Ballmer Sends Wakeup Call to Staff

Puneet writes "An MSNBC article outlines details of how the world's biggest software company seems to be facing a technology gap. Steve Ballmer, chief executive officer of Microsoft, sent a memo across the company basically saying that with no immediate breakthroughs in technology coming, and with the Linux computer operating system and a batch of other open-source programs biting at its heels, Microsoft will have to do a better job of persuading customers it has something they need. . Microsoft must "improve business consistency" so that customers are not hit with unexpected - and unwanted - changes. Also covered by Forbes but in lesser detail."

Comments

Re:Huzzah!

[Chief Crazy Chicken]

Linux and all of its branches like BSD

Timeline of GNU/Linux and Unix

Note particularly:
1980: Bell Labs finally shows interest in BSD Unix
-and-
1991: 05Oct: linux 0.02, first mention of directory-name 'linux' on netnews

Maybe ballmer should read more

[Anonymous Coward]

When you compare POSIX native thread in the next release of Linux and this article by Chris Brumme about AppDomains it's obvious the issues with distributed transaction on windows platform has serious problems. In Brumme's article, he discusses why creating new threads is heavy weight and diificult to scale. Read his other articles, they are very informative. Distributed transactions don't necessarily require threading, but without a robust threading implementation, solving the problem is that much harder. Not only that, doing complex distributed transactions requires a robust Object Persistence manager, which isn't available from microsoft. There are third party tools for .NET that do Object Persistence management, but it's not nearly as mature as several Open source apps.

There are several important differences between how .NET handles dynamic runtime loading of classes and how java does it. .NET requires a separate AppDomain, which means it has a higher overhead. Using a separate AppDomain is only needed if you need to unload/reload an assembly at runtime. Although java classloaders are difficult to grasp for many programmers, it provides a better way of handling dynamic loading. I won't bother going into the details of how dynamic loading works. Tomcat has plenty of examples of how it is done for each webapp.

When did .NET fail?

[OrangeGoo]

Sorry, I haven't seen a failure of .NET. I'm just curious where you're looking. I work for the US Army Corps of Engineers, and we use the heck out of .NET and everyone loves it. There is some Java development here, too, but most of our new stuff is in C# (which is, of course, essentially a Microsoft-ized Java).

I haven't heard any complaints from people who use .NET on a regular basis. Personally, I think it's great.

Re:.NET failed?

[kahei]

Yeah, I know. I am STILL regularly explaining to people what the hell .NET is. Microsoft could have said:

'.NET is a runtime environment and set of libraries for programs written in a bytecode called IL. There are some developer tools that compile languages like C# to IL, and there are some high-level services like ASP.NET implemented in .NET'

What they said was, I believe:

'.NET is all about XML. .NET *is* XML'

This is part of what they got for putting Steve Ballmer in charge.

So as a PR thing, yeah, totally mishandled. But for providing solutions, it's very good -- I'd use it over Java whenever possible, and so would several ex-Java people I know.

Re:One of Microsoft's strong points

[drix]

they realized they were too late in jumping on the Internet bandwagon, they admitted it and started development on a browser to compete with Netscape.

You can be damned sure that if Netscape had a stranglehold on the desktop OS market in the mid 90s, I would be writing this post on some incantation of Mozilla right now and not IE 6. MS's eventual triumph in the browser wars had nothing to do with its capacity to innovate and really not even that much to do with it's ability to play catch-up, which it isn't even that good at. I mean, anyone with half a brain will tell you that feature-, speed-, and stability-wise, Mozilla 1.4b rocks anything that MS has ever created, and this after years of Netscape, Inc.'s atrophy and braindeath, to boot. So basically, their ability to bundle and integrate the browser and OS saved their butts from a "war" that by all means they should have lost, and not their introduction a superior, albeit late-to-market, product.

The bedrock of MS's business model has always been the fact that, no matter how much they fuck up in other sectors, at the end of the day they still own the OS market. You mentioned the WordPerfect/Lotus episode, which is another good example of this. WP is arguably superior to Word to this day. How did MS extricate themselves from this particular snaggle? Why, code hooks into the OS to improve performance, of course.

I have a hard time seeing how that dubious technique is going to save them this time. What are they going to do, bundle their operating system with their operating system? :) Make Windows perform better by integrating it with Windows? Once the foundation of their business model begins to erode, the emperor has no more clothes. This is a scenario that, until Linux, they haven't really been faced with before, and it's going to obligate them to take a long, hard look at the very core of their corporate philosophy, culture and business model. Institutional momentum being what it is at one of the world's largest companies, I pray with renewed hope these days for the eventual death and destruction of MS :)

Re:One of Microsoft's strong points

[expro]

WordPerfect had been told for years by Microsoft that OS/2 was the future for corporate work. They had a great OS/2 version representing a much larger investment with much fewer framework problems, but which ultimately didn't have enough potential market to even justify release of a final supported version after Microsoft pulled the fast one shifting from OS/2 to Windows for corporate use forcing app developers to use greatly-degraded facilities, which Microsoft had been practicing at a bit longer.

Having to suddenly deal with all the Microsoft "innovations" of Win 16 resulted in a result that was, by comparison with Microsoft's efforts, crap. Sure, Microsoft was better at dealing with the sudden shift and limitations of their own monstrosity (or perhaps you would like to be using it today). This is characteristic of Microsoft's strategy of adding bumps to the road for other developers, leveraging their control of the OS against applications developers.

Re:What else are they supposed to do?

[alexandre]

A linux for school distro was just relaesed here in
Québec:

http://www.edulinux.org/spip/

it's based on mandrake 9.1 and add some better local french support and many useful tools needed in university or college ... everyone should try and make that in their own state/province :-)

Re:What else are they supposed to do?

[ThogScully]

When Linux can come up with a browser that works as well on every site I use

I can't believe you think IE is a better browsing experience than Mozilla. If you do, it's likely that you just haven't used Mozilla. Mozilla's features and standards compatibility are so much more advanced than IE that you've debunked your own post in this one statement alone.

comes up with 100% Office clone (and I do mean 100%)

Trying not to sound like a conspiracy theorist here, but you're already too reliant on MS to notice anything else. This isn't a possible goal. There's no way that reverse engineering the Microsoft proprietary formats can be as quick and accurate as Microsoft designing them. By using MSOffice, you are supporting the very formats which have already locked you into Microsoft for as long as Microsoft remains proprietary (read forever).

Realistically, if MSOffice could open OpenOffice files, I would really have no need to use it again. I don't notice the features it lacks and like many of the features it has over MSOffice and I assume I'm not alone in that. My only qualm is that I have to save files in MSOffice formats and confirm that they look right in MSOffice before sending them to someone using it.
-N

IE's lunch has been well and truly eaten, then...

[leonbrooks]

Try a few of these in konqueror:

audiocd:/ (yes, put an audio CD in your first CD drive before you click)

fish://luscious@your.fave.ssh.server

smb://nearestdozebox/c

There are plenty of others.

Re:Microsoft "Producing" software ?

[nathanh]

No? Who did they steal BOB from? Word? Excel? Visual Studio? Visual Basic? .NET? Windows NT/XP/2000? XBox?

Well, the word "steal" isn't quite correct, but if you say "bought" or "borrowed" then all of your examples fall flat. Let's go through them all:

BOB was a rather poor reimplementation of a software product called Lemon Dog written by the Inner Workings company. Microsoft's character in Bob was Rocky Dog, if you recall. Inner Workings sued Microsoft over that one (and settled out of court I think).

Visual Basic was originally called Ruby. It was written by the Cooper Software Company and bought by Microsoft in 1991.

Windows NT was largely written by Dave Cutler's crew who was "bought" en-masse from Digital. It's little wonder that NT and VMS share a fair number of similar design concepts; there are many in-depth articles discussing the similarities. The user interface is actually CDE-based; lots of people seem to forget that CDE was both a standard co-developed by Sun, IBM and Microsoft as well as an implementation that seems now to be associated mostly with UNIX systems. Many other parts of Windows NT were co-developed with IBM as part of the OS/2 project before Microsoft decided to go their own way with Windows.

I don't think anybody could seriously argue that .NET is anything other than a direct ripoff of J2SE and J2EE. Even the architectural diagrams are the same. When you go to a .NET Developer course the instructors start with "basically it's just like J2EE".

Visual Studio is ill-defined. Visual C is just Lattice C bought from Lattice Technologies. Visual Sourcesafe was bought from OneTree Software. The user interface for Visual Studio (the editors and syntax checkers, etc) are obviously a ripoff of Visual Cafe from Symantec; even the name lacks any imagination.

Xbox is a PC masquerading as a console. It was most certainly a Johnny-come-lately to the console market. It was most certainly a "me too" response from Microsoft. I don't see how you can possibly argue that the Xbox is an example of Microsoft creating something new.

Word, Excel, and all the office suites. Well those things are more an evolution than anything else. You can certainly argue that their first versions were ripoffs of existing products (eg, WordPerfect, 1-2-3, Visicalc) but there is little resemblance to those original products in the modern versions. PowerPoint was bought from ForeThought Technologies; and it shows, because PowerPoint has never properly integrated into Office.

But to claim that they've created nothing new is pure ignorance.

I'm sure they have created something new (they are a huge company with 1000s of projects) but I am honestly struggling to find an example. Nothing you've listed is all that convincing. Even their original flagship products - MS-DOS and Basic - were either bought or "borrowed". Microsoft is a good repackager and reimplementor, but they have never been a good creator of software or ideas.