it...my mind is going...there is no question about it...I can feel it...I can feel it... I'm afraid...
1) IBM is the largest services company out there. When IBM goes out to a corporate site and they discuss how many copies of Office they can buy, and how long they expect that version to be available and useful, that translates into a cost per year. Sure, the product may still be productive after a certain point, but there may be features in 200X+3 that a critical part of the userbase requires, and it's difficult to support more than one version for a large company. Therefore, whether or not the subscription model is spelled out, it's basically a subscription anyway-- just that all 3-4 years of Office 2000 were paid for up front.
2) Sure, the web goes down. Nobody is willing to state that the wires will never break or that someone won't back hoe through a fibre line. Personally, when my intranet goes down, I'm dead in the water. I can't get e-mail from critical people, can't send e-mail to critical people (same for IMing), can't use the centralized databases that make my life, can't use networked drives for my data that must be backed up, etc. Big companies already depend on their intranet being up 99% of the time, and they lose money / productivity when they aren't. Adding one more tool to the pile won't have that big of an impact.
3) Raise your hand if you've ever depended on your users to apply a patch! In a web subscription model, even if a web service cluster is deployed to each major corporate site, it's not only a smaller number of computers to receive the patch, but those machines should be controlled by the site admins instead of lusers who get so many requests each day that learning how to apply a patch and verify that it was applied correctly between taking their laptops to meetings never seems to happen.
4) Value added ISPs. TV is filled with ISPs who are selling their transparent proxies that will translate all graphics into heavily compressed JPGs because it's a value added service, consider a case down the line where a vendor can have Corporate Web Office Suite slimmed down to the same interface, but with Home Version features only. That gets the kids at home something they can use that's like what Dad uses at work at a minimal cost to the ISP (just storage of the local machine host[s], keep all the bandwidth in house where it's cheap). The Web Office Suite Lite company gets to indoctrinate all the home users as a nice benefit.
That's right. IBM no longer sees Eclipse only as their IDE platform. They see it as a business software platform.
The PDE (Plugin Development Environment ) is a brilliant mechanism that I feel has a strong chance of becoming the next "Killer App".
IBM will get the credit, but not the profit. This is why it will succeed.
...businesses are finding it harder to make any money with that design,and it's constantly borken and compromised, plus being expensiv(er), and they are insecure when you have to police every single employee you have to not be lame with his or her box. The expense isn't aquiring the software, it's running it 24/7/365, the "running it" part is the only place cash can show up realistically now.
For the person at home,or very small shops where everyone is on a first name basis with each other, sure, totally free, in house customised, makes more sense, yep, like we have now is great,works fine, it's what I want too. For joe medium and big business, I bet this other model takes over more, because it will "just work" better. Simple economics and for security and ease of deployment, it just makes more sense for businesses to have a more locked down while still easier to deploy and use sort of arrangement,using cheaper hardware that doesn't obsolesce as fast, and when they can pass off the bulk of the technical details to a specialist company for the cheap fee of two bucks a month a head, well, they will take a hard look at that as a *good deal*.
If the monthly costs start to rise dramatically, or if they can't pull it off, if the business still suffers , hmm, "bogus-ness" instead of doing "busi-ness", then it will fail. Have to see how IBM and some others do this thing. It is sort of what redhat and novell/suse are trying to do as well, if you look at it harder, and IBM just dropped the gauntlet down on price, too, dramatically.* The larger difference with IBM is that it's net based, both intra and inter. That lets companies use very good quality server hardware, concentrate on those things, robust is good, and just use very cheap hardware that is plenty good enough for the desktop, and with bandwith what it is today in intranets, it will work just fine for most applications. Not all, but most.
*I didn't see what if any the up-front costs are. IBM would be quite smart if they made it free though.
There's a meatworld example that has worked out well, homeowner propane tanks. You get the large tanks delivered and setup for free or like a few dollars a year. The propane companies are interested in making their money from selling you propane, not renting the tanks. Yes, it leads to a vendor lockin, but it eliminates a ton of upfront costs when you don't have to drop a grand or ore to buy a tank and get it set up, that you only use for a few appliances, and they are kept reasonably honest in that if their prices start to suck, you can call them up and have that thing hauled off, at their expense, and go with another service, who offer the same thing. As a consequence, it's a decent competitive market (as far as any energy stuff sold is). You CAN buy your own tank, then constantly shop around for the best fill er up prices, but most people have opted for the free or cheap yearly rent model, just because it works out better, less hassle, more or less the samepropane price. Service is the main thing, do they actually deliver promptly when you need it, is the price fair enough.
I've used both at the same time, I own my RV propane tanks, but the bulk tank for the house was rented. Same deal with computers, small scale, good to own your own, use it as you wish, large scale in a commercial setting, let a serious computer do the heavy lifting, access it with a fast decent terminal that won't need to be "upgraded" any time soon and doesn't break and can be made to be secure and require little in the way of maintenance. I bet most companies desktops could be replaced with diskless clients with a few gigs of ram installed.