What we really need is Microsoft to allow removal of any and all programs that are not basic for an operating system. Yes, even Internet Explorer.
On the post-antitrust Windows XP, Internet Explorer can be removed quite easily. So can Outlook Express and MSN Messenger and even Windows Media Player: go to Add/Remove Programs, click the Add/Remove Windows Components and unselect what you will.
The catch? This installs the _apps_, not any libraries that the app uses. And IMO this is the correct thing to do, since these libraries (URLMON, WinInet, MSHTML, etc) are part of Windows' API and are used by many 3rd party apps.
In fact, the only thing I can't see an obvious way of uninstalling is Windows Movie Maker. Hmm...
In an ideal world. In the real world people leave, new people join, managers still want stuff done yesterday. Now, this happens in the auto industry too, but there they have the laws of physics giving them a reality check everyday. In the software business, any combination of new hardware, new configurations and new business environments may invalidate your older assumptions. Now how will you ensure that imperfect humans pass on their knowledge _exactly_ in the face of such change?
> as long as software can ignore all the product liability laws
Well, the software for the space shuttle is built to some pretty exacting standards (and still has bugs, but that's beside the point) but most customers are not ready to either pay for that kind of quality or wait for that kind of time.
*shrug* I have an old Celeron 366Mhz with 192MB RAM on XP -- it mainly runs email and IM for my parents and does a good job of it. Definitely no 'continuous disk grinding'. OTOH I did a clean install (I prefer those to 'upgrades') on that.
If I absolutely *had* to upgrade, I'd still not do it until I was sure the partition sizes and swapfile locations were good for XP, _and_ that my volumes were defragmented before I started. I've had to do this on a few occasions (NT->2000 and 2000->XP) and so far the experience has been quite good.
Given that everything from the Linux kernel to SSH to Apache to Firefox has had buffer overruns, I'd be wary of describing their authors as 'unprofessional'.
Rather, buffer overflows are trivial to avoid in class assignments (and indeed, small projects). It's when the project grows larger, gets split into multiple program units and gets multiple authors that you really start scratching the surface of industrial strength development (something the armchair developers on/. have never really experienced).
To top it all, code that is 'safe' can often be made 'unsafe' by running it under circumstances the authors never intended: there's a whole class of overflow attacks that use code/data injection to crack even supposedly secure programs (and no, not even Java/C# is safe from this).
Yes, but for anyone not using grep for Windows or Google/MSN Desktop Search, there is a Tweak UI setting that switches XP's search back to look _exactly_ like Win2k's search.
(TweakUI's a good download to have around anyway.)
Keep telling yourself that. People like you are _just_ the people F/OSS folk need to figure out why they haven't been able to even scratch the enterprise messaging market (dominated by IBM's Notes/Domino and MS' Outlook/Exchange).
What perf issues? especially with SP2? XP for the most part works better than Win2k on the same hardware.
It's also the last OS from Microsoft that actually treated users like they were using a computer instead of dumbing things down. (In WInXP: control panel "lite", stupid road blocks if you want to browse the file
And these do help novices. People who don't like them can switch them off. Experts and deployers can create scripts to install XP with all of these disabled.
Anime, if packaged and presented well, can be _quite_ palatable to American audiences. There was these two brothers from Chicago who made a live action anime flick in 1999, complete with sexy leather-clad chick and heavy philosophical overtones, to the point where people were coming back to see the damn thing _and_ saying 'wtf' at the same time...
Ah, the level of discourse on Slashdot. Great-Grandparent inflates Apple's importance in the PC harware biz. Grandparent makes an eminently sensible point about how USB products could more rationally be explained by Microsoft's decision to ship USB with Win98 (and remember hardware vendors get wind of new Windows features far ahead of the customer, via WinHEC)... and the parent counters with his best-est argument-- the freaking _color_ of the things.
Honestly, I've heard more substantial arguments from 6-year old girls.
As to your question:
> tell me why 90% of USB peripherals were a translucent shade of "Bondi Blue" for a solid year
Because no one in the PC world gives a rat's ass about how things _look_ as long as it works with Windows and is cheap (this is true even now). Because 90% of these things are made by Taiwanese firms in the same industrial park. Because it's just easy to make a cheapass knockoff of copy a popular design. Because it doesn't take much time to put a plastic shell together compared with the time it takes to make sure that your circuit boards and firmware correspond to what Microsoft and Intel have been shouting from the rooftops for the past 18 months (at WinHEC and the Intel equivalents).
Here's a clue: if USB adoption had been driven by Apple, the first cheapass OEM units wouldn't have showed in the market for _at least_ 6 months as they scrambled to get their electronics _and_ software/interop act in place. The fact that they could respond almost immediately to Apple's look-and-feel meant that they were already ready with the actual electronics and interop, and would've shipped them in beige as usual if Apple hadn't come in and made fruit the flavor of the season.
From Engadget's live coverage of the WWDC: 10:33am PDT - [Steve Jobs] As a matter of fact, this system I've been using here... the keynote's been running on a P4 3.6GHz all morning
I hate dreadfully touchpads too-- maybe I have sensitive fingers but I hate the feel of my skin on the pad they use. And while I can control trackpoints (which is what IBM calls them IIRC) quite precisely I have a pretty bad time getting a touchpad-based cursor to move exactly where I want.
This is not a license that I would like to accept.. But it's not the case that this license violates Section 7....I want to reassure people that it's not an issue
I think it's pretty clear that while he's differentiating his personal opinion from the actual legal situation.
There must be something in the license that made him say that. MS has showed many times in the past that they do not want to work with the GPL even going as far as calling it "a cancer".
Ah, the stance of the true slashbot: when faced with anything that disturbs your worldview, step back and vent against the Man, the actual situation be damned.
No, but I play one on TV and have a college education!
Wow, you went to college and can read? Awesome. Me, I just cobble letterblocks together in preschool.
If you could read it, you would _know_ it is incompatible with the GPL.
Basically, you are saying you can't explain why, but if I could just read the fine licenses, I'd magically understand? Hmm, perhaps you're thinking about Section 7, especially this:
For example, if a patent license would not permit royalty-free redistribution of the Program by all those who receive copies directly or indirectly through you, then the only way you could satisfy both it and this License would be to refrain entirely from distribution of the Program.
Well, considering that Microsoft _is_ allowing royalty free distribution to all, there is no problem that I see. Unless, of course, you plan to sue Microsoft, in which case it possible that you may be asked to stop distributing your program. Note that this does not affect the rights of your distributees (including their right to distribute) if they did not sue Microsoft, since they have a perpetual royalty-free right to use, with the do-not-sue reservation.
Does this make the license compatible with the GPL? No. However, as is clear from my original post, GPL'd code can legally use these formats because the rights of the distributees are not affected. Don't take my word for it, take Eben Moglen's:
Microsoft issued the schemas under a royalty-free license that had some wondering if they could work with GPL software. "Microsoft may have patents and/or patent applications that are necessary for you to license in order to make, sell, or distribute software programs that read or write files that comply with the Microsoft specifications for the Office Schemas," according to the accompany patent license agreement. Given that GPL software developers have refused to work with patent-encumbered code, the anxiety was familiar.
"I don't think the alarm is justified," the FSF's pro bono counsel Eben Moglen told us last night. "This is not a license that I would like to accept; Microsoft is saying we might have some patents. But it's not a problem if Microsoft is making it available to everyone to make use and sell.....
Moglen says that this doesn't prevent a free software programmer creating derivative works, because the schema is being licensed from Microsoft.
"Section 7 of the GPL says if you're restricted by a judgement or a license you've accepted, and they're incompatible, then you can't distribute your software under GPL. But it's not the case that this license violates Section 7."
"I want to reassure people that it's not an issue".
It's on the Channel9 video, complete with a somewhat blurry demo. Also theseURLs.
It sounds like it's more of a indexable structured text document (for lack of a better term) like that used in the PDF format. (PDF uses a stream of textual "objects" that can contain encoded binary data. Each object is indexed in an offset table at the end of the file. Offset tables override one another, thus allowing objects to be rewritten and overridden without rewriting the entire file.)
That's pretty much it. Objects can indeed be rewritten and sequences (like Powerpoint's slide order) altered without rewriting the file. The indexable 'structured text' objects are really XML. The 'offsets' are really directory entries in a zip container.
The only thing far-fetched is your inability to understand that Microsoft is moving to an XML format, not another binary format.
DOCX isn't an XML format. It's an open compound document format that happens to use XML to store some of its data and (much more importantly) the relationships of the pieces w.r.t one another, and ZIP to bind it together. If they ever do analytics you can bet your britches they'll add binary blobs (mostly indexes+precomputed tables) to XLSX, like they're already going to do for Excel macros, with some descriptor XML for describing the relationships between the various blobs).
And they'll do this easier and faster with their in-house format than knocking on OASIS' doors to get a dialect to describe dynamic relationships ratified.
This ties right into my previous point that teams need to ship on time accept no dependencies, a point you blithely ignore. (And it amazes me that with your 'many years as a software engineer' you fail to see this point. What did you work on-- the ISO 7-layer network model? That was a sterling example of design by committee all right.)
Once you decide to go compressed XML, you're locked into a linear format that decides many of those tradeoffs for you.... If your argument was going to hold water, Microsoft wouldn't take that path.
Compressed XML is irrelevant here. And in what sense is it a 'linear' format? On the contrary, XML's nested-ness makes it very inefficient sometimes: people still use it, of course, when the interop benefits outweigh the disadvantages. And Microsoft's focus on Office-as-groupware/enterprise software clearly provides enough benefit to outweigh the disadvantages of using XML in office.
XML does not remove your responsibility to make tradeoffs. Your document hierarchy is decided by the design decisions you make, and those are guided by the tradeoffs and wins you choose. My team prototyped and discarded two different models for representing a GUI workspace in XML before choosing a third that was a balance between comprehensibility and speed.
> Try again. Need I remind you of the still unfinished WordML spec
The eWeek article is FUD. The Office 2003 reference schemas have been available for quite a while now on a no-charge, perpetual grant basis. Considering that the journalist based his story on OASIS sources, this isn't surprising.
And re your long and sordid history of Microsoft's innovation, I'd suggest laying off the tabloid headlines. Businesses around the world acquire people and new technology. I'd say Microsoft has done more work on its acquisitions than (say) IBM has. No sane person could compare Word 1 and Word 11 with a straight face, which is more than can be said of Domino 5 and 7. COM and Wolfpack came as DEC was selling assets to stay afloat. Cutler left and joined MS because there was no OS future at Digital -- and yet comparing the NT kernel to VMS is ridiculous, like comparing apples and oranges.
Word 2003 does implement Reveal Codes -- sort of. Place your caret on some formatted text, choose View|Task Pane, and choose the 'Reveal Formatting' task pane.
I know, I know, WordPerfect fans will say this isn't the same thing at all. But given the limitations of their run-based model, at least it's better than nothing.
Okay, name some features that require a proprietary format. Go ahead, take your time. I'll wait.
There are certain CS types who have a very abstract view of code and data. For them, data is data, however it's laid out, and therefore all formats are alike (it's just a row of numbers, for chrissake, there's only one way to store that, let's create an ISO-standard(tm) format). They are also the ones who usually say, there's no big deal about electronics, I could make a computer from water-pipes if I had to (sure they could, but how useful would it be?).
Then there are those who've actually created sophisticated formats, and know that each format (even the XML ones) are born of thousands of tiny design decisions and tradeoffs until what you have is something that bestows a market benefit upon your product.
You asked for features: here are two. High end analytics (logical future path for Excel). Unstructured storage (logical extension for anything Office related, esp if WinFS finally ships).
Is this so farfetched? Are there existing products that use their proprietary format to an advantage? Take a look at IBM's Redbrick and Domino sometime. Domino is a particularly beautiful example of how the app's storage format has helped it carve out a niche for itself in the marketplace.
DOCX: Requires an expensive Microsoft license to use.
I thought the whole point about this is that DOCX and friends are freely licensed? There's nothing now to stop someone from writing an OO plugin that reads and writes DOCX.
And if they truely *are* the same at a technology level (which may be true), that kind of undermines your whole argument about proprietary formats, doesn't it?
I can almost picture you there: the hobby coder for whom the world revolves around glacially moving mailing lists and IRC channels. Software Engineering doesn't work that way. To ship in 2006, Microsoft would have to start work in 2003 (immediately after Office 2003's launch). And in 2003 there'd be no guarantee that OASIS would finish their work by 2005. On the other hand, an internal Microsoft effort could be estimated and planned for. Any product team knows this: do not accept dependencies if you want to ship on time.
Incidentally, this is the same reason KDE and GNOME reimplement each other's work so much, and why despite all the mailing lists and talk of shared HIGs they differ so much in silly little ways: the business of software development carries its own inertia. If you can't stand it, don't go there.
Frankly I despise people like you, who in their blind hatred of Microsoft would turn software into precisely the kind of overregulated field that telecom has become today. The day an ITU-like org takes over software standards (and God knows the EU and the UN are trying) is the day innovation in this field will die.
An AC replied to me saying that MS has the right to revoke if you sue them. *shrug* That sounds more like lawyerese CYA than malevolence, but as I said, IANAL.
Are you a lawyer? If not, I'll take your assertion that it's incompatible with the GPL with a grain of salt. Note that RMS' patent diatribes have not stopped others (Nokia, for example) from offering their patents for use in GPL projects.
Now, if the GPL specifically contained a section proscribing patented formats, then your claim would have merit. But fortunately for the rest of us, there is now such thing (although IANAL).
Fact: As Lotus fans everywhere will tell you, "It's the groupware and the workflow, stupid". Well, the person who popularized that phrase (Ray Ozzie, one of Lotus' founders IIRC) is now a director at MS.
Fact: Business love office but _hate_ the islands of data it creates.
Fact: The office suite market is becoming a commodity (OpenOffice). It's not there yet but MS can see the writing on the wall. In such a situation, it's not the individual pieces but how the pieces play together. Opening up the Office file formats will make Office a much better 'enterprise app' (hate that phrase) because it 'plays well with others' (hate that phrase too).
Fact: Govts, particularly the EU, but are _demanding_ open formats vocally. Ditto businesses, because of Sarbannes Oxley.
There is no dogma at Microsoft (apart from Windows uber alles). If tomorrow the market asked for Open Source loud enough, they'd probably sell Open Source solutions. So it's no suprise really that Microsoft would give them what they asked for.
And it allows MS to set the standard in the market that everybody follows. There is a value to that.
And, funny-- I didn't see anything in this about them opening up Outlook's datastore. Would've been neat for open-source calendars if they could've read Outlook's calendar store.
I am aware of the value of standards. However, the capabilities and of character-sets and toilet-paper dimensions can be described with a precision that you cannot apply to word-processors or spreadsheets (if you think Excel already does everything a person can want, you obviously haven't seen some of the analytics software used in financial firms).
Standardizing wordprocessors/spreadsheets when their very capabilities are evolving is more than madness, it's bureaucratic control-freakery of the worst order.
Attempting to evade standards only hurts everyone.
Considering that DOCX is equivalent to what OASIS offers, I'm not sure how Microsoft is 'evading' standards. It sounds more like a bunch of vendors with fractional marketshare is griping about how the big 800 pound gorilla isn't playing at their party.
> We're talking about a file format for a word processor and a spreadsheet.
Yeah. Word processors and speadsheets. Simplicity itself. Man, those programmers at WordPerfect/WordPro must have been dumb _and_ never heard of hex editors, to this day they can't accurately render or save Office docs.
An average GUI wordprocessor/spreadsheet has more complexity in it than emacs, measured by the number of ways a user can interact with it. To this, add differing program models (WordPerfect used a XML-like DOM model for storing data, Word used a Run-based model where blocks of text were tagged with attributes (this was why Reveal Codes took so long to implement in Word)).
Now try to come up with a common format and you'll end up with a limited, common-denominator format like RTF that everyone interprets a little differently (e.g., Lotus' and MS' interpretation).
I'd say using a interoperable native format is better than developing a watered down, common-denominator format.
Slashdot Mirror
What we really need is Microsoft to allow removal of any and all programs that are not basic for an operating system. Yes, even Internet Explorer.
On the post-antitrust Windows XP, Internet Explorer can be removed quite easily. So can Outlook Express and MSN Messenger and even Windows Media Player: go to Add/Remove Programs, click the Add/Remove Windows Components and unselect what you will.
The catch? This installs the _apps_, not any libraries that the app uses. And IMO this is the correct thing to do, since these libraries (URLMON, WinInet, MSHTML, etc) are part of Windows' API and are used by many 3rd party apps.
In fact, the only thing I can't see an obvious way of uninstalling is Windows Movie Maker. Hmm...
> All programming is small projects
In an ideal world. In the real world people leave, new people join, managers still want stuff done yesterday. Now, this happens in the auto industry too, but there they have the laws of physics giving them a reality check everyday. In the software business, any combination of new hardware, new configurations and new business environments may invalidate your older assumptions. Now how will you ensure that imperfect humans pass on their knowledge _exactly_ in the face of such change?
> as long as software can ignore all the product liability laws
Well, the software for the space shuttle is built to some pretty exacting standards (and still has bugs, but that's beside the point) but most customers are not ready to either pay for that kind of quality or wait for that kind of time.
*shrug* I have an old Celeron 366Mhz with 192MB RAM on XP -- it mainly runs email and IM for my parents and does a good job of it. Definitely no 'continuous disk grinding'. OTOH I did a clean install (I prefer those to 'upgrades') on that.
If I absolutely *had* to upgrade, I'd still not do it until I was sure the partition sizes and swapfile locations were good for XP, _and_ that my volumes were defragmented before I started. I've had to do this on a few occasions (NT->2000 and 2000->XP) and so far the experience has been quite good.
Given that everything from the Linux kernel to SSH to Apache to Firefox has had buffer overruns, I'd be wary of describing their authors as 'unprofessional'.
/. have never really experienced).
Rather, buffer overflows are trivial to avoid in class assignments (and indeed, small projects). It's when the project grows larger, gets split into multiple program units and gets multiple authors that you really start scratching the surface of industrial strength development (something the armchair developers on
To top it all, code that is 'safe' can often be made 'unsafe' by running it under circumstances the authors never intended: there's a whole class of overflow attacks that use code/data injection to crack even supposedly secure programs (and no, not even Java/C# is safe from this).
I don't use the MS search at all - too slow.
Yes, but for anyone not using grep for Windows or Google/MSN Desktop Search, there is a Tweak UI setting that switches XP's search back to look _exactly_ like Win2k's search.
(TweakUI's a good download to have around anyway.)
> You lose about 5-15% in some situations with SP2 v XP SP1
Yes, mostly because of software DEP checks. These can be disabled so you _can_ if you wish get a marginal speed boost at the price of security.
Keep telling yourself that. People like you are _just_ the people F/OSS folk need to figure out why they haven't been able to even scratch the enterprise messaging market (dominated by IBM's Notes/Domino and MS' Outlook/Exchange).
the performance issues of WinXP with SP2.
What perf issues? especially with SP2? XP for the most part works better than Win2k on the same hardware.
It's also the last OS from Microsoft that actually treated users like they were using a computer instead of dumbing things down. (In WInXP: control panel "lite", stupid road blocks if you want to browse the file
And these do help novices. People who don't like them can switch them off. Experts and deployers can create scripts to install XP with all of these disabled.
Anime, if packaged and presented well, can be _quite_ palatable to American audiences. There was these two brothers from Chicago who made a live action anime flick in 1999, complete with sexy leather-clad chick and heavy philosophical overtones, to the point where people were coming back to see the damn thing _and_ saying 'wtf' at the same time...
Ah, the level of discourse on Slashdot. Great-Grandparent inflates Apple's importance in the PC harware biz. Grandparent makes an eminently sensible point about how USB products could more rationally be explained by Microsoft's decision to ship USB with Win98 (and remember hardware vendors get wind of new Windows features far ahead of the customer, via WinHEC)... and the parent counters with his best-est argument-- the freaking _color_ of the things.
Honestly, I've heard more substantial arguments from 6-year old girls.
As to your question:
> tell me why 90% of USB peripherals were a translucent shade of "Bondi Blue" for a solid year
Because no one in the PC world gives a rat's ass about how things _look_ as long as it works with Windows and is cheap (this is true even now). Because 90% of these things are made by Taiwanese firms in the same industrial park. Because it's just easy to make a cheapass knockoff of copy a popular design. Because it doesn't take much time to put a plastic shell together compared with the time it takes to make sure that your circuit boards and firmware correspond to what Microsoft and Intel have been shouting from the rooftops for the past 18 months (at WinHEC and the Intel equivalents).
Here's a clue: if USB adoption had been driven by Apple, the first cheapass OEM units wouldn't have showed in the market for _at least_ 6 months as they scrambled to get their electronics _and_ software/interop act in place. The fact that they could respond almost immediately to Apple's look-and-feel meant that they were already ready with the actual electronics and interop, and would've shipped them in beige as usual if Apple hadn't come in and made fruit the flavor of the season.
For all the 13-year olds on /. who think they're funny, here's where the word monad really comes from.
From Engadget's live coverage of the WWDC: 10:33am PDT - [Steve Jobs] As a matter of fact, this system I've been using here... the keynote's been running on a P4 3.6GHz all morning
I hate dreadfully touchpads too-- maybe I have sensitive fingers but I hate the feel of my skin on the pad they use. And while I can control trackpoints (which is what IBM calls them IIRC) quite precisely I have a pretty bad time getting a touchpad-based cursor to move exactly where I want.
Go on, quote him completely.
...I want to reassure people that it's not an issue
This is not a license that I would like to accept.. But it's not the case that this license violates Section 7.
I think it's pretty clear that while he's differentiating his personal opinion from the actual legal situation.
There must be something in the license that made him say that. MS has showed many times in the past that they do not want to work with the GPL even going as far as calling it "a cancer".
Ah, the stance of the true slashbot: when faced with anything that disturbs your worldview, step back and vent against the Man, the actual situation be damned.
Wow, you went to college and can read? Awesome. Me, I just cobble letterblocks together in preschool.
If you could read it, you would _know_ it is incompatible with the GPL.
Basically, you are saying you can't explain why, but if I could just read the fine licenses, I'd magically understand? Hmm, perhaps you're thinking about Section 7, especially this:Well, considering that Microsoft _is_ allowing royalty free distribution to all, there is no problem that I see. Unless, of course, you plan to sue Microsoft, in which case it possible that you may be asked to stop distributing your program. Note that this does not affect the rights of your distributees (including their right to distribute) if they did not sue Microsoft, since they have a perpetual royalty-free right to use, with the do-not-sue reservation.
Does this make the license compatible with the GPL? No. However, as is clear from my original post, GPL'd code can legally use these formats because the rights of the distributees are not affected. Don't take my word for it, take Eben Moglen's:
> I'd love to know your source
It's on the Channel9 video, complete with a somewhat blurry demo. Also these URLs.
It sounds like it's more of a indexable structured text document (for lack of a better term) like that used in the PDF format. (PDF uses a stream of textual "objects" that can contain encoded binary data. Each object is indexed in an offset table at the end of the file. Offset tables override one another, thus allowing objects to be rewritten and overridden without rewriting the entire file.)
That's pretty much it. Objects can indeed be rewritten and sequences (like Powerpoint's slide order) altered without rewriting the file. The indexable 'structured text' objects are really XML. The 'offsets' are really directory entries in a zip container.
The only thing far-fetched is your inability to understand that Microsoft is moving to an XML format, not another binary format.
.... If your argument was going to hold water, Microsoft wouldn't take that path.
DOCX isn't an XML format. It's an open compound document format that happens to use XML to store some of its data and (much more importantly) the relationships of the pieces w.r.t one another, and ZIP to bind it together. If they ever do analytics you can bet your britches they'll add binary blobs (mostly indexes+precomputed tables) to XLSX, like they're already going to do for Excel macros, with some descriptor XML for describing the relationships between the various blobs).
And they'll do this easier and faster with their in-house format than knocking on OASIS' doors to get a dialect to describe dynamic relationships ratified.
This ties right into my previous point that teams need to ship on time accept no dependencies, a point you blithely ignore. (And it amazes me that with your 'many years as a software engineer' you fail to see this point. What did you work on-- the ISO 7-layer network model? That was a sterling example of design by committee all right.)
Once you decide to go compressed XML, you're locked into a linear format that decides many of those tradeoffs for you
Compressed XML is irrelevant here. And in what sense is it a 'linear' format? On the contrary, XML's nested-ness makes it very inefficient sometimes: people still use it, of course, when the interop benefits outweigh the disadvantages. And Microsoft's focus on Office-as-groupware/enterprise software clearly provides enough benefit to outweigh the disadvantages of using XML in office.
XML does not remove your responsibility to make tradeoffs. Your document hierarchy is decided by the design decisions you make, and those are guided by the tradeoffs and wins you choose. My team prototyped and discarded two different models for representing a GUI workspace in XML before choosing a third that was a balance between comprehensibility and speed.
> Try again. Need I remind you of the still unfinished WordML spec
The eWeek article is FUD. The Office 2003 reference schemas have been available for quite a while now on a no-charge, perpetual grant basis. Considering that the journalist based his story on OASIS sources, this isn't surprising.
And re your long and sordid history of Microsoft's innovation, I'd suggest laying off the tabloid headlines. Businesses around the world acquire people and new technology. I'd say Microsoft has done more work on its acquisitions than (say) IBM has. No sane person could compare Word 1 and Word 11 with a straight face, which is more than can be said of Domino 5 and 7. COM and Wolfpack came as DEC was selling assets to stay afloat. Cutler left and joined MS because there was no OS future at Digital -- and yet comparing the NT kernel to VMS is ridiculous, like comparing apples and oranges.
Word 2003 does implement Reveal Codes -- sort of. Place your caret on some formatted text, choose View|Task Pane, and choose the 'Reveal Formatting' task pane.
I know, I know, WordPerfect fans will say this isn't the same thing at all. But given the limitations of their run-based model, at least it's better than nothing.
Okay, name some features that require a proprietary format. Go ahead, take your time. I'll wait.
There are certain CS types who have a very abstract view of code and data. For them, data is data, however it's laid out, and therefore all formats are alike (it's just a row of numbers, for chrissake, there's only one way to store that, let's create an ISO-standard(tm) format). They are also the ones who usually say, there's no big deal about electronics, I could make a computer from water-pipes if I had to (sure they could, but how useful would it be?).
Then there are those who've actually created sophisticated formats, and know that each format (even the XML ones) are born of thousands of tiny design decisions and tradeoffs until what you have is something that bestows a market benefit upon your product.
You asked for features: here are two. High end analytics (logical future path for Excel). Unstructured storage (logical extension for anything Office related, esp if WinFS finally ships).
Is this so farfetched? Are there existing products that use their proprietary format to an advantage? Take a look at IBM's Redbrick and Domino sometime. Domino is a particularly beautiful example of how the app's storage format has helped it carve out a niche for itself in the marketplace.
DOCX: Requires an expensive Microsoft license to use.
I thought the whole point about this is that DOCX and friends are freely licensed? There's nothing now to stop someone from writing an OO plugin that reads and writes DOCX.
And if they truely *are* the same at a technology level (which may be true), that kind of undermines your whole argument about proprietary formats, doesn't it?
I can almost picture you there: the hobby coder for whom the world revolves around glacially moving mailing lists and IRC channels. Software Engineering doesn't work that way. To ship in 2006, Microsoft would have to start work in 2003 (immediately after Office 2003's launch). And in 2003 there'd be no guarantee that OASIS would finish their work by 2005. On the other hand, an internal Microsoft effort could be estimated and planned for. Any product team knows this: do not accept dependencies if you want to ship on time.
Incidentally, this is the same reason KDE and GNOME reimplement each other's work so much, and why despite all the mailing lists and talk of shared HIGs they differ so much in silly little ways: the business of software development carries its own inertia. If you can't stand it, don't go there.
Frankly I despise people like you, who in their blind hatred of Microsoft would turn software into precisely the kind of overregulated field that telecom has become today. The day an ITU-like org takes over software standards (and God knows the EU and the UN are trying) is the day innovation in this field will die.
An AC replied to me saying that MS has the right to revoke if you sue them. *shrug* That sounds more like lawyerese CYA than malevolence, but as I said, IANAL.
Are you a lawyer? If not, I'll take your assertion that it's incompatible with the GPL with a grain of salt. Note that RMS' patent diatribes have not stopped others (Nokia, for example) from offering their patents for use in GPL projects.
Now, if the GPL specifically contained a section proscribing patented formats, then your claim would have merit. But fortunately for the rest of us, there is now such thing (although IANAL).
The license is a perpetual grant and cannot be revoked. http://www.microsoft.com/Office/xml/faq.mspx
Simple:
Fact: As Lotus fans everywhere will tell you, "It's the groupware and the workflow, stupid". Well, the person who popularized that phrase (Ray Ozzie, one of Lotus' founders IIRC) is now a director at MS.
Fact: Business love office but _hate_ the islands of data it creates.
Fact: The office suite market is becoming a commodity (OpenOffice). It's not there yet but MS can see the writing on the wall. In such a situation, it's not the individual pieces but how the pieces play together. Opening up the Office file formats will make Office a much better 'enterprise app' (hate that phrase) because it 'plays well with others' (hate that phrase too).
Fact: Govts, particularly the EU, but are _demanding_ open formats vocally. Ditto businesses, because of Sarbannes Oxley.
There is no dogma at Microsoft (apart from Windows uber alles). If tomorrow the market asked for Open Source loud enough, they'd probably sell Open Source solutions. So it's no suprise really that Microsoft would give them what they asked for.
And it allows MS to set the standard in the market that everybody follows. There is a value to that.
And, funny-- I didn't see anything in this about them opening up Outlook's datastore. Would've been neat for open-source calendars if they could've read Outlook's calendar store.
I am aware of the value of standards. However, the capabilities and of character-sets and toilet-paper dimensions can be described with a precision that you cannot apply to word-processors or spreadsheets (if you think Excel already does everything a person can want, you obviously haven't seen some of the analytics software used in financial firms).
Standardizing wordprocessors/spreadsheets when their very capabilities are evolving is more than madness, it's bureaucratic control-freakery of the worst order.
Attempting to evade standards only hurts everyone.
Considering that DOCX is equivalent to what OASIS offers, I'm not sure how Microsoft is 'evading' standards. It sounds more like a bunch of vendors with fractional marketshare is griping about how the big 800 pound gorilla isn't playing at their party.
> We're talking about a file format for a word processor and a spreadsheet.
Yeah. Word processors and speadsheets. Simplicity itself. Man, those programmers at WordPerfect/WordPro must have been dumb _and_ never heard of hex editors, to this day they can't accurately render or save Office docs.
An average GUI wordprocessor/spreadsheet has more complexity in it than emacs, measured by the number of ways a user can interact with it. To this, add differing program models (WordPerfect used a XML-like DOM model for storing data, Word used a Run-based model where blocks of text were tagged with attributes (this was why Reveal Codes took so long to implement in Word)).
Now try to come up with a common format and you'll end up with a limited, common-denominator format like RTF that everyone interprets a little differently (e.g., Lotus' and MS' interpretation).
I'd say using a interoperable native format is better than developing a watered down, common-denominator format.