Hmm... FWIW, I just upgraded from VMWare Server Beta (whatever version was around last week) to the full release, and it seems noticeably faster when writing out from the Windows VM to the Linux host via Samba. It's still not quite as fast as doing it locally - perhaps a 50% performance hit? - but acceptable. Maybe it's the deactivation of the debugging lines, maybe I did something different this install. I'll have to try playing movies again.
Regarding #3 (modern h/w is fast enough for almost any app), I thought the same. What I discovered was that while the CPU efficiency is fine, writing out of the VM to the host (e.g. Samba share through a local-only subnet, VMWare's virtual folders) is pretty slow. This made it annoying to work with large files.
Also, playing movies inside a VM doesn't work that well. My best result was running VLC inside a Windows VM, but there was noticeable flicker. Music was okay, but tended to skip a little when doing other activities.
In the end, I decided to use my host OS (Ubuntu) as normal, and run the Windows VM for key Windows apps (like my employer's remote access software). I also intend to use a VM to replace my headless Debian server running on an old K7 (not used for workstation/desktop type activities).
Amusing anecdote, I'd swear the virtualized Windows runs faster than the Windows host OS it replaced. It's probably the cleanliness of the registry, etc.
The above is all on a P4 2.8 (Prescott) with 3GB RAM, btw.
I like and use Eclipse TPTP as well. I've used OptimizeIt before and fiddled with JProbe before settling on Eclipse TPTP.
Subjectively, I liked JProbe's interface more, but they only supported Java 1.4 at the time, whereas OptimizeIt and Eclipse TPTP supported 1.5 (which I was using).
I find Eclipse TPTP a little finnicky to set up though. I've set it up two or three times, and I must not be understanding the instructions properly - they're very detailed and specific, but it doesn't go smoothly for me.
Using language as the sole or even main criteria probably isn't as useful as one might think.
FYI my recent background, I spent two years doing C# (.NET 1.1), and the two years prior to that in Java (1.4), and some Java 1.5 most recently. In all cases it's mostly server-side stuff (messaging, database, etc.) - can't stand the vagarities of UI subjectiveness.
It's probably more interesting what industry you're targeting. e.g. I see a lot more Java used for heavy duty stuff than C#, in financial software; but.NET has a better enforced security/licensing framework if you're selling software components.
I've also noticed that Java has more interesting 3rd party stuff (due to maturity? popularity?) that you can learn good ideas from or leverage: e.g. Hibernate, all of Apache's stuff, JBoss, Spring, Struts, etc. A lot of these are (being?) ported over to the.NET world, but some haven't. Also, Java's maturity means googling may be a little more fruitful.
From a tools perspective, I must regretfully insist that Eclipse still has the edge over VS. Even if I accepted that VS2005 finally caught up last month when it was released (I haven't explored it thoroughly), Eclipse's plugin architecture and maturity mean that there's a lot of very useful add-ons for Eclipse.
For UI, I'd have to give it to C# (on Windows). People have expectations of what UI's supposed to be like, and that's Windows. C#'s UI API closely aligns with Win32, whereas Java's Swing... SWT looks nice, never used it, no idea how much work it is.
If language must be discussed...
Java has the "feel" of a precise, theoretically correct, and somewhat obstinate authour. Some things are tedious to do, but there's an overall consistency - there's a certain exacting style, and if it's inconvenient, that's a small price for consistency/"purity".
C#, OTOH, has a feel of pragmatism, and occasionally a little bit of hurry (or parallel development without enough collaboration?). Sometimes the APIs aren't quite consistent (I like comparing C# collections to Java collections for this), and there are a lot of convenience features/concessions. C#'s more modern, so there's built-in support for commonly accepted notions like Properties, Events, and function pointers.
In an effort to be useful, I'll try to make some recommendations. Given the contextual void, these recommendations are suspect, but anyway...
For strictly academic, go Java, then C#. Java's maturity means a larger knowledge base, and Java will probably teach you better style, and you'll appreciate the conveniences that C# gives whilst having learned to avoid abusing it. It's like knowing some of the pain of memory allocation in C to appreciate a GC, and understand memory well enough to know the GC's limitations.
For front-end heavy apps, or pretty apps, or very custom apps, go C# (on Windows). C# generates better looking UIs, and if you're going to do a lot of custom work, you may as well do it in a convenient language. VS's excellent support for "drag/drop of components" style of development is very useful here.
For server-side heavy software or bigger projects, go Java. In my experience, I've found the less UI or business flow, and the more processing-centric the problem, the better I can leverage libraries, and the more net value I get, despite the overhead cost of using other people's software. Bigger projects also means you might want to buy/get more software targeted at development or deployment, for which Java has a larger # of players (e.g. development tools, middleware platforms, etc.).
For cross-platform, go Java. I'm not familiar with Mono, but I think.NET cross-platform is a mirage. IIRC, Mono doesn't support the same UI classes as C#/Windows. Also IIRC, for some classes, like System.GC, some methods are part of the standardized spec, some aren't (q.v..NET Standard Library Annotated Reference?). Certainly you can transfer lan
My impression wasn't a mass market appeal, but that the movie was a summary of what Whedon had in mind for the series. Given that he only had a couple hours to tell it all, it's understandably superficial and rushed.
As I imagine it, Firefly was shut down while Whedon was still holding many cards close to his chest. The movie felt like those cards were being shown and explained, albeit in a hurried fashion that didn't do as much to build up interest and satisfaction like the t.v. episodes.
So perhaps Wash was slated to die in a better written way, perhaps in episode 45, but we never got close enough to even see the hint coming.
I used to use Trillian (now GAIM). They support the "normal" protocol for MSN and Yahoo, for the basic messaging functions anyway.
However, the "proper" MSN and Yahoo clients also support modes that work when only outgoing traffic to an HTTP port is allowed. I believe MSN switches automatically, for Yahoo it's a separate Connection mode.
My experience with the alternative clients is that they don't support that HTTP protocol. I have to use a SOCKS proxy and configure Trillian and GAIM to connect.
I do software development for a living and a hobby (no t.v.). So I spend a lot of time on a keyboard.
I've been interested in Dvorak for awhile, and I keep a printout of the layout and hotkeys active to flip between QWERTY and Dvorak for practice. I do 100 WPM on the former (measured) and about 20-30 WPM om the latter (guesstimate). I did some typing practice with Dvorak, similar to what one does for QWERTY in typing class, but only got as far as the home row and never got around to doing the rest. I'm still a dabbler in Dvorak.
I find it's not particuarly hard to switch if you don't think about the layout too much. Just like when I type QWERTY, it works better if I think about the words and the fingers and letters take care of themselves. I find that for both Dvorak and QWERTY, if I start thinking about the keyboard layout, I confuse myself. Of course, it's much more pronounced with Dvorak.
Dvorak and vim don't mix well for me. Mentally, vim is more like another keyboard layout that I've learned, rather than atop QWERTY. I think even if I get good at Dvorak, I'll have to use QWERTY for vim. So Dvorak will be more useful for "mass" typing, like a Word document. Relearning hotkeys will also be a pain, so I'll probably always keep a really easy and fast hotkey (e.g. Ctrl-Shift-1) to toggle keyboard layouts.
Expanding into the broader RSI topic, I've done some research and experimentation in that area, too. My original wakeup call was shooting pains from my elbow to the pinkie, while just walking around far away from any computer.
Originally, I found that a split keyboard (original Microsoft) helped. Then pain started coming back (just the wrist area).
I've tried various wrist support products, and I found the best one I've used is a gel-like product that deforms from your pressure on it and stays mostly deformed when you take your hand off. (http://www.gaylainc.com/WristRst.html)
My current opinion is that the Aeron chair is good but pricey, the Maxim keyboard is excellent, and the keyboard tray is a necessity.
The Aeron chair is unusual, because you need to sit "just so" for it to feel comfortable. "Just so" happens to be good for you, so it's like continuously being encouraged to sit properly. If you're stubborn, it won't ever feel comfortable (I often still don't sit properly). The mesh fabric is quite nice for keeping your backside cool.
The Maxim keyboard is very comfortable. The adjustable split isn't a big deal, once you get what you want (they recommend a certain setting anyway). However, the keys have a short depth (discourages the habit of killing a key and having the excess force bounce back on you through your fingers) and are quite sensitive (I've tried without success to depress it gently without actually registering a keystroke). The key layout is still familiar (unlike some other keyboards which shrink the backspace, align the key columns differently, etc.). Best keyboard I've used (out of about a half dozen Microsoft and Logitech keyboards personally, as well as old cheap ones at university computer labs).
It's unfortunate, but any employee is automatically assumed to represent the company to some degree by the external public. It shouldn't be so, but the public mind isn't that discriminating.
Consider the blogs by technical employees of big name companies like Microsoft or Google. Even if it is just their own personal opinion, I'm sure the execs would be upset if the blogger proclaimed such-and-such upcoming product sucks, and rightfully so: it'd be all over/. in a second that some MS employee is making negative comments about the next version of their product. Do you think everyone would carefully consider that these are personal opinions and shouldn't be assumed to be the company's view?
Or to phrase it differently, the mass mind loves and believes anecdotes, even if they make no sense in the larger context. A single data point (one employee in a company) will have a disproportionate influence.
I wonder if Google is letting it remain easy to make spam blogs w/ Blogger in order to get more data samples, to fine-tune their filters? i.e. replicate the internet problem in the small, with controllable parameters.
After all, why run through the entire gamut of blog styles and presentation formats, when you can just examine content-only from your own servers.
I would definitely regret ignoring blogs. I'm a software developer, and getting into the head of other software people and trying to divine their reasoning and thought patterns is enlightening for me. It does take some hits and misses to find people you're interested in learning from.
For me, blogs are a substitude for the old mentor-student pairing of older craft-type professions. I can't find anyone to just sit down and teach me the "tricks of the trade", but at least I can read about a few.
It's also like how people say learning many languages helps you with the ones you already know (human and programming languages). So learning how someone else attacked a problem is very instructive.
I think a mistake in this argument is treating the OS for the x86, the XBox, and portable devices as the same.
While they may have the same roots, they in no way replace one another. If everyone had an XBox 360, that would diminish the gaming PCs of the world, but that would be small compared to the corporate OSs in the world. Also, people tend to bring home what they're used to, so it's unlikely they'd eschew a regular PC for surfing the web and email for a web-enabled XBox.
I don't think it's so much about Microsoft trying to sever their ties with x86, so much as they are trying to leverage what they have experience in (OS software) into other markets.
Dominating most of a market limits your growth potential. Intel and Apple are doing the same. They all want to grow beyond their existing captive markets, and are willing to risk a little shake up with old friends and enemies to get it.
Depends on how static the background window is. I found it useful for similar reasons as you cite, but not so useful if I'm tail'ing some log.
The constant change with alpha blending on does shoot up the CPU noticeably.
It's also noticeable if you have it set so that a window being moved around still repaints as it moves (as opposed to just getting a border outline that moves). Same problem, having to frequently alpha blend a lot of pixels quickly.
I think that going this route is a reasonable way to try to break into the mainstream. The non-technical computer buyer is likely to recognize the Intel trademark, and associate that with other computers he's seen.
Of course, Windows has even higher recognition, but that's not really compatible with Apple.
An OSX box with an Intel Inside sticker is perhaps more reassuring to your non-technical consumer than without. Apple can capitalize a little on the sunk costs of years of Intel marketing for free.
That's true, they're obligated to "play nice". However, in all fairness, nearly all countries do this kind of thing to varying degrees. Protectionist policies, tariffs, subsidies, etc. Banning a country's product outright is just blatant, but not the only way to squeeze out foreigners.
It's a hard choice (as one might imagine running a country would be). Countries theoretically engage in trade when there's a mutual advantage in doing so (you produce this for cheaper than I can, or opportunity cost, so here's money for it), but the money exits the country. So is it better to get the best bang for your buck, or take something worse knowing that at least the money goes on to feed your citizens and stimulate the economy?
IMO, the firefox mouse gestures are fine (I use all-in-one), but I have to stick to the basics (e.g. close window, refresh). I use Opera at home and Firefox at work, and while they have some overlap in the default gestures, they're not a perfect match.
I do like being able to see the gesture I draw with Firefox, though.
I would actually think it's an economic decision. Unlike more tangible products, software can be made anywhere. The money the Chinese government hands over goes directly into the foreign economy.
On the other hand, if it goes to a Chinese software company (or company based in China), the money spent on the software will go to local employees, who will spend it buying local groceries, etc. Lots of Chinese citizens get income from that one government expenditure, i.e. the multiplier effect.
OTOH, while I can undertand a preference for Chinese companies, a blanket policy banning foreign companies seems a bit silly, considering the maturity of the Chinese industry. Extra consideration for being local, but still looking at one's needs, seems more appropriate.
If your company runs an internal DNS server/relay, won't they be able to log the sites you're going to? They may not be able to see the content, but they might wonder why you're trying to hit a banned site so often.
I know, what I meant was that the HTML for Google's version of its page for Palm is already different, so why not just drop the prefetch hints for that version of the page?
So if I go to google.com/palm and make a query, why should the search result page that I get back have prefetch hints?
We already A) know it's someone who wants low bandwidth, and B) generate specialized HTML for Palm browsers anyway.
If you're on a low bandwidth connection, what about using www.google.com/palm ? Presumably someone at Google has/will make the connection that if you're going through there, prefetching probably isn't a good idea.
Well, if they provide *certain* OS capabilities (e.g. filesystem) wouldn't that go a long way in getting everyone's information?
Obviously not implemented as a remote filesystem with nothing local, but what about (as something that's been done before) a system to sync certain files, say C:\Documents And Settings or/home?
If you can already get your email via the web and search it, wouldn't it be nice to access your key files (securely) on the web and search that as well (a la Google Desktop)?
And talk about safety of data with their system of redundancy.
How does automatic versioning of all your files sound to you?
I think they could do a lot of interesting things that could supplant existing OS functionality without *truly* (i.e. in the technically precise sense) replacing an OS.
Slashdot Mirror
Hmm... FWIW, I just upgraded from VMWare Server Beta (whatever version was around last week) to the full release, and it seems noticeably faster when writing out from the Windows VM to the Linux host via Samba. It's still not quite as fast as doing it locally - perhaps a 50% performance hit? - but acceptable. Maybe it's the deactivation of the debugging lines, maybe I did something different this install. I'll have to try playing movies again.
Regarding #3 (modern h/w is fast enough for almost any app), I thought the same. What I discovered was that while the CPU efficiency is fine, writing out of the VM to the host (e.g. Samba share through a local-only subnet, VMWare's virtual folders) is pretty slow. This made it annoying to work with large files.
Also, playing movies inside a VM doesn't work that well. My best result was running VLC inside a Windows VM, but there was noticeable flicker. Music was okay, but tended to skip a little when doing other activities.
In the end, I decided to use my host OS (Ubuntu) as normal, and run the Windows VM for key Windows apps (like my employer's remote access software). I also intend to use a VM to replace my headless Debian server running on an old K7 (not used for workstation/desktop type activities).
Amusing anecdote, I'd swear the virtualized Windows runs faster than the Windows host OS it replaced. It's probably the cleanliness of the registry, etc.
The above is all on a P4 2.8 (Prescott) with 3GB RAM, btw.
I like and use Eclipse TPTP as well. I've used OptimizeIt before and fiddled with JProbe before settling on Eclipse TPTP.
Subjectively, I liked JProbe's interface more, but they only supported Java 1.4 at the time, whereas OptimizeIt and Eclipse TPTP supported 1.5 (which I was using).
I find Eclipse TPTP a little finnicky to set up though. I've set it up two or three times, and I must not be understanding the instructions properly - they're very detailed and specific, but it doesn't go smoothly for me.
Using language as the sole or even main criteria probably isn't as useful as one might think.
.NET has a better enforced security/licensing framework if you're selling software components.
.NET world, but some haven't. Also, Java's maturity means googling may be a little more fruitful.
.NET cross-platform is a mirage. IIRC, Mono doesn't support the same UI classes as C#/Windows. Also IIRC, for some classes, like System.GC, some methods are part of the standardized spec, some aren't (q.v. .NET Standard Library Annotated Reference?). Certainly you can transfer lan
FYI my recent background, I spent two years doing C# (.NET 1.1), and the two years prior to that in Java (1.4), and some Java 1.5 most recently. In all cases it's mostly server-side stuff (messaging, database, etc.) - can't stand the vagarities of UI subjectiveness.
It's probably more interesting what industry you're targeting. e.g. I see a lot more Java used for heavy duty stuff than C#, in financial software; but
I've also noticed that Java has more interesting 3rd party stuff (due to maturity? popularity?) that you can learn good ideas from or leverage: e.g. Hibernate, all of Apache's stuff, JBoss, Spring, Struts, etc. A lot of these are (being?) ported over to the
From a tools perspective, I must regretfully insist that Eclipse still has the edge over VS. Even if I accepted that VS2005 finally caught up last month when it was released (I haven't explored it thoroughly), Eclipse's plugin architecture and maturity mean that there's a lot of very useful add-ons for Eclipse.
For UI, I'd have to give it to C# (on Windows). People have expectations of what UI's supposed to be like, and that's Windows. C#'s UI API closely aligns with Win32, whereas Java's Swing... SWT looks nice, never used it, no idea how much work it is.
If language must be discussed...
Java has the "feel" of a precise, theoretically correct, and somewhat obstinate authour. Some things are tedious to do, but there's an overall consistency - there's a certain exacting style, and if it's inconvenient, that's a small price for consistency/"purity".
C#, OTOH, has a feel of pragmatism, and occasionally a little bit of hurry (or parallel development without enough collaboration?). Sometimes the APIs aren't quite consistent (I like comparing C# collections to Java collections for this), and there are a lot of convenience features/concessions. C#'s more modern, so there's built-in support for commonly accepted notions like Properties, Events, and function pointers.
In an effort to be useful, I'll try to make some recommendations. Given the contextual void, these recommendations are suspect, but anyway...
For strictly academic, go Java, then C#. Java's maturity means a larger knowledge base, and Java will probably teach you better style, and you'll appreciate the conveniences that C# gives whilst having learned to avoid abusing it. It's like knowing some of the pain of memory allocation in C to appreciate a GC, and understand memory well enough to know the GC's limitations.
For front-end heavy apps, or pretty apps, or very custom apps, go C# (on Windows). C# generates better looking UIs, and if you're going to do a lot of custom work, you may as well do it in a convenient language. VS's excellent support for "drag/drop of components" style of development is very useful here.
For server-side heavy software or bigger projects, go Java. In my experience, I've found the less UI or business flow, and the more processing-centric the problem, the better I can leverage libraries, and the more net value I get, despite the overhead cost of using other people's software. Bigger projects also means you might want to buy/get more software targeted at development or deployment, for which Java has a larger # of players (e.g. development tools, middleware platforms, etc.).
For cross-platform, go Java. I'm not familiar with Mono, but I think
My impression wasn't a mass market appeal, but that the movie was a summary of what Whedon had in mind for the series. Given that he only had a couple hours to tell it all, it's understandably superficial and rushed.
As I imagine it, Firefly was shut down while Whedon was still holding many cards close to his chest. The movie felt like those cards were being shown and explained, albeit in a hurried fashion that didn't do as much to build up interest and satisfaction like the t.v. episodes.
So perhaps Wash was slated to die in a better written way, perhaps in episode 45, but we never got close enough to even see the hint coming.
I used to use Trillian (now GAIM). They support the "normal" protocol for MSN and Yahoo, for the basic messaging functions anyway.
However, the "proper" MSN and Yahoo clients also support modes that work when only outgoing traffic to an HTTP port is allowed. I believe MSN switches automatically, for Yahoo it's a separate Connection mode.
My experience with the alternative clients is that they don't support that HTTP protocol. I have to use a SOCKS proxy and configure Trillian and GAIM to connect.
I do software development for a living and a hobby (no t.v.). So I spend a lot of time on a keyboard.
I've been interested in Dvorak for awhile, and I keep a printout of the layout and hotkeys active to flip between QWERTY and Dvorak for practice. I do 100 WPM on the former (measured) and about 20-30 WPM om the latter (guesstimate). I did some typing practice with Dvorak, similar to what one does for QWERTY in typing class, but only got as far as the home row and never got around to doing the rest. I'm still a dabbler in Dvorak.
I find it's not particuarly hard to switch if you don't think about the layout too much. Just like when I type QWERTY, it works better if I think about the words and the fingers and letters take care of themselves. I find that for both Dvorak and QWERTY, if I start thinking about the keyboard layout, I confuse myself. Of course, it's much more pronounced with Dvorak.
Dvorak and vim don't mix well for me. Mentally, vim is more like another keyboard layout that I've learned, rather than atop QWERTY. I think even if I get good at Dvorak, I'll have to use QWERTY for vim. So Dvorak will be more useful for "mass" typing, like a Word document. Relearning hotkeys will also be a pain, so I'll probably always keep a really easy and fast hotkey (e.g. Ctrl-Shift-1) to toggle keyboard layouts.
Expanding into the broader RSI topic, I've done some research and experimentation in that area, too. My original wakeup call was shooting pains from my elbow to the pinkie, while just walking around far away from any computer.
Originally, I found that a split keyboard (original Microsoft) helped. Then pain started coming back (just the wrist area).
I tried out the Smart Glove (http://www.imakproducts.com/products/smart_glove. htm) and got good results from that (it also helps with using laptops, where other ergo products can't be used).
I've tried various wrist support products, and I found the best one I've used is a gel-like product that deforms from your pressure on it and stays mostly deformed when you take your hand off. (http://www.gaylainc.com/WristRst.html)
Once I had the money to, I decided that just for the chance of improvement, it was worth sinking the money into possible hoopla. So I bought a couple Kinesis Maxim keyboards (http://www.kinesis-ergo.com/max-spec.htm) and an Aeron chair (http://www.hermanmiller.com/CDA/SSA/Product/0,159 2,a10-c440-p8,00.html), as well as a keyboard tray from Kinesis (I also have a contoured Logitech mouse, if that makes a difference).
My current opinion is that the Aeron chair is good but pricey, the Maxim keyboard is excellent, and the keyboard tray is a necessity.
The Aeron chair is unusual, because you need to sit "just so" for it to feel comfortable. "Just so" happens to be good for you, so it's like continuously being encouraged to sit properly. If you're stubborn, it won't ever feel comfortable (I often still don't sit properly). The mesh fabric is quite nice for keeping your backside cool.
The Maxim keyboard is very comfortable. The adjustable split isn't a big deal, once you get what you want (they recommend a certain setting anyway). However, the keys have a short depth (discourages the habit of killing a key and having the excess force bounce back on you through your fingers) and are quite sensitive (I've tried without success to depress it gently without actually registering a keystroke). The key layout is still familiar (unlike some other keyboards which shrink the backspace, align the key columns differently, etc.). Best keyboard I've used (out of about a half dozen Microsoft and Logitech keyboards personally, as well as old cheap ones at university computer labs).
It's unfortunate, but any employee is automatically assumed to represent the company to some degree by the external public. It shouldn't be so, but the public mind isn't that discriminating.
/. in a second that some MS employee is making negative comments about the next version of their product. Do you think everyone would carefully consider that these are personal opinions and shouldn't be assumed to be the company's view?
Consider the blogs by technical employees of big name companies like Microsoft or Google. Even if it is just their own personal opinion, I'm sure the execs would be upset if the blogger proclaimed such-and-such upcoming product sucks, and rightfully so: it'd be all over
Or to phrase it differently, the mass mind loves and believes anecdotes, even if they make no sense in the larger context. A single data point (one employee in a company) will have a disproportionate influence.
I wonder if Google is letting it remain easy to make spam blogs w/ Blogger in order to get more data samples, to fine-tune their filters? i.e. replicate the internet problem in the small, with controllable parameters.
After all, why run through the entire gamut of blog styles and presentation formats, when you can just examine content-only from your own servers.
I would definitely regret ignoring blogs. I'm a software developer, and getting into the head of other software people and trying to divine their reasoning and thought patterns is enlightening for me. It does take some hits and misses to find people you're interested in learning from.
For me, blogs are a substitude for the old mentor-student pairing of older craft-type professions. I can't find anyone to just sit down and teach me the "tricks of the trade", but at least I can read about a few.
It's also like how people say learning many languages helps you with the ones you already know (human and programming languages). So learning how someone else attacked a problem is very instructive.
I think a mistake in this argument is treating the OS for the x86, the XBox, and portable devices as the same.
While they may have the same roots, they in no way replace one another. If everyone had an XBox 360, that would diminish the gaming PCs of the world, but that would be small compared to the corporate OSs in the world. Also, people tend to bring home what they're used to, so it's unlikely they'd eschew a regular PC for surfing the web and email for a web-enabled XBox.
I don't think it's so much about Microsoft trying to sever their ties with x86, so much as they are trying to leverage what they have experience in (OS software) into other markets.
Dominating most of a market limits your growth potential. Intel and Apple are doing the same. They all want to grow beyond their existing captive markets, and are willing to risk a little shake up with old friends and enemies to get it.
Depends on how static the background window is. I found it useful for similar reasons as you cite, but not so useful if I'm tail'ing some log.
The constant change with alpha blending on does shoot up the CPU noticeably.
It's also noticeable if you have it set so that a window being moved around still repaints as it moves (as opposed to just getting a border outline that moves). Same problem, having to frequently alpha blend a lot of pixels quickly.
I think that going this route is a reasonable way to try to break into the mainstream. The non-technical computer buyer is likely to recognize the Intel trademark, and associate that with other computers he's seen.
Of course, Windows has even higher recognition, but that's not really compatible with Apple.
An OSX box with an Intel Inside sticker is perhaps more reassuring to your non-technical consumer than without. Apple can capitalize a little on the sunk costs of years of Intel marketing for free.
That's true, they're obligated to "play nice". However, in all fairness, nearly all countries do this kind of thing to varying degrees. Protectionist policies, tariffs, subsidies, etc. Banning a country's product outright is just blatant, but not the only way to squeeze out foreigners.
It's a hard choice (as one might imagine running a country would be). Countries theoretically engage in trade when there's a mutual advantage in doing so (you produce this for cheaper than I can, or opportunity cost, so here's money for it), but the money exits the country. So is it better to get the best bang for your buck, or take something worse knowing that at least the money goes on to feed your citizens and stimulate the economy?
IMO, the firefox mouse gestures are fine (I use all-in-one), but I have to stick to the basics (e.g. close window, refresh). I use Opera at home and Firefox at work, and while they have some overlap in the default gestures, they're not a perfect match. I do like being able to see the gesture I draw with Firefox, though.
I would actually think it's an economic decision. Unlike more tangible products, software can be made anywhere. The money the Chinese government hands over goes directly into the foreign economy.
On the other hand, if it goes to a Chinese software company (or company based in China), the money spent on the software will go to local employees, who will spend it buying local groceries, etc. Lots of Chinese citizens get income from that one government expenditure, i.e. the multiplier effect.
OTOH, while I can undertand a preference for Chinese companies, a blanket policy banning foreign companies seems a bit silly, considering the maturity of the Chinese industry. Extra consideration for being local, but still looking at one's needs, seems more appropriate.
If your company runs an internal DNS server/relay, won't they be able to log the sites you're going to? They may not be able to see the content, but they might wonder why you're trying to hit a banned site so often.
I know, what I meant was that the HTML for Google's version of its page for Palm is already different, so why not just drop the prefetch hints for that version of the page? So if I go to google.com/palm and make a query, why should the search result page that I get back have prefetch hints? We already A) know it's someone who wants low bandwidth, and B) generate specialized HTML for Palm browsers anyway.
If you're on a low bandwidth connection, what about using www.google.com/palm ? Presumably someone at Google has/will make the connection that if you're going through there, prefetching probably isn't a good idea.
Well, if they provide *certain* OS capabilities (e.g. filesystem) wouldn't that go a long way in getting everyone's information?
/home?
Obviously not implemented as a remote filesystem with nothing local, but what about (as something that's been done before) a system to sync certain files, say C:\Documents And Settings or
If you can already get your email via the web and search it, wouldn't it be nice to access your key files (securely) on the web and search that as well (a la Google Desktop)?
And talk about safety of data with their system of redundancy.
How does automatic versioning of all your files sound to you?
I think they could do a lot of interesting things that could supplant existing OS functionality without *truly* (i.e. in the technically precise sense) replacing an OS.