“There has been a lot of attention in the last few weeks about the discrepancy in the desktop background appearance between the control panel and Windows Photo Gallery. This is something that we have been tracking internally for quite a while, so I thought I would clue in everyone on the issue here and how we are addressing it for RTM.
When will we have native .png wallpaper support?
I don’t know about you, but when I’m generating some of my own wallpapers (computer-generated, not digital photographs), I tend to get better results with PNG images as oppposed to JPEG images.
What exactly is stopping you from using pngs as backgrounds? I use them quite often on XP.
If you read the article you’ll see that even JPEG files get converted to BMP for use as a wallpaper. So you’re not using a PNG as a wallpaper, your PNG is being converted into a BMP.
When Microsoft invents an 99% PNG compatible format, you’ll get a native 99% .png wallpaper support.
When Microsoft invents an 99% PNG compatible format, you’ll get a native 99% .png wallpaper support.
They wouldn’t ever. They’d just define a new format (say XMG (eXtensible Microsoft Graphic) ) and it will be propritary, buggy, low performance and not do what people want. Well that’s what experience leads me to believe anyway
Stephen
“support for JPEG wallpapers is a new feature in Vista. In previous versions of Windows, only BMP images could be used for wallpapers.”
So how am I using a JPEG file for my background right now in XP? Is it silently getting converted to BMP behind the scenes?
Though both JPG and PNG are supported types for wallpapers in XP, IIRC the formats are converted to BMPs internally because that API only supports BMPs. I’ll post a link if I can dig up the info.
//Though both JPG and PNG are supported types for wallpapers in XP, IIRC the formats are converted to BMPs internally because that API only supports BMPs.//
This is way behind the times, isn’t it?
KDE will not only support PNG & JPG for wallpaper, as of version 3.4 (or later) KDE supports SVG wallpapers.
http://dot.kde.org/1103326589/
http://dot.kde.org/1106477233/
Apparently Windows still has a great deal of catching up to do.
Why did this comment get modded down? It is surely on topic.
Edited 2006-10-30 02:36
The article was a bit unclear on that, which is unfortunate, because it’s pretty much the crucial piece of info that’s needed to understand why the article describes a difficult task. Although, given that they allowed such a bubblegum-and-bailing-twine solution to remain in place for so long, I have a hard time feeling much sympathy.
//The article was a bit unclear on that, which is unfortunate, because it’s pretty much the crucial piece of info that’s needed to understand why the article describes a difficult task. Although, given that they allowed such a bubblegum-and-bailing-twine solution to remain in place for so long, I have a hard time feeling much sympathy.//
It isn’t a difficult task. There are at least three other Windows managers that have far better support for wallpaper formats than the kludges in Windows.
There seems to be something political in this. First lack of PNG support, now lack of SVG support throughout Windows.
PNG is an ISO standard (ISO/IEC 15948:2003):
http://en.wikipedia.org/wiki/Png
SVG is an open standard promoted by W3C consortium:
http://en.wikipedia.org/wiki/Svg
Windows has very poor support for both.
We seem to be seeing a pattern here.
It seems like Ogg vorbis support in WMP revisited.
One has to ask why Windows devs refuse to support open standards.
It isn’t a difficult task. There are at least three other Windows managers that have far better support for wallpaper formats than the kludges in Windows.
Yes, I’m aware of that – I’m posting from BeOS, which has had the ability to use standard image file formats as backgrounds for at least the 7-8 years I’ve been using it (and possibly longer).
The “difficult taks” I was referring to was the task described in the article – attempting to clean up Vista’s handling of background images before the release date.
“The “difficult taks” I was referring to was the task described in the article – attempting to clean up Vista’s handling of background images before the release date.”
Supporting decent graphics file formats is not a difficult task. It has already been done, and the code to support it is open.
http://www.w3.org/Graphics/SVG/
http://svgi.org/
http://en.wikipedia.org/wiki/Libpng
http://www.libpng.org/pub/png/libpng.html
http://www.libpng.org/pub/png/libpng-1.2.5-manual.html
A lot of help is available.
One suspects the real problem (for something like libpng) is this attribute that it has: ‘OS: cross platform’. One suspects this is the only reason why Windows doesn’t support it.
It seems that only Windows developers are shy of supporting open standards in graphics and multimedia formats.
Is it because of a desire to lock-in Windows users to non-open formats, one wonders?
Supporting decent graphics file formats is not a difficult task.
No, supporting decent file formats is not a difficult task. But that’s not what I was referring to, as that’s not the task the article describes.
//No, supporting decent file formats is not a difficult task. But that’s not what I was referring to, as that’s not the task the article describes.//
Yes it is. The article talks about the kludges they had to include because the API for the Windows background supports only .bmp format. They had problems scaling .bmp to the desired range of screen size & resolution. Such problems are best addressed by including better support for other formats, and not by the kludge approach they did take which excludes such support for other formats within the API.
As the very first post in this thread asks … “When will we have native .png wallpaper support?”
Not with Vista, it would seem.
Please note, including support for open formats across Windows would not be difficult at all. There is ample support for designing such support. The reason why Windows lacks support for formats such as these is that Microsoft have deliberately left it out.
Think about that for just a moment.
Edited 2006-10-30 06:13
Please note, including support for open formats across Windows would not be difficult at all. There is ample support for designing such support. The reason why Windows lacks support for formats such as these is that Microsoft have deliberately left it out.
So you’ve seen the current implementation, know the areas that such a change would impact, and know that there aren’t any other items that are of a higher priority?
Yes it is. The article talks about the kludges they had to include because the API for the Windows background supports only .bmp format.
Eh? You’re disagreeing with me, and then essentially re-stating my point… Which is that the article does not describe supporting decent file formats, it describes awkward workarounds that were necessary solely because they painted themselves into a corner in the first place. AKA, what I was originally referring to as a “difficult task.”
As the very first post in this thread asks … “When will we have native .png wallpaper support?”
Not with Vista, it would seem.
Please note, including support for open formats across Windows would not be difficult at all. There is ample support for designing such support. The reason why Windows lacks support for formats such as these is that Microsoft have deliberately left it out.
Think about that for just a moment.
Riiiight. Oh the children, think of the humanity, et cetera, so on and so forth.
One has to ask why Windows devs refuse to support open standards.”
Because they didn’t “invent” them? 🙂
OGG/Vorbis support, standard RockRidge extension to ISO 9660, PostScript for printer output… there are so many standards around the computer world, but most of them you’ll not be able to find in MICROS~1 products (by default). Sad…
//Because they didn’t “invent” them? :-)//
I don’t think so. Windows does include a lot of things that Windows didn’t invent, including for example the very concept of a mouse-driven GUI.
//OGG/Vorbis support, standard RockRidge extension to ISO 9660, PostScript for printer output… there are so many standards around the computer world, but most of them you’ll not be able to find in MICROS~1 products (by default). Sad…//
You are of course correct. For that you got modded down. I have modded you back up again for a bit of balance.
Edited 2006-10-30 03:47
“//Because they didn’t “invent” them? :-)//
I don’t think so. Windows does include a lot of things that Windows didn’t invent, including for example the very concept of a mouse-driven GUI. “
Yes, I know that, but maybe some MICROS~1 fanboys have to be remebered the truth.
With this comment I liked to state the fact, thatt you hardly can find open standards inside MICROS~1 products (by default). That would give the user the possibility of using data produced on a non-“Windows” machine – or even transfer their “Windows” generated data to another platform. It seems they can’t resist to block such things.
But to come back to the topic: I like the concept of user (!) configuring the look of the GUI. Surely, this concept isn’t new, but it may be a way for “Vista” to gain more acceptance among the users who don’t like the default GUI settings and themes.
It would be great to have a “TweakUI” included in Vista as it is known to the obsoleted “Windows” types (95, 98 etc.). Why should only the programmers ad Redmond have “cool” GUIs while the end user have to use the childish one? 🙂
“You are of course correct. For that you got modded down.”
Wow… that’s discussion culture… remebers me to… no, I won’t say. 🙂
It seems like Ogg vorbis support in WMP revisited.
One has to ask why Windows devs refuse to support open standards.
No one gives a damn about Ogg Vorbis or SVG outside of the Slashdot crowd. There are tens of thousands of file formats out there. Why should Microsoft waste its time supporting those that no one uses?
//No one gives a damn about Ogg Vorbis or SVG outside of the Slashdot crowd. There are tens of thousands of file formats out there.//
Your question is misplaced.
The thing that Ogg Vorbis, PNG and SVG all have in common is that they are open and cross-platform formats. They are all perfectly good formats, and arguably better than competing formats.
//Why should Microsoft waste its time supporting those that no one uses?//
More to the point – why doesn’t Microsoft support these formats? IMO the answer is perfectly clear – Microsoft have a huge aversion to supporting any formats that could possibly give their end users a cross-platform “out”, an escape-route if you will away from Microsoft platforms.
Put it this way, and ask yourself a far more relevant question as an end-user of Microsoft software – “Why is Microsoft so reluctant to let my documents created on Windows be saved in any sort of cross-platform open format”?
Edited 2006-10-30 03:58
PNG is supported in Vista. Can’t be sure about SVG.
If someone wants to get either of these working, there’s an extension mechanism within Windows. For Ogg Vorbis, you just need a DirectShow Filter (just like DivX or ffdshow). And for SVG, there’s something called WIC (Windows Imaging Components), through which you can write a codec to decompress any image format you want.
As far as I can tell, you guys want Micros~1 to spend money on supporting standards that harm their business in real ways even after they have opened up APIs for third parties to implement them. If Ogg Vorbis and SVG support were such big deals for you, why wouldn’t you write or pay someone to write open-source decoders for windows?
//PNG is supported in Vista.//
Versions of Internet Explorer up to and including 6 do not support native alpha-channel transparency. Internet Explorer 7.0 has addressed the PNG-rendering issues, and now displays them correctly. AFAIK, that is about the extent of it. PNG files still can’t be used fully for backgrounds, and aren’t fully supported in Office.
//Can’t be sure about SVG. //
http://forums.microsoft.com/TechNet/ShowPost.aspx?PostID=602714&Sit…
“Unfortunately, neither Vista nor IE7 will include SVG support.”
//As far as I can tell, you guys want Micros~1 to spend money on supporting standards that harm their business in real ways even after they have opened up APIs for third parties to implement them.//
These formats are International Standards and Web Standards. Only Microsoft does not support them. On the face of it, this has the strong appearance that it is Microsoft’s deliberate policy that open formats (especially emerging ones) will not be supported natively by Windows or Office. On the face of it, this has the strong appearance that Microsoft is actively trying to make it so that such open formats do not succeed.
//If Ogg Vorbis and SVG support were such big deals for you, why wouldn’t you write or pay someone to write open-source decoders for windows?//
There are open-source codecs for Ogg Vorbis for Windows. The problem is that they are not included with Windows, and the default message that appears in WMP when you try to play such a file gives one the impression that Windows cannot support this format. This is not the case, Windows can support the format, it is just that Microsoft refuse to.
http://windowsxp.mvps.org/ogg.htm
The existance of a standard does not obligate MS or anyone else to support it. There are plenty of examples of MS creating, supporting, and participating in the devlopment of standards, but they don’t have to support every standard that appears, especially as others have pointed out, there are extensibility points for others to do so.
The existance of a standard does not obligate MS or anyone else to support it.
Even if there’s a demand for it?
Yes. Even if there’s demand, support is at their discretion. Demand is just one factor that would determine whether investment in that area (possibly at the expense of another) is justified.
//The existance of a standard does not obligate MS or anyone else to support it. //
So they admit they painted themselves into an ugly corner in respect of the backgrounds, {“This is something that we have been tracking internally for quite a while”}, admitted it was important to them to have it right, {“Can you say “major embarrassment”?}, knew that their competition had good support for decent formats in this area, knew that there was a lot of help out there for them to also include the same support, yet they refuse to do that and go with a kludge instead?
When just supporting the standards was viable and would have solved their problem? The phrase “dog in the manger” comes strongly to mind here.
As I said before, the real question one should ask oneself is this: “More to the point – why doesn’t Microsoft support these formats?”.
//There are plenty of examples of MS creating, supporting, and participating in the devlopment of standards, but they don’t have to support every standard that appears,//
he SVG standard in conjunction with SIML in particular is very capable.
http://en.wikipedia.org/wiki/Svg
“SVG allows three types of graphic objects:
* Vector graphic shapes (e.g. paths consisting of straight lines and curves, and areas bounded by them)
* Raster graphics images / digital images
* Text
Graphical objects can be grouped, styled, transformed and composited into previously rendered objects. Text can be in any XML namespace suitable to the application, which enhances searchability and accessibility of the SVG graphics. The feature set includes nested transformations, clipping paths, alpha masks, filter effects, template objects and extensibility.
SVG drawings can be dynamic and interactive. The Document Object Model (DOM) for SVG, which includes the full XML DOM, allows straightforward and efficient vector graphics animation via ECMAScript or SMIL. A rich set of event handlers such as onmouseover and onclick can be assigned to any SVG graphical object. Because of its compatibility and leveraging of other Web standards, features like scripting can be done on SVG elements and other XML elements from different namespaces simultaneously within the same web page. An extreme example of this is a complete Tetris game implemented as an SVG object, found here. (The link requires an SVG enabled browser.)
If storage space is an issue, SVG images can be saved with gzip compression, in which case they may be called “SVGZ files”. Because XML contains verbose text, it tends to compress very well and these files can be much smaller. Often however the original vector-file (SVG) is already smaller than the rasterised version.”
This allows for all sorts of interesting possibilities in web pages and indeed even embedded into documents & presentations. I mean, even a complete tetris game. Animations, all sorts of cool things.
These are only feasible, however, if they have wide support. That necessarily means Office and IE7. If those applications don’t support the standard (which they don’t) then the standard will fail because of lack of use. Then Microsoft will bring in their own proprietary “standard” *cough* *VML* *cough* which only Windows supports, and which has been rejected by standards bodies, charge everyone royalties for using it and/or refuse to allow an open-source implementation.
http://en.wikipedia.org/wiki/Vector_Markup_Language
// especially as others have pointed out, there are extensibility points for others to do so.//
Not good enough. There should be support for consensus standards built into the OS to provide support throughout the OS and major applications.
The only winners form Microsoft’s refusal to support standards are Microsoft. The losers are end users.
So why is there this seeming support for Microsoft?
I strongly suspect astroturfing here.
Edited 2006-10-30 09:01
So you’re saying that MS should support whatever is out there whether or not it aligns with other technology investments or would divert resources from other (possibly higher priority) projects? That’s rediculous.
Also, how was MS going to support SVG over VML in Office 2000 or IE 5 when both products shipped before SVG?
Concerning wider format support, has it not occurred to you that MS has spent their time and resources for this release on a more flexible, more secure codec architecture (WIC) and that they would rather build new functionality (such as a major rev of the desktop wallpaper API) on that platform rather than the old architecture? Native JPEG support seems to be basically an interim addition that provided performance benefits for other technologies like SuperFetch while being a fairly low cost addition.
I strongly suspect astroturfing here.
I smell a standard talking point reply whenever reality doesn’t fit one’s assumptions (especially in topics involving Microsoft).
//So you’re saying that MS should support whatever is out there whether or not it aligns with other technology investments or would divert resources from other (possibly higher priority) projects? That’s rediculous.//
No, I am saying that Microsoft should support open & unencumbered interoperability standards that are agreed by independent standards bodies, rather than their go-it-alone proprietary “standards” that are less functional and aimed purely at lock-in to Windows platforms.
Of all the major software vendors, Microsoft is the only one with a “one platform, strictly no interoperability” policy.
It isn’t at all difficult to support these open standards. Every other software vendor has no trouble at all supporting them. They would all be of tremendous benefit to Microsoft’s end customers.
End customers are becoming increasingly insistent on interoperability and standards compliance. Only Microsoft (of all software vendors) refuse to cater to that end-customer need.
http://business.newsforge.com/business/06/10/20/1621200.shtml?tid=1…
“The IDABC drafted something called the European Interoperability Framework, which serves as a guideline to member states and EU bodies. It identifies the core requirements of administrations, which are: availability and reliability, security, accessibility, sustainability (including availability over the long term), independence from vendor lock-in, value for money (including the cost of software implementation and licensing), scalability, and re-usability. By accounting for these comprehensive requirements, the EIF was a strong statement in support of the kinds of open standards that the free software community favours. Namely, a standard should be adopted and managed by a non-profit organisation and be open to all interested parties. It should be fully published, with the specifications being made available free of charge. Any intellectual property contained within the standard, or irrevocably connected to it, should be made available on a royalty-free basis. Finally, there should be no constraints on the re-use of the standard.”
If Microsoft continue with their lock-in policies, then they could well find themselves increasingly ineligible to be considered as a software provider.
//I smell a standard talking point reply whenever reality doesn’t fit one’s assumptions//
It is your assumption that Microsoft get to dictate the market (and not the end customers) that is increasingly failing to fit the reality.
Microsoft only think they are in control. Microsoft’s continued insistence on “my way or the high way, no standard interoperable formats for You!” will increasingly see them excluded from even bidding in the future.
Edited 2006-10-30 11:35
Oh, BTW, OOXML misses out too.
“Any intellectual property contained within the standard, or irrevocably connected to it, should be made available on a royalty-free basis.”
There are a significant number of dependencies built in to the OOXML proposed standard that fall afoul of the requirement above. OOXML has many dependencies on other Windows platform components “irrevocably connected” to OOXML which most decidedly are not open nor available on a royalty-free basis.
This seems very much to be the pattern for Microsoft. One way or another they try to lock people in via formats that only Microsoft platforms can support.
This just isn’t going to wash.
Edited 2006-10-30 12:04
There are a significant number of dependencies built in to the OOXML proposed standard that fall afoul of the requirement above. OOXML has many dependencies on other Windows platform components “irrevocably connected” to OOXML which most decidedly are not open nor available on a royalty-free basis.
I suppose you are referring to OOXML’s support for COM objects. ODF also supports COM objects, the COM standard has been available for licensing from The Open Group for years, and COM is not required for implementing or interoperating with either standard. COM is also not a Windows-specific technology. Microsoft themselves have implemented it on Windows, Mac, and Unix, as have others.
OOXML is licensed on a royalty-free basis. You can use the CNS or OSP provisions — your choice.
“We are giving potential implementers of Ecma Office Open XML the ability to take advantage of either the CNS or the OSP, at their choice. Microsoft had already stated that it offers an irrevocable covenant not to sue (CNS) to anyone wishing to implement the formats. We understand that some may prefer the new OSP, which we’d like to facilitate.”
…
The Open Specification Promise is a simple and clear way to assure that the broadest audience of developers and customers working with commercial or open source software can implement specifications through a simplified method of sharing of technical assets, while recognizing the legitimacy of intellectual property.”
http://www.microsoft.com/interop/osp/default.mspx
Entities like these don’t seem to have a problem with it:
Apple
Barclays Capital
BP
The British Library
Essilor
Intel
NextPage
Novell
Statoil
Toshiba
United States Library of Congress
Edited 2006-10-30 13:08
//Entities like these don’t seem to have a problem with it//
However there are far more non-lapdog people who do have a problem with it:
http://en.wikipedia.org/wiki/Standard_office_document_formats_debat…
There is a huge list of reading for you via the links at the bottom of the above page.
http://www.robweir.com/blog/
That was a good one, hey! Microsoft are going to standardise their legacy bugs!
However there are far more non-lapdog people who do have a problem with it:
Oh, So Apple and Novell are MS lapdogs? Please!
I guess Red Hat is as well because they basically endorsed the OSP. The main entities that have a problem with OOXML are those trying to push ODF as being the only standard anyone would need. The links substantiate that.
That was a good one, hey! Microsoft are going to standardise their legacy bugs!
If you did research, you’d know it isn’t an MS bug. It originated with Lotus 1-2-3 and was retained for compatibility.
Oh no! ODF doesn’t even have a standard representation for formulas and can’t even interop with the apps that support it because they all use different ways of saving that (and other) data.
http://software.newsforge.com/article.pl?sid=05/09/09/192250&from=r…
“That was a good one, hey! Microsoft are going to standardise their legacy bugs!”
First, the bug is Lotus’s bug that Excel incorporated for compatibility. BTW, Lotus is run by ODF-champion IBM, how about that?
Anyway, read
http://blogs.msdn.com/brian_jones/archive/2006/10/25/spreadsheetml-…
and you’ll see how trivial and inconsequential that problem is. It was dealt with by the ECMA committee, and after reading the above blog entry, I’d say they made the correct decision regarding it.
No, I am saying that Microsoft should support open & unencumbered interoperability standards that are agreed by independent standards bodies, rather than their go-it-alone proprietary “standards” that are less functional and aimed purely at lock-in to Windows platforms.
And they do this in many areas.
Of all the major software vendors, Microsoft is the only one with a “one platform, strictly no interoperability” policy.
Unsubstantiated conjecture.
It is your assumption that Microsoft get to dictate the market (and not the end customers) that is increasingly failing to fit the reality.
I have neither assumed nor asserted that MS dictate the market. I have asserted that they (and any other company or entity) have the right to choose which standards they support. Just because a standard exists does not make its implementation mandatory for anyone.
Microsoft only think they are in control. Microsoft’s continued insistence on “my way or the high way, no standard interoperable formats for You!” will increasingly see them excluded from even bidding in the future.
Once again you make this assertion without backing it up. Just because you say it does not make it true. Microsoft has both created and supported many industry standards over the years.
//And they do this in many areas. //
Name one.
//Unsubstantiated conjecture. //
Microsoft Office is available for exactly how many platforms? IE7 is available for exactly how many platforms? ActiveX is available for exactly how many platforms? DirectX is available for exactly how many platforms? Visual Basic is available for exactly how many platforms? .Net is available for exactly how many platforms? Visual C is available for exactly how many platforms? Win32 API is available for exactly how many platforms?
What “cross-platform application support” tools does Microsoft supply?
//I have asserted that they (and any other company or entity) have the right to choose which standards they support. Just because a standard exists does not make its implementation mandatory for anyone.//
This is not correct. You are absolutely mistaken here.
I would use as an example from the telecommunications industry the telephone signalling standard, or alternatively the television broadcasting standard. These standards are open and unencumbered, so that any vendor might produce a telephone handset and expect it to work (interoperate) with a PABX from another vendor. Similarly a television set from any vendor should be expected to work with any VCR, any DVD player and any television station broadcast transmitter from any other vendor. In these markets, interoperability with products from different manufactures is mandatory. In fact, the interoperability standards are mostly set and enforced by independent not-for-profit bodies.
This interoperability between products from different manufacturers is the very basis of competition in these markets.
Microsoft, OTOH, seem to think that an interoperability standard such as the way that their desktop clients interoperate with their servers, or the way that digital documents are saved, should be the exclusive “right of Microsoft to support”. They seem to think it is perfectly legitimate for themselves to keep an interoperability standard as a trade secret. This just is not so. Behavior such as this is antitrust, and it is illegal, and Microsoft have been convicted for such illegal practices more than once.
//Once again you make this assertion without backing it up. Just because you say it does not make it true. Microsoft has both created and supported many industry standards over the years.//
Just because Microsoft does something or comes up with something or implements something in some particular way does not make it a standard. Standards are not set by just the one corporation, however much Microsoft might want that to be the case.
Oh, I have backed this up all-right. Well and truly.
For example, such a thing a European Interoperability Framework really does exist …
http://ec.europa.eu/idabc/en/document/3473
Microsoft software does not comply.
http://ec.europa.eu/idabc/en/chapter/5883
I can back this up all day. I have provided infinitely more links than you have, I would also point out.
//Just because you say it does not make it true.//
Right back at you, kiddo.
Edited 2006-10-30 13:26
Name one
Web Services
Microsoft Office is available for exactly how many platforms? IE7 is available for exactly how many platforms? ActiveX is available for exactly how many platforms? DirectX is available for exactly how many platforms? Visual Basic is available for exactly how many platforms? .Net is available for exactly how many platforms? Visual C is available for exactly how many platforms? Win32 API is available for exactly how many platforms?What “cross-platform application support” tools does Microsoft supply?
So when proven wrong about their support of industry standards, you are now trying to say that they have to make all of their products cross-platform? You’re just moving the goalpost.
This is not correct. You are absolutely mistaken here.
I would use as an example from the telecommunications industry the telephone signalling standard, or alternatively the television broadcasting standard. These standards are open and unencumbered, so that any vendor might produce a telephone handset and expect it to work (interoperate) with a PABX from another vendor.Similarly a television set from any vendor should be expected to work with any VCR, any DVD player and any television station broadcast transmitter from any other vendor. In these markets, interoperability with products from different manufactures is mandatory. In fact, the interoperability standards are mostly set and enforced by independent not-for-profit bodies.
This interoperability between products from different manufacturers is the very basis of competition in these markets.
It is you who are mistaken. I don’t see mobile or wi-fi phone makers being forced to interop with landline signaling or PBX. They even have multiple incompatible standards within their industry. For TV broadcasts, there are multiple incompatible standards depending on the region. There are multiple, incompatible standards that various service providers use for their broadcasts. HDTVs and DVD players may use a number or interference or encryption techniques to prevent interop with VCRs or other recording devices.
Microsoft, OTOH, seem to think that an interoperability standard such as the way that their desktop clients interoperate with their servers, or the way that digital documents are saved, should be the exclusive “right of Microsoft to support”. They seem to think it is perfectly legitimate for themselves to keep an interoperability standard as a trade secret. This just is not so. Behavior such as this is antitrust, and it is illegal, and Microsoft have been convicted for such illegal practices more than once.
There are now and have always been multiple extensibility points for the integration of third-party file systems, networking protocols, and other technologies. Microsoft has also provided (again, for years) support for several non-MS protocols that non-Windows machines could use for interop. They haven’t been convicted of anything. They have committed no criminal offence. Try again.
Just because Microsoft does something or comes up with something or implements something in some particular way does not make it a standard. Standards are not set by just the one corporation, however much Microsoft might want that to be the case.
Standards can be from one company. Wide adoption of such technologies makes them de facto standards. There were many companies before MS and many since that have created de facto standards. It is only you who has asserted that an MS technology is standard simply by invention. Neither I nor MS have asserted this. You still ignore that MS has participated with standards organizations. You keep ommiting this by choice as it deflates your argument that they only use their own tech.
Oh, I have backed this up all-right. Well and truly. For example, such a thing a European Interoperability Framework really does exist …
http://ec.europa.eu/idabc/en/document/3473
Microsoft software does not comply.
Once again, saying it does not make it true. I see nothing in the EIF that would exclude MS technologies. CLI technologies, OSP technologies (WS, OOXML, soon XPS) and more would all be usable under that framework.
When just supporting the standards was viable and would have solved their problem?
What do you mean, exactly? JPEG is a “standard” and Vista does support it. But merely supporting that didn’t solve the problem, because the “problem” is not what formats are supported as much as it is that different components (Windows Photo Gallery (or whatever it’s called), IE, and Desktop Control Panel) were supporting whatever formats they supported in different ways. The problem is the apparent lack of an API that all these apps could call into that would be flexible enough to do whatever they wanted, but that problem would exist regardless of what picture formats are supported.
“I strongly suspect astroturfing here.”
You say this so often, that it makes me think it’s your cop-out response. You bring it out to automagically discredit any argument that’s made against your anti-MS tirades. A pretty pathetic debating tactic.
Oh please.
Go look at .NET’s graphics support and you’ll see that PNG is not only supported, it’s the preferred graphics file format.
If what you say were true, Microsoft wouldn’t support png, gif, jpeg, mp3, aac, h.264 (the latter two are natively supported by Zune), etc, since Microsoft didn’t invent those either.
And to answer your last question, “Why is Microsoft so reluctant to let my documents created on Windows be saved in any sort of cross-platform open format?”, how does Windows prevent saving documents in cross-platform formats? Does OpenOffice not work on Windows now? Does Windows block all attempts to save or load RTF (a cross-platform format created by Microsoft)? How about ASCII, can Windows apps no longer load/save ASCII files? Give me a break.
Considering Microsoft is supporting gazillions of old unused formats, it’s weird that Microsoft do not support formats used by millions everyday.
Especially when they point to millions of users complaining about the scaling goof.
it’s weird that Microsoft do not support formats used by millions everyday.
A) Do you have any evidence that Ogg Vorbis and SVG are “used by millions everyday”? If so, are a substantial number of these “millions” going to be using Windows or sending these files to users of Windows?
B) Do other operating systems support these formats out of the box? Does Linux (not value-added distributions such as Ubuntu or Debian, but Linux proper)? Does OS X[1]? Does OpenBSD? Does FreeBSD? Does NetBSD? All of those operating systems require 3rd party components to view/play such content. That is precisely the same case as with Windows.
[1]Leopard should have SVG support when it is released next year but the current version of OS X, Tiger, does not.
See answer here (must have been mentally sleeping):
http://osnews.com/permalink.php?news_id=16340&comment_id=176893
A) Yes and yes and perhaps (unfortunately). A few million users are nothing in reality. Consider the total number of XP-users and take something like .01% of that.
B) What is Linux proper? Are we talking the kernel only, or the contents of a standard distribution, or a minimal LinuxFromScratch installation? There is no such thing as Linux proper.
And yes, OpenBSD, NetBSD and FreeBSD supports Ogg Vorbis and SVG out of the box.
And yes, it’s done through 3rd party components. However, the issue is not whether support is through 3rd party components. The issue is whether or not to ship with support for these standards out-of-the-box, either through homewritten codecs and libraries or through 3rd party codecs and libraries. Windows do not ship with such support, while Linux, *BSD, Syllable and several extreme hobbyist OS’es do.
Windows does have support for open standards (incl. SVG – to some extent), but not out of the box. Personally I don’t think they can include 3rd party codecs without getting in trouble (I can almost already hear certain companies whining about it).
A) Yes and yes
Evidence?
B) What is Linux proper? Are we talking the kernel only?
Yes. That’s what Linux is, isn’t it?
And yes, OpenBSD, NetBSD and FreeBSD supports Ogg Vorbis and SVG out of the box.
I’m running OpenBSD right now. I have FreeBSD on my other machine. Please tell me how I can play Ogg Vorbis files and watch SVG animations “out of the box”, i.e., without installing additional packages or ports.
Edited 2006-10-30 17:01
When people are talking about linux they are usually talking about one of the many linux-distributions one can download.
When you talk about XP you do not talk about the NT kernel, but the entire XP installation.
The Windows kernel can play very little, and the same goes for the Linux kernel
I will not answer the other questions until we have established whether we’re discussing the entire system or merely the kernel.
I will not answer the other questions until we have established whether we’re discussing the entire system or merely the kernel.
We could argue about what consistutes a base “Linux system”, but let’s ignore Linux instead.
I’m unclear as to how that clarification has any bearing on my requests for:
a) Evidence that SVG and Ogg Vorbis files are used daily by millions of people (a substantial number of whom use Windows) per your assertion.
b) Instructions as to how to view/play such files using the OpenBSD/FreeBSD/NetBSD base systems per your second assertion. The base systems in these cases do not include packages/ports/repositories, but rather the files included in standard installations (for OpenBSD, this is the kernel, base39.tgz, comp39.tgz, etc39.tgz, game39.tgz, man39.tgz, misc39.tgz, xbase39.tgz, xetc39.tgz, xfont39.tgz, xserv39.tgz, and xshare39.tgz).
As I stated earlier:
I’m not answering further questions until we have a clear definition on what we are discussing.
Windows, Linux, *BSD is all irrelevant, if we aren’t discussing the entire system.
Apparently you are also talking about the Linux kernel when talking about Linux, but the entire system when talking about Windows and *BSD.
And no, I will not ignore one system, just because you deliberately attempt to create confusion.
It’s either all of them or none at all.
And no, I will not ignore one system, just because you deliberately attempt to create confusion.
I take back my Linux statement after you pointed out the ambiguity and now you refuse to back up your claims about SVG/OGG usage and *BSD support out of spite? Is that what you’re saying?
No. I’m not saying that.
In the post where I pointed out the (deliberate) ambiguity in your posts, I also stated that I wouldn’t answer any more questions until we had a definition on “Linux proper”.
We have no such definition, and therefore I will not answer further questions. I stated that several posts ago and have nothing to do with *BSD. It’s merely a matter of cleaning up deliberate ambiguity on your side.
However, if you really want it, then okay.
Let’s get a definition on “out-of-the-box” since *BSD’s usually are shipped much the same way as gentoo and therefore do not exist as an “out-of-the-box” installation.
When is it “out-of-the-box” supported, and when is it supported through “Windows Update”/ports/portage-like update system and when should it be considered as 3rd party support.
Considering the distribution nature of gentoo and *BSD I consider them to have “out-of-the-box” support since portage/ports _are_ the installation.
I don’t consider the libraries for SVG and Ogg Vorbis support as 3rd party libraries on (GNU/)Linux and *BSD as the companies and communities around them are handling the packaging (and to some extent also the development).
Ogg Vorbis support is no more 3rd party on (GNU/)Linux and *BSD than support for IE is 3rd party in Windows.
Considering the distribution nature of gentoo and *BSD I consider them to have “out-of-the-box” support since portage/ports _are_ the installation.
Either you know very little about *BSD (I don’t see how Gentoo is relevant) or you’re out-and-out spreading FUD.
No one who uses or develops Free/Net/OpenBSD would honestly claim that applications such as Opera or Eclipse are part of either the base system or the standard installation. Heck, the ports tree itself is a separate, non-automated, post-installation download on OpenBSD.
And you still haven’t delivered the promised evidence for SVG/Ogg being used by “millions everyday,” including a substantial number of Windows users.
And I’m not going to until I receive clarification on the matters earlier discussed.
I’ve made that very clear all the way.
RE[15]: JPEG support is new?
By dylansmrjones (1.35) on 2006-10-30 20:42:46 EST in reply to “”
And I’m not going to until I receive clarification on the matters earlier discussed.
I’ve made that very clear all the way.
——————————-
RE[8]: JPEG support is new?
By eMagius (1.51) on 2006-10-30 12:00:43 EST in reply to “RE[7]: JPEG support is new?”
A) Yes and yes
Evidence?
B) What is Linux proper? Are we talking the kernel only?
Yes.
——————————-
You’ve got nothing to back up your outlandish claims. At least have the balls to admit you were just spreading FUD.
Yea like that Neo-Nazi’s gonna confess…the only thing his “master race” has are lies and murder.
I am not a neo-nazi.
Where do you have such information from?
I am as far from nazism as anyone can possibly get. One more of those claims, and it’ll have severe consequences. If necessary I’ll go as far as sueing OSN in order to prevent you from spreading more lies about me.
Do I make myself clear?
(And yes, that is also to the OSN-staff.)
I am not spreading FUD.
In post by me on 2006-10-30 18:28:17 CET in reply to “RE[8]: JPEG support is new?” I asked for clarification on what we were discussing. It turned out you were talking about the Linux kernel and the ENTIRE Windows system.
You also wanted to include *BSD and despite the fact that there is no “out-of-the-box” version of these (with exception of DesktopBSD and PC-BSD – both based on FreeBSD) you still want me to analyze them as such.
Before I can do that we need a common definition, instead of me using my definitions and you using your definitions.
A few posts ago I wrote this:
“Let’s get a definition on “out-of-the-box” since *BSD’s usually are shipped much the same way as gentoo and therefore do not exist as an “out-of-the-box” installation.”
You didn’t come up with your view on the definitions, while I make an attempt to answer your questions. However, with no common definition I am bound to fail to express my view in a way you can comprehend, and vice versa.
In order to get a constructive dialogue working, I suggest we develop a common definition on these issues before we enter further discussions.
A few posts ago I wrote this:
“Let’s get a definition on “out-of-the-box” since *BSD’s usually are shipped much the same way as gentoo and therefore do not exist as an “out-of-the-box” installation.”
…
You also wanted to include *BSD and despite the fact that there is no “out-of-the-box” version of these (with exception of DesktopBSD and PC-BSD – both based on FreeBSD) you still want me to analyze them as such.
As I pointed out earlier, that’s definitely not the case. Here, I’ll link you to the OpenBSD installation guide, so you can try installing OpenBSD yourself: http://www.openbsd.org/faq/faq4.html . It’s a small OS and installation is pretty speedy (usually less than five minutes from start to finish), so I’ll wait.
Ready?
Now you see that OpenBSD (and FreeBSD and NetBSD) are very much complete systems right out of the box. Have I installed additional software on my OpenBSD machine? Sure, I’ve installed MySQL, PHP, elinks, GNU screen, and a few other applications. But that’s just like how I’ve installed additional software on the Windows and OS X machines in my lab.
Again, OpenBSD is complete and functional right out of the box (as are FreeBSD and NetBSD), including a web server, several text editors, the X Windowing System, compilers, a web browser, games and more; these are all part of the base system.
There’s no need to create a definition of an “OpenBSD system” — it’s what you get when you install OpenBSD. OpenBSD is OpenBSD.
you still want me to analyze them as such.
I’ve asked you to cite evidence for your claims that SVG and Ogg Vorbis are used by “millions everyday” including a significant number of Windows users. The whole “definition of *BSD/Linux” has nothing to do with your ability to provide such a reference.
IIRC in win 9x, jpg wallpapers were only possible with “Active Desktop.” I don’t remember anything like that for XP though.
PNG would be nice,in particular for computer generated images as the OP mentions. Hard edges and solid colors especially are best served with a lossless format. So are we sure Vista can’t do png? I find that rather hard to believe.
Of course Vista can make PNGs into wallpapers (it works the exact same way as BMPs and JPGs. But it seems to convert them internally into JPEGs, the same as pre-Vista Windows convert any wallpaper into uncompressed BMP.
So this wouldn’t be a real feature :-(.
Having every picture converted into a _lossy_ format is a really bad idea!
//PNG would be nice,in particular for computer generated images as the OP mentions. Hard edges and solid colors especially are best served with a lossless format. So are we sure Vista can’t do png? I find that rather hard to believe.//
For “drawings” (ie. ‘computer generated images’ as opposed to digital photographs) for wallpapers, a scalable vector image format is better than a raster format. With a scalable vector image format you can get a small file size and good quality and at the same time achieve independence from screen resolution and size.
http://en.wikipedia.org/wiki/Computer_wallpaper#Formats
“Some desktop systems, such as Mac OS (version 8.6 or later), KDE (version 3.4 or later), and GNOME, support vector wallpapers (PICT in Mac and SVG in KDE and GNOME). This has the advantage that a single file may be used for screens of any size, or stretched across several screens, without loss of quality.”
Edited 2006-10-30 02:34
mac os 9 can use more image formats
png jpg bmp gif pict psd qtif sgi tga tiff +more
//mac os 9 can use more image formats
png jpg bmp gif pict psd qtif sgi tga tiff +more//
Does the tetris game work for you?
http://www.croczilla.com/svg/samples/svgtetris/svgtetris.svg
Imagine that – an animated game implemented in an image format!
Edited 2006-10-30 11:47
works without problems
The technical reason behind all this is that Microsoft’s Operating System is really an outdated mess; the amount of ugly Hungarian Notation C++ code and weird API calls they have there is amazing.
The thing is getting better, which means that they have made a tremendous effort to make Vista work.
This wallpaper problem shouldn’t be a problem if things were designed “more or less” in a modern way.
Then you can argue that Microsoft loves *not to* make use of Open Standards all the time… something that is really a pity; Windows Vista could be a nice and very useful desktop, if they only played better with the rest of the crowd.
*sigh*
//The technical reason behind all this is that Microsoft’s Operating System is really an outdated mess; the amount of ugly Hungarian Notation C++ code and weird API calls they have there is amazing.//
All the more reason to shun Vista then, one would logically surmise.
//Windows Vista could be a nice and very useful desktop, if they only played better with the rest of the crowd.//
But they don’t, and so it isn’t nice at all.
BTW, have you read what Microsoft think are reasonable terms for people to use their new software?
http://www.regdeveloper.co.uk/2006/10/29/microsoft_vista_eula_analy…
Unbelievable hubris, is it not?
BTW, have you read what Microsoft think are reasonable terms for people to use their new software?
http://www.regdeveloper.co.uk/2006/10/29/microsoft_vista_eula_analy…..
Unbelievable hubris, is it not?
As usual the Register gets it wrong. The best they could come up with is a benchmarking clause that has been around for years (check any version of .NET), is not specific to Vista, and is similar to clauses in several non-MS products. And Virtualization is not banned. You need an additional license for the VM instance unless you are running Ultimate Edition (which allows you to use the same license for the installed and VM instances) or Enterprise (which gives you the right to run 4 seperately licensed VM instances on the same computer).
//As usual the Register gets it wrong.//
You think this story was the Register?
http://www.channelregister.co.uk/2006/10/29/microsoft_vista_eula_an…
http://www.hexus.net/content/item.php?item=7075
http://www.techdirt.com/articles/20061019/102225.shtml
http://www.theinquirer.net/default.aspx?article=35108
http://www.informationweek.com/industries/showArticle.jhtml?article…
http://www.windowsitpro.com/Windows/Article/ArticleID/93904/93904.h…
http://www.windowsitpro.com/mobile/pda/Article.cfm?ArticleID=93937&…
http://weblog.infoworld.com/gripeline/archives/2006/10/a_vista_of_l…
http://www.informationweek.com/blog/main/archives/2006/10/what_do_y…
http://blogs.zdnet.com/hardware/index.php?p=118
http://weblog.infoworld.com/virtualization/archives/2006/10/the_tru…
http://blogs.zdnet.com/hardware/index.php?p=128
http://www.dabcc.com/article.aspx?id=2816
http://www.bitsofnews.com/content/view/4264/44/
Microsoft has a lot bigger PR disaster on its hands than just the Register.
Edited 2006-10-30 13:42
Microsoft has a lot bigger PR disaster on its hands than just the Register.
Whatever. There are other news orgs and actual users that don’t have a problem. Many of the articles about the horrors of WGA/activation are pure speculation and mirror what was said about XP prior to its release, yet it is now the dominant client OS. Many of the virtualization stories were countered with clarifications shortly after they were published. Ed Bott has several non-alarmist articles on ZD.
Most shouting gloom and doom make up a minority of the customer base (some don’t even use Windows, or at least claim not to do so), shout doom and gloom with each release, then move on to something else after launch. It’s a cycle that repeats with every release. This pretty much sums it up:
http://www.istartedsomething.com/20061017/dont-need-windows/
Of course XP is the dominant Client OS (what else, considering the weakness of controlling the market mechanisms). But that doesn’t change the fact that many users are sick and tired of the MS EULA and WGA.
The WGA is even illegal in certain european countries.
Your lies about being a libertarian are become more barefaced with every post. I trust that “ethnic cleansing” program is coming along well, mein fuhrer?
Not again… *sigh*
. o O ( Eggman on the run )
I’m assuming Windows Vista has some generic API for opening and transforming images like CoreImage in OS X.
I can’t work out from the article, but it seems like the reason images needed to be converted to a bitmap is so that it can be loaded onto the background of the graphics subsystem — or as a background on the root window if it works like that.
As I see it, if they’ve loaded the JPEG/BMP/PSD/PNG or whatever with their image transform API and have cropped, high quality resized, etc. and now have the image in memory ready for the background — uncompressed I should imagine at this stage (this image could be compressed and cached to somewhere on the hard drive for next bootup etc.).
Since the desktop is composited, then surely it could just be assigned to a/some texture(s) on whatever object is representing the root window and uploaded to VRAM. As it’s a texture managed by hardware, the software never has to deal with redrawing again and thus can free the memory and not deal with costly/slow transfers from RAM to VRAM. Any shaders could also be applied on the video card.
Software itself such as IE, display control panel, Microsoft’s photo/image viewers, Aunt Joe’s random image viewer freeware, could use a generic API that specifies the location of an XML file describing the image and how it should be transformed (take a look at GEGL for an example of what I mean by this: http://pippin.gimp.org/gegl/gallery/index.html, It’s possible MS already have some API supported description of this .. WPF/XAML?). Whenever the registry entry or whatever pointing to the XML file describing the wallpaper is changed, explorer or some such fires up the code to perform the change of wallpaper.
That way, the only time there will be any messing around with the image in memory is when the wallpaper is changed or at logon/boot. The image will be transferred to video card in the same way it has ever been and there will be no more disk I/O than before. If a sudden change of wallpaper isn’t ‘aesthetic’ enough, there’s probably something they could implement in DirectX to fade between the textures after they’re both in VRAM.
Am I missing the point entirely?
I think you’ve got it. The CoreImage-like API is Windows Imaging Components. There is also a hardware-accelerated image pipeline for doing all this scaling and stuff (which is used by the Photogallery–mentioned on the Shell blog). Problem is, this stuff needs time to be well-tested and baked in, so IE and Office, and others are still using the old GDI+ way of doing things. Also, the security implications of new image processing code are serious. I don’t expect the new systems to be used until at least next release (Vienna), if then.
Adding to the above, the main issue isn’t the API used for displaying background images. It’s the different processing steps that each application uses before displaying the image.
Each path (Explorer, Photo Gallery, Control Panel, IE, etc.) performed their own processing steps and had their own scaling algorithms that they used prior to displaying the image on the desktop. Control Panel’s JPEG scaling algorithm was not of sufficient quality and produced visible artifacts in some instances. The other paths each used different scaling implementations so they didn’t suffer from this issue. The fix was to use the same scaling algorithm that the Control Panel path used for bitmaps, converting the JPEGs to 24-bit before sending them through the bitmap scaler.
“””There has been a lot of attention in the last few weeks about the discrepancy in the desktop background appearance between the control panel and Windows Photo Gallery.”””
I don’t happen to use Windows. And I’m not thinking about using Vista. But if I did happen to use Windows, and if I were thinking about moving to Vista, THIS WOULD STILL SEEM LIKE THE NUMBER ONE MOST USELESS TOPIC EVER POSTED ON OSNEWS.
Even worse than Eugenia’s old rant about how the corners of the buttons in Gnome were not slightly rounded.