openSUSE Hobbled By Microsoft Patents

kripkenstein writes "openSUSE 10.2 no longer enables ClearType (which would improve the appearance of fonts). The reason given on the openSUSE mailing list for not enabling it is, 'this feature is covered by several Microsoft patents and should not be activated in any default build of the library.' As reported on and discussed, this matter may be connected to the Microsoft-Novell deal. If so, Novell should have received a license for the Microsoft patents, assuming the deal covered all relevant patents. Does the license therefore extend only to SUSE, but not openSUSE?"

Comments

Prior art

[Ph33r th3 g(O)at]

Steve Gibson pointed out decades-old prior art that would invalidate the Cleartype patent (if our patent system weren't corrupt) several years ago.

Re:Prior art

[gweihir]

Steve Gibson pointed out decades-old prior art that would invalidate the Cleartype patent several years ago.

Indeed he did. Not that the idea itself merits a patent anyways. It is pretty obvious and shopuld not be patentable in the first place.

Re:Prior art

[morgan_greywolf]

Gibson is exactly right. When I first saw sub-pixel rendering (aka 'ClearType') explained, I remembered programming graphics on an Apple IIe, and you had a difference between even and odd pixels that forced you to draw lines in a way that is exactly the same as how ClearType works.

I could claim prior art if I could just get those damned 5.25" floppies to read in anything. Of course, this was common practice back in the day, so maybe some old Apple II programmers out there can come up with AppleSoft BASIC code or something.

Re:Prior art

[0123456789]

There's a difference between being able to find prior art for something, and being able to afford to go to court to defend yourself against a patent infringement lawsuit. Sadly, the gulf between the two positions is pretty wide. Maybe there should be an appeal process for patent awards? If you can show that a patent affects you in some way, and shouldn't have been granted for some reason (eg prior art), you could appeal against the patent award and attempt to get it rescinded in a quicker and cheaper process than a full-on court case?

Re:Prior art

[tomhudson]

The funny thing is I just installed OpenSUSE 10.2 alpha 3 and the fonts look better than ever; if this is how they look without cleartype, who needs it?

Re:Prior art

[pikine]

I'm afraid the decades-old Apple II and IBM PC is not prior art. Pixels are either on or off for Apple II and IBM PC's CGA displays, so they apparently don't (and can't) care too much about color fringing. Sub-pixel font rendering on LCD screen deals with 256 shades for each sub-pixel, and the emphasis is on how to adjust sub-pixel brightness to reduce color fringing.

This is explained in Steve Gibson's Turning Theory into Practice. Sub-pixel font rendering is not the same as sub-pixels on CGA displays. The ideas are related, but the plumbing is different.

Perhaps I'm misleading in saying that CGA is not prior art of ClearType. I haven't actually read the patents of ClearType, so I obviously cannot tell; I'm basing my claim solely on Steve's webpage alone.

Re:Prior art

[kobaz]

I couldn't agree more.

I've never found cleartype to be helpful either, I much rather not have cleartype as on every single display device I've enabled it on it looks like crap. I've tried it on high and low end crts and high and low end lcds, it all looks much better (and more readable) without cleartype.

Re:Prior art

[dre80]

I entirely agree - OpenSUSE 10.2 (final) has the best-looking fonts I've seen in any distro, ever. Easily. I've yet to have the time to find out why, but it really is a beautiful thing.

Re:Prior art

[tomz16]

Any time I've turned on cleartype on a fresh install of windows, my first impression has always been that it just made fonts look "blurrier", for lack of a better word.

However, after using it for a day or two, turning it off is absolutely painful. IMHO, it really DOES make text MUCH easier to read on an LCD.

-Tom

Re:Prior art

[Anonymous Coward]

Prior art doesn't have to be identical - patents are supposed to be for something novel and nonobvious*, not for a logical progression given technological development. i.e. the fact display subpixels didn't have 256 brightness levels back in the day doesn't mean that it isn't blitheringly obvious that IF THEY ONE DAY DID, similar techniques would naturalyl apply.

* Yes, in practice the USA grants patents for basically anything regardless of merit, but the USA sucks.

Re:Prior art

[spitzak]

Holy confusion, batman!

That was completely misleading. The use of 256 levels for making antialiased fonts is really old and has nothing to do with this.

What sub pixel rendering does is make the pixel represent *more* than 256 different possible combinations of the fg and bg colors, where combinations are how an fully opaque edge falls into the square the pixel represents. Exactly how many is unclear, it is not 256^3, but I think it is 3*256 for the case of an antialiased vertical straight edge of an object significantly larger than a pixel (it goes down as the edge gets less vertical or if another edge gets within a few pixels of it). This is done by using hardware quirks so that various of the 256^3 possible colors are interpreted by the viewer as different coverages.

The Apple technique made a pixel represent more than the 2 possible coverages you mentioned. This is done by using hardware quirks so that the some of the 4 (16? i dunno) possible colors are interpreted by the viewer as different coverages.

They certainly are related, but I don't believe enough to invalidate Microsoft's patent. The difference is that the "hardware quirks" are vastly different.

Re:Prior art

[Sleepy]

I remember using anti-aliased fonts in "Spectrum 512", a terrific graphics package for the Atari ST series computer. With it, you could optimize images for TV. If you could not afford an Amiga, this was a very impressive way of generating titles and graphics for video. The Atari ST had a composite NTSC output port and the horsepower to drive it.

I remember the same technique used on the Atari 8-bit computers, in the monochrome "Graphics Mode 8" level. By offsetting the *placement* of pixels, you could accomplish new colors. The effect was used (but less noticable) in lower-res modes, which also had more color, but you could choose between "smoothing" OR "more colors".

Note that these were COMMON techniques on computers that hooked to TVs, and also to CGI monitors. It was especially useful on the Apple 2c computer which had a monochrome high res mode like the Atari 8-bit.

I see a lot of posts here confusing these old techniques with "anti-aliasing". This is NOT the technique used -- anti-aliasing simply blends the difference between 2+ contrasting pixels. These old techniques did not do that - they CALCULATED the placement of pixels to take advantage of "display artifacts" - and generate perceived resolutuins (or color range) higher than the hardware was intended to deliver.

It would be difficult for someone else to argue Apple's prior art if Apple is un-interested in the battle, so this patent is not going anywhere.

Re:Prior art

[statusbar]

Well, if AppleSoft BASIC did it with HPLOT, wouldn't that mean that Microsoft had prior art? Or did HPLOT do something different than you are describing?

--jeffk++

Re:Prior art

[Anne Thwacks]

I have just thrown away the manual for my NEC P5 printer, but I think it did the same thing too! I bought it when Windows 1.0 came out.

If no, then something close enough to make the idea clearly obvious to anyone who is not actually stupid. And there's HPs "RET" or whatever it was called to fake up 1200x1200 resolution on printers that could only actually deliver 300x300.

I should think the photo-type setter people were effectively doing this in the 60's.

Re:Prior art

[Locutus]

and now you see how Microsoft will use its "IP" to fight OSS. Just by having billions in the bank and flooding the patent office with false patents so they can threaten most OSS projects out of existence. Hey, maybe I should file a patent on THAT business model. ;-/

LoB

Re:Prior art

[Anonymous Coward]

How about the Sinclair ZX80's block graphics?

http://www.8bit-museum.de/docs/sinclair6r1.htm

That's a 1981 German version of a prior UK article. Interestingly it also claims that the PET has a similar graphics mode?

Re:Prior art

[Anonymous Coward]

I think the argument is that the improvement Cleartype provides is not a "Non-Obvious" improvement that any member of the trade could have come up with. Clear type fails on Non-obviousness, it's a improvement on older techniques. I think any programmer worth his salt who saw the crude version on the Apples or CGA displays would have gone "Hmm, I wonder if I can do a better job on modern LCDs".

Re:Prior art

[Hymer]

Hmmm... I installed OpenSuse 10.2 and the first impession was "Dear god, what have they done... those fonts sucks..." it was much worse than my SuSE 9.3 on the same laptop (A31p with 1600x1200 screen) then I changed the font from Arial (why do I have a poor MS font as the default system font ?) to Luxi Sans... and everything is smooth again...

Re:Prior art

[someone300]

I just looked at a screenshot of "Cleartype" on OS X (From http://www.microsoft.com/typography/ClearTypeFAQ.m spx). It looks revolting. I use Linux and OS X with subpixel rendering enabled. The first three lines look OK, but it starts to look blurry after that.

Screenshot of my OS X system: http://img248.imageshack.us/my.php?image=picture3r p7.png

If you zoom in, you can see I'm definitely using subpixel rendering. I get the odd blurry looking font on my OS X system, but nowhere near the sort of stuff I see on other people's Windows laptops.

Re:Prior art

[tomz16]

The text in your screenshot DOES look blurry past the first line or two... Going to the microsoft page ALL of the cleartype RGB text looks crystal clear on my Dell 2001FP
with cleartype enabled. Same on my IBM x40 laptop.

My guess is that your OSX computer is scaling the image in some weird way that doesn't quite line up with your physical LCD pixels.

-Tom

Re:Prior art

[rbanffy]

The Apple II did it in a completely different way.

When you set the high bit of a byte in the high-res screen, all 7 pixels in that byte get shifted half-pixel to the right (maybe left - it's been a long time).

It's like you had a 192x560 pixel screen but you always had to plot two pixels side by side.

Re:Prior art

[AJWM]

The Apple II didn't have a CGA. Colors were generated by dot-timing the luminance signal into an NTSC composite monitor (read, TV), faking out the color decoding in the monitor. Subpixel rendering was done by reversing that to choose the appropriate color to generate the desired dot timing. On a monochrome monitor these showed up as higher-resolution dots than the nominal pixels in display memory.

Go look at the circuit diagram for an Apple II, for pete's sake. It's not that complicated, maybe a dozen or so 74-series chips plus the memory and CPU.

Clear type uses exactly the same idea -- pick the color to activate the desired combination of R, G and/or B stripes in the LCD pixel -- i.e. activate the desired sequence of horizontal dots by color choice.

Re:Prior art

[Pharmboy]

Just by having billions in the bank and flooding the patent office with false patents so they can threaten most OSS projects out of existence.

Hey, maybe I should file a patent on THAT business model. ;-/

Sorry, Microsoft can prove prior, invalidate your patent, then reword and patent it themselves.

Re:Prior art

[mikael]

remember the same technique used on the Atari 8-bit computers, in the monochrome "Graphics Mode 8" level. By offsetting the *placement* of pixels, you could accomplish new colors.

This method of Artifacting is described in great detail in the Atari Archives.

Re:Prior art

[jZnat]

A lot of people out there (inexperienced with typography of course) prefer Arial over any other font, and it baffles me as well. I'd rather use circa-1990 X bitmap fonts than deal with Arial. Arial is a ripoff of Helvetica with a few cosmetic changes that just make it look worse so that it doesn't look good on paper (the goal of Helvetica IIRC) nor on screen (the goal of Arial in the first place).

Re:Prior art

[DaveG, the Quantum P]

I have thought the same. In fact, I've found it so blurry it's made me go cross-eyed, so I switch it off too.

Re:Prior art

[Kadin2048]

I looked at the screenshot, and both of the Microsoft samples in it look terrible. However, from what I can tell, the rest of the text on your screen looks fine. Oddly though, when I look at the same thing on my Windows machine, I don't see the same blurriness in the MS samples. Maybe whatever Microsoft is trying to do with the samples to make them look better just backfires when rendered on an OS X machine. (Maybe it's a 72dpi vs 96dpi thing?)

FWIW, I've tried the Cleartype thing on a few MS laptops and have never been very impressed with it. I still find the text on my Mac laptop, with just the regular subpixel rendering, equally if not more pleasant to read.

Slightly offtopic: It's bugged me that the average resolution of laptop displays has seemingly decreased over the past few years, at least in terms of what you can get without spending a bundle, but I guess that's just the buying habits of the unwashed masses in action -- people seem to prefer larger screens and don't care (or at least don't buy; I suspect they probably do care later on) about resolution, so the major manufacturers have cut corners on it.

Re:Prior art

[Master of Transhuman]

Of course, the answer to that is that everyone ignores Microsoft patents and LETS them try to take EVERYBODY to court - including countries that don't give a rat's ass about US patents.

Microsoft can try to be SCO and build its business on the backs of lawyers, but it's not going to work, however much Bill G might dream of it.

People forget that IBM holds more patents on everything than anyone and is making billions off Linux - as well as having the best lawyers on the planet. If Microsoft tries such a thing, IBM will hand Bill his head.

Threats are threats only. It's only when you submit that you lose.

Re:Prior art

[ncc74656]

Well, if AppleSoft BASIC did it with HPLOT, wouldn't that mean that Microsoft had prior art? Or did HPLOT do something different than you are describing?

HPLOT, as a command that took advantage of features specific to the Apple II, may have been coded by someone at Apple, not Microsoft. Apple took one of the 6502 ports of Microsoft BASIC and added support for things such as graphics and cassette I/O. The only thing I'm not 100% clear on is who wrote those system-specific bits. While Microsoft could've done the whole thing, I think it's more likely that there would've been a provision for OEMs (such as Apple) to add system-specific bits...not unlike today, where an OEM starts with a Windows install CD and then throws a bunch of drivers at a new machine.

Re:Prior art

[bhtooefr]

Apple did it, and was kicking themselves for using MS BASIC, because it was... sucky... in the extensibility respect, but they did it anyway, because they had to.

However, the point is moot - Integer had the same commands, and it was written by Woz from scratch.

And, it was a hardware "feature" that Woz had implemented in the color circuitry.

Re:Prior art

[angulion]

You surely mean OpenSUSE 10.3 alpha 3?

10.2 final is old..

Re:Prior art

[ConceptJunkie]

Reminds me of the infamous HAM mode on the Amiga implemented in Photon Paint, a copy of which I purchased almost 20 years ago.

The idea was to implement 12-bit color using only the 6 bits per pixel available. You had constraints on which colors could be adjacent to each other.

Ah, those were the days.

Re:Prior art

[VisceralLogic]

I use font smoothing on OS X, which if I recall is based on PDF display technology. It works the same on CRTs and LCDs, not being based on sub-pixel access.

MS's ClearType, on the other hand, is based on sub-pixel access. I don't like it because it gives the letters funky tinges of color (at least on my Dell LCD), which I guess is due to each sub-pixel being a particular color (red, green, or blue), not the full spectrum mix available to an entire pixel.

The result is that font smoothing on OS X does not affect the color, so I leave it on. ClearType does affect the color, which I do not appreciate, so I leave it off.

Re:Prior art

[statusbar]

No, integer basic did not have hi-res mode:

    http://www.landsnail.com/a2ref2.htm

However I'm not certain that the apple II hires mode is even relevant to this "Intellectual Property" problem.

--jeffk++

Re:Prior art

[TheRaven64]

OS X does do sub-pixel AA by default. If you don't believe me, turn on control for zooming, hold down control and use the mouse wheel to zoom in to some text. You will see that your 'black' text is actually multicoloured. The thing that makes text (and vector art) look nice on OS X is a little algorithm that nudges lines that are close to a pixel boundary onto the boundary.

Re:Prior art

[VisceralLogic]

You're right... I was wrong. I've had font smoothing on ever since QuickDrawGL on System 7, and never noticed a change moving to an LCD screen.

It makes me wonder why MS even bothered patenting their crap, since it looks so horrible compared to OS X.

Re:Prior art

[Anonymous Coward]

Most people also don't properly configure cleartype. It is very dependent on the orientation of the lcd panel and how much imphasis is placed on this difference.

Re:Prior art

[bhtooefr]

Quoted from Apple II History Chapter 3:

On the motherboard were two empty sockets that were available for the user to plug in their own ROM chips. The $D000-$D7FF space was most often used by a plug-in ROM chip sold by Apple, known as "Programmer's Aid #1." It contained various utilities for Integer BASIC programmers, including machine language routines to do the following: ...

• Handle hi-res graphics from BASIC, including code to clear the hi-res screen, set colors, plot points and lines, draw shapes and load shapes from tape.

True, that isn't PART of Integer BASIC.

But... from Chapter 16:

To fully appreciate this narrative, you need to know a little about an old Integer BASIC program, APPLEVISION. This was found on the DOS 3.2.1 System Master disk, and was a fun little display that showed off the use of hi-res graphics. It began by creating a simple line drawing of a room, with a picture on the wall ("HOME SWEET HOME") and a television set. On the screen of the TV appeared a man who danced to the tune of "Turkey In The Straw", which sounded on the built-in speaker. It ran repeatedly, until the user interrupted the program. It was fascinating at the time, since there was nothing in the program text that showed off exactly how the hi-res effects were accomplished. (my emphasis)

Re:Prior art

[statusbar]

Right, I had an apple ][ before there was DOS: newspaper article

And at that point, Microsoft's AppleSoft had hires graphics and integer basic did not. So AppleSoft had prior art, at least with regards to this.

One could always use pokes to change mode and fill memory, and the applevision program worked by poking a 6502 program into memory which performed the animation and music...

--jeffk++

Re:Prior art

[Hooded One]

Sounds like your display has one of the less common subpixel arrangements. Most displays (I believe) line them up as RGB from left to right, but it's possible to have the order reversed, or to have either order from top to bottom. Try playing around with the ClearType Tuner (the web version requires IE, or there's a downloadable version that installs in Control Panel) and see if one of the other configurations helps. It also works better if your display is set to its native resolution, but you probably already knew that.

Re:Prior art

[Hooded One]

Ok, looking at the ClearType FAQ, it seems Windows doesn't support the vertical arrangements, but they're probably even rarer. Hopefully you don't have one of those.

Re:Prior art

[runningduck]

Very odd. I checked out the image on both my Linux and Mac computers and the Mac displays the center two lines blurry in both Safari and Firefox. Safari and Firefox both display the image fine outside of the web page. http://img.microsoft.com/typography/faq/brg.gif It look fine when I display my Linux Firefox on my Mac screen via X.

There seems to be some issue interpreting the image as either 444x210 or 444x211. The web page specifies the image's demensions as 444x210 but the actual dimensions are 444x211.

So is the Mac the only system that is rightfully distorting the image as the HTML dictates?

Re:Prior art

[bhtooefr]

Well, it's all a moot point anyway as to which versions of BASIC had the hires manipulation commands - the implementation of subpixel rendering is in the hardware. :)

Re:Prior art

[someone300]

You have it right, I believe. Maybe it's something to do with the APIs for resizing images on the different systems. Different resampling algorithms can cause strange things like this.

Re:Prior art

[SenorCitizen]

Vertical arrangements are rare, until you rotate your display by 90 degrees and XP's ClearType goes b0rk. I hear Vista also supports vertical pixel arrangements, but as is typical, they won't be available in XP. But in any case, ClearType is *the* feature that made me upgrade from Windows 2000. On an LCD, subpixel antialiasing is the only way to go, IMO.

Novell is the Judas Goat.

[140Mandak262Jamuna]

I think Novell has become an wholly owned subsidiary of MSFT and is being used for the express purpose of setting up precedents and creating more and more FUD. I have seen a version of anti-aliasing and sub-pixel addressing way back when in, of all places, grc.com.

Re:Novell is the Judas Goat.

[Woy]

Novel became the Mr. Hands of the Linux world.

Re:Novell is the Judas Goat.

[Colin Smith]

Judas was a goat?

Anyway. They are very helpfully pointing out the patents which Microsoft says apply to Linux...

Re:Novell is the Judas Goat.

[duncanmhor]

Judas Goat - used at an abbatoir to lull animals into a false sense of security.

Re:Novell is the Judas Goat.

[elrous0]

See, I'm always telling people I learn something new every day on /.

Re:Novell is the Judas Goat.

[arivanov]

The apple version is forced by the sillies of their videoram technical implementation. It is also not the earliest prior art.

There is a much older prior art, more specifically the sub-pixel version of bresenheim algorithm described in "Fundamentals of Interactive Computer Graphics" which is general, with full mathematical description to accompany it and predates Apple 2. IIRC (I do not have the book in my new house) the book explicitly mentions it as related to fonts and describes subpixel font rendering.

Re:Novell is the Judas Goat.

[Anonymous Coward]

see Terry Pratchett - Feet of Clay

Re:Novell is the Judas Goat.

[krakass]

Next you're going to tell me Saint Peter was really a rabbit and that the pope isn't supposed to be a man.

Re:Novell is the Judas Goat.

[couchslug]

"I think SCOvell has become an wholly owned subsidiary of MSFT and is being used for the express purpose of setting up precedents and creating more and more FUD."

Fixed it for ya.

Re:Novell is the Judas Goat.

[mangu]

Next you're going to tell me Saint Peter was really a rabbit and that the pope isn't supposed to be a man.

Well, about this last one there have been some controversies

Prior art?

[140Mandak262Jamuna]

I have not been to GRC.com for a long time, I quickly grabbed the URL and posted it here in another thread. Looks like that site cites a long list of prior art. Makes the OpenSUSE's decision even more suspect.

Re:Prior art?

[Anonymous Coward]

Not only that, linux/X.org subpixel rendering works in a somewhat different, rather more general way to ClearType. Any geometric object can be subpixel rendered and antialiased, whereas microsoft's method implements separately for each graphics/font object kind. EVEN IF microsoft patents were to hold up in court, there'd be a good case wouldn't cover the technique used in linux/X.org. I think this is indeed an attempt to sow misleading precedent by microsoft - Novell AND openSuse should be considered corrupt and abandoned.

Well, that's it then.

[Anonymous Coward]

No more Suse Linux on my servers. I know that subpixel rendering has no impact on server applications, but I now consider that distribution rogue.

Re:Well, that's it then.

[Delkster]

Out of curiousity, do other major distributions enable this either? In other words, is this news at all?

A page on the FreeType project site says:

Finally, many Linux distributions seem to distribute a patched version of FreeType 2 with the bytecode interpreter activated, unlike to the sources we distribute.

However, I've previously been under the impression that most distributions would ship at least without some features covered by patents. On the other hand, it's not only MS who owns patents that concern subpixel rendering, and I don't know who owns what, so that's why I'm left wondering if someone else actually knows.

Re:Well, that's it then.

[jZnat]

Ubuntu does enable it in libfreetype2, but you need to enable subpixel rendering in the options. Subpixel font hinting doesn't work on CRTs (you get a rainbow effect of sorts), so it's not the best thing to assume. Although, it might be able to detect if you're on an LCD properly and enable it.

It does enable normal antialiasing by default, however.

Re:Well, that's it then.

[deanoaz]

>>> 'Yes, firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.'

What would firefox be showing you, that was worth looking at, if there were no women?

Re:Well, that's it then.

[Master of Transhuman]

If you read TFA, you'll find it's more likely that the openSUSE developers are morons than they are "rogue". The article points out that the Novell deal SHOULD cover this situation but apparently the openSUSE people don't understand that.

It's also pointed out that Red Hat Fedora does more or less the same thing, as quoted below:

"Let's see what Red Hat does when comes to FreeType2 and patent concerns, by looking in the freetype.spec defaults:

RHEL5 (freetype-2.2.1-16.el5.src.rpm) and FC6 (freetype-2.2.1-10.fc6.src.rpm) only take care to disable the Apple-patented BCI:
# Disables patented bytecode interpreter. Setting to 0 enables
# the bytecode interpreter.
%define without_bytecode_interpreter 1

Fedora 7 however, just like openSUSE 10.2, also disables the Microsoft-patented subpixel hinting:
# Disable patented bytecode interpreter and patented subpixel rendering.
# Setting to 0 enables them.
%define without_bytecode_interpreter 1
%define without_subpixel_rendering 1

While it's not clear to me why Red Hat's Enterprise Linux 5, recently released, still doesn't care about some Microsoft patents (usually, Red Hat is very cautious when comes to patents), it's also very curious that the patent covenant between Novell and Microsoft, which is supposed to cover openSUSE's possible patent infringements, is still ignored by the openSUSE team, even in cases where it could have been useful to them!

Re:Well, that's it then.

[laughing rabbit]

Hear, hear!

Re:Well, that's it then.

[wolverian]

Ubuntu and Debian indeed to enable the bytecode interpreter, choosing to ignore patents that have never been enforced.

Re:Well, that's it then.

[Champ]

My Ubuntu experience is to the contrary. It does subpixel rendering, but it does it poorly and without the bytecode hinting interpreter. Installing patched font rendering packages improves the situation dramatically, but that's not part of the default install.

Now it is clear

[javilon]

Novell is the new SCO

Re:Now it is clear

[TangoCharlie]

Your off-the-cuff remark is more true than you think. Don't forget that SCO (that is Caldera)
got the unix rights from Novell. The whole SCO vs Novell issue will result in Novell getting
those rights back. The result? A linux company "owning" unix. We've been here before!! Novell
OpenSCO here we come....

Re:Now it is clear

[A_Non_Moose]

Novell is the new SCO

Has netcraft confirmed this?

Re:Now it is clear

[toriver]

Don't forget that SCO (that is Caldera) got the unix rights from Novell.

Yes, but doesn't that make Novell the old SCO? *frowns*

anti-aliasing makes me need glasses

[stokessd]

That might be a good thing(tm). In many cases I prefer non anti-aliased fonts. I have a nice LCD with a DVI connection for a clear picture, then I'm supposed to fuzzy it up? Anti-aliasing lakes me think I need glasses in many cases.

Sheldon

Re:anti-aliasing makes me need glasses

[daeg]

That's why you can turn it on and off. :-) I know a friend that has sensitive vision can't stand it, either, he says all the letters have a blue halo.

Re:anti-aliasing makes me need glasses

[140Mandak262Jamuna]

I find that cleartype is easy on my eyes while browsing websites etc with lots of text meant to be read by humans. Code, OTOH, looks horrible in cleartype. When I have to tell, single quote from double, where braces are very important, where I have to tell zero from o, two from zee, ell from one, bah... ClearType makes a mess.

Re:anti-aliasing makes me need glasses

[Anonymous Coward]

It's zed, not zee, in English.

Re:anti-aliasing makes me need glasses

[ericrost]

only if you're from Canada, eh.

Re:anti-aliasing makes me need glasses

[xoyoyo]

Depends on the quality of the implementation and the quality/density of the screen. Ironically, the inventor's implementation is poor.

On a 72dpi LCD attached to a PC running Windows the effect is obvious (and hideous) all the glyphs have red and blue fringes. Turning ClearType off is the first thing I do on a Windows box after disabling the Windows XP theme.

On my 100dpi+ MacBook Pro I had to use the zoom function to confirm that it was using sub-pixel anti-aliasing. Even on my second monitor it's acceptable, and that's a cheap low density screen.

Apple have spent some time getting font anti-aliasing right: the initial AA in OS X looked like someone had just applied gaussian blur to the whole screen. Now it actually does what it's supposed to do, which is reduce eye fatigue.

On the other hand, once we get our long promised 300dpi screens monitor resolution will be the same as paper and we can dump kludgy hacks like ClearType.

Re:anti-aliasing makes me need glasses

[Anonymous Coward]

may be it is a zed in English, but it is zee in American.

Re:anti-aliasing makes me need glasses

[KiloByte]

Depends on the font. With the older MS fonts, ClearType looks outright ugly and unreadable. But if you use their newest set, ClearType is pretty much a must -- and it kicks ass. On LCD, that is. The regular gray-scale AA done by Microsoft doesn't work right -- but with TrueType, the new MS fonts look great both on LCD (sub-pixel AA) and on CRT (gray-scale AA).

I admit to them: Consolas really pwns Bitstream/DejaVu Mono. Just don't try it without working antialiasing. With MS Courier/MS Lucida, forget about AA.

Re:anti-aliasing makes me need glasses

[Anonymous Coward]

I know Americans are bad at geography, but there are more English speaking countries than just Canada and the United States of Where's-That?

Re:anti-aliasing makes me need glasses

[jimstapleton]

I think a lot of it is monitor quality too..

My old Hitachi was a nice LCD in regards to image quality, and it looked great with clear type.
However the backlight died (and the response time was a bit low), and now I have a cheap Samsung - The letters have halos on them with clear type.

So, monitor quality is a big part of it, not just the rendering technology, though both are important.

Re:anti-aliasing makes me need glasses

[CastrTroy]

More like, unless you live in the United States.

Re:anti-aliasing makes me need glasses

[Anonymous Coward]

If you haven't realized the legibility improvement that anti-aliasing provides by now, then yes, you do need glasses.

Re:anti-aliasing makes me need glasses

[mattr]

IIRC different displays may have different order of R,G,B component pixels which may require a reversed antialiasing pattern (as if the screen was flipped upside-down). Though the effect is subtle it also shows a red and/or blue fringe. Though that may not be what you are talking about.

Re:anti-aliasing makes me need glasses

[Anonymous Coward]

Perhaps if people bothered to tune it for their specific display, it might not look so dodgy? Microsoft provide a tool to do so...

Re:anti-aliasing makes me need glasses

[RalphTheWonderLlama]

Yeah, I HATE clear type. I want my text to be razor sharp, not fuzzy. The damn thing keeps turning back on by itself too!

Re:anti-aliasing makes me need glasses

[mashade]

Who's Zed?

Zed's dead, baby. Zed's dead.

[long live zee]

Re:anti-aliasing makes me need glasses

[jonsmirl]

If the color fringing is obvious you may have a BRG screen instead of a RGB screen. There is a control panel setting to change this. You may even have a screen with vertical RGB/BGR stripes instead of horizontal.

Click on the microscope photos at the bottom of the Wikipedia article. They'll show you what is going on.

Re:anti-aliasing makes me need glasses

[xoyoyo]

Nope, I know the screen is horizontal RGB. It's just that ClearType sucks and I notice these things. No matter how good the monitor ClearType always looks to me like you're looking at the screen through a cheap pair of binoculars. Fiddling around with the ClearType Tuner improves matters somewhat but only near the limit where ClearType is basically turned off.

There are three parameters at work here: one is the preference of the user (and my preference, as a result of my years of work on print I suspect, is for crisp graphics with minimal anti-aliasing); one is the implementation of the anti-alising scheme (and Microsoft's sucks and has always sucked ever since it appeared in PocketPC, which is where I first came across it); and three is the quality of the monitor. The only one you can really do anything about is the quality of your monitor.

Anti-aliasing began as a technology to avoid artefacts in rendering especially on low resolution displays such as the CRTs available when the technology was first developed (early 70s). Now that we are seeing TFT displays with resolutions higher than 110 ppi AA becomes less significant. The first laser printers with an acceptable print quality had a resolution of 300dpi. Once we get up to that we won't need Microsoft's moire fringes anymore.

Re:anti-aliasing makes me need glasses

[jonsmirl]

Sub-pixel rendering only works on LCDs. There is no way to maintain registration between the scanning beam and the shadow mask on CRTs so it can't be used there. Full pixel based anti-aliasing does work on CRTs.

I use sub-pixel rendering under Linux. My screen is about 90 DPI. I need a magnifying glass to see it working or I have to be six inches from the screen. Maybe the Linux version isn't as obnoxious. I took the wikipedia pictures with a 60x microscope.

Re:anti-aliasing makes me need glasses

[dramenbejs]

Yes, I still remember my friends story:

He and his girlfriend was in USA on holidays and knowing nothing about american laws, he was hugging and kissing his girl in public.
Than a policeman came and a portion of their talk went like this:

P: Where are you from? (He asked when he heard their accent)
F: From Prague.
P: What?
F: Prague, Czech republic.
P: What?
F: Um.... Its in Europe.
P: What?

F: Eh?!?!?!

So my friend thinked a bit and than said (with gesticulation):
Ohio, Pennsilvania, New York, water, water, water, water, water, water, EUROPE!

Re:anti-aliasing makes me need glasses

[xoyoyo]

I wasn't claiming that sub pixel works on CRTs, just that anti-aliasing in general was designed to deal with rendering on low resolution devices (actually just on lower resolution devices than the source material, but I'm trying not to get distracted). And ditto, sub-pixel on my mac is good enough that I have to use hardware zoom to see it. On PCs it's lousy, which is the point I was trying to make in the first place.

Re:anti-aliasing makes me need glasses

[Anonymous Coward]

While technically correct, he probably should have put New Jersey after Pennsylvania (btw, not Pennsilvania). I mean if you're going to talk shit, at least spell it right.

Re:anti-aliasing makes me need glasses

[shutdown -p now]

It largely depends on the fonts. All monospace fonts normally seen on Linux distros look pretty bad when anti-aliased, in my opinion (yes, this includes Vera & DejaVu), so there I highly prefer Terminus. On Windows, though, the new Consolas looks excellent, and is overall the best monospaced font for coding, in my opinion, with deliberate and very clear distinctions between "0" and "O", "1" and "l" etc.

Re:anti-aliasing makes me need glasses

[dramenbejs]

Pardon the typo. I have never been in any english speaking country (including USA). I am writing OK, don't I?

You mean you can spell correctly every state in former Soviet Union, in azbuka of course? :-D (to draw a parable). You would make a fine exception :).

I am sorry to hurt your feelings, but don't you think that some of the cops are just too stupid?

PS: Is it possible to write an azbuka to slashdot post? How?

It's FreeType for a start!

[DrMindWarp]

This is complete nonsense written by someone that is clearly clueless and forwarded by an editor that is equally clueless. This is a FreeType library setting for compiling programs (not ClearType!). It is the same for every Linux distribution as it is the default setting for the development library. It has never been enabled by default.

Re:It's FreeType for a start!

[segedunum]

That doesn't answer why it was OK to have this enabled before, and has then somehow become a big no-no.

Re:It's FreeType for a start!

[Movi]

Yes, indeed Cleartype sucks, and ive not even known you can have it under linux. For an ever better font setup you can enable BCI in freetype and have freetype display font quality on par with Mac OS X (which nobody can dispute displays the best quality). For example ubuntu people can download debs with prepatched freetype here http://ubuntuforums.org/showthread.php?t=343670&hi ghlight=feisty+fonts.
The standalone patches are here http://david.freetype.org/lcd/

Of Course it isn't covered...

[Darundal]

...which looks worse? The free OS not having the feature enabled by default due to an admission that it is Microsoft patented tech, or having it covered in the agreement so that nobody sees that there "is" Microsoft patented tech in linux. Guess which one 'ol Bally just loves.

Suse vs Open Suse

[Anonymous Coward]

I put a store-bought version of Suse on my daughter's machine and everything was good. We upgraded her mobo and downloaded the 64 bit version of Open Suse. There are a myriad of niggling little details that don't quite work the same. The commercial version of Suse was a joy. The other one isn't. We're switching to Ubuntu.

Re:Suse vs Open Suse

[PrescriptionWarning]

I'm heading in the same direction towards Ubuntu, away from openSuse.

I've installed openSuse 10.2 on no fewer than 3 machines I use, and although for the most part they each work fine for my day to day activities, each one of them can no longer open yast2 unless accessed via command prompt as root, the auto-update is broken on each, and since I use Gnome, the VNC server for each is completely fubar. I'm now just waiting for the next version of Ubuntu to be completed here in the next couple of weeks.

Re:Suse vs Open Suse

[MollyB]

I just last week switched from SuSE 9.1 to Ubuntu 6.06 (Dapper Drake) on my laptop. The old OS always crashed or froze, and was a tempermental beast that chronically corrupted the Reiser FS. The Ubuntu install was less than 20 minutes (although downloading of package upgrades took an hour on DSL) and has been running superbly 24/7 since boottime. I hope you are happy making the change; I most certainly am!

BTW, if you are a Windows person who is looking for a friendly Linux distro, this is for you. You can run Ubuntu from the CD to try it out before installing, and it is the very epitome of "user-friendly." You don't even have to edit config files, if that seems daunting. Take the plunge--you won't regret it.

Re:Suse vs Open Suse

[beef623]

Heh, I went the other way not too long ago. I started with Ubuntu, but the more familiar I got with linux, the more I hated Ubuntu. I finally switched to openSuse a few months ago(mainly because it was the only distro I could get running on my laptop at the time) and fell in love with it. I finally completely shed myself of Ubuntu when the box I had at home wouldn't let me even log in anymore. I haven't really looked back.

I will say that I don't like a lot of the defaults in 10.2, especially the main menu it has by default. Hadn't really noticed the font thing though.

Re:Suse vs Open Suse

[Master of Transhuman]

When you get there, you will let us know about all the niggling problems with Ubuntu, right?

Trust me - there will be some. I run Kubuntu and run into them occasionally. Just the poor separation of root and normal user will have you irritated daily. After forgetting to key in "sudo" before every command when you want to do things as root for the millionth time, you'll be doing "su -" like me, if not actually logging on as root (temporarily anyway).

Novell has more manpower than Ubuntu to test their releases. You'll notice that, too. Although the reason I didn't switch to SUSE last fall was because they screwed up their software update process due to poor testing.

Re:Suse vs Open Suse

[Anonymous Coward]

Moving to a 64-bit OS was your biggest mistake.

Exactly

[Anonymous Coward]

Cleartype is just sub-pixel AA which existed long before MS ever used it for font rendering. Bytecode type hinting is patented by (IIRC) Apple, it is usually disabled in Freetype and and an alternative (auto-hinting) method used instead.

Apples and oranges, the bug reporter is confused or trolling.

Re:Exactly

[Anonymous Coward]

First of all, nobody said it's ClearType, except for the moron who inserted into the FreeType2 code the following comment: "Uncomment the line below if you want to activate sub-pixel rendering (a.k.a. LCD rendering, or ClearType) in this build of the library. Note that this feature is covered by several Microsoft patents and should not be activated in any default build of the library."

Secondly, BCI is something not enabled by default. The sub-pixel hinting is (was) always available by default in all the distros, you could use it from the KDE Control Center, from the GNOME Control Center or using font.conf settings.

Someone is so hasty to comment on Slashdot, that he didn't bother to read it carefully.

Well, democracy à la Slashdot is usually the power of the mob.

Re:Exactly

[Anonymous Coward]

So nobody said it was sub pixel rendering except that is exactly what is alleged in the bug report that was linked to from the article?

Plus I said that BCI was disabled by default in freetype and that auto-hinting was used instead.

Someone is so hasty to comment on Slashdot, that he didn't bother to read it carefully.

So was this an attempt to berate this humble AC or an admission of guilt?

Re:Exactly

[WhiteWolf666]

Please read the bug report.

"KDE ignored the setting for sub-pixel anti-aliasing, it seems to be permanently off".

This isn't about the BCI. This is about sub-pixel font anti-aliasing.

Frankly, this hobbles Linux compared to Windows/OS X.

Re:Exactly

[Anonymous Coward]

Second AC:

So nobody said it was sub pixel rendering

WTF? No-one said any such thing:

First AC:

First of all, nobody said it's ClearType

Note, no mention of sub-pixel rendering.
You do know that ClearType and sub-pixel rendering aren't the same thing, right? Sub-pixel rendering is a general concept, ClearType is the specific name of Microsoft's implementation.

Re:Exactly

[djurban]

ClearType is more than just enabling subpixel anti-aliasing.

Its whole point is the wizard that allows the user to choose not only between whether he has BGR or RGB device (first step) but also checks the gamma correction in the second step - thats why most user se the big difference in before and after. There is more than one thing involved there. Actually it is a bit more complicated.

A good read would be: http://research.microsoft.com/~jplatt/cleartype/si d2000.pdf

Re:Exactly

[Anonymous Coward]

You do know that ClearType and sub-pixel rendering aren't the same thing, right? Sub-pixel rendering is a general concept, ClearType is the specific name of Microsoft's implementation.

So cleartype is sub-pixel rendering and sub-pixel rendering is not sub-pixel rendering? The Comment was made in the context of the discussion about FT_CONFIG_OPTION_SUBPIXEL_RENDERING as per the linked bug report. Specifically the only thing we are interested in about cleartype is Microsoft RGB sub-pixel rendering (as covered by the cleartype patent). When he says cleartype, I read sub-pixel rendering because that's what this discussion is about.

I strongly suspect you as your "First AC".

Get it were M$ got it

[Anonymous Coward]

As far as I know Microsoft licensed Acorn Risc Os technology for several font description and rendering features from Acorn when that company still existed. Acorn's Risc Os including the font technology is now owned by Pace Ltd and/or Risc Os Ltd. If the disputed technology is related to those licenses then there's a way to overcome this issue.

The Acorn Risc Os technology developed by Sophie Wilson was building on a Swiss university project on font rendering: scaffolds, hinting, sub pixel anti-aliasing etc, 1983 or so. Acorn already had done work on font rendering before Risc Os appeared.

Ernst

It's only the filtering

[oergiR]

AFAICT, subpixel rendering is not disabled, only the 5-tap filter that's supposed to reduce colour fringes. See http://www.grc.com/cttech.htm. Apparently this is one of the things Microsoft has patented, and I haven't seen any "prior art" for this specific technique. In my humble opinion disabling the filter is not much of a loss as it just makes fonts look fuzzier.

Re:It's only the filtering

[b0z0n3]

So, Microsoft has patented blurring text?

I don't want to pay M$ everytime I have a couple of beers....

Re:It's only the filtering

[strredwolf]

No, that dates back to the 1800's, so there's prior art there.

Hidden warning

[b1ufox]

This as it seems, is yet another legal puns MS has up its sleeves.MS struck a deal with Novell months back, which obviously created a fury among free software zealots.Now this seems to be a Red signal for Linux users, who uses OpenSuse or any other free Linux distribution, as it implies IMO _you_ being a non SUSE(and means even OpenSUSE i guess) users are infringing on MS's so called intellectual property.

Is this the start of the hide and seek of infringement legalities?

Lets hope SUSE understand this can be just the beginning. Novell people should put in some thinking into not getting pawned once again by MS.

Whatever i am better off without them on my Edgy Ubuntu machine. :)

~psr

Freetype library is GPL

[mysidia]

As a result, if you hold a license for a patent that is required to redistribute/sell Freetype (or any piece of software covered by the GPL), then, to comply with the GPL you have two options you must EITHER: (1) not distribute the software, OR (2) the patent license must permit anyone's free use

The relevant GPL section is the preamble To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all. , and under Section 7 of the GNU General Public License: 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.

This means for instance, that Novell would not be free to provide users of SuSE the benefit of a patent license to use a certain feature of a GPL'ed library or software program, and deny that feature to openSuSE users.

Re:Freetype library is GPL

[mwvdlee]

According to the Microsoft point-of-view (which Novell is following), the specific code contains patents and the person who released this code with a GPL license had no right to do so. This person would not have been allowed by the patent holder to release his code in GPL.

Entertaining the possibility that Microsoft's patent is indeed valid and without prior art, the code should no longer be considered GPL licensed and thus only those entities with a valid license would be allowed to distribute it.

So from Microsoft/Novell's point of view, which assumes the patent claim is correct, it makes sense.

Re:Freetype library is GPL

[Anonymous Coward]

Entertaining the possibility that Microsoft's patent is indeed valid and without prior art, the code should no longer be considered GPL licensed and thus only those entities with a valid license would be allowed to distribute it. If the code should no longer be considered GPL licensed, how can SuSE distribute it at all? Only acceptance of the GPL allows redistribution.

Re:Freetype library is GPL

[ThePilgrim]

What happens if I add this code in the UK, where s/w patents are invalid?

Does that mean I can't distribut my code under the GPL, or do I have to put in geographical if statments.

Why does this remind me of the 32/128 bit encription fieasco all those years ago?

Re:Freetype library is GPL

[NullProg]

Entertaining the possibility that Microsoft's patent is indeed valid and without prior art, the code should no longer be considered GPL licensed and thus only those entities with a valid license would be allowed to distribute it.

IANAL but view this article: http://www.cnn.com/TECH/computing/9812/08/cleartyp e.idg/

SuSE Linux has never distributed Cleartype fonts, its always been a separate download.

Enjoy,

Re:Freetype library is GPL

[squiggleslash]

As long as you don't have a patent agreement that would restrict downstream licensees, you can license under the GPL. The GPL's anti-patent language really only kicks in if you have some agreement with some third party which means that the rights granted by the GPL wouldn't apply to people you distribute the code to. It is not your job to enforce patents (or other effects of foreign laws) in the code you distribute.

Not only do you have no such agreement, but it's hard to see how you could even be party to such an agreement given software patents do not apply to you.

For someone importing your code to a country where software patents apply, the onus is on the patent holder to find them and challenge their use of the code.

Re:Freetype library is GPL

[ThePilgrim]

So in this case I could happly distribut the code with the flag switched on for the M/S patented code and it does not become a problem untill someone tries to import it into the US.

Humm... Have I just descovered a way for Novell to get MS patented code into the Open Source code base?

Re:Freetype library is GPL

[mysidia]

If the code were no longer distributable under the GPL, then Novel would have no license, and it would be copyright infringement to attempt to distribute the library.

Typical 'Bend Over' Novell

[segedunum]

There's obviously quite a bit of prior art to Cleartype, but Novell as an open source company does not want to stand up and defend itself and its software from it (as well as Red Hat actually). I rather suspect both Red Hat and especially Novell are using the non-issue of patents to try and give their so called enterprise distributions an actual selling point.

The question really is, why was it deemed OK to enable it before, and suddenly it has become a big deal where it is disabled?

Additionally, there seems to be some confusion of the Microsoft/Novell deal. The patent agreement would not be legal with the terms of the GPL, rather Microsoft gave a covenant not to sue to Novell's customers and promised to be nice to OpenSuse's users. Whether that would cover this, I don't know.

Same with fedora

[Anonymous Coward]

Fedora disables the truetype bytecode interpreter and subpixel hinting features of the freetype library. There is an "autohinting" system used instead, but in my experience it looks much worse, compared to recompiling freetype with the patented features enabled.

Can we now use the GPL?

[marcosdumay]

The GPL is very clear on one point: if you know your software infringes on some patent, you can't distribute it, even if you have a deal with the patent holder enabling you to do that*. Can Novell now be prosecuted? Is that code GPLed (it seems to be KDE, so it probably is)?

* Unless that deal is extended to everybody that touches the code.

Re:Can we now use the GPL?

[Kjella]

The GPL is very clear on one point: if you know your software infringes on some patent, you can't distribute it (...) * Unless that deal is extended to everybody that touches the code.

Sorry, but it just doesn't. It says that if you have a patent or patent license, you can't distribute it unless that deal is extended to everybody that touches the code. That's exactly what the GPLv3 and MS/Novell deal is about, a "patent indemnification" which acts, talks and walks like a patent license but in legal terms isn't. In any case, better to let the courts handle it than self-censorship. With all the fuzzy and invalid patents around almost nobddy would get anything distributed.

Re:Can we now use the GPL?

[pak9rabid]

Can Novell now be prosecuted?

prosecuted implies criminal law. This, as well as all other issues dealing with licensing issues, all fall under civil law. Therefore no one is prosecuted, but rather sued by the license holder

never so

[Deternal]

As far as I can read, it has never been enabled. It needs to be enabled at compile time, which the ansvar to the linked bug report clearly states by c&p of the relevant info from the FreeType lib.

This is a complete non-issue and has been known for a while. It predates the Novell/MS agreement.

Re:never so

[oergiR]

Mod parent up.

The main developer of FreeType decided to disable the filter in September. The Novell deal was later and had nothing to do with this.

Re:never so

[MMC Monster]

The question is, with the Novell/MSFT deal, can SuSE (not openSuSE) now enable it at compile time?

Re:never so

[Anonymous Coward]

That makes no sense. FreeType developers are good, whereas Novell/SUSE are bad. How could this decision be made by FreeType developers then? There must be another explanation.

Re:never so

[SillyNickName4me]

The question is, with the Novell/MSFT deal, can SuSE (not openSuSE) now enable it at compile time?

Short answer, no.

Long answer:

When they do enable it at compile-time and distribute the result, then they are still bound by the GPL which means that anyone receiving those binaries has the right of getting the source code and distributing both binaries and source (including the patented technology in compiled form). While I am not familiar with all the details of this deal, I do believe that this is not the case. Since that would mean additional restrictions on top of the GPL, this would not be legal for Novell to do.

Re:never so

[Anonymous Coward]

No, it has been enabled by default for many years in the freetype library, and in fact there wasn't even a compile-time option to disable it.

However, about six months ago, a compile-time option was added to control whether sub-pixel rendering is included, and it was set to be disabled by default. It isn't so much a matter of this distribution specifically disabling it, but other distributions either explicitly enabling it or still using older versions of the library.

Re:never so

[MMC Monster]

If they distribute unaltered binaries with source, that is no different than Novell distributing them (think: torrents). If they make modifications of the source or binaries, they can't distribute them anyway, as that would infringe on Novell's trademarks (think: the recent issues with various linux distributions distributing modified versions of Firefox without changing the name/icons).

Or is it now illegal to download SuSE via torrent? (I really don't know. Is it available for purchase, only?)

Re:never so

[Anonymous Coward]

On the contrary.
The Novell deal was _announced_ later and may have something to do with this.

Re:never so

[SillyNickName4me]

If they distribute unaltered binaries with source, that is no different than Novell distributing them (think: torrents).

For all I know this is correct.

If they make modifications of the source or binaries, they can't distribute them anyway, as that would infringe on Novell's trademarks (think: the recent issues with various linux distributions distributing modified versions of Firefox without changing the name/icons).

You cannot use the name or logo because it is no longer what the name and logo stand for. You can however distribute a renamed and modified version without any additional restrictions.

Or is it now illegal to download SuSE via torrent? (I really don't know. Is it available for purchase, only?)

They cannot both comply with the GPL and forbid you from redistributing it. They can however stop you from modifying it and representing it as the unmodified version.

The issue you are pointing at is that neither the name or the logo of firefox is covered by the GPL (same with for example Redhat, or even Debian). Possible cleartype code integrated or linked to a GPLed program however is covered by the GPL.

Re:never so

[novus ordo]

I think the main issue here is that the "patent covenant" they signed up on is useless. At least from a practical standpoint.

I have a question...

[purpleraison]

I am not familiar with the history of the clear type patents etc., but I do know that no version of Windows has ever had font-smoothing until Vista rolled around. Whereas Linux and Macintosh has used smoothed pixels for as long as I can remember.

Is this clear type, and if so -- What has changed?

Re:I have a question...

[mwvdlee]

Subpixel font smoothing has been in WinXP since the start.

Re:I have a question...

[TheNetAvenger]

I am not familiar with the history of the clear type patents etc., but I do know that no version of Windows has ever had font-smoothing until Vista rolled around.

Win 3.1 had 256 Palette Anti-Aliased fonts in 1992, provided by drivers from ATI and MS. (Basically as long as TrueType has been in Windows, Anti-Aliasing has been available.)

Win95/Win98 - Via IE4/Plus Pack and on Win98, Windows9x also had anti-aliased font technology. The 'Font Smoothing' setting was not enabled by default, so a lot of idiots never thought it existed.

Win2k - Also had Font Smoothing (Anti-Aliasing), even though again it was not enabled by default.

WinCE was the first OS to ship with ClearType (The MS PDA OS)

WinXP - Shipped with 'Font Smoothing' Enabled with 'ClearType' available as an option that users could turn on. PS ClearType is more than Color based Font AA, as the patents on it are dealing with font hinting filters that keep the text from having bleeding colors or blurry edges. Hence why fonts on different multiple color background properly AA with no color halos or greyscaling.

Vista - Ships with ClearType Enabled by Default. The Vista Font Engine is now 3D GPU assisted, but produces basically the same ClearType results as XP. WPF applications in Vista do get slightly newer and higher quality ClearType rendering.
(Vista drivers are inherently smarter so that non-typical LCD pixel order and CRTs automatically use the best setting for the Monitor, where in XP, the user had to 'Fine Tune' this setting themselves.)

Whereas Linux and Macintosh has used smoothed pixels for as long as I can remember.

Is this clear type, and if so -- What has changed?


Um, Macs didn't get native Font Anti-Aliasing until System 8.5, released in 1998. (Several years after it was available in Windows) There were a few third party utilities that enabled Anti-Aliasing on Macs prior to this, but they were not universally used or supported throughout the OS and applications, where on Windows, all TrueType fonts were anti-aliased no matter what application or part of the OS used them. (An example of one these Mac utilities was "SmoothType")

As for Linux, Font Anti-Aliasing has not always been available either, especially before it was available on Windows, since Linux did not yet 'fully' exist.

So I have no idea why you think Windows never had AA Fonts, but since I have been using it since 1992, I can pretty much assure you that you are very mistaken.

Re:I have a question...

[Anonymous Coward]

Yup, I agree with TheNetAvenger...
          I can't speak for Win 3.1 w/ ATI drivers, because I did not have an ATI card back then. But Win95+ had some type of antialiased fonts.

          Also, Linux certainly didn't have nice fonts all along. I've used Linux since 1994 (starting with Slackware). Distros of that era didn't support truetype at all -- it used a bunch of 75 and 100DPI bitmapped fonts of various point sizes. Then for a good while, truetype fonts were supported but they looked pretty awful. X only supported black-and-white fonts, so the truetype would just get converted to a bitmap for the X server to use. I'd say I got non-crappy looking fonts starting around 2000, possibly a bit later. Once the support showed up, it was just so smooth people think it's been in Linux all along 8-).

Nothing New for OpenSuse

[Prototerm]

I gave up on Open Suse when the 10.0 version came out, and they started removing stuff from the standard release. They first took out support for the nVidia drivers, then some of the wireless drivers, forcing me to find and install them both manually. So, whenever they get a little antsy about something, they remove it. As much as I really like Suse, I prefer something that just works out of the box, and doesn't make me jump through hoops just to use my own computer.

If I wanted to do *that*, I'd install Vista!

Re:Nothing New for OpenSuse

[lmb]

Are you in fact complaining that openSUSE complies with the license of the Linux kernel? That is rather amazing.

Re:Nothing New for OpenSuse

[Anonymous Coward]

As much as I really like Suse, I prefer something that just works out of the box, and doesn't make me jump through hoops just to use my own computer.

Than why are you using Linux at all?

Re:Nothing New for OpenSuse

[Anonymous Coward]

And at really ALL SUSE versions (Enterprise Desktop, Enterprise Server, Real Time...) do the same. There isn't any "standard release" with support for Atheros/madwifi, nVidia propietary drivers, or any other propietary kernel module...

Re:Nothing New for OpenSuse

[watermel0n]

I use SuSE Linux since 6.2 version and currenlty use OpenSuSE 10.2 . No problem with nVidia driver just add http://download.nvidia.com/opensuse/10.2 to YAST install sources. It will complain that "module license 'NVIDIA' taints kernel" but works perfectly.
OpenSuSE is a good distro, only this Novell/MS relationship annoys me a little bit.

Re:Nothing New for OpenSuse

[Anonymous Coward]

If you install one repository you get the nvidia drivers through yast as usual, and if you have an intel (or most non broadcom) wireless it works out of the box.

This is completely clean -

[lmb]

openSUSE does not ship code which is known to infringe patents or IP, so the patents either get invalidated (lengthy and expensive) or the code disabled / removed. This policy is not affected by the NOVL/MSFT deal at all; quite the contrary, it has always been Novell/SUSE's policy to not ship such code.

Just like openSUSE doesn't ship infringing Linux drivers, or Debian not shipping certain licenses.

What the heck is the fuzz about?

Re:This is completely clean -

[metamatic]

openSUSE does not ship code which is known to infringe patents or IP

Funny, my copy includes Mono.

Re:This is completely clean -

[Anonymous Coward]

Now you just need to point to a patent or IP that Mono infringes...

Re:This is completely clean -

[seandiggity]

You really don't understand the problem of software patents if you think openSUSE (a huge distribution with thousands of packages) doesn't infringe on any patents. If all patents were actually enforced, software development would virtually come to a halt.

Re:This is completely clean -

[L'homme de Fromage]

This one seems to cover Microsoft's ADO.NET and parts of ASP.NET, neither of which are covered by the ECMA standard for C#/.NET. There are lots more. The Mono project has admitted that ADO.NET, ASP.NET and Windows.Forms are patented extensions by Microsoft, and that they may have to remove those at some time (which would render Mono basically useless for most Windows developers and even some Linux developers).

Note to Ron Hovsepian

[bl8n8r]

Ron, If you want to know how similar arrangements have panned out, you may be interested in viewing the graphic video: Boa Constrictor Eats Bird Alive

This is what I like about Linux

[GFree]

I think this makes for a pretty good example of one of the strengths of something like Linux. If you find your distro moving into directions you don't like, you can leave pretty easily and try another distro. They're all Linux, just wrapped up differently, and so if a distro decides to pull some shit like this, they'll only be hurting themselves because there's no real lock-in to any one distribution.

Microsoft are trying to cripple Linux using traditional methods, but all they can really cripple is openSUSE due to the Novell partnership. It's not like MS can take over EVERY SINGLE DISTRO, particularly the homegrown stuff. A good example of the power of choice I think.

Re:This is what I like about Linux

[Aequo]

Do you have any examples of how openSUSE has been crippled by Microsoft? It has already been pointed out further up that this article was _clearly_ either written by someone trying to spread FUD or by someone who just isn't very knowledgable (subpixel hinting is a freetype setting that the freetype developers themselves suggest disabling for distros). It is quite funny to see so many people jumping on the bandwagon, attempting to find 'omgz evil' in Novell because they made a business deal with Microsoft; obviously a deal that turns out to have done them more bad than good in the eyes of the community.

Re:This is what I like about Linux

[Anonymous Coward]

You're absolutely right! This issue is a load of cr*p. Unless open source is your religion, there is no motivation to react to opensuse's decision on this. Since it takes about an hour to switch your distro to some other FREE OS, there is no need to be loyal to any of them. When opensuse stops delivering what I want (patent free or not), I'm off to the next, and the next, and the next.......

Re:This is what I like about Linux

[kripkenstein]

Do you have any examples of how openSUSE has been crippled by Microsoft? It has already been pointed out further up that this article was _clearly_ either written by someone trying to spread FUD or by someone who just isn't very knowledgable

The openSUSE mailing list (and source code) states directly that a certain feature is disabled because it is covered by Microsoft patents. Were it not for those patents, the feature would be enabled (or given as an option during runtime, etc.). So, something to do with Microsoft (their patents) is crippling openSUSE (who do not enable a feature based on those patents).

Now, it may be that this is not new, and has existed previously as well. It was certainly news to me, and apparently to other readers here. Hopefully this matter can be cleared up. Yet, in essence we do have a case of openSUSE hobbling itself due to Microsoft patents, that part can't be denied, I don't think, contrary to what you are saying here. Or perhaps I didn't understand your point.

Clear-Type replacement

[Kim0]

They could just use this replacement, which is not patented:
http://oyhus.no/SubLCD.html

Re:Clear-Type replacement

[novus ordo]

I'm wondering if this is being used in in freetype?

the openSUSE team did the right thing

[w_albright]

IMHO, they did the right thing. One of openSUSE's goals is to be completely open source software (hence the 'open' in 'openSUSE'). Even if they may have the right to use them due to the MS/Novell patent deal, they do not want the distro encumbered with non-OSS software in the default install. Fedora 7 also disables this feature.

If you want a distro protected (encumbered) by MS patents, buy SUSE Enterprise.

Re:the openSUSE team did the right thing

[metamatic]

I look forward to SuSE removing Mono (and hence GNOME) from OpenSuSE.

Re:the openSUSE team did the right thing

[houghi]

Well, make one. It is not that hard to make a SUSE based distribution

        * Download the ISOs you desire
        * If you need extra RPM's, download them.
        * Run makeSUSEdvd -i [any other parameter]
        * Run rembrand -b "#FF00FF" -R /tmp/pink_rpms *.rpm
        * Browse through Branding Overview and check for any left-overs.
        * Edit, remove and change anything you desire.
        * Run makeSUSEdvd -C

You can configure your YaST any way you like. Remove Mone (and GNOME) and have it work still as a normal SUSE installation with choices you desire. Add MPlayer and codecs to your DVD. Add all those programs that you would want and are not available by standard.

Re:Patent or Copyright?

[smallfries]

Nobody is claiming to have invented, or patented, a font. At least read the summary if you're not going to read the article.

Correction. AA support in Windows.

[burnttoy]

Windows has had AA text in the following formats.

  Right Click (or Right Menu Key) -> Properties -> Settings Tab -> Tick "Smooth Edges of Screen Fonts".

WinXP - ClearType fonts supported (at least on Pro) - get a control panel applet from msdn/microsoft.com to change settings. HW support via alpha blending.

WinXP Tablet Edition - Support of 90 degree rotation e.g. aliasing in Y instead of X (screens mounted portrait)... I think I'm right on this.

Vista - more of the same I guess!

YMMV - It's been a while since I mucked with Windows GDI Drivers.

Aren't there any 'Open Sauce' law students...?

[Anonymous Coward]

Who are doing patent and IP law courses?

Aren't there any professors looking for a practical law project?

I thought there were people falling over themselves to get a decent CV to present to practices when they graduate. What could be better than 'worked on the "university of X vs Microsoft" case?

Of course, you would actually set up a small company to do the sueing to limit your exposure. The Uni would fund that as part of the course expenses. It would be several classes worth of cheap training, win or lose.

Patent.

[KingSkippus]

It's not the fonts that are in question, it's the method by which they're rendered on a screen.

Slashdot ate my comment!

[burnttoy]

Windows has had AA text in the following formats.

Win95,3.1, probably 98 etc - none at all! Just 1bpp

98SE,ME - these have support for 2 bit per pixel transparency masks as part of the GDI device driver. I can't remember how to turn on this feature but the Win2K method is shown below.

NT4 - no support - just 1bpp text.

Win2K - Same as 98/ME, 2 bits per pixel transparency. Try Desktop (Win+D), Right Click (or Right Menu Key), Properties, Settings Tab, Tick "Smooth Edges of Screen Fonts".

WinXP - ClearType fonts supported (at least on Pro) - get a control panel applet from msdn/microsoft.com to change settings. HW support via alpha blending.

WinXP Tablet Edition - Support of 90 degree rotation e.g. aliasing in Y instead of X (screens mounted portrait)... I think I'm right on this.

Vista - more of the same I guess!

YMMV - It's been a while since I mucked with Windows GDI Drivers.

Re:Slashdot ate my comment!

[shutdown -p now]

Win95,3.1, probably 98 etc - none at all! Just 1bpp

Wrong. Plain (non-subpixel) was available in Win95 already, if you had Microsoft Plus! for Windows 95 installed.

Also, GDI+ (which WinForms uses) has its own antialiasing algorithm, and WPF has yet another. So it's even more diverse.

Re:Slashdot ate my comment!

[burnttoy]

Gotcha. Couldn't remember if it was the 95 Plus Pack or 98SE.

Novell - Just brilliant

[HangingChad]

It really doesn't matter if this is related to the patent deal with Microsoft or not. The damage is done by the mere perception that Novell is aligned with Redmond.

This whole deal is to IT was Iraq is to foreign policy: A bad idea implemented without a clear exit strategy.

Unless the goal was to drive users to Ubuntu. In that case it's a brilliant plan.

Re:Novell - Just brilliant

[bwalling]

The damage is done by the mere perception that Novell is aligned with Redmond.

I won't disagree with that statement, but that's no excuse for this ridiculous story posted to Slashdot. For all of the griping around here about other companies' FUD, this is basically pure FUD itself. Alas, it's not an isolated case. It's too bad so many people read this site - it's a very poor source of information if you just scan the front page.

Re:Novell - Just brilliant

[pair-a-noyd]

Absolutely correct.

I have been a Suse fan and PAYING Suse CUSTOMER for several years. Until recently. This MicroSuse crap has pissed me off and turned me off.
I totally abandoned M$ products several years ago (I have about 100+ PC's here) because I got sick and tired of the constant problems. I hate M$. I hate everything about M$, I hate all their products, I hate their tyranical and predatory business tactics, I hate their strong arm threats, I hate their anti small business models. M$ only cares about M$ and increasing their profits. If global domination and total tyrany is required to achieve those goals then they embrace that with a zeal. Ballmer and Gates are scumbags and I hope they both burn in hell for eternity. I do not want any M$ products in my home or business.
I now despise Novell/Suse with the same contempt that I despise M$. My problem is now migrating my systems from Suse to Gentoo. And I don't want Mono or Beagle or Open Office with M$ compatibility on my machines.
I WANT my all of my computers to be as incompatible with M$ as possible.

Novell/Suse, you betrayed us. This sin can not ever be forgiven, we will hold it against you forever.
Please cease and desist your attempts to "poison the well" and keep your code out of the Free Linux world.
You fucked up and now you must pay for it. Die already will you?

I know this will get modded down as flamebait or troll but I don't care, you are hearing the voice of ONE very pissed off Suse customer. Novell will never see another red cent from me.

Blurred text dates back to Babylon

[argent]

The Code of Hammurabi already had sections regulating brewers and taverns. Anyone got an earlier precedent than blurred cuneiform?

Cleartype

[Frozen Void]

I have win98 and i see not difference between font smoothing on/off.
I bet it all depends on the font,and only serif fonts like Times New Roman could be enjoying this feature.

Re:Cleartype

[Anonymous Coward]

Of course. lol.

I for one...

[Anonymous Coward]

welcome our old poorly rendered overlords.

Re:I for one...

[Tablizer]

I for one patent the welcoming of overlords. Now pay up!

Um, didn't Linux already fix this?

[JetScootr]

I seem to recall about 10 years ago font copyrights, etc, and the ClearType issue came up regarding Linux. The question then was whether it was OK to do *something* like this, or include fonts, etc, in OSS files and/or SW. Anyone remember the details?

Ditto.

[Anonymous Coward]

SUSE has been my favorite distribution for many years, and this patent deal really hadn't affected my opinion of it until now. Because until now, the patent deal had not adversely affected the quality of the software. Consider yourself written off, my dear SUSE. Perhaps I will find greener pastures in Ubuntu territory.

Ubuntu failed on my laptop, OpenSuSE worked.

[Anonymous Coward]

I bought a new Toshiba Satellite M105 series laptop in December 2006 and repartitioned the hard drive so I could dual-boot XP Home and Linux. I've been a longtime Slackware user and decided to give Ubuntu (6.10 "EdgyEft") a try since there's so much hubbub about it these days. The boot-and-run-from-CD was certainly impressive so I decided to install to my hard drive. The installation never would complete. It would always get some percentage done (different amount every time) and then hang. Even with a full T1 (1.5Mbps) Internet connection, Ubuntu never would successfully install to this machine. It did not just "Simply Works" at all. :-(

So I then downloaded a boot iso for OpenSuSE 10.2 and did an over-the-net install, and the install went flawlessly. Everything works perfectly on this laptop, including the ethernet, WiFi, dvd burner, sound, touchpad, usb and firewire interfaces.... literally all hardware devices worked perfect the first time. I was stunned, because with my past experiences installing Slack on a laptop, getting Linux to work 100% on a laptop has always been a bitch. In my case, OpenSuSE 10.2 just "Simply Works" where Ubuntu wouldn't even install. Maybe when FeistyFawn gets released, I may give it another try.

Re:Ubuntu failed on my laptop, OpenSuSE worked.

[MollyB]

Wow. I'm truly sorry for your situation. I haven't the faintest idea why we have had such opposite experiences. This is a Jetbook circa 2005 and a 80GB drive with no other partitions. Since you are acquainted with Slackware, your knowledge is far far greater than mine regarding possible reasons. I only can offer this from personal experience, and it is blatant superstition on my part, but I believe computers are sneaky; they gauge the knowledge of the operator and all problems lie just outside this boundary. My apologies also to folks who took my advice at face value; they would be wise to consider your post.

Again, no problems for this particular hardware/Ubuntu combo. Your mileage may vary...

Re:Ubuntu failed on my laptop, OpenSuSE worked.

[Billhead]

I've always had problems installing from the live-CD, if I intend to do an installation I download the "alternate" ISO.

Re:Ubuntu failed on my laptop, OpenSuSE worked.

[Assassin_for_Atari]

Actually for anyone looking for a "install it and just watch it work" distro, I have been referring people to "Linux Mint" its pretty much ubuntu but with flash and all the media codes installed. They also have a great "start menu" and other utilities that make in a Drop in solution!

Re:Ubuntu failed on my laptop, OpenSuSE worked.

[Master of Transhuman]

When I switched to Kubuntu last fall, the stupid CD installer wouldn't let you exit the change mount points screen. So I had to use the alternate text installer.

This means the ENTIRE INSTALL PROCESS WAS NEVER TESTED!

That's just braindead for any distro.

People who think Ubuntu is somehow magic are going to be sorely disappointed. It's just another distro and makes numerous mistakes in design and implementation and lack of adequate testing.

Re:Ubuntu failed on my laptop, OpenSuSE worked.

[stony3k]

I had the same problem, but the installer did work for most people. My bet is that they did not have the resources to test the installation on all possible configurations. It worked for most configurations, but you and I were unlucky in that it didn't.

However, there's one way to great way in which we can help Ubuntu work better in future. Users like you and me need to get out and test the alpha/beta versions on our systems and when we find an issue, we need to report it, file a bug, etc. Complaining on slashdot will not make it work better.

Re:Ubuntu failed on my laptop, OpenSuSE worked.

[Master of Transhuman]

While I agree that Ubuntu doesn't have the manpower to adequately test, this particular bug wasn't an issue about testing "all configurations" - this was BASIC. I mean, if you don't test the screen that changes mount points, what else wasn't tested? It had to be that NOTHING in the install was tested! Did they just test a default install and didn't even exercise that screen at all? That's not a "test" in any sense of the word. Allowing that bug to slip through pretty clearly shows that the entire install process was not tested.

There's no way around it - they screwed up big time.

I'd like to be able to test distros, but I don't have the time or the disk space at the moment. In any event, there do seem to be plenty of people who like to do that - which makes one wonder again how such an obvious bug as the one I cited slipped through the cracks.

Mark my words...the legal war cometh...

[Anonymous Coward]

M$FT is simply using openSUSE here as a tool to prove that Linux violates patents (I don't believe this to be true). The more that Novell disables due to Microsoft patents, the more you can be sure that such disabling will be pointed to in the major legal war coming soon. This is the equivalent of troop movement...small pockets of M$FT troops are being moved into strategic areas near Linux's borders...they're hiding out in farmhouses and covered up with bales of hay, but they are indeed there. It's only a matter of time before they jump up, yell "Righteousness and truth be damned...", and begin the attack.

Fathers, gather your weapons now. Board up your windows and doors. Make sure your cellars are concealed and livable, so that your wives and daughters will be comfortable. The War to End All Wars is upon us, and it's our obligation to join the cause and fight the dread beast Microsoft. I fear for our ancestors if we do not.

The first step? Abandon openSUSE. They have been compromised, tainted and brainwashed beyond recovery. Best to put a bullet in their brain and save whatever honor is left in their history. Any among us who refuse to do so should follow the same fate.

Re:Mark my words...the legal war cometh...

[Spasmodeus]

The War to End All Wars is upon us, and it's our obligation to join the cause and fight the dread beast Microsoft. I fear for our ancestors if we do not. Why, is Microsoft building a time machine to go back and kill your great grandpa?

Christ, it's like a Community Theater production of Les Linux Miserables around here.

Novell hasn't received any patent licenses

[ClosedSource]

"As reported on and discussed, this matter may be connected to the Microsoft-Novell deal. If so, Novell should have received a license for the Microsoft patents, assuming the deal covered all relevant patents. Does the license therefore extend only to SUSE, but not openSUSE?"

The Microsoft-Novell deal only protects their customers, not the companies themselves. If Novell violates a MS patent, they can be sued by MS; the deal doesn't change that scenario in any way.

Fork Them!

[Anonymous Coward]

IANAL and all that, but does this patent have any validity in the European Union? If not, fork them and distribute it from Europe. Someone tell me if this is feasible or not.

OpenSuse Fonts

[TheNinjaroach]

I've been using the free SUSE since 9.2 and I've always thought it was one of the best looking desktops, especially with their fonts. It gets better and better with each version, but with 10.2 I think they've stepped back a notch. This article highlights exactly the reason why, I couldn't quite figure it out before.

As crappy as Novell is behaving lately, I really need to switch in protest. Unfortunately, they will force it on me when the distro completely collapses under the pressure of their bad decisions.

Re:OpenSuse Fonts

[tomhudson]

I skipped 10.1 (stuck with 10.0 at home for the longest time, and still use 10.2 at work), so maybe I'm just judging from the improvement against 10.0.

As for the patent deal, I'm inclined to cut them some slack. We don't know the exact terms (its more than likely cross-licensing for stuff that has nothing at all to do with linux, and MonkeyBoy is just fudding everyone; if it were true, why doesn't he show it? Esp. since Novell has denied point blank that there is ANY linux patent problem ...)

We've been through similar storms with Sun, for example (the $8m license), but it all turned out okay in the end - don't forget where we got OpenOffice, and Java. Novell is a big contributor in terms of code ... if they can suck(er) $$$ out of Microsoft, and we get better interoperability, its worth a bit of confu [s|d] ion.

... and lets not forget the heavy lifting Novell is carrying in SCO vs Novell ...

They make a good product. They may have miscalculated the maneuvering by microsoft, and the community being so ready to believe the worst (without proof ... ) but I think we're better off with them than without them.

It is about precedents

[WindBourne]

This is headed for a major court battle. In fact, it should be obvious that SCO is just the first. They hope that they can use Novell dropping this, to convince a jury that the Linux community "admits" that we are at fault.

I expected this of MS. They are the dirtiest company on this planet but they do know how to keep themselves alive. But I am still trying to figure out WHY Novell is partaking of this? At best, it will have only a short-term advantage. But I have dropped SUSE and will not be suggesting them to others. Kind of sad. Novell was a good company.

Re:It is about precedents

[advocate_one]

corporate critters only have an eye for the next quarterly returns... as long as they can make their killing and successfully cash out their stock options, they don't care about the stockholders really... they don't intend to be around when the crap hits the proverbial fan. They'll be off raiding another company

Re:It is about precedents

[DevoPhl]

> This is headed for a major court battle. In fact, it should be obvious that SCO is just the first.

I said this back when SCO sacrificed itself to try and take down Linux. Linux has been treading on thin ice for years. The programming landscape has been filled with copyrights and patents for decades and to try and develop an operating system, not to mention applications that don't use them is next to impossible. Lets face it, it tough to write 5 lines of code without violating some sort of a patent.

Today, you can only make a have baked OS and have half baked applications before you enter the patent and copyright mine field. Linux might be able to exist at the server level thanks to Unix's legacy but for it to make any progress on the desktop, its going to have to start licensing these expensive technologies. Many of these technologies are reserved by companies like Microsoft and you can bet Linux will not get a free ride licensing them. Look at what SCO wanted to do to Linux. They intentionally priced their SCO-Linux license so that Linux would become the most expensive small server OS by a wide margin. SCO wasn't going to just license Linux, they were going to bury Linux.

I think this is what Microsoft wants to do. I'm sure Apple and others are in the same boat. The idea is to restrict certain technologies to the point where Linux looks like Windows 3.1 in functionality and thus preserve their desktop monopoly.

In my opinion, open/free Linux is still 5-10 years behind Windows and OSX in terms of desktop functionality and these hurdles aren't going to allow the gap to close any time soon. On the other hand, enterprise versions of Linux, that can license these technologies, are so expensive, they really aren't an alternative to Windows on the desktop.

If it weren't for IBM backing Linux, it would be dead today. Microsoft, SCO and others would have patent-infringed Linux back to the stone ages. Clearly IBM's presence has kept a lot of this from happening. But as OSS and Linux become more popular, its clear the lawsuits will start coming fast and furious. We've only seen the beginning of this.

Re:It is about precedents

[twistedcubic]

In my opinion, open/free Linux is still 5-10 years behind Windows and OSX in terms of desktop functionality ...

Yeah, a friend of mine has a Windows 95 machine, and I tried it out recently. Man, I was BLOWN AWAY. I've been wasting productivity on my Gentoo and Ubuntu machines, when all this time Windows 95 just makes it much easier. I mean, it looks and functions better, and everything just works. My friend showed me how to take all my LaTeX files (100 or so) for my book, with all the revision history in Git, and convert it all to Microsoft Word. Wow! Productivity SQUARED! After watching my reaction, my firend slapped me on the head (really hard) and exclaimed, "It's the APPS, stupid!"

Re:It is about precedents

[Anonymous Coward]

Novell did a very big mistake when they went to bed with micro$oft. They "sold the farm", actually, gave it away.
I was a dedicated SUSE 9 and SUSE 10 user and software developer, but a month ago I tossed SUSE out the window because of their stupidity. This IS just the start. In 5 years, they'll be dead. Like I said, big mistake.

Re:It is about precedents

[drinkypoo]

Today, you can only make a have baked OS and have half baked applications before you enter the patent and copyright mine field.

This realization is precisely what the patentmongers fear. Because if this realization gets out in the world it will do them irreparable damage - the argument [in court] will be that it is utterly impossible to develop software because of the patent system. Admitting that you have a problem is the first step on the road to recovery...

In my opinion, open/free Linux is still 5-10 years behind Windows and OSX in terms of desktop functionality and these hurdles aren't going to allow the gap to close any time soon.

Your opinion isn't worth much. There is one place, and one place only, where Linux is behind. That area is in drivers. Oh sure, Linux might support more hardware than any version of Windows - I wouldn't have any trouble believing that. But the problem with Linux is that the support for some hardware raises numerous problems which you simply don't have to deal with on Windows.

Aside from that, can you name one kind of task that's easier on Windows than, say, on Ubuntu Feisty? I really will settle for just one clear example.

Re:It is about precedents

[WindBourne]

I think that General Configuration is easier on Windows from a GUI POV. In both GNOME and KDE (and all the distros), we have LIMITED GUI config capabilities. I that apple is awesome on this, and MS is simply better than Linux, not awesome (but I can not really judge MS as I use it as little as possible).

Now, with that said, I prefer working in /etc. It is very easy to work with and understand. But, when I try to explain to my father why he wants squid AND how to configure it, it is impossible. I simply do not put him on squid (really a shame). Now, if somebody could explain HOW to properly update windows via their KV DB ....

Re:It is about precedents

[Master of Transhuman]

"In my opinion, open/free Linux is still 5-10 years behind Windows and OSX in terms of desktop functionality"

Here we see a truly stupid Microsoft shill.

Even Bill G must be cringing at how stupid this comment is.

He is correct about ONLY ONE thing is this entire post - IBM can hang Microsoft any day of the week in a patent war - and IBM has the best lawyers in the business and is making billions off Linux.

Linux and OSS have nothing to fear from Microsoft except stupid FUD tricks believable only by morons like this.

Re:It is about precedents

[Anonymous Coward]

There is no way that Microsoft can enforce their patent pool against Linux without suffering a massive backlash of counter-patents. Come on, you are talking about suing all the governments in the planet, and Microsoft knows that. Its to late to sue anything, just to costly for corporate image. Their only objective right now is to slow down the beast, they know they cant stop it because the free software model is way superior to the IP model. Period.

Re:It is about precedents

[Brandybuck]

I think that General Configuration is easier on Windows from a GUI POV.

In what way? I would put them both on the same level. I'm including OSX in here as well. Some have more options, knobs and sliders, and some of less. But the ease of use stays roughly the same. In terms of hardware specificity, Windows has the edge, but in terms of *general* configuration, there's not much difference. The simplicity of GNOME and OSX are offset by the frustration of missing functionality, which evens out to the slightly more complex Windows and KDE configurations that offer more control.

p.s. I'm being overly fair, of course. Objectively, the Windows control panel has severe usability problems. It's only user familiarity that causes people to assume that it's easy.

Re:It is about precedents

[drinkypoo]

Now, with that said, I prefer working in /etc. It is very easy to work with and understand. But, when I try to explain to my father why he wants squid AND how to configure it, it is impossible. I simply do not put him on squid (really a shame). Now, if somebody could explain HOW to properly update windows via their KV DB ....

Webmin can administer squid.

On OSX you can use SquidMan.

WPM claims to administer Squid.

I'm currently installing a new program (March 1 '07) called Internet Cache Setup that claims to configure squid with a wizard. I guess I have to restart firefox, so I'll come back and let y'all know how it worked...

Re:It is about precedents

[drinkypoo]

Okay I'm back after fixing a bug in internet cache setup. Look for a line like this:

echo "visible_hostname $HOSTNAME" >> $SQUID_CONF;

Replace it with this:

echo "visible_hostname `/bin/hostname`" >> $SQUID_CONF;

This makes it work on Ubuntu Feisty. Relying on environment variables is always a bad idea (the author was not setting $HOSTNAME, but counting on it to be set for him.)

However it still sucks; all it does is let you turn it on, share your cache, or set multicast. It doesn't even let you set the cache size!

moving on... wpm is designed for redhate and uses cgi scripts. It couldn't find apache on Ubuntu.

Because there is no documentation I will skip it. I'm not interested in writing someone else's tool for them today.

I have an OSX system next to me but I don't want to mess with it, so I'm not testing SquidMan.

Next I guess I'll have to try webmin. There is a debian package, I'm installing it now. I'm on Ubuntu Feisty but so far every .deb I've installed has worked. For that matter, the two RPMs I converted with alien worked fine, too (iscan, to support the Epson 3170 Photo, and the plugin for that specific scanner model.)

Anyway, that digression aside, when webmin is installed on Ubuntu you have to start it manually: sudo /etc/webmin/start. Connect with https (https://localhost:10000) and log as root. Squid is under "servers" but you have to configure your module path - there will be a link there (for some reason webmin expects squid to be installed in /usr/local.) If you want full reports you must install both squid and calamaris.

Change the following settings for Ubuntu (Feisty):

Full path to squid config file: /etc/squid/squid.conf
Squid executable: /usr/sbin/squid
Full path to PID file: /var/run/squid.pid
Full path to squid cache directory: /var/spool/squid
Squid cachemgr.cgi executable: /usr/share/webmin/squid/cachemgr.cgi
Full path to squid log directory: /var/log/squid

At this point webmin from the debian package will manage squid on Ubuntu.

Re:It is about precedents

[mangu]

In my opinion, open/free Linux is still 5-10 years behind Windows and OSX in terms of desktop functionality

In my opinion, the fact that a comment like this got (Score:4, Insightful) is a sad comment on the sorry state of affairs of the Slashdot moderating system...

Re:It is about precedents

[DevoPhl]

I think you missed my point. The desktop is all about the applications. Open Source Software has really only been working in the desktop application realm for about 5-8 years and it has a lot of catching up to do to get to where Windows is today. There are some positive signs in applications like OpenOffice, Evolution and Gimp. But there are so many areas that OSS either has no application or the software is in its infancy, thus lacking what is needed to be truly usable.

You notice I said open/free as well. You can only go so far with free. If Linux is to compete with Windows and MacOSX, it needs a better set of proprietary applications. Lets face it, fully supporting audio, video and other proprietary areas will require software manufacturers to either start porting their applications into Linux (not likely to happen soon) or for Linux to start licensing these technologies and incorporating them into OSS applications. Either way, it won't be free. I would pay $75 for a desktop Linux distribution that included proprietary codecs, drivers, etc so that my Linux desktop could play a variety of music, DVDs, etc without using marginal and/or illegal add-ins.

I have a Fedora Core 6 system at work and have a similar setup at home. I've been a programmer in Unix for over 20 years and a systems administrator for over 15. I use Linux for as much as I can use Linux. I would love to see my Linux computer completely replace my Windows computer but without the full suite of "usable" applications, I will have to have a Windows computer. In this sense, I believe Linux is still 5-10 years behind Windows.

Re:It is about precedents

[drinkypoo]

I would pay $75 for a desktop Linux distribution that included proprietary codecs, drivers, etc so that my Linux desktop could play a variety of music, DVDs, etc without using marginal and/or illegal add-ins.

Illegal? Feh. I consider it an act of civil disobedience (and let's not have the conversation about whether that requires that you are likely to be arrested for it, thankyou) to play a DVD on my PC without paying a licensing fee. Which is pretty fucking amazing when you think about it. How the fuck did it become illegal to utilize a product which you've purchased?

Re:It is about precedents

[mrchaotica]

In what way?

In Windows and Mac OS, you can configure everything from the one single place. With Linux, you can only configure GNOME-related stuff in the GNOME control panel, KDE-related stuff in the KDE control panel, other stuff only by digging through /etc, the kernel by using menuconfig or xconfig, etc. From the perspective of a Windows or Mac user, it's rather haphazard.

Re:It is about precedents

[Brandybuck]

It's also a case of "ignorance makes bliss". As a long time Unix user, when I go to Windows or Mac, I feel very constrained by the configuration dialogs. If I want to do something outside of the Redmond/Cupertino boxes, I have to start tweaking obscure registry settings, download buggy and untrustworthy utilities, etc. Windows and OSX users think their dialogs let them do everything, simply because they don't know anything else exists outside of the dialog's border.

Agreement

[Anonymous Coward]

MS didn't agree not to sue novell, but to not sue their customers, didn't they?

Mod this up!

[spitzak]

It is brilliant! And it certainly is different than Microsoft's implementation and not covered by any patent.

In fact it is a good deal simpler, it seems to reduce the color fringing naturally without having the calculation for one pixel effect nearby ones. It also only requires 2x horizontal resolution of the original rendering, rather than 3x.

From discussion here it sounds like what Microsoft patented is a filter, similar to their "font smoothing" filter, that turns a 1x resolution image into the colored cleartype image. That would mean it covers very little, and will be as obsolete as font smoothing in a year. However I kind of doubt this, I was under the impression that they were rendering at 3x horizontal resolution and then using this filter to put that in the lcd pixels without color artifacts. Anybody know?

In either case, rendering at 2x resolution and not having to do any filtering makes this be excluded from the Patent. Also looks like the results are better for very tiny fonts, too, mostly because there are only two color fringes, not 3.

This is a good thing

[iminplaya]

I don't care if every computer in the country gets nuked to enforce IP law. I'm for whatever it takes to wake people up, force their hand, and resolve the issue once and for all. Maybe then we'll know the public's true opinion on the matter. They're for it now because they're not aware of its effects. Incidences like these will help clear things up for them.

Re:This is a good thing

[Anonymous Coward]

Yawn,

Oh, did you say something?

Re:This is a good thing

[iminplaya]

Yeah, I said your mama's so fat, she uses her boogers for bowling balls. Go back to sleep.

Those Patents should expire soon (already have? )

[geohump]

When were the cleartype patents filed? Patents only last 17 years.

the only solution is to drop SuSE

[Anonymous Coward]

Every Novell product became tainted the day they went in bed with Microsoft. Though SuSE was a very stable and mature distro, now the only solution to avoid potential troubles is to drop it immediately and migrate to a safer one.
Debian is better for servers, Ubuntu for desktops, RedHat for corporate drones and Gentoo for experimenters: there's enough choice for everyone.

Re:the only solution is to drop SuSE

[deanoaz]

If Novell is forced to punt, after all the work they put into Linux, will anyone else be willing to risk tying their future to Linux afterward?

I've wondered, ever since Novell announced their plans to migrate to Linux, if the true roadblock to success wouldn't be the OSS community itself. There is a whiff of elitism in the OSS ranks that seems to want Linux to always be reserved for 'the right people'. Every time I start to get over that feeling I'm forced to look at a MAN page and it all comes back. There is a contingent inside Linux that does not want it to be user-friendly, or common in the business world, or used by for-profit companies.

If the OSS community torpedoes Novell, what do you think that will do to the long term prospects of Linux becoming an accepted alternative to Windows in the Enterprise? I'm sure that many do not care or would be happy to see Novell and then Linux fail in the competitive market.

Re:the only solution is to drop SuSE

[sjwest]

I've wondered, ever since Novell announced their plans to migrate to Linux

That was 'Pre Ron' (new ceo) remember now Ron Hovispan recently called all Novell products (that compete with exchange/ad recently as 'crap' in conjunction with the ms pr department. Now

If the OSS community torpedoes Novell, what do you think that will do to the long term prospects of Linux becoming an accepted alternative to Windows

Novell where early and then too late to the party. - Linux will be there because it scales on old pc's and in the money argument. Fud might make it tricky for it managers who have replaced ibm for microsoft (eg not getting fired) but a day will come.

The question is: What has Novell done for you recenlty ?, - be honest this is why Redhat lives and why Novell's ceo has to admit that all its windows/server products are rubbish compared to ms's products in return for a cashflow from Redmond.

Im an ex Suse user (9.2). - Novell have zero creditbility. I imagine Novells ceo will recommend they use Vista at Novell soon one as Ron does not like chairs being thrown() with those words 'what have you done recently for microsoft'.

Nice Headline!

[davevr]

openSUSE hobbled by patents??

yeah, right, and my access to your wallet is hobbled by those pesky anti-mugging laws.. geez.

Windows font rendering is crap anyway, go FreeType

[oddityfds]

I always find font rendering in Windows to be really bad anyway, whether on a CRT or LCD, with or without ClearType. The only decent setting is with all antialiasing turned off. Maybe that's why so many people seem to dislike antialiasing.

FreeType - with no patent-infringing algorithms enabled - looks much better than Windows and MacOS X on both high- and low-resolution LCD displays and with subpixel rendering enabled.

Novell Boycott?

[Anonymous Coward]

How's that going so far?

Mod parent up +5 insightful, please.

[drdanny_orig]

I couldn't have said it better myself. Which is why I didn't.

Licenses.

[miguel]

Just to clarify the thesis of the post.

Novell has not received any licenses to any patents, and neither has SUSE, nor OpenSUSE.

The Microsoft-Novell agreement is about not suing customers over any potential patent infringement.

Since OpenSUSE is a community effort, and it is used by people that might not be customers of Novell, removing code that is known to infringe on a patent is the correct thing to do (same policy applies to Mono).

SuSE always did this even before Novell

[alexfromspace]

As long as I can remember, and I have used SuSE since version 7, SuSE always provided some non-free programs and packages along with their distributions. These non-free programs and packages were not enabled/installed by default, and the user had to manually select them for install. All this does is make sure the user is informed about any possible implications for using such non-free programs and packages, and it is up to the user to take necessary steps to ensure that the user is legally protected in case they chose to use any of these programs and packages. This is nothing but a precaution for the interest of the users of SuSE and also to prevent SuSE from getting any cease and desist letters.

No patents in Europe!

[Anonymous Coward]

Novell has huge deals in Europe. Wouldn't it make sense to make a special european version, which includes patented technologies.
Software patents can not be obtained, or enforced in Europe so this would make sense.

Very Similar != Same in court eyes

[Tablizer]

The courts seem to favor the patent defendant (such as MS here) when issues of "obviousness" come to light. It seems it has to almost be exactly the same thing for a judge to overturn it, not merely very similar. A similar issue appeared in the GM (or was it Ford?) break-pedal case. Obviousness was pretty well demonstrated, but it was not deemed obvious *enough*. However, that case is still on appeal and climbing its way up the courts and may affect this case also because the level of obviousness is the key issue. Apple II using a very similar technique may not be enough. It may have to be pretty much exactly the same thing with the exact same look to qualify. (I think this is stupid, but that is another issue.)

This Is Kinda Cool.... Really...

[Anonymous Coward]

props to another (U|X|K|EDU)buntu user. Can't wait for Feisty to dual boot some laptops, too.

back on topic, it seems that we will now find out what patents msft claims are in linux.

if it goes into suse but not opensuse, then msft is making a claim that it is theirs. otherwise, it looks like everything else is good.

so far, it seems dancing stevo's patent claim on linux amounts to cleartype fonts.

good riddance clear type fonts.

it will be fun to see what else msft withholds from opensuse due to alleged patent issues.

just in case you've been playing too much online poker to understand what i'm saying, msft has started showing you their pocket cards.

Native resolution not common

[Tablizer]

My guess is that your OSX computer is scaling the image in some weird way that doesn't quite line up with your physical LCD pixels.

Most people do not use native resolution anyhow in my experience. Native resolution is too hard to read for most people because the default fonts are too small under it. There are some OS settings to allegedly scale fonts up, but in practice it seems poorly implementated or not consistent across vendors such that most op for non-native resolution. It is too much of a headache to set right. Plus, some find Clear-Type kind of blurry or "rainbowee" anyhow even if you do have them set right. Most admins seem would rather enable it on request rather than create builds with it on by default.

Re:Native resolution not common

[tomz16]

Most people do not use native resolution anyhow in my experience. Native resolution is too hard to read for most people because the default fonts are too small under it. Most people suffer from a serious disease called stupid... Enlarging the fonts in windows is trivial and works seamlessly in my experience. Just change Display Properties -> Font Size : Large Fonts

-Tom

Re:Native resolution not common

[Chandon Seldon]

Wait a second... there are people out there who run LCDs at non-native resolution because "the default fonts are too small"? That's disgusting - I didn't know that Windows sucked that much - hopefully Vista has a resolution-independent UI now.

Further, No wonder there are arguments about HDTV not looking better than DVDs. Many people are apparently just utterly oblivious to picture quality.

Re:Native resolution not common

[WNight]

Try changing it though. 90% of stuff works, but very frequently options boxes and such won't properly display anymore because their overall dimensions are hardcoded, pushing the buttons and text off the bottom of the screen.

This is why we shouldn't give designers pixel-perfect layout. They always end up requiring it.

Re:Native resolution not common

[tomz16]

I use medium-large fonts on all of my PC's depending on the pixel pitch of the display. I have NEVER had a problem with any dialog displaying incorrectly in any software application (and i've easily used hundreds of different applications). Having programmed a few GUI's myself, I definitely see where the potential for mishaps exists, but it's simply a non-existent problem in my experience.

Re:Native resolution not common

[Tablizer]

Try changing it though. 90% of stuff works, but very frequently options boxes and such won't properly display anymore because their overall dimensions are hardcoded, pushing the buttons and text off the bottom of the screen.

This is generally my experience also. Seems nobody tested them well outside of the default settings. This is both within Windows and vendor apps that may have their own rendering conventions/engine.

This is why we shouldn't give designers pixel-perfect layout. They always end up requiring it.

Auto-flow layouts have their own tripping points and gotcha's. I am not sure there is any perfect solution. It is very difficult to design non-trivial GUI's that can span a wide range of resolutions.

Re:Native resolution not common

[Ankur Dave]

Windows (and Mac, and Linux) have always had DPI settings, but most users don't know about them.

Re:Native resolution not common

[Chandon Seldon]

A DPI setting is basically irrelevant if the UI is primarily made up of bitmapped elements, or if the primary unit of measure for layout is "pixels". This is a serious issue with most of the GUI interfaces that are available, and even for the web.

Hopefully, implementing 3D interfaces like Aqua and Aero has caused OS developers to use more resolution independent vector graphics, but this class of issue will still be significant until a really good solution is implemented (maybe per-window DPI settings with hardware scaling).

Re:Native resolution not common

[Bert64]

The font should always be the same physical size regardless of resolution...
The computer is able to work out, by querying the screen (although some internal laptop panels don't seem to support it) the physical size of the sceeen, and thus compute the correct DPI.
Remember that font sizes are specified in "points" rather than pixels. 72 points should always equal one inch. If this is not the case on your screen, then your DPI is set wrongly.
All recent versions of X11 and OSX can correctly determine the DPI if your monitor supplies the necessary information. If it does not, then at least under X11 you can use the Displaysize directive and specify the width/height of the screen in milimeters:
                DisplaySize 285 204
Like so...
Aparrently the ability to work out screen DPI was supposed to be a much hyped feature of vista too, but it seems to just have the old fixed options of 75 and 96 DPI, regardless of what the monitor actually supports.
Also, a lot of windows apps break quite badly if you change the DPI.

Re:Native resolution not common

[Corwn of Amber]

Then you have pushed the resolutions to what, 10% above or under the norm? Oh. You use Windows in English. All software is written in English, so the sizes of dialog boxes and other GUI elements are calculated on English lengths. French uses 50-75% more chars to store the same meaning, so GUI elements are out of proportions when you change font sizes.

Re:Native resolution not common

[tomz16]

French uses 50-75% ... Easy fix.... stop using French...

As long as it's outside the U.S., sure.

[Kadin2048]

So in this case I could happly distribut the code with the flag switched on for the M/S patented code and it does not become a problem untill someone tries to import it into the US.

This is (IMO) correct. And this is the situation not just for the font-rendering, but for all other software-patent issues as well. You can write software that violates software patents and license it under the GPL all you want; the problem only occurs when someone in a jurisdiction that does enforce software patents wants to use it.

Since a very large percentage of the Linux userbase and developerbase are in the U.S., most software is basically designed around the legal situation in the U.S.

There have been at some times, Linux distros maintained entirely outside the U.S., which have taken a much freer approach to patent-encumbered software. IIRC Mepis may be one like this, and there are probably others. Technically they're illegal to download in the U.S. but that doesn't seem to be stopping anyone.

I suppose if you bought some rackspace in a "data haven" jurisdiction like Sealand, you'd be free to write code without regard to any sort of non-technical barriers; you could produce a distro that did everything that people would want it to do, right out of the box, without licensing or tithing to anyone. It would just be illegal pretty much everywhere, and I suspect if it got popular enough, the big monopolies would lobby to have your internet connection blockaded.

Agreed..., but Ubuntu is still evolving

[Anonymous Coward]

People who think Ubuntu is somehow magic are going to be sorely disappointed. It's just another distro and makes numerous mistakes in design and implementation and lack of adequate testing.

Agreed. But it's only fair to give Ubuntu another chance when the next release comes out very soon now. Like any other distro, it has its growing pains since it is still a fairly recent distro. It took SuSE until version 7.3, (some may argue that it took as long as until version 8.0) to get well-tested and most of the major kinks ironed out, and now at version 10.x is getting very finely tuned, but at the expense of quite a bit of bloat. Ubuntu, while still a long way from perfect, has had the distinction of getting an awful lot of new folks introduced in a very short timeframe to a very useable lean and trim desktop Linux with ample eye candy in a very non-painful, shallow learning curve way, and it'll continue to get refined more and more to a finely polished shine with each subsequent release version.

Re:Agreed..., but Ubuntu is still evolving

[Master of Transhuman]

I don't think you can compare versions of Linux from several years ago to the current state of distros. It's more appropriate to compare existing distros with each other. It's pointless to compare the latest Mandriva with Mandriva 9.

Ubuntu is based on Debian. While it is obviously an attempt to make a "release often" version of Debian and thus foregoes much of Debian's famed stability, it still has plenty of solid distros to draw on to do that.

The problem is "Not Invented Here" Syndrome, which exists with most distros. Everybody thinks they can do better than the next guy - and they can't if they don't borrow heavily from the next guy.

Ubuntu started out with an advantage - Shuttleworth's money. This wasn't a one-man distro with a couple maintainers who work weekends and evenings. This was an attempt to make a commercial-level distro on a par with Red Hat and Mandriva and SUSE. Thus they should be held to the same higher standard as a commercial distro.

ClearType patent invalid...

[micromuncher]

Subpixel rendering was covered by Apple][ patents decades before. Why SuSE would bother disabling ClearType is beyond me.

http://www.grc.com/ctwho.htm

http://edition.cnn.com/TECH/computing/9812/08/clea rtype.idg/

In order for prior art to matter...

[Rimbo]

There are two things that must exist.

1. It must be documented.

I present to you the Beagle Bros Big Tip Book for the Apple ][, by Bert Kersey, ISBN-10: 0553342800, ISBN-13: 9780553342802, Publisher: Bantam Books - 1986. (I forgot the page number, but it's in there, complete with how it works and an example program.)

As everyone knows, the Apple ][ graphics system was 128 pixels wide -- in color. But in monochrome, one could get 256 pixels wide. What the above book details is a way to get 512 pixels wide on any standard Apple ][ -- no 80-column adapter or 128KB necessary.

It's the exact same algorithm ClearType uses.

2. The parties must give a damn.

Novell just signed their soul away to Microsoft, demonstrating that they don't give a damn. This is just the first step in the suckyzation of Novell's free Linux offal^H^Herings. This sort of thing has been around for decades, so why hasn't anyone challenged the patent? Because nobody gives a crap, that's why. Everybody's using Fedora or Ubuntu if they're going free, anyway.

Re:Prior art? Sounds like a job for Pubpat.

[Jaywalk]

It sounds like this is exactly the sort of patent that Pubpat was formed to combat. From their About page:

The Public Patent Foundation ("PUBPAT") is a not-for-profit legal services organization that represents the public's interests against the harms caused by the patent system, particularly the harms caused by undeserved patents and unsound patent policy. PUBPAT provides the general public and specific persons or entities otherwise deprived of access to the system governing patents with representation, education and advocacy.

It sounds like this patent has already been proven invalid and is being actively enforced. What more do you need?

Not a controlled experiment

[bill_mcgonigle]

I put a store-bought version of Suse on my daughter's machine and everything was good. We upgraded her mobo and downloaded the 64 bit version of Open Suse. There are a myriad of niggling little details that don't quite work the same. The commercial version of Suse was a joy. The other one isn't. We're switching to Ubuntu.

You changed three variables (hardware, 64-bit, suse versions) with a sample size of 1 and concluded that one of the variables was causative. Sorry, invalid conclusion.

Not that I can imagine why any OSS developer would waste time on *SUSE products at this point. Ubuntu, Fedora, Nexenta can all use help.

Poor Reporting/No Update

[Anonymous Coward]

If the editor had bothered to actually read one of the articles submitted, this issue also affects Fedora Core 7. So is this a Novell/MS patent agreement issue? No, but posting it that way brings out the Slashbots. I wonder if Bruce Perens will come in this discussion and troll some more today.

The article author even updated his page to reflect what he has learned, "Courtesy of Max Spevack, I could get some relevant info concerning this issue. As it looks like, the subpixel hinting that is turned off by default now is a new feature in FreeType 2.3, and this explains why the building option to turn it off is only needed starting with Fedora 7, or openSUSE 10.2. The flag is also there so you could easily negate it and rebuild the package with the new, improved subpixel hinting on. Some more details on the patents in FreeType2, from David Turner of the FreeType project, here."

Will Slashdot issue update the main page? Unlikely. This place is circling the toilet as far as I'm concerned. Maybe the next story should be about how Micheal Bay is "ruining" the Transformers.

novell feels the heat

[MaoTse]

Today they changed the status of the 259718 bug - see bugzilla.novell.com.
If you're not an employee you won't have access ;-)

Novell explains its position

[blowry]

Hi: I'm a PR guy at Novell. I've posted Novell's explanation of this situation on our blog here: http://www.novell.com/prblogs/?p=318 Thanks. - Bruce

Not Quite

[Champ]

The Apple II family had 280 x 192 graphics; each row was 40 bytes of 7 single-bit pixels. The eighth bit per byte, as others have pointed out, is a position shift: when high, it shifted each of the other seven pixels one half-pixel to the right. Because of NTSC chrominance timing, this changed the color of those pixels (usually Green->Orange, and Purple->Blue).

There were some odd consequences of this. You usually couldn't have adjacent green and orange pixels, unless they were represented in separate adjacent bytes. Each byte's seventh pixel, if the half-pixel shift was in effect, would also be truncated to a half-pixel in length if the adjacent byte to the right was unshifted. So occasionally you'd get brown when you expected orange, or dark blue when you expected blue, unless you knew of the phenomenon and compensated for it manually.

Weird stuff. I learned programming on these machines. Lots of really fun, poorly documented "features" were there to play with. I too fondly remember the Beagle Bros. tools.

Re:Not Quite

[Rimbo]

280, 256, whatever. Ya talk at me like I never programmed an Apple ][ in HGR mode before. :)

What patent deal???

[Anonymous Coward]

The reason the "patent deal with Microsoft" does not seem to apply here - IS REAL Simple --

It was not a patent deal - No patent rights were exchanged!!!!!!!!

The deal was to not sue each others customers for patent violatons - but rather to take up such claims only the offending company.

Your screen does not use subpixel rendering

[adah]

Screenshot of my OS X system: http://img248.imageshack.us/my.php?image=picture3r p7.png

If you zoom in, you can see I'm definitely using subpixel rendering. I get the odd blurry looking font on my OS X system, but nowhere near the sort of stuff I see on other people's Windows laptops.

I zoomed in your screen shot, and regretfully you were not using subpixel rendering. The ClearType fonts (as in the Microsoft sample text) have coloured pixels in even black text, while the black text in your address line is complete black and white.

Re:Your screen does not use subpixel rendering

[someone300]

Try again. It's less coloured than the Microsoft font, but it is definitely coloured, as eyedropper in Photoshop reveals.

Re:Your screen does not use subpixel rendering

[adah]

Try again. It's less coloured than the Microsoft font, but it is definitely coloured, as eyedropper in Photoshop reveals.

No, it simply is not. Check my zoomed image of your original one:
http://wyw.dcweb.cn/temp/ClearType.png
Compare the text in the top and that in the bottom.

Re:Your screen does not use subpixel rendering

[someone300]

The text in the address bar and Firefox widgets is indeed not subpixel rendered, however, the web page text is. The comparison I was making was between the web page text (rendered by OS X) and the image on the page (rendered by Microsoft's ClearType). The web page text is rendered by my system using subpixel rendering and is what I am referring to, since most text on my system is rendered that way. Firefox's GUI widgets are probably custom rendered or something, but Finder's widgets do indeed use subpixel rendering.

Re:Your screen does not use subpixel rendering

[adah]

The text in the address bar and Firefox widgets is indeed not subpixel rendered, however, the web page text is. The comparison I was making was between the web page text (rendered by OS X) and the image on the page (rendered by Microsoft's ClearType). The web page text is rendered by my system using subpixel rendering and is what I am referring to, since most text on my system is rendered that way. Firefox's GUI widgets are probably custom rendered or something, but Finder's widgets do indeed use subpixel rendering.

True. And it really puzzled me a little. However, checking your screen shot carefully again, I think OS X is not using ClearType (though subpixel rendering is used), since I can tell clearly that even the font display in Firefox is fuzzier than the image below (sorry ;-)).—Though it is really strange that only the first three lines in the image in your screen shot are clear, while in my browser all is good: http://wyw.dcweb.cn/temp/ClearType_MS.png