Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
The Internet Businesses Google Microsoft

Changing Climates for Microsoft and Google 393

ReadWriteWeb writes "Weather metaphors abound as this article looks at the evolving software environment — and in particular the competition between Microsoft and Google. Milan says that while Google enjoys relative dominance on the Web platform today, two fissures exist that will force them to move. The first is Microsoft's ability to use the exact same HTML based strategy as Google (like Microsoft's current Live initiative); and secondly Microsoft leapfrogging the current environment by solving rich application installation/un installation and enforcing an acceptable contract regarding what rich apps can do on a user's machine. Unfortunately for Google, Microsoft is a lot closer to solving these two issues than people think. Microsoft has the best virtual machine with .NET, the best development tool with Visual Studio and the best access to developers with their MSDN programs. And they have a notion. Steve Ballmer himself has started touting the exact strategy they need — Click Once and Run."
This discussion has been archived. No new comments can be posted.

Changing Climates for Microsoft and Google

Comments Filter:
  • What MS Doesn't Have (Score:5, Interesting)

    by nate nice ( 672391 ) on Monday December 04, 2006 @12:51PM (#17100494) Journal
    They have .NET which is greta and all, but for the Web they leverage ASP.net which is still a dinosaur of an idea.

    Google has GWT, which only about 100 people on Earth get right now. Google has an understanding for the Web, Web applications and how users should interact in the World Wide Web far surpassing MS's "reactive" method of toolkit design.

    I see two companies. One which is using old methods, not innovating or developing new ideas and assuming stability in something as fast moving and cutting edge as the WWW. I see another company challenging old ideas (relatively old anyways) and proving the WWW is more than Web Pages and stateless client/server communication.

    I see a company that think they get this but only see flashy UI's as the means to the end here. I see another company that understand the UI is just a view to this new idea that the Web is a series of intercommunicating applications users can access from anywhere.

    But then, I don't expect many people, especially a monolith who's made their fortunes through brute strength rather than new ideas, to see this until it's apparently obvious. The search for the holy grail of the Web's next "killer app" is right in front of peoples faces.
  • by rjdegraaf ( 712353 ) on Monday December 04, 2006 @12:52PM (#17100508)
    You can try for yourself:

    on www.google.com search for 'microsoft':
    Results 1 - 10 of about 393,000,000

    on search.live.com search for 'google':
    google page 1 of 751 results

    I like my search results 'unbiased', so I choose google.
  • Re:google is (Score:3, Interesting)

    by somersault ( 912633 ) on Monday December 04, 2006 @12:53PM (#17100540) Homepage Journal
    It didn't start off that way, well as far as I remember, but still.. what's not to respect about a company that does its job well, and in an unobtrusive (compared to all the crappy flash ads and banners we have these days) way? Not that I usually look at the google ads either. In fact I think they're being blocked completely now with an ad blocker..
  • by ummit ( 248909 ) <scs@eskimo.com> on Monday December 04, 2006 @01:09PM (#17100746) Homepage
    We're not talking about "will get it right ... introducing some practically workable mechanism for allowing only trustworthy code", We're talking talking about a model laid out in .net 1.0 and refined in 2.0 about a year ago.

    Neither of us will convince the other on this point, so I won't try.

    If, a year or two from now, .net 2.0 (or whatever version it's up to by then) is stable and secure, I will say, "Shit, I was wrong."

    I ask only: if, a year or two from now, there is some undreamt-of new "impossible" attack against or subversion of the idea, such that people are clicking once and getting pwned all the time, you do the same.

  • by Rik Sweeney ( 471717 ) on Monday December 04, 2006 @01:10PM (#17100766) Homepage
    Microsoft has the best virtual machine with .NET

    If it's the best then why doesn't it work on a Mac or Linux?
  • by hiroller ( 994761 ) <dvan_cuyk AT hotmail DOT com> on Monday December 04, 2006 @01:18PM (#17100880)

    I'll concede on the third allegation which I interpreted as the denial of access to the source code. This is one of the reasons that I have Linux running on my home box since I like to know how things tick on the inside. But I develop with M$ at work and I wanted to point a few things:

    "Microsoft has the best virtual machine with .NET,"
    Nope.
    Actually, I don't know if I could say that it is the best ever but it is a damn good virtual machine! It can run as well or even better of its equivalent JVM http://www.gotdotnet.com/team/compare/Benchmark_re sponse.pdf [gotdotnet.com].
    "the best development tool with Visual Studio"
    Nope.
    Bar none, VS is the best development tool that I have used. M$ V$ 2005 alone is amazing and while it oversimplifies things, I like it b/c it makes me tremendously more productive which is great b/c now I have more time to read Slashdot at work!

    Just b/c it's made by M$ doesn't mean that it is a horrible product. The company itself makes some really shady ethical decisions but there are a lot of developers working for M$ just like us who want to release a great product.

  • by SashaMan ( 263632 ) on Monday December 04, 2006 @01:27PM (#17101012)
    I originally thought your comment was trolly, but it actually turns out to be true, especially if you compare all possible cases:

    Google searching "microsoft": 39,500,000 results
    Google searching "google": 52,800,000 results
    MSN searching "microsoft": 80,139,835 results
    MSN searching "google": 648 results

    I can understand leaning a little more one way or the other, but 648 versus 52 million? Give me a friggin break.
  • by morgan_greywolf ( 835522 ) on Monday December 04, 2006 @01:35PM (#17101178) Homepage Journal
    Wrong. I said "Arbitrary files" not "any files". Go look up "isolated storage" - it allows a partially trusted app to read and write files, while ensuring that the only app that it is capable of messing with is itself. And what's so bad about remote servers? It works for gmail.
    This is yet more argument from ignorance.


    Maybe. But when you have an OS where major parts of the GUI subsystem run in ring 0 with many, many bugs in that subsystem, making installation of a trojan or a worm or other malware a simple matter of exploiting those security bugs, I don't exactly get that 'warm and fuzzy' feeling about One-Click installs of applications from the Internet, an inherently untrustworthy network.
  • by Almahtar ( 991773 ) on Monday December 04, 2006 @01:59PM (#17101524) Journal
    I worked on a codebase of several hundred thousand lines over the summer, all on VS 2005. I was excited to see what all the buzz was about for myself. Intellisense constantly lied about which functions call others (example: it told me a function called itsself while I was looking at it - all 5 lines of it - and it clearly did not...). It constantly jumped to forward declarations when I asked where definitions were. The toolbar buttons would change their placement on occasion, and I would have to put them back. The program they provide for browsing the Windows API documentation frequently crashed (especially when left minimized for a half hour or so).

    One of my favorite "features" was when I would tell Visual Studio to close and it would decide what I really meant was "update your intellisense then close". Great. With a project that size updating intellisense took about 2 minutes. I don't need intellisense updated right now, because I can't use it if you're closed. Just close.

    The real clincher, though, was the "crash-on-debug" error that started plaguing the office. When you tell VS to "build and debug" it would build the program and then seg fault immediately. That's a serious pain with a large project because it takes a few minutes to load it again. To debug, you'd have to build the program then run it manually and then manually attach the process for debugging. This bug would strike staff at random, and the only solution was to do a complete rebuild of the entire project, non-distributed. This could take hours.

    With the amount of talent in that office and the amount of frustration at that crash, we could have just fixed the bug ourselves and saved a lot of time if the product in question was open source, but it wasn't.

    Visual Studio has cost that company a lot of money in wasted man hours.
  • Re:google is (Score:4, Interesting)

    by jdray ( 645332 ) on Monday December 04, 2006 @02:10PM (#17101720) Homepage Journal
    I actually like the ads in GMail. They're context-sensitive, so when I'm chatting with one of my friends about the latest hare-brained idea (rocket boosters based on parafin/lox, personal VTOL aircraft, etc.), we get an on-going catalog of mostly-related products. Some of them have been very useful, and gotten us past some difficult engineering problems.
  • by truthsearch ( 249536 ) on Monday December 04, 2006 @02:12PM (#17101752) Homepage Journal
    I've written a fair amount of .net code to run within a private corporate network. I've never written anything to be run through the browser or "from the internet". But I was calling COM objects with no security limits. Does .net "off the internet" prevent COM or Win32 calls? Because if not a few lines of code can control the entire windowing system, and probably worse.

    In my experience .net is just a poor copy of java. This discussion is like the mid-1990's all over again. Virtual machines, sandboxes... it wasn't popular last time. Why should this time be any different?
  • by plopez ( 54068 ) on Monday December 04, 2006 @02:19PM (#17101888) Journal
    Look at their market cap, the reconizability of the brand, the domination of the desktop, the desire for college CS students to get hired there, and who doesn't know about Bill Gates' multi-billion dollar charity? That's not anything to be respectful of?>/i>

    Market cap, obtained via a illegal and uethical business practices. So no. I have more respect for Warren Buffet as he earned his money. Wealth does not automatically bring respect.

    Destop domination, no. Obtained through illegal and unethical means.

    The opinion of college students? No, they usually have nothing in thier experience to foster a good opinion. Though some are sharp and if they come from a non-trad background (older, ran thier own business, prodigies etc.) I might listen.

    Gate's charity? Obtained through illegal and unethical means. Therefore it is blood money. Just because someone is wealthy and donates to charity does not mean they should be respected. Respect should be reserved for those who work hard and obtain thier goals via legal, ethical and moral means.

    But that's just my opinion.
  • by 0xABADC0DA ( 867955 ) on Monday December 04, 2006 @02:59PM (#17102488)
    Uh, right, so to show CLR running "as well or even better of its equivalent" you compare two entirely different web architectures, J2EE and .NET. Kind of like comparing performance of OpenOffice to MS WordPad and concluding that g++ is much slower than MS c++.

    To benchmark these is complicated as .NET gets a lot of its speed by directly calling C functions, whereas in Java pretty much everything is implemented actually as Java code. This is of course a tradeoff since .NET contains a LOT more functions that can potentially be cracked because they are not run as 'managed' code. In the end it's basically a wash on performance; on actual virtual machine operations Java is in general significantly faster and on API calls .NET is usually quite a bit faster.

    DotNET has a couple technical shortcomings when it comes to performance, one being the impossibility of fast interpretation of bytecodes (the instructions depend on the argument types so can't be easily dispatched). Another is using "real" generics, which they thought would improve speed by avoiding some box/unbox operations but it also leads to type explosions and, slow instanceof and casts (for example you could use so much memory just due to instantiated types that CLR has to constantly throw away older code and re-JIT, not to mention poor use of cache).
  • Re:Here's a test... (Score:4, Interesting)

    by OldeTimeGeek ( 725417 ) on Monday December 04, 2006 @03:35PM (#17103032)
    Many companies don't run anything except for Windows on their desktops (mine included).

    Many, but not all.

    The company that I work for has some very important customers that don't, and I'd rather spend the time making sure that we worked without regard to operating system than being in the position of having to tell them that we're not interested enough in their business to make our site work for them.

    Besides, who knows what the future will bring? Fifteen years ago, if someone told you that you should start developing for Microsoft NT/AS because Novell wouldn't be a factor in the NOS business, would you have believed them?

  • by recursiv ( 324497 ) on Monday December 04, 2006 @07:10PM (#17106198) Homepage Journal
    I turned off adult content filtering in the options:


    google
    Page 1 of 65,601,473 results


    Very strange

Say "twenty-three-skiddoo" to logout.

Working...