Slashdot Mirror

Er, we've already got a great (at least in the sense of "large") Web infrastructure with market buyin and lots of developers, which is totally cross-platform. Distributed X apps have none of that. If you haven't seen any good AJAX drawing programs yet, that means you should encourage people to develop one, not discourage it. Especially since encouraging X development for the same benefit is obviously a doomed adventure.

I was using X more as an example of a dumb, supposedly obsolete system for network-transparency which can still do things which are impossible with AJAX. Like, erm, drawing circles quickly, setting pixels to particular colours without creating a miniature new document section and so on.

As traditional graphical user interface APIs go, AJAX is basically shit, as it's a simple document markup system with scripting grafted on. Its main, colossal advantage is that everyone's got the client software already - there's no X server, Java virtual machine or VNC client to download. But for someone designing this hypothetical drawing program, they'd be writing it without the aid of any modern graphics APIs. I'd be utterly amazed if someone produced a fast, usable and cross-platform equivalent of (say) Inkscape using AJAX.

Improving existing web applications? Brilliant! Making possible interesting new ways of submitting textual data or retrieving graphical data? Definitely. Recoding existing, offline desktop applications in a way that isn't a gigantic hack? Not really...

Now there was an innovative company - typically for a UK computer company, they were far better at developing tech than at selling it. The exact converse is what seems to bring success in the world, sadly...

In case you are using an object-oriented programming language, you may ask yourself if SQL really is a good choice. SQL is not object-oriented, it's use is usually neither typesafe nor compile-time checked in OO languages, it can't be refactored from an object-oriented IDE and it does not follow OO principles.

Food for thought:

Native Queries
http://www.db4o.com/about/productinformation/white papers/

DLINQ
http://msdn.microsoft.com/netframework/future/linq /

Being somewhat involved in the VoIP(/telecom) world I think Software Isolated Processes is a poor name for their new technology. SIP as in Session Initiation Protocol is the de facto standard for VoIP communication, hell even Microsoft has adopted it

This Tony hoar guy looks nuts. I mean, look at that... was he wearing his pajamas, or did he stole his wife sticking ? And why did he put this protograph on his bio ? Even the worse geek would take more care of what he gives to see. He must of an ancient, forgotten species. The uber alpha geek, or the Meta Punk Geek Granny ! Anyway, i like his pants.

If you want to get an idea of some cool SQL improvements, check out the
http://msdn.microsoft.com/netframework/future/linq /
LINQ (Language Integrated Query) project for c# 3.0. Some cool stuff tht i never really thought about.
 
    For example, their select statements go backwords ie from table, select column1, n2, n3 etc... Seems kinda wacky at first, but it makes sense since you really should know what table your'e selecting from before you specify the columns.

ex.

public void Linq3() {
        List products = GetProductList();

        var expensiveInStockProducts =
                from p in products
                where p.UnitsInStock > 0 && p.UnitPrice > 3.00M
                select p;

        Console.WriteLine("In-stock products that cost more than 3.00:");
        foreach (var product in expensiveInStockProducts) {
                Console.WriteLine("{0} is in stock and costs more than 3.00.", product.ProductName);
        }
}

Silly href. research.microsoft.com

went to the site http://research.microsoft.com/error.aspx?aspxerror path=/os/singularity/default.aspx as mentioned in the story & was greeted with an aspx server error... the irony is just too great!

I'm only replying to the parent so that this post is high up the screen.

Look at page 31 of this PDF. Microsoft publish benchmark statistics showing Linux (and FreeBSD) to be better than Windows.

ftp://ftp.research.microsoft.com/pub/tr/TR-2005-13 5.pdf

• Tony Hoare (quicksort, CSP, Hoare logic, among other things)
• Leslie Lamport (TLA+, the bakery algorithm, the byzantine generals problem, LaTeX, among other things)
• Simon Peyton-Jones (Haskell, many things FP)
• Luca Cardelli (many things in type theory and PL semantics)
• Eric Meijer (many things FP)

I dislike MS production software and business practices as much as the next guy. But don't make the mistake of underestimating MS Research just because you dislike MS.

• Tony Hoare (quicksort, CSP, Hoare logic, among other things)
• Leslie Lamport (TLA+, the bakery algorithm, the byzantine generals problem, LaTeX, among other things)
• Simon Peyton-Jones (Haskell, many things FP)
• Luca Cardelli (many things in type theory and PL semantics)
• Eric Meijer (many things FP)

I dislike MS production software and business practices as much as the next guy. But don't make the mistake of underestimating MS Research just because you dislike MS.

• Tony Hoare (quicksort, CSP, Hoare logic, among other things)
• Leslie Lamport (TLA+, the bakery algorithm, the byzantine generals problem, LaTeX, among other things)
• Simon Peyton-Jones (Haskell, many things FP)
• Luca Cardelli (many things in type theory and PL semantics)
• Eric Meijer (many things FP)

I dislike MS production software and business practices as much as the next guy. But don't make the mistake of underestimating MS Research just because you dislike MS.

• Tony Hoare (quicksort, CSP, Hoare logic, among other things)
• Leslie Lamport (TLA+, the bakery algorithm, the byzantine generals problem, LaTeX, among other things)
• Simon Peyton-Jones (Haskell, many things FP)
• Luca Cardelli (many things in type theory and PL semantics)
• Eric Meijer (many things FP)

I dislike MS production software and business practices as much as the next guy. But don't make the mistake of underestimating MS Research just because you dislike MS.

• Tony Hoare (quicksort, CSP, Hoare logic, among other things)
• Leslie Lamport (TLA+, the bakery algorithm, the byzantine generals problem, LaTeX, among other things)
• Simon Peyton-Jones (Haskell, many things FP)
• Luca Cardelli (many things in type theory and PL semantics)
• Eric Meijer (many things FP)

I dislike MS production software and business practices as much as the next guy. But don't make the mistake of underestimating MS Research just because you dislike MS.

can't be slashdotted already surely, not the great microsoft, for http://research.microsoft.com/os/singularity/ i get the following error message -

Server Error in '/' Application.
Runtime Error
Description: An application error occurred on the server. The current custom error settings for this application prevent the details of the application error from being viewed remotely (for security reasons). It could, however, be viewed by browsers running on the local server machine.

Details: To enable the details of this specific error message to be viewable on remote machines, please create a tag within a "web.config" configuration file located in the root directory of the current web application. This tag should then have its "mode" attribute set to "Off". ...

PDF of the Microsoft paper

This is actually really cool stuff. Here is the pdf. Hopefully the ftp server is still up. ftp://ftp.research.microsoft.com/pub/tr/TR-2005-13 5.pdf

Maybe a good chance if you are interested!

We are hiring! If you are interested in a full-time Researcher position, please email a C.V. or resume, a research statement, and the email addresses of three reference letter writers to Galen Hunt. You may also email copies of two publications you feel represent your best work. Minimum education requirement for a Researcher is a Ph.D. in Computer Science or equivalent.

To facilitate our hiring process, we strongly encourage interested fulltime researcher candidates to submit their application materials as soon as possible and preferably by February 15, 2006.

In evaluating candidates, we pay particular attention to demonstrated qualities of research taste, innovation, and first-hand system building. We value highly a proven research track record as demonstrated by strong publications in top venues.

If you are an exceptional Ph.D. candidate interested in a research internship, please use the MSR Internship Application.

Microsoft is an equal opportunity employer and supports workforce diversity.

Maybe a good chance if you are interested!

We are hiring! If you are interested in a full-time Researcher position, please email a C.V. or resume, a research statement, and the email addresses of three reference letter writers to Galen Hunt. You may also email copies of two publications you feel represent your best work. Minimum education requirement for a Researcher is a Ph.D. in Computer Science or equivalent.

To facilitate our hiring process, we strongly encourage interested fulltime researcher candidates to submit their application materials as soon as possible and preferably by February 15, 2006.

In evaluating candidates, we pay particular attention to demonstrated qualities of research taste, innovation, and first-hand system building. We value highly a proven research track record as demonstrated by strong publications in top venues.

If you are an exceptional Ph.D. candidate interested in a research internship, please use the MSR Internship Application.

Microsoft is an equal opportunity employer and supports workforce diversity.

Considering the problems their web server is having though, here's the link to the technical paper (In PDF Format) so you can skip the website altogether. Singularity PDF

Singularity is a research project in Microsoft Research that started with the question: what would a software platform look like if it was designed from scratch with the primary goal of dependability? Singularity is working to answer this question by building on advances in programming languages and tools to develop a new system architecture and operating system (named Singularity), with the aim of producing a more robust and dependable software platform. Singularity demonstrates the practicality of new technologies and architectural decisions, which should lead to the construction of more robust and dependable systems.

Where have I hear that before?.

A key aspect of singularity is ... Software-Isolated Processes, which encapsulate pieces of an application ... and provide information hiding, failure isolation, and strong interfaces. All code outside the kernel executes in a SIP.

That sounds alarmingly like a closed address space. Microsoft in 2005 is giving us what UNIX had over 30 years ago. thx u sir!

SIPs are closed object spaces, not address spaces

I stand corrected! What's the difference?

Two Singularity processes cannot simultaneously access an object.

Ruh ruh! So we're going back to the model where two processes can't open a handle to the same file? You mean ... like .. er... DOS?

A process cannot dynamically open or generate code.

So like ... you can't run Perl on it. Or shared libraries. #include <dlfnc.h> is a thing of the past.

SIPs are created and terminated by the operating system, so that ... resources can be reclaimed.

Yeah uhhh... the runlib library for, say, C executables in UNIX does this. And has for a long time. Like, since Gerald Ford or something.

I'd go on but I'm just making fun of them. If you read through their overview document there's actually some very good ideas in there, and knowing a few programmers from W2K, I can tell you that they do employ some top-notch talent there.

It seems http://research.microsoft.com/os/singularity/ collapsed into a singularity due to overwhelming /. traffic...

Although I agree that IE7 will slow down the growth of firefox I doubt it will really diminish the current market share. Slightly better support for web features will cut a little bit out of the "it only works in IE" problem, and most of the new stuff in IE7 is in firefox, or a plugin. In the end a lack of new features combined with any possible previous bad IE experiences will keep the new FF users where they are.

In the future I see technologies like xul as "where things are going".(and yes microsoft has one too) HTML+Javascript is only going to get us so far, and although using the javascript to dynamically update pages can do some (comparatively) awesome things, it does not have the feel of a real solution to it. So if I were a web developer[1] I would start playing with these tools now to be ready in ~3-5 years when they become a preferable alternative.

[1] I am not a web developer. If you are one, and I sound like I don't know what I am talking about, that is probably because I don't.

I suppose I am the only person in the whole world who finds the ipod physical interface totally "the suck" and the software unintuitive

Maybe not the only one, but certainly a member of a very small minority.

I thought the original jog wheels were cool just because they were retro, smooth and elegant

Yep, until you got a bit of sand in the wheel. That was No Fun at All. Apple switched to a touch pad to improve reliability.

what must the worst one be like?

Heh. Kind of like this.

-jcr

Here's the MS support article describing this problem:
http://support.microsoft.com/?scid=kb;EN-US;Q82028 6

They say, "A user interface that was designed to look good on a 96-DPI monitor may not look as good at higher resolutions. Text and graphics that are small at 96 DPI may appear much smaller at 200 DPI. When the number of pixels-per-inch increases, the size of each pixel decreases. If you double the density of the pixels, the size of the text may be halved so that the text is no longer readable. As a result, Web pages that specify pixel sizes for containers and text appear half their size, and the layout around them is adjusted accordingly.

WORKAROUND
Internet Explorer version 6 and later versions solve these problems by proportionally adjusting the scale on screens that have higher resolution.

Scaling is not a perfect solution. Embedded Microsoft ActiveX Controls, binary behaviors, and other elements that use Microsoft Windows Graphics Device Interface (GDI) calls do not scale well or do not scale at all. The GDI does not perform automatic scaling based on the density of the display.
. . .
Native support for high-DPI monitors will be included in the next major operating system release from Microsoft.

*shrug*

Take it how you please; but even in the 'next major operating system from Microsoft', you'll need to tell it what size your monitor is. If it cannot find that out correctly using a list of supported monitors and DDC information, you'll have to enter it manually.

Which is exactly how SuSE works, now.

What Are the Advantages of Running the Osa.exe File?

The Osa.exe file initializes the shared code that is used by the Office XP programs. When you use the Osa.exe file to initialize shared code, the Office XP programs start faster. If the Office programs, instead of Osa.exe, initialize the shared code, the programs take longer to start.

In the future, if you want to rant about misinformation, you might want to make sure that your information is correct.

And what is google groups exactly?

A way to search *text*

As for indexing content on usenet, this has been done for quite some time. The more comprehensive (scariest) approach seems to be taken by the folks at Microsoft, evidence enough of why real names/email addresses should never be used. Also, there's sites like this one that people can use if they're so inclined. Whatever.

Personally, I think most web interfaces suck for searching, but only slightly less than downloading 500K headers in a binary group looking for something ... interesting. What would be ideal if Easynews et al. could just offer a big text dump of any group that could be grepped locally without the clicking and advancing page by page nonsense. Either way, publicity is Not a Good Thing for usenet, which has for years been A Really Good Thing. It's easy enough for anyone to create a new group and migrate there with everyone else (ad infinitum), but that's hardly a welcome idea.

Microsoft is no fan of third party kernel patches, for the reasons you cite. Unfortunately for us (and them), enough existing software does this to make it impossible to change in Win32. But... the good news is this "feature" (unauthorized kernel patching/hacking) is generally not possible in the 64 bit versions of windows. http://www.microsoft.com/whdc/driver/kernel/64bitP atching.mspx [microsoft.com] outlines this. Will they get upset at Sony over this? Probably not since there's other (semi-legitimate) software that does this as well, and they made the deliberate decision to leave the "feature" in windows.

I looked at your company's website, and it looks like you are pretty entry level... in any case I doubt you are an employee (other then a random consult, probably for Ferdi up at the high school, hardly a state employee).

In response to your 1-7 bullets.

1.) - That simply is crap. Word Pad (and hundreds of other editors) can open pretty much all MS Word files (Macros aside, and they wouldn't be used in public documents). Formatting may be lost, but you the content is still there. MS Docs are large because they offer backwards compatibility for a few generations. Not to mention 2003 opens all the documents from 97-on. You are spreading disinformation here.

2.) - Romney appointed this position. "elected representatives" is wrong. 1 person, the governor, that is accurate. The very nature of my statement "appointed" implies an elected official put them there. However, the CIO is not accountable to the people, they are accountable to an elected official. Nobody knows who appointed this yahoo, look at yourself. You didn't even know. There is no accountability to the people.

3.) - No. What happened is the CIO made a bad decision, and due to checks and balances state officials stepped in. What if the CIO woke up this morning and decided to move the state back to DOS? Pretty stupid, huh? Well it would be important to allow the elected officials to step in and stop it. This is a similar case.

4.) - Clearly you haven't followed this case. For one, it isn't referring to any random state computer. Secondly, tens of thousands are not 125,000 (10,000 are not even being talked about for THIS upgrade anyways). Thirdly, the majority of the registry machines run their own software and don't need any office, a pdf viewer at most, courts aren't in this discussion (the rmv isn't either for that matter), and frankly your list is uniformed and just FUD. Your incorrect generalizations, your numbers, are all just pulled out of your ASS.

5.) - This is one of your more stupid points. NT 4 is over a decade old. The handful of machines using it would be upgraded soon anyways. Not only are there very FEW, but even FEWER that are actually used as a desktop where people need office solutions. Win98 is in the same boat, and most of those got upgraded to Win2k so incredibly fast it's not even funny. Win98 was very unstable, and certainly hasn't been left around for a decade. Anyone who is using 98 is going to be upgraded soon anyways, and frankly is so neglected they probably wouldn't get the open format software to begin with. But more specifically to the absurdity of your point.

Here are the hardware specs for OO and MS Office 2k3. The hardware requirements are virtually identical.
http://www.openoffice.org/dev_docs/source/sys_reqs _20.html
http://www.microsoft.com/products/info/product.asp x?view=22&pcid=3c3bd1bb-5595-4512-bcca-f764770e1d7 1&type=req

6.) - I disagree, but let's take it as accurate just for fun. It still doesn't even out her fake numbers. Nice try though.

7.) - People are familiar with Acrobat. They use it all day at work, and it's all over the internet TODAY. PDF's are just as common (if not more so) as word files. It is a proprietary format owned by Adobe, just like MS owns the .doc format (actually MS's format is more open). Almost nobody in the general public has heard of your open format crap. After everyone's bout with spyware they are hesitant to download software. Maybe not 2 years ago, but today people are... certainly not some random program that they have never heard of before.

As far as elected officials getting their ear bent. MS gave Romney more money in the last Election then Obrien. The money trail stops at the CIO's boss' front

Freedom and choice are great, but the fact is that many people often just want the choice made for them in as sensible a way as possible.

Well, then those people should ask someone who will make the choice for them.

Here or here

And the problem with installing the compilers is..?

After all, you can always download them for free.

Want to save thousands of dollars on MSDN? [macrocosmictech.com]

Why are you charging $17 for this link?

FUD
http://support.microsoft.com/kb/290144/

I quote:
What is the Osa.exe file?

The Office Startup Assistant (Osa.exe or OSA) is a program that improves the performance of Office programs. . . .

What Are the Advantages of Running the Osa.exe File?

The Osa.exe file initializes the shared code that is used by the Office XP programs. When you use the Osa.exe file to initialize shared code, the Office XP programs start faster. If the Office programs, instead of Osa.exe, initialize the shared code, the programs take longer to start.
Back to the top Back to the top

Can I Remove the Osa.exe File?

You can safely remove the Osa.exe file without causing the Office XP programs to fail. However, if you remove Osa.exe, you no longer benefit from the performance advantages that are provided by running Osa.exe. Also, the Office Shortcut Bar (OSB) may no longer start automatically, if you configured the OSB to start when Windows starts. (See the notes for the command-line switches later in this article.)

Don't forget the indexing services; I'm sure they load random shared DLLs, and they do NOT show up in the task manager. I quote:
Microsoft's Indexing Service
  Starting with Office XP, Microsoft has included a new "fast searching" feature (parasite) which may cause your computer's hard disk to run continuously.
  Apparently, this feature is implemented via Mosearch.exe and Mosdmn.exe, neither of which shows up in the task manager. As with findfast, don't just delete these files. Instead, follow the (very confusing) instructions provided in OFFXP: Hard Disk Runs Continuously After You Install Office XP (Q282106) to disable it. Unfortunately, this feature must be disabled for each installed Office XP application.

Also, I believe the various DCOM stuff that was optional in earlier MS systems, and comes standard in newer OSes preloads a shared office code.

Additionally, MS Office is 'prefetched', meaning that the DLL's are organized on disk for optimal loading speed. This happens with all windows apps.

You can do this on Linux, too. In SuSE 10.0, which has OpenOffice.org preloading set by default, OpenOffice.org loads in 5 seconds, on a mediocre celeron. On my system it times to around 3.4 seconds.

Also, you didn't bother to read the benchmark I linked. OpenOffice.org write uses less ram than MS Word, takes less time to startup (when neither are preloaded), and has a much smaller HD footprint.

Check out http://www.microsoft.com/business/executivecircle/ content/casestudydetail.aspx?csID=14967. I work for a largish health system in the midwest. We're currently piloting the Azyxxi image system.

Just today, I had the opportunity to talk with a doc who saw the system for the first time. He about crapped himself with glee. Since the image viewing component is a set of crApctiveX controls, it seems likely that we'll be rolling out a patient-viewable feature in the next year.

Get your doc as a client and push things like this on him. You know what his clients want because YOU ARE A CLIENT!

These activex components work in WINE, so don't whine about IE only :P

This is nonsense. Are you a Mac user? This "security model" you describe seems to be the update policy of OS X. Office 98 was a Mac product.

Microsoft's support lifecycle policy is clear: Business and Development software (including Windows 2000/XP Pro and Office) get five years of mainstream support and an additional five years of extended support. Extended support includes security updates, paid support, and Support/Knowlege Base information.

You say "Microsoft will immediately stop patching Windows 2000" when Vista is released. Wrong. Microsoft will continue to patch Windows 2000 until June 30, 2010. Microsoft will continue to patch Office 2000 until June 30, 2009.

This is nonsense. Are you a Mac user? This "security model" you describe seems to be the update policy of OS X. Office 98 was a Mac product.

Microsoft's support lifecycle policy is clear: Business and Development software (including Windows 2000/XP Pro and Office) get five years of mainstream support and an additional five years of extended support. Extended support includes security updates, paid support, and Support/Knowlege Base information.

You say "Microsoft will immediately stop patching Windows 2000" when Vista is released. Wrong. Microsoft will continue to patch Windows 2000 until June 30, 2010. Microsoft will continue to patch Office 2000 until June 30, 2009.

This is nonsense. Are you a Mac user? This "security model" you describe seems to be the update policy of OS X. Office 98 was a Mac product.

Microsoft's support lifecycle policy is clear: Business and Development software (including Windows 2000/XP Pro and Office) get five years of mainstream support and an additional five years of extended support. Extended support includes security updates, paid support, and Support/Knowlege Base information.

You say "Microsoft will immediately stop patching Windows 2000" when Vista is released. Wrong. Microsoft will continue to patch Windows 2000 until June 30, 2010. Microsoft will continue to patch Office 2000 until June 30, 2009.

as a normal internet application....
you should be thinking Indigo (fancy .Net remoting) instead.
http://msdn.microsoft.com/msdnmag/issues/04/01/Ind igo/default.aspx
Do you really think they will let their core application suite(Office) run in a web browser on linux? *yeah right*

as a normal internet application....
you should be thinking Indigo (fancy .Net remoting) instead.
http://msdn.microsoft.com/msdnmag/issues/04/01/Ind igo/default.aspx
Do you really think they will let their core application suite(Office) run in a web browser on linux? *yeah right*

You are a pretty smart guy. Too bad you can't get basic things like gender down. "He" is a "She". HER name is Linda. You would have learned that if you read the first sentence of the article. Next time read the article you are posting about before actually posting. :)

"That is just FUD Microsoft Office 2003 XML schema license."

You are correct it was not coherent. In my haste I forgot to put a "." and typed the link incorrectly. It was a link to FAQ for the MS open xml format (not patent free, but free for anyone to use without royalties.) In other words you can use it, but can't change it on them. I think you will find it quite educational. It should have read like this.

"That is just FUD. http://www.microsoft.com/Office/xml/faq.mspx"

Anyways, you must not be following the case closely. The argument originally made was over fear that the ms format wouldn't be readable in 10 years. Not any of the nonsense you tout. It was today we saw a change and the cost came into play, but sadly she was disingenuous.

Her numbers for Office at FULL PREMIUM PRICE (without the bulk/volume and government discounts) would give her 125,000 copies of MS Office... sorry to disappoint her, but there just aren't that many office employees under her jurisdiction. Her MS numbers included a brand new top of the line premium computer + undiscounted premium MS Office licenses, but not for an Open Format solution. As we just discussed she also added a few extra licenses that weren't needed... 125,000 vs. a few thousand. Hmmmmmmmmm.

That's sweet and all, but intellectually dishonest. Office 12 will run on any XP machine out today. Not to mention they would be upgrading the computers as part of the IT cycle anyways. She was spouting FUD. Your side will never make headway against MS until you stop lying about numbers and make a case based on merits, not on fake facts. Memory hog anyone? That wouldn't be Excel with that title would it? Nope it wouldn't. OO all the way baby!

"Who cares if it is a de-facto standard."

The answer to that would be the vast majority of citizens in the Commonwealth of Massachusetts who use the MS software and are happy with it (I'm one of them). The typical person does not want to go search the web for some "free" software they had never heard of just to read a government document. Not to mentions people's hesitation to install it and have it clutter up their computer with another program.

The state CIO's job is to serve the people, not the OSS agenda. It's an appointed position and not elected... in other words not accountable to the people. That is probably why we see a unilateral decision like this. And today, regrettably for Linda, is why the elected officials in the state feel the need to squash this on behalf of the citizens of MASS.

Sorry to break up that little run of sarcasm to point out ahead of time; Almost every reply that I would automatically get to this post about the licensing of Microsoft Office 2003 XML schema license is to a great extent FUD. There is not really a terrible patent issue, all licensees get royalty-free rights to all Microsoft patents to allow using the schema freely. Granted this is GPL-incompatible since anyone distributing the result has to accept the schema license, and also the schema license has a BSD-style advertisement clause. These are hardly the earth-shattering taking-away-our-freedom-lets-sing-with-RMS issues though.

One notable thing to point out here is that OpenDocument actually has a similar IP issue, notice how you get a royalty-free license from Sun for the IP in that format?

Overall this is making an awful lot of noise considering that people are supposedly getting "liberated" from a fairly reasonable product here, despite it being from Microsoft.

That is correct. You use a to block disable all CD autoplay.

The point I was making was that autoplay on USB drives is another potential vector. I personally don't disable it for that reason. I disable it more because I consider it a dead light bulb feature.

Once every three days, by default, Windows XP will perform a partial defragmentation and adjust the layout of the disk based upon current use. The files to be moved are written in the file Layout.ini (found in the Prefetch directory under the System Root directory).

The listed cost for a Microsoft MSDN Universal subscription is $2799/year. I don't know exactly how much was actually paid for it in my case, as my employer pays mine and they bought a site-licensed arrangement for the approximately 20 developers where I work. In any event, it's *substantially* less than the total cost of all of the OS versions and application software included, and it's very useful to be able to get the pre-release versions of different packages that are often available.

I will direct you to this article:
Kernel-mode drivers that extend or replace kernel services through undocumented means (such as hooking the system service tables) can interfere with other software and affect the stability of the operating system. For x86-based systems, Microsoft discourages such practices but does not prevent them programmatically, because doing so would break compatibility for a significant amount of released software. A similar base of released software does not exist for x64-based systems, so it is possible to add this level of protection to the kernel without breaking compatibility.

As usual, it's the bane of all software corporations today -- legacy crap that some programmer shat out in his sleep.

Possible to recover the system without a GUI if needed?
Yes

Modular kernel which can be stripped of unneccesary features?
Yes

Does what you tell it to reliably and regularly rather than suddenly deciding one day to do things totally differently for no apparent reason?

Not sure about that one. XP's better, mind. That being said, I got fed up of waiting. Linux doesn't have that problem, and so far neither does MacOS.

Possible to recover the system without a GUI if needed?
Yes

Modular kernel which can be stripped of unneccesary features?
Yes

Does what you tell it to reliably and regularly rather than suddenly deciding one day to do things totally differently for no apparent reason?

Not sure about that one. XP's better, mind. That being said, I got fed up of waiting. Linux doesn't have that problem, and so far neither does MacOS.

http://research.microsoft.com/mesh/ Got an Windows XP PC and a wireless card? Join the mesh. Lets make a shore to shore wireless mesh from house to house, from school to school, from library to library, from cafe to cafe. Manifest Destiny. Let it spread all across the land becaues the air waves belong to US. An Internet that is in the air between us all. We dont need SBC. We dont need Comcast. They wasted their money on the land lines. It is time for Americans to stop selling our birth rights!

Vewy intwesting. Yeah, here's an MSDE - Express comparison:

http://msdn2.microsoft.com/en-us/library/ms165672. aspx

BTW, "embedded" a DB like this is not goot for memory. MSDE eats up 10's of megs of memory over time with just a simple schema. I've seen it at 200mb mem usage for a DB with a 2mb dataset.

For embedded DB work, people should take a look at Firebird.

Here's the Microsoft article on it: http://www.microsoft.com/technet/security/bulletin /MS04-028.mspx