Windows Azure Offers Developers Iron-Clad Lock-in 227
snydeq writes "Microsoft's move to the cloud is certain to create a whole new kind of developer partner, Fatal Exception's Neil McAllister writes. But as much as Microsoft ISVs will likely go along with the shift to Windows Azure to keep revenue streams going, the kind of lock-in they will experience will be worlds away from what they face today. Rather than being able to ignore the new version of a key framework, developers will have no other option than to update their code to suit Microsoft's latest platform. That kind of lock-in will leave customers in the lurch, subject to their vendors' bottom lines, as ISVs that can't afford to rework code to keep up with Microsoft's latest platform will begin dropping services, and customers will have little choice but to accept the new terms of service their vendors send along."
Re:So that explains.. (Score:5, Interesting)
I can't imagine customers putting up with this sort of thing for very long, especially in a business environment.
Oops - you didn't pay... your entire business goes dead.
Open source never looked so good.
(and apparently the new Linux version just out shows how the gap is rapidly shrinking)
Re:Don't worry, Miguel will fix it (Score:5, Interesting)
We won't see v1.0 until Microsoft releases Azure v2.0, though.
Re:No serious enterprise customers will adopt this (Score:2, Interesting)
"Two things are infinite: the universe and human stupidity; and I'm not sure about the universe." -- Einstein
Yeah, okay, maybe you're right. ;)
Windows Azure Offers Developers Iron-Clad Lock-in (Score:5, Interesting)
... as opposed to, say, Google App Engine.
Re:Like iPhone (Score:3, Interesting)
for starters, Android is an open platform. Android dev kits are completely free (no developer program membership fee). and Google's distribution agreement, which is far less draconian, only applies if you want to distribute your application through Google. but developers are free to distribute their application themselves.
That's the very reason we abandoned Windows (Score:3, Interesting)
Because you spend so much time serving the Microsoft machine. Not just licensing, product activation and the time and resources that takes, but the constant upgrade cycles, new languages, new versions of the frameworks, security patches that break things...it's all freaking insane.
We scrapped all that. Servers, desktops, dev tools, everything and migrated our development environment and desktops to open source. We can scale for the cost of hardware, our dev tools are simple, don't take all day to install and don't hog all your system resources. We use a lot of command line and prefer it. While you're still installing VisualStudio and getting through registration, we're already working.
Our ROI is off the scale, we have more cash, spend more time actually working and we're turning out systems in time frames that would be the envy of any development shop. We use open source in business and our business works. I came out of a big Windows shop and we blow away anything they're doing with a fraction of the personnel.
So now MS wants to take elements from several product lines, put it in a blender, then lock developers into their way of doing things. Gosh, let me think about that...no.
If Microsoft offered real value, simple licensing terms, and provided products that actually contributed to our enterprise environment without being a dickish pain in the ass, we'd probably have a place for their products in our mix. But right now, no freaking way. Anything MS touches turns to crap. Their products are slow, complicated and bloated and we get by just fine without them.
Re:Like iPhone (Score:3, Interesting)
The fee Microsoft charges for MSDN is a pittance; that's not really an issue. Android is a different market altogether.
The telcos have their own dev market.
Oracle has its own dev market, as does Microsoft, VMWare, and dozens of others.
That doesn't mean I agree with what you have to do to get Microsoft's thunderstorm cloud, but to make it rain, you'll have to spend money and time somewhere. My preference would be in an open environment with lots of choices. But even LAMP is a committment choice-- it just has an open source concept that I personally like to live with. MSDN enforces a discipline that takes a different kind of investment with a different kind of developer and a different potential market.
There are lots of choices in this world; I'm not choosing this one for these and other reasons.
likely developers won't be forced forward (Score:3, Interesting)
Likely you'll leave your Microsoft applications running on the platform version they're developed for while Microsoft may host many platform versions side-by-side. It's not unlike Google maps where developers can choose the API version their application runs with 1.x, 2.x, etc. Microsoft might be evil but they're not stupid, and they've been creating develop tools and frameworks for a very long time. They won't alienate their developers so there's no sense to assume a fearful stance because of a Slashdot submission like this.
Re:Windows Azure Offers Developers Iron-Clad Lock- (Score:4, Interesting)
...which has been reimplemented as open source.
It only took maybe a week after it was launched, too.
The only reason you'd be "locked in" to Google's service there is if you depend on them hosting your app for free. I call that a fair trade.
Is this any different than Gooogle App Engine? (Score:4, Interesting)
I haven't delved deep into the workings of either... but is the Azure/Microsoft lockin any different than lockin would be in writing apps for Google's App Engine?
Exactly like OS X. (Score:5, Interesting)
I'm usually the first to bash Microsoft. I'm usually the last to defend them. I do think they deserve every bit of flame they get.
But this is just getting stupid...
Apple did exactly the same thing with OS X. I'm talking about the initial launch -- OS X was a completely backwards-incompatible change from OS 9. In fact, there were major architectural changes -- like the introduction of such modern features as protected memory -- which would have made it pretty much impossible to maintain pure backwards compatibility and do everything they wanted to do.
So they said "fuck it", switched to a completely different architecture, and wrote an emulation/virtualization system called Classic.
One thing which I know I've heard described for Windows 7 was the ability to run an older version (like Vista) in a virtual machine. You know, kind of like Classic. The only difference would be if Microsoft wanted to charge you for the license -- and I hope they aren't that stupid.
I (and others) have frequently disparaged Microsoft for their bloated, crufty, undocumented (or under-documented, or mis-documented), and downright weird APIs. I know that before I heard about this change (which isn't news, by the way, it's been on Slashdot before), I figured I would do exactly the same thing if I was in Microsoft's shoes. Don't even try to support the old APIs -- just start entirely from scratch, build a compatibility layer, and tell people to upgrade.
One more thing, and then I'm done: What the fuck does this have to do with lock-in? What, did you think Win32 was open? It's only portable thanks to Wine, and Wine never has, never will, never can catch up and support every single app.
If you're going to be locked in anyway, why not be locked into something newer and (presumably) cleaner?
If it's not clean, that's another argument. But this strategy is not about lock-in.
End rant.
Re:Well... (Score:3, Interesting)
Have you ever taken a good look at Microsoft's frameworks, such as MFC? Abstracting that sh*t out is _hard_. Much of your design is buried in non-code proprietary "resource" files which do not provide the facilities for layering or abstraction. In Windows you can dip below the high level OO APIs and program to the lower level C libraries (Win32), that's how Qt, FLTK, and other frameworks do it. It costs you productivity, but it lets you write portable code. Most programmers aren't going to think that far ahead - they'll take the easy way which Microsoft puts front and center in their documentation and certification classes. Low level programming may not even be _possible_ in Azure; you may only get the high level "easy" APIs which prevent abstraction for portability. Then your only option is to write emulation libraries for other platforms which can run the same designs as Azure, assuming that patents and terms of service agreements don't disallow it.
Apple and Microsoft are like peas in a pod (Score:3, Interesting)
What Apple did with OSX, Microsoft did with Vista and Azure. But as I recall OSX Classic mode couldn't run all legacy Mac programs just as Vista's Win32 Legacy mode cannot run all Legacy Windows and DOS programs.
I recall Mac OSX had the Basilisk 2 [online.fr] emulator to run Classic Mac 68K programs that OSX Classic mode couldn't run.
Windows Vista uses VMWare or Virtual PC to run XP and under in Vista for Legacy Windows and DOS programs.
But it is ironic that Amiga, Inc. when it wrote AmigaOS 3.1 found a way to run the old 68K and PowerPC AmigaDOS/Workbench 1.X and 2.X programs under it without too many problems, and even gave legacy rights to a group to create an open source version of AmigaOS 3.1 called AROS [sourceforge.net] Amiga Research OS that can run on i386 and PowerPC systems and have built in emulation for 68K Amiga code based on UAE with their own version of Kickstart in AROS with backwards compatibility.
Amiga got it right, Microsoft and Apple didn't, for solving Legacy Software problems.
Re:Have faith. (Score:1, Interesting)
http://marketshare.hitslink.com/report.aspx?qprid=10 [hitslink.com]
Sorry to disappoint you.
Linux seems to be doing great though. In about five years it should totally surpass Windows 2000.
Re:Vuze? (Score:3, Interesting)
When I think of Windows Azure, I tend to think of This Game [wikipedia.org].
Bill Gates and Microsoft are part of the New Alliance, and put Azure tattoos on the developers so that they can secretly control them and lock them into working for the New Alliance.
Re:Windows Azure Offers Developers Iron-Clad Lock- (Score:2, Interesting)
Re:Vuze? (Score:4, Interesting)
Realistically in tight economic times development will stagnate, companies will stick with what they have for as long as they can and only change when they are forced too and then that change will be targeted at long term solutions, where they have the greatest control over outlays and future investment cycles.
Really? Because that's not the trend I'm seeing - at all.
As a hosted application provider, I'm finding our clients squeezing their belts, left, right, and center. They're nervous, they're scared, they're jumpy. And it seems that the more jumpy and scared they get, the more contracts we are signing, left, right and center!
See, our product is designed to cut costs by automating compliance to legal requirements. It's a hosted application, and many of our new contracts view us as a way to eliminate the cost of maintaining home-brew stuff that's low quality and costly to maintain. Our products, on the other hand, are comprehensive, well funded, and reasonably priced.
I guess you could say that there is some lock-in with our product, because although we don't want to hold anybody hostage, we're not giving away our source code, either. We certainly wouldn't hesitate to turn over our client's data on demand, (they can click-to-download most of it without ever consulting us) but our clients aren't generally the coding type, and the marketplace for our wares is almost a niche.
I believe that a good business is truly a relationship between the business and its customers. When the business truly considers the needs of its potential customers, and works to meet those needs in an efficient, professional, and competent manner, the customers really won't mind returning the favor. In our case, we almost let our customers outsource their worries to us, and we work hard to make sure that we deliver.
The result? Rapid growth, and customers who rave!
Re:Don't worry, Miguel will fix it (Score:3, Interesting)
I attended PDC this year. Miguel gave a talk about / demo of Mono and its future on Wednesday evening to an enthusiastic standing-room only audience. At one point, a Microsoft leader in the back (couldn't see who it was) yelled out, "Microsoft loves Mono!".
More interesting was "The Future of C#" talk earlier in the week, where the future of C# was revealed to be... Python. No kidding. C# 4.0 will include a static type of "dynamic" (the jokes just write themselves), which uses Jim Hugunin's (of Jython / IronPython fame) Dynamic Library Runtime (DLR) to do duck typing et. al. C# 5.0 virtualizes the C# compiler so that an executing C# program can compile and execute C# code on the fly. I thought the C# developers were going to break their hands in wild applause as the speaker demonstrated a command line written in a prototype C# 5.0 from which statements executed dynamically - like you get when you type "python" at a command prompt. It was a bit surreal; these guys need to get out more. :-)
The punch line is that Miguel demonstrated the use of the DLR and compiler virtualization in the current version of Mono.
I know his reputation isn't sterling in these parts, but there's certainly nothing wrong with his coding skill. His was also one of the most entertaining talks at PDC (and the only one with the demos hosted on Linux).
Re:so what? (Score:2, Interesting)
No, you're wrong. The reason why Microsoft is dominant, especially in the office is simply because MS Office has become dominant. Every office uses it. 3rd party applications are not 100% compliant, and never will be. When MS Office is no longer the de facto standard, then things will change.
I had high hopes for ODF, until it was politically killed by deliberately political interference in the state of Massachusetts. Then there was the deliberate, and illegal interference at the ISO voting (and no government seems to be pulling aside their local ISO group and arresting them for falsities - go figure). It's amazing what shit loads of money and heavy ties to political figures can do.
Dave