I borrowed a Typematrix 2030 for a while and really liked it - I've been watching their website for months waiting for them to get stock back in, since they say they're currently sold out.
Actually, the one thing that wasn't perfect for me was the moved caps-lock: I like to remap caps-lock as an extra control key (like the Sun keyboards), and the double-height shift key on the 2030 means I can't do that. (I actually still used the remapped caps-lock with the 2030, because its top-center position was less of a penalty-zone than the real control keys).
I do believe it's quite likely that Microsoft instigated the SCO lawsuit stuff. I was just wondering if there was an irrefutable smoking gun - it's frustrating to have to keep guessing. Looking around a bit further, along with what spisska said above, I guess this is pretty close to damning evidence of involvement, at least, if not instigation?
I've wondered in the past why multi-core/multi-processor systems usually seem to have a power-of-two number of cores. This quote is interesting:
Besides, it's very rare for users to need an odd number of
processors (in any of the parallel codes I've seen at least). Most
parallel problems are able to work in parallel by decomposing some sort
of domain (be it physical, a mathematical matrix, etc.), and this
decomposition usually happens in more than one dimension (this generally
is done to optimize computation vs. communication). So generally worst
case are prime numbers of nodes and the best cases are powers of two.
So perhaps it's a convention borne of parallel computing algorithm design? But there could be a more fundamental SMP architecture reason - anyone know?
the owning companies have never expressed any desire to ban fansubs
That's not quite true: Anime company Media Factory ('Kimi Ga Nozomu Eien', 'Akane Maniax') asked animesuki to stop listing fansubs of their works, even though there were no US-licenced versions.
Thanks - interesting article. But the last line reads: "In the end, regardless of ethics, or motive, fansubs are technicaly illegal." - that doesn't seem to support your argument, or have I misunderstood?
My vote goes for camE - command line only, but it comes with some pre-canned config files that make it easy to set up. It can do timestamps, pretty antialiased/alpha-blended text, keep an archived copy of each image - and it talks scp as well as ftp for somewhat more secure uploading.
For hardware, I've had good luck with the Philips/pwc cameras (there was a time when they were only supported by a binary module, but the free replacement now works well enough for webcam use).
The nicest way that I can think of immediately is to have your bookmarks online, stored on some service or other. I don't know how it's done these days, but old netscape used to store bookmarks as an HTML file anyway, so you could possibly even, say, visit your bookmarks from another person's browser to show them something (assuming you don't mind making them public).
A friend was recently trying to get his several firefoxes to read and edit the same bookmarks file by symlinking/shortcutting the standard location to a single place on a network share, but he ended up giving up (something about the windows side, iirc).
turnitin.com seem to be a pretty scummy group, all up - full of talk about how they're making the world a better place, while keeping a database of every student paper and web site so a machine can tell academics whether the students have written good papers or not.
My university started using them this year, for undergrad classes. The problem seems to be the same as putting metal detectors on school entrances, for example - it proves to the students that you don't trust them one inch. That can't be healthy, psychologically.
Anyway, if you don't want to help with turnitin.com, you can put
User-agent: TurnitinBot Disallow:/
in your robots.txt file. As a good character reference, their user-agent used to be 'SlyBot' until they realised that this was giving them a poor image.
Weird X could do translucent windows back in 2000 - and this is real translucency, where 'covered' windows keep updating visibly, even through multiple layers of cover.
I actually ran this for a while, until it drove me crazy:).
Thinking out loud: the thing about 'must include non-alpha' is that it essentially forces the users to pick non-dictionary words. That's good all by itself. Sure, some of them will just use 'password1' or whatever, which is still dictionary-able (but not much *more* so, since they're probably going to pick the word they always choose anyway and just add a number). And with many users, you'll get stuff that's somewhat hard to do a dictionary attack on, like 'jack4betty' or 'y311ow'. Does this make any sense? I mean, I can see how suboptimal use provides no further protection, but is it likely to reduce the keyspace much in a real world scenario?
I think the parent is probably referring to some of the pictures on google's early hardware photos page, courtesy of the wayback machine. If so, the lego never necessarily went into `production', it was just when they were messing around.
Also, Microsoft often seem reluctant to go after people for interface copying - remember fvwm95, a window manager for X which cloned (at the pixel level, mostly) the windows 95 look'n'feel?
Well, back in '98 or so when I first visited x.org, it was the website of the `x consortium', which seemed to be a steering group with the various companies that ship an X server implementation as members, and Xfree86 as an `honorary member'. It seems like x.org is the same group, or at least related.
I'm not an expert on this, but my understanding is that quite a lot is outdated about X.
Some examples (feel free to correct me or add more):
lack of alpha blending
lack of native understanding of antialiasing, shading etc in the drawing primitives
insufficient speed to do things like play video (video etc usually seem to need to do some kind of direct hardware access to play smoothly)
X forgets a window's contents when it's covered over - try moving a large window over the front of a browser window quickly and see the slow redraw going on behind.
flickery movement - I don't know if this is a double-buffering type problem or what, but it seems to be difficult to animate large primitives without them flickering as they move.
no abstraction of GUI toolkits: this one's debatable, but in the early (athena etc) widget sets, buttons were boxes or circles with text in them. Now they're big, coloured, shaded items that animate when clicked, but all of this must be sent over the X protocol as a series of bitmaps with some font information. I think it's time that some higher-level information was sent (for example, abstracting an entire widget set so it can be rendered client side).
There are fixes for some of this stuff (see the Xrender extensions, for example) but they will only really be extensions - Xrender degrades gracefully into standard X, which is nice, but this sort of cruft will keep building up.
So, the graphics aren't completely vector based, but the icons are - in that screenshot, the icons in the Icon Catalog and the one at top left are vector-based. The three icons at top-right are bitmaps though (they're minimised applications, which always seem to get bitmap icons).
One interesting thing is that 4DWM recently got antialiasing on the SVG icons, which looks pretty sweet.
OK, so the number one thing I notice about my university compared to the jobs I've had is the lack of rules:
There's no fixed hours - it depends on the job, obviously, but lots of people seem to show up pretty casually.
Zero dress code - eccentricity is praised rather than condemned, and no-one bats an eyelid if you wander around barefoot in heavy metal T-shirts and bright blue hair.
Self-motivated work - there aren't any bosses prowling the cube-farm looking for slackers, so you have to have self-discipline to get anything done
These are all observations as a research student working with and around employees at my university, so I may be somewhat inaccurate.
I appreciate your point - that it's pretty useless to try and do something like that - but yeah, I do think it's kind of sad if something fun or cool can't be done because of some long-irrelevant decision to keep secrets from people.
Your IBM PC won't run linux because the hardware doesn't have the functionality. In my imaginary example, Froonix would run if only the drivers had been shared with the community. It's slightly different.
Thanks, you've clarified your opinions here...
Essentially I think we both understand each other but we're coming from different perspectives - you're justifying the corporate decisions to keep drivers proprietary, while I'm really just trying to show why users really are worse off with proprietary drivers from a usefulness standpoint.
"If by releasing that as open source it could compromise their business model, then they are making a sound business decision."
I agree. But I'm not really talking about good business models here, I'm talking about hardware I might buy or use. I now try not to buy components that will mess me around one day if/when I change my OS, and that includes Nvidia and Intel's stuff.
"I do know of at least one disadvantage to Open Source drivers."
I meant something slightly different here, again from the user's perspective: The user is not disadvantaged in any way by the drivers he uses being open, all other things being equal. She can be disadvantaged in certain situations by the drivers being closed, however. Clearly, NVidia write the best drivers for their cards, and Intel the best for Centrino. My point is that it'd be even cooler for the users if those drivers were also open, for the reasons outlined before.
"...those advantages aren't as important as protecting the commercial viability of Nvidia..."
So - this one's interesting: It seems like you're arguing that NVidia's profit margins need to be protected so that good graphics hardware will exist (i.e. my user-centric perspective). This can be argued both ways, but consider the possibility that if NVidia's trade secrets were suddenly all revealed to the planet, better hardware from multiple companies (and more cutthroat competition) might result? Bad for NVidia, good for graphics card progress in general? I don't know about this one, but I think maybe it's not so cut-and-dried as you suggest.
"What is this 'cool' thing that someone may want to do with a 50 year old laptop..."
Sorry, my example maybe wasn't realistic enough.
Intel and Linux needn't be dead, of course - it could be more like the network card I had, where Centrino is a retired product and Linux 3.0 no longer has the same driver hooks. I can easily imagine that happening within the useful life of a Centrino machine. Or, say, OpenBSD today. For the general case of proprietary drivers: I have a webcam with proprietary Phillips drivers for Linux. I can grudgingly live with this. But recently I wanted to make it work on a Mac... no dice. And there's nothing I can do about it, except wait for the kindly Linux hacker who signed an NDA with Phillips to finish writing his MacOS X drivers.
So I've argued a couple of points:
Closed drivers can cause problems for users - they're just inherently less useful than the same code in open form
Users are sad when they have problems
So this is why I'm sad about the Centrino drivers probably being proprietary. I'm not actually differing much in opinion from you, it's more an emotional response. We both understand why corporations often keep their drivers proprietary, but it bothers me, and I hope I've explained why.
It's sad because Intel could have released open drivers, and (possibly) it's now less likely that anyone will bother trying to write open implementations.
Open drivers are better than proprietary ones because there are no disadvantages, and there exist advantages such as allowing Linux distributions to contain them by default, the kernel to include them, and any potential bugs to be fixed by concerned users, rather than fretfully waiting for the landlordesque 'owner' to fix it for you.
Proprietary software is still OK, still useful, and still (as CowboyNeal said) better than nothing. However, one of the most exciting things about Linux is the spirit of open charity and freedom embodied in sharing and cooperating. When someone releases linux driver code which is not open, it saddens those who like the usual open nature of Linux drivers.
Imagine, years from now, that Intel has died, and Linux has died. You find an old 'centrino' laptop, and want to make it do something cool. You find a cool operating system, let's call it 'Froonix', and boot it up. But when you try out certain features of the laptop, they don't work. There's some message about 'proprietary drivers' that couldn't be ported from Linux, a predecessor of Froonix. It's sad, right?
Keeping drivers proprietary, which usually means they are tied to one hardware and software platform forever, is shortsighted and irresponsible... I've thrown out a PCMCIA network card, for example, because its drivers were only written for windows 95. The company has no interest in writing any more (it's a discontinued product) and yet the still-useful hardware cannot work with any current OS. Because of some ridiculous desire to keep an obsolete trade secret or two.
You may not agree, of course (from your post it looks like you don't), but can you understand why it saddens certain people?
Hmm... alas, my MS natural keyboard doesn't have the front feet. That sounds like a nice feature.
Coolpad: Thanks, I didn't know about that feature! I've only ever seen photos of it being used at a sharp towards-the-user angle, which makes my wrists ache just thinking about it. I approve much more of it now...
I bought a laptop recently. Here's what I find myself using or carrying with me:
Ethernet cable
TV- and Audio- breakout cables, for watching anime at friends' houses
Digital Camera adapter: I have a USB card-reader, as it's faster and easier than my digicam's serial interface, but the sync cable would be ok too.
Kensington lock
In addition, here's what I'd kinda like to have but don't:
Second power adapter: It'd be super-sweet to have a power adapter at home and at my desk at uni - between those two places where I spend the majority of my time, I'd never have to think about running out of power, and never have to carry a powerbrick with me. I currently just leave the power adapter at home (I don't run the laptop constantly all day, so there's enough juice for a workday's use) and it's so much nicer just being able to unplug and go, rather than winding up power cables to take with.
pocket usb2/firewire hard disk: Technically this isn't a laptop-specific item, but I think it's a good match: They're a great no-nonsense backup solution that goes with you when you travel.
GPRS or equivalent phone: We're just-almost getting to the point now where you can realistically use your laptop and phone to have always-on networking. I haven't researched this area, but it's exciting...
Some kind of USB-serial adapter: Lots of smaller laptops are `legacy-free' (hah), including lacking serial/parallel ports. People often miss this, so if you need to use those devices, get an adapter:)
Car power adapter. Makes road trips possible without going crazy looking for power in motels and the like.
Slashdot Mirror
It's a bit out of date, but this Matt Cutts blog entry claims that the toolbar doesn't feed URLs into the web search index.
I borrowed a Typematrix 2030 for a while and really liked it - I've been watching their website for months waiting for them to get stock back in, since they say they're currently sold out.
Actually, the one thing that wasn't perfect for me was the moved caps-lock: I like to remap caps-lock as an extra control key (like the Sun keyboards), and the double-height shift key on the 2030 means I can't do that. (I actually still used the remapped caps-lock with the 2030, because its top-center position was less of a penalty-zone than the real control keys).
I do believe it's quite likely that Microsoft instigated the SCO lawsuit stuff. I was just wondering if there was an irrefutable smoking gun - it's frustrating to have to keep guessing. Looking around a bit further, along with what spisska said above, I guess this is pretty close to damning evidence of involvement, at least, if not instigation?
Is there any strong evidence that Microsoft was actually behind the SCO incident?
Thanks - interesting article. But the last line reads: "In the end, regardless of ethics, or motive, fansubs are technicaly illegal." - that doesn't seem to support your argument, or have I misunderstood?
My vote goes for camE - command line only, but it comes with some pre-canned config files that make it easy to set up. It can do timestamps, pretty antialiased/alpha-blended text, keep an archived copy of each image - and it talks scp as well as ftp for somewhat more secure uploading.
For hardware, I've had good luck with the Philips/pwc cameras (there was a time when they were only supported by a binary module, but the free replacement now works well enough for webcam use).
I use it like this:
It gives a warning about accessing the drive via
The nicest way that I can think of immediately is to have your bookmarks online, stored on some service or other. I don't know how it's done these days, but old netscape used to store bookmarks as an HTML file anyway, so you could possibly even, say, visit your bookmarks from another person's browser to show them something (assuming you don't mind making them public).
A friend was recently trying to get his several firefoxes to read and edit the same bookmarks file by symlinking/shortcutting the standard location to a single place on a network share, but he ended up giving up (something about the windows side, iirc).
turnitin.com seem to be a pretty scummy group, all up - full of talk about how they're making the world a better place, while keeping a database of every student paper and web site so a machine can tell academics whether the students have written good papers or not.
/
My university started using them this year, for undergrad classes. The problem seems to be the same as putting metal detectors on school entrances, for example - it proves to the students that you don't trust them one inch. That can't be healthy, psychologically.
Anyway, if you don't want to help with turnitin.com, you can put
User-agent: TurnitinBot
Disallow:
in your robots.txt file. As a good character reference, their user-agent used to be 'SlyBot' until they realised that this was giving them a poor image.
Weird X could do translucent windows back in 2000 - and this is real translucency, where 'covered' windows keep updating visibly, even through multiple layers of cover.
:).
I actually ran this for a while, until it drove me crazy
Your wish is granted!
Homosexual Robot Cop, courtesy of the coolest comic ever, Bob the Angry Flower.
Thinking out loud: the thing about 'must include non-alpha' is that it essentially forces the users to pick non-dictionary words. That's good all by itself. Sure, some of them will just use 'password1' or whatever, which is still dictionary-able (but not much *more* so, since they're probably going to pick the word they always choose anyway and just add a number). And with many users, you'll get stuff that's somewhat hard to do a dictionary attack on, like 'jack4betty' or 'y311ow'.
Does this make any sense? I mean, I can see how suboptimal use provides no further protection, but is it likely to reduce the keyspace much in a real world scenario?
I think the parent is probably referring to some of the pictures on google's early hardware photos page, courtesy of the wayback machine. If so, the lego never necessarily went into `production', it was just when they were messing around.
Also, Microsoft often seem reluctant to go after people for interface copying - remember fvwm95, a window manager for X which cloned (at the pixel level, mostly) the windows 95 look'n'feel?
Well, back in '98 or so when I first visited x.org, it was the website of the `x consortium', which seemed to be a steering group with the various companies that ship an X server implementation as members, and Xfree86 as an `honorary member'. It seems like x.org is the same group, or at least related.
- lack of alpha blending
- lack of native understanding of antialiasing, shading etc in the drawing primitives
- insufficient speed to do things like play video (video etc usually seem to need to do some kind of direct hardware access to play smoothly)
- X forgets a window's contents when it's covered over - try moving a large window over the front of a browser window quickly and see the slow redraw going on behind.
- flickery movement - I don't know if this is a double-buffering type problem or what, but it seems to be difficult to animate large primitives without them flickering as they move.
- no abstraction of GUI toolkits: this one's debatable, but in the early (athena etc) widget sets, buttons were boxes or circles with text in them. Now they're big, coloured, shaded items that animate when clicked, but all of this must be sent over the X protocol as a series of bitmaps with some font information. I think it's time that some higher-level information was sent (for example, abstracting an entire widget set so it can be rendered client side).
There are fixes for some of this stuff (see the Xrender extensions, for example) but they will only really be extensions - Xrender degrades gracefully into standard X, which is nice, but this sort of cruft will keep building up.So, the graphics aren't completely vector based, but the icons are - in that screenshot, the icons in the Icon Catalog and the one at top left are vector-based. The three icons at top-right are bitmaps though (they're minimised applications, which always seem to get bitmap icons).
One interesting thing is that 4DWM recently got antialiasing on the SVG icons, which looks pretty sweet.
- There's no fixed hours - it depends on the job, obviously, but lots of people seem to show up pretty casually.
- Zero dress code - eccentricity is praised rather than condemned, and no-one bats an eyelid if you wander around barefoot in heavy metal T-shirts and bright blue hair.
- Self-motivated work - there aren't any bosses prowling the cube-farm looking for slackers, so you have to have self-discipline to get anything done
These are all observations as a research student working with and around employees at my university, so I may be somewhat inaccurate.I appreciate your point - that it's pretty useless to try and do something like that - but yeah, I do think it's kind of sad if something fun or cool can't be done because of some long-irrelevant decision to keep secrets from people.
Your IBM PC won't run linux because the hardware doesn't have the functionality. In my imaginary example, Froonix would run if only the drivers had been shared with the community. It's slightly different.
james
So I've argued a couple of points:
- Closed drivers can cause problems for users - they're just inherently less useful than the same code in open form
- Users are sad when they have problems
So this is why I'm sad about the Centrino drivers probably being proprietary. I'm not actually differing much in opinion from you, it's more an emotional response. We both understand why corporations often keep their drivers proprietary, but it bothers me, and I hope I've explained why.It's sad because Intel could have released open drivers, and (possibly) it's now less likely that anyone will bother trying to write open implementations.
Open drivers are better than proprietary ones because there are no disadvantages, and there exist advantages such as allowing Linux distributions to contain them by default, the kernel to include them, and any potential bugs to be fixed by concerned users, rather than fretfully waiting for the landlordesque 'owner' to fix it for you.
Proprietary software is still OK, still useful, and still (as CowboyNeal said) better than nothing. However, one of the most exciting things about Linux is the spirit of open charity and freedom embodied in sharing and cooperating. When someone releases linux driver code which is not open, it saddens those who like the usual open nature of Linux drivers.
Imagine, years from now, that Intel has died, and Linux has died. You find an old 'centrino' laptop, and want to make it do something cool. You find a cool operating system, let's call it 'Froonix', and boot it up. But when you try out certain features of the laptop, they don't work. There's some message about 'proprietary drivers' that couldn't be ported from Linux, a predecessor of Froonix. It's sad, right?
Keeping drivers proprietary, which usually means they are tied to one hardware and software platform forever, is shortsighted and irresponsible... I've thrown out a PCMCIA network card, for example, because its drivers were only written for windows 95. The company has no interest in writing any more (it's a discontinued product) and yet the still-useful hardware cannot work with any current OS. Because of some ridiculous desire to keep an obsolete trade secret or two.
You may not agree, of course (from your post it looks like you don't), but can you understand why it saddens certain people?
Hmm... alas, my MS natural keyboard doesn't have the front feet. That sounds like a nice feature.
/james
Coolpad: Thanks, I didn't know about that feature! I've only ever seen photos of it being used at a sharp towards-the-user angle, which makes my wrists ache just thinking about it. I approve much more of it now...
- Ethernet cable
- TV- and Audio- breakout cables, for watching anime at friends' houses
- Digital Camera adapter: I have a USB card-reader, as it's faster and easier than my digicam's serial interface, but the sync cable would be ok too.
- Kensington lock
In addition, here's what I'd kinda like to have but don't:Hope this helps!