Several people have already posted comments asking why the world needs yet another Linux distro.
I wrote a Slashdot comment explaining why Ubuntu is interesting. Click here to read it.
A comment by Doc Ruby states that Ubuntu is not package-compatible with Debian. I said otherwise in my comment linked above, but I haven't checked it out for myself yet so I'm probably wrong.
Take a gasoline powered leaf blower. Scale it up so that it makes a lot of wind. Now build about five of these and attach them to a car-sized vehicle.
If you bring one of those home, I don't want to be your next-door neighbor.
To be fair, Moller is proposing an engine that would be more efficient (and thus quieter) than the cheap two-stroke engine in a leaf blower. But he still needs a boat-load of thrust per engine to lift the car, its fuel, and its passengers and their luggage, plus extra power to handle engine failure.
The last time I looked at Moller's web site (years ago) it said something hand-wavy about active noise cancellation. It sounded good at the time but now I'm dubious; moving as much air as this thing will move, how can you ever get it really quiet? (And how much will the noise cancellation gear weigh?) Noise cancellation works great if you are sitting in a predictable spot, but how can you cancel the noise of the flying car for all listeners in all directions?
I'd love flying cars, with safe autopilots please, but I think that unless someone invents antigravity they won't be practical.
On Earth, anyway... A Miracle of Science showed flying cars inside the domed cities of Luna. With one-sixth the gravity of Earth, flying cars might be more reasonable.
P.S. I'd like to plug A Miracle of Science. It's a nifty hard-science SF story that nonetheless has a sense of wonder. It's one of my favorite web comics.
we have a pretty damn effective electricity transport system already in place.
Well, to get power to your house, yes. But we don't have a really practical way to run cars with it. Current electric cars have limited range, and when the batteries wear out you have to buy new ones and safely dispose of the old ones.
If BMW has managed to make a practical hydrogen car, good for them and I hope they get rich. But I'm dubious.
are they adding content which legitimately justifies the price they are charging on its own
They are. That's the whole idea of "Twisted Flicks".
They show the movie without sound, and they take suggestions from the audience. "Okay, I need the name of a place. Sedro Wooley. Okay, now I need a profession. Dressmaker." And so on.
Then improv comedy guys then use the suggestions from the audience to make something up on the fly. They have to think on their feet and make funny stuff up rapidly. What you are really paying for is the improv.
Really, they can use any movie to do this. But there is potentially extra humor value if the movie is well-known; if the movie is itself part of pop culture, then sly pop-culture references might be easy to slip in.
I'm not a lawyer, but I don't think these guys can hide behind the parody protection laws. They aren't really looking to specifically make fun of the movie, they just need a starting place to hang their improv humor.
They are showing a movie in a theatre, so they need permission from whoever owns the movie.
You have a hydrogen leak, and someone walks in and flips on the light switch...
I hope that the vent system will have a little pilot light or sparker or something, and will burn up the wasted hydrogen. Maybe even run the waste hydrogen through a fuel cell and charge a storage battery?
Except for your enclosed garage scenario, even unburned waste H2 should be safe, because it's lighter than air, so it will disperse quickly. If it settled to the ground in a pool, that would be bad. But even gases like propane, which can settle in a pool, rarely explode because they dissipate quickly. But yeah, contain them and ignite them and you are having a bad day.
I hope, if they're going to do this, they're at least going to have the sense to perfume the hydrogen, like they do natural gas
I don't think this will work. The whole point of the liquid hydrogen is that it is much much much more compact than gaseous hydrogen. Is there a scent you could add, that would be potent enough you could smell it when one milliliter of liquid hydrogen vaporises into one liter of gas? And would this potent smell chemical cause problems for the fuel cells, or make nasty chemicals when burned in the engine?
We have absolutely no way of getting hydrogen, outside of fossil fuels, that doesn't use up more electricity than we put into it. I've never heard of any way even proposed to get said hydrogen.
The FA proposes to use solar power to get the hydrogen out of water. This will work. You can put large solar plants in places where the sun shines a lot, and transport the hydrogen to where you need it.
The problem is that I'm dubious about liquid hydrogen as the way to transport it. Suppose you build a bunch of solar collectors out in, say, Mexico... or maybe you make solar collectors that just float on the ocean in an out-of-the-way place. Great. But can you keep it cold enough to still be liquid all the way from there, to the local fuel station where the customer buys it? Will you lose a bunch to boil-off as the liquid warms, or perhaps burn up a bunch to run refrigerators to chill it? Will you lose so much that it drives the price up? In short, is this really practical?
The other problem is that if this becomes really popular, I'm not sure solar can scale up enough to provide all the hydrogen people will want. Well, there's always the nuclear option.
I keep hoping someone will invent some magical fuel tank that will somehow really lock the hydrogen in. Maybe little microcells that each store a small quantity or something.
Our biggest problem isn't getting energy, it's getting energy exactly where you want it. If hydrogen becomes practical, you could put a power plant just anywhere and use hydrogen to bring the power wherever it's needed. And the hydrogen part wouldn't pollute anything.
The little "wires" connecting things inside a chip are called "traces", and are the thinnest things you can draw on a chip. So "trace size" is used as a yardstick to compare chips for how small their various parts can be. (You will often see the word "process", as in "90 nanometer process" instead of "90 nanometer traces"). The smaller the trace size, the less power dissipation. Both of these new chips have a small trace size, although the articles didn't say exactly how small.
If you make a chip, and then you improve your chip-making technology so that you can draw thinner traces, you can perform a "die shrink": you produce a similar chip design, using the smaller traces, and the whole chip takes up a smaller amount of space. You can buy some graphics cards that are basically an older GPU with a die shrink; they dissipate very little heat and are inexpensive. A GeForce4 MX (budget card) was basically a GeForce 3 with a die shrink, IIRC.
Chips are made from silicon wafers. One whole wafer is "fabbed" (made in a chip fabrication plant), then cut up for individual chips. The more chips per wafer, the cheaper each chip is. (This is all the more true because flaws can happen during the fabbing process; if one flaw means one dead chip, then more chips per wafer means a similar number of flaws results in a lower percentage of dead chips made, and thus lower costs.)
A smaller trace size makes it easier to push the clock rate higher. But GPUs are definitely clocked lower than CPUs, so that helps them dissipate less heat. If you are pushing a Pentium 4 at 3 GHz, and the GPU is only doing 0.7 GHz, clearly that helps the GPU dissipate less heat.
Smaller trace sizes make it harder to make the chip work right; the smaller the traces, the more problems you might have electrically (I don't fully understand the details). Also, you need to be more careful with cooling; a hot chip with a tiny die size needs a really good heatsink, and there is less margin for error. The old, relatively large chips like the 486 family were easy to cool in comparison with today's chips.
Summary:
Smaller trace size means less heat Lower clock rate means less heat
To me the big deal is that I'm getting some CDs in the mail that I can give away to people that I don't have to burn.
That's not a big deal to me, because my CD burner works well and I buy 100-packs of CDs. I can burn as many CDs as my friends will ever need.
What's a big deal to me is a distribution that will be good for newbies, but still 100% free software, so I'm allowed to give away CDs. Since I use Debian, I want a Debian-based distro that is both 100% free and newbie-friendly; it looks like this is it.
It looks to me that Debian is cracking under it's own weight.
I don't agree. Debian works about the same as it always has. The people, like me, who use Debian are all as happy as we have always been.
My point was that Ubuntu is a great complement to the Debian project. Nothing is perfect for everyone; Debian isn't perfect for everyone. Ubuntu will be better for many people.
Debian seems to have too many packages, too many developers
These go together. Most of the Debian developers have a very small number of packages to manage, which is why they are willing to do it for free.
In my opinion, that might make Debian work.
I wouldn't put it this way, since Debian works quite well for me. But we agree that Ubuntu will be a good thing for many people.
Well, OOo is now a major component of a modern, powerful Linux desktop, so it should be taken into account.
My point is that it does no good to complain to the GNOME developers that OOo is slow. The GNOME developers do not develop OOo.
Re: WinXP, yes, out of the box it's a pig. But just defaulting to classic theme, turning off system restore, disabling the theming and file-cataloging background processors and it's very fast. Even just about usable in 64M!
Well, if you run IceWM, and use Rox Filer for your file manager, and run AbiWord and Gnumeric instead of OOo, I'll bet you could be happy in 64M. You would still be "using GNOME" (the libraries anyway) to run your GNOME apps.
Or you could just buy some more RAM and run full throttle GNOME, which is my choice.
Who cares about Yet Another Distro? What's different about this one?
It's Debian, in a friendly wrapper, free and for free.
You can get Debian in a friendly wrapper by buying Xandros, or Linspire. They include nonfree software, and Linspire hooks you in to a software distribution scheme that costs a minimum of $50 per year.
You can get Debian free if you are a Linux expert. Get a Debian installer and have fun. However, Debian has 10,000 packages, and you need to know enough to pick and choose which ones you want. Ubuntu has 1,000 packages, and they have made default choices for you. (Want something Ubuntu doesn't offer? Grab it from the main Debian distribution; it will work.)
Also, Debian comes in three major branches: stable, unstable, and testing. Stable is really stable, but only updates every two years or so. Unstable updates daily but can be unstable. Testing updates automatically from unstable when the packages appear stable (a week goes by without major bugs posted against the unstable package, IIRC). Ubuntu on the other hand is promising a six-month release cycle; if you use Ubuntu, you should have a nice stable system, but you will get new packages much more often than if you use Debian Stable.
Ubuntu will occupy a similar niche to Fedora, but Red Hat makes all the decisions for Fedora while Ubuntu will have a community process.
The closest distro to Ubuntu is probably Bruce Perens's UserLinux project. But UserLinux is focused squarely upon business, whilst Ubuntu seems to be more focused on individual users.
Ubuntu should preserve all the things I like best about Debian, while being more friendly to newbies and offering a much fresher stable release. There isn't another distro quite like it.
I'm downloading it now and I look forward to trying it out.
when are they going to concentrate on performance and memory usage? Right now it's _terrible_ - just as bad as Windows XP.
Old rule of thumb: first you make it work, then you make it faster.
GNOME has been following that rule. Nautilus used to be slow, but in recent versions it has become quite snappy.
GNOME has also done their homework; their architecture is pretty much sound. They don't add bloat just because it would make their life easier in some way. For example, GNOME doesn't have its own file open/file save dialogs; it just uses the GTK ones. And DBUS will let them do all sorts of cool stuff (like knowing when a CD is inserted into a drive) without polling the hardware.
OpenOffice.org is slower and heavier than MS Office.
I can't argue with that. But what does that have to do with GNOME? OOo has nothing to do with GNOME. The "GNOME Office" would be AbiWord, Gnumeric, and other GNOME native apps, not OOo.
OOo started out huge and slow. The OOo guys are doing good work on it (compare OOo with StarOffice 5.x sometime!). But the GNOME apps are faster and lighter than OOo.
I mean, I can run Office, IE and Outlook together SMOOTHLY on a WinXP box with 128M RAM.
This does not match my experience. I helped a friend who was running XP on a 128MB computer and it wasn't smooth. Maybe, once everything was started up, and the system had settled down, maybe then. Maybe. (But then run something and the disk thrashes. 128MB isn't enough for XP.)
One thing to realize, however: the people here like to ridicule MS, but they have some good software, and in particular their compilers are good. Their compilers make tighter, faster code than GCC does, and this is a huge advantage when trying to hit a target like "run a bunch of software in 128MB".
There are LOADS of people with 64 and 128M boxes out there who can't run a modern, desktop Linux effectively
Are there really? Last time I checked, RAM was cheaper than Windows XP. I would encourage those people to buy a 256MB stick of RAM and speed up their systems.
GNOME is working on cool new features. They do work on speed and size to some extent, and GCC keeps getting better, and RAM is still cheap. All the new computers, with 2 GHz processors and tons of RAM, will run GNOME very well. In short, I think the GNOME guys are doing a fine job and they do not need to change their priorities at all.
All my data is XORed against itself before it is written to disk.
What a waste of valuable CPU cycles! Here's a speedup that does the same thing much faster:
/* implement "XOR data with itself" security algorithm */ /* but cleverly don't actually use XOR */ /* don't forget to null-terminate encrypted data! */
int CopyWithL337XORSecurity(char *in, char *out) {
int length;
the realist in me says that once Firefox really takes off, we can look forward to people finding security exploits in it too.
But it will still be better.
Internet Explorer is designed to install software on your computer (or else Windows Update wouldn't use it). Firefox is just an application and doesn't have the ability to install software. This, alone, makes a huge difference for spyware.
Firefox does have the ability to install Firefox extensions, which are little scripts that can make Firefox do things. We will need to be vigilant to make sure that there isn't any script-based spyware.
However, I don't think you can get Firefox to install an extension without popping up a dialog box asking for permission, so it's still better than IE.
I don't know if this will work for you, but it's cheap and not risky.
Shut down and unplug the equipment. Crumple up sheets of newspaper, and pack them into the open space of the equipment. Seal it up and give it a day or two, then pull out the newspaper and discard it.
Why do I think this might work? I once bought a used refrigerator. It had been stored with the door closed and the power off, and its inside smelled very bad. I scrubbed the insides with cleaner and let it run for a few days, and it still smelled bad. Someone told me to try stuffing it full of newspaper, close it up, and leaving it for a day. I didn't really think it would work, but it did; the odor was gone. Just gone.
I think that the way it works is just that the fibers of the paper soak up the smell.
Now, if the meat-smelling equipment has actual deposits of meaty chemicals (pork fat or whatever) then I'll bet you will need to wash the equipment with alcohol or something to get rid of it. But if you just have odors soaked into various plastic things, this might work, just as it worked on my old refrigerator.
If you try it and it does work, do please let me know.
The real silver bullet is an effective system of negative feedback. When the schools do a bad job, they need to be punished, and when they do a good job, they need to be rewarded. A simple idea.
Simple, yes, but hard to do in real life. Teachers' unions, educational bureaucracies all the way up to the federal level, politicians making promises... all of these things can complicate the school system to the point where incompetence isn't punished, nor excellence rewarded. And attempts to use standardised tests to guarantee that kids are taught well, just mean that teachers will wind up "teaching the test".
The best thing you can hope to do is to allow parents to move their kids around to the best schools. This will not, itself, fix the problem instantly; but it will introduce an element of feedback into the system. Over time, this will inevitably force the schools to improve.
If a restaurant has poor food, people will take their business to other restaurants. It doesn't matter what kind of union the cooks have, it doesn't matter what kind of promises politicians might have made, etc. If the customers vote with their feet, the better restaurants will prosper and the worst ones will have to close. The same thing would happen with schools, but it would take longer (people eat several meals per day, but they would probably leave their kids in any particular school for at least a few months before deciding to move the kids somewhere else).
I have debated this issue in the past with some people who claimed that parents must not be trusted to choose schools for their kids. That's lunacy. There will be a few bad parents, but by far most parents really want what is best for their kids. The parents and kids together are the best judges of how well a school is serving them.
Note that middle-class and upper-class parents already have some freedom to pick schools; I know my parents, whenever we moved, would carefully consider what the schools were like, and they would only move someplace where the schools were decent. The poorest people, who are trapped in the bad part of town (no money to move somewhere else), those are the ones who really want school vouchers.
By the way, public school systems spend a lot of money per student. The vouchers are generally for less than the public school system would have spent on a student. If a student takes a $3000 voucher and goes to a private school, that is usually a net profit for the public school. In my state, the average per-student spending is $9,454 per year.
Your ideas about grammatical rules is slightly primitive: it's *hard* to check grammar.
My idea was a bit hand-wavy, I admit, but I still think it could work. For coding work, you need experienced software guys; but if you set up the project right, you can have language contributions from non-geeks. Start with existing natural-language tools like NLTK. If you need a dictionary, import one, such as Wiktionary. Set up an easily-parsed format for the rules, and let people submit rules.
The result would be about as good as the Word grammar checker; still brain-dead, still offering comments about things that might be wrong because it can't be sure, but by golly a grammar checker.
You figure it's ten years of work. If you can take advantage of work that has already been done (NLTK), and get help from many people (writing rules), it could shape up much quicker than that.
I looked at all the screenshots, and nothing on there jumped out and bit me and yelled "Windows! IE!" I have no idea what FlipmodePlaya is complaining about.
It looks to me like it's just the GNOME 2.x that I know and love, with subtle, very incremental bits of polish. FlipmodePlaya, perhaps you could be a bit more specific?
P.S. I'm really looking forward to some of the new features, specifically Volume Manager and the new MIME handlers. GNOME 2.8's MIME features won't just be easier to use than previous GNOME versions--they will actually be easier to use than Windows's application association system.
I want a PDA that looks exactly like a Palm Tungsten T2, but has the full-size screen like a T3. (To say the same thing another way, I want a Palm Tungsten T3 with buttons and 5-way navigator exactly like a T2. But I want more battery life than the T3 currently gets.) The PDA screen should be visible in bright light (a "transflective" screen like the T2 and T3 already have).
To go along with the PDA, I want a cell phone with Bluetooth. The cell phone will be clipped in a holster on my belt, just like my ancient StarTac is now. With a hands-free earphone/microphone, and using the PDA to dial via Bluetooth, I could make phone calls without removing the phone from the holster. The phone should be a folding phone, about the same size as my StarTac. (Perhaps a little bit smaller. I don't ever seem to wish that the StarTac were smaller; it seems like the perfect size. But some people might like it better if it were smaller.)
The phone, by virtue of its relatively large battery, should also double as a WiFi relay, talking to the PDA with Bluetooth. If I'm in a place with WiFi I can use that to connect to the Internet, and if not I can use the cell phone.
I also want a Bluetooth modem gadget. I wouldn't carry this everywhere I go, but it would let me connect to the Internet where there is no WiFi but there is a phone jack. For total nirvana, this gadget should also have an Ethernet jack, so that it can double as an Ethernet connector. (You can already carry a small Bluetooth USB dongle, and use that to take advantage of an Ethernet network -- as long as there is a PC there, for which you have permission to plug in a USB gadget and permission to install appropriate drivers. An actual Ethernet jack would be easier.) Again for total nirvana, the modem gadget should use the same exact batteries as the cell phone, and be rechargeable with a Palm travel charger (or you should be able to remove the battery and use the cell phone travel battery charger). The modem gadget should also serve as a WiFi relay (although since I carry a cell phone all the time, I really want the WiFi in the cell phone).
I don't actually want the cell phone to be a modem or Ethernet relay, unless they really could do that somehow without making it bigger or heavier. Maybe with optional plug-in attachments.
I'd also like a Bluetooth keyboard for the PDA, with a little easel stand that will let you conveniently use the PDA in either portrait or landscape mode.
For use in the landscape mode, I want an ssh client that can toggle its display between proportional font and fixed-width font. Most of the time, the proportional font would be just fine, and would allow fitting more text on a line with a bigger font size; for times when you need to look at "ASCII art" (such as ncurses programs that draw boxes on the screen) you can toggle back into fixed-width.
I'd like a nice digital camera that has a motion video recording mode, and uses an SD/MMC card (same as the PDA). The PDA should have software for viewing photos and video from the card.
I'd like a GPS that uses Bluetooth to talk to the PDA. You guessed it -- it should use the same battery as the cell phone.
As long as I'm dreaming, I'd like a "sled" that you can dock the PDA into, which would have one or two cell phone batteries docked into it (and could charge the PDA from them), and a CompactFlash slot that would work with all CompactFlash devices. Come to think of it, this could be an easy way to get Ethernet, phone modem, and WiFi with just one gadget!
I'd like the sled to have another slot for new and interesting PDA sled devices, such as an autoranging Volt-Ohm meter, various and divers data collection gadgets, maybe even an oscilloscope (not that I personally need that one). I always thought the Visor's "Springboard" expansion slot was a great idea; this is an improved version. Springboard was too tightly coupled to the particular microcontroller the Visor used; this should be more universal. Maybe just US
Writing a program to check grammar is hard. You will notice that it will frequently flag some bit of grammar and say that it may be wrong, with an explanation of how to tell. Obviously that means it will give you false positives, where it complains about something that is perfectly correct. In fact, the better your grammar in the first place, the more often the checker is wrong, and the more false positives you will have to ignore. (The worse your grammar, the more useful you will find the grammar checker, too.)
If you think the Word grammar checker is so bad, tell me something: what grammar checker is better? MS bought Grammatik and bundled that with Word, and Grammatik is the best grammar checker I have ever used. (It's still brain-dead, with lots of false positives; it's just that I have never used anything better. I'll admit I haven't used many grammar checkers anyway.)
Here's a web page, from a company that sells a grammar checker called Grammar Slammer. I've never used it. But this page talks about what you can reasonably expect a grammar checker to do for you, and it's worth reading:
In short, Word's grammar checker is kind of dumb, but I don't think it's really much different from other available grammar checkers.
P.S. What would it take to make a free, open source software grammar checker? When I think about it, it actually seems very doable. You need an "engine" that can read in the text, do some parsing, and apply rules; and where a rule matches, pop up a dialog. Then you just need a whole bunch of rules, and an open source development process would allow many people to contribute rules. Potentially, once the engine is done, you will get dozens to hundreds of rules and it will quickly become just as good as the proprietary, closed-source grammar checkers.
You could easily make multiple rule sets: English (American), English (Queen's), French, Klingon, etc.
Heck, how about English (Klingon): "There is no honor in passive voice! Rewrite NOW!"
Word's on-the-fly "auto-formatter" will detect when the user does your 1. and 2., and will convert the text into appropriate formatting.
3. through 5. would require that they replace "Clippy" with "Cujo" or perhaps "Ted" (as in "Bundy").
Actually, bad jokes aside, a really styles-driven word processor would force people to make decent-looking documents, if they would actually use such a word processor. Take a look at Yeah Write.
This was just a rant. It isn't really worth your time to RTFA. Here, I'll summarize it for you:
Something is wrong with Word, as currently installed on Dvorak's computer. He would rather describe the symptoms in detail than fix it by, say, reinstalling Word. Direct quote: "I suppose I should reinstall Word, but other people have told me they have the same problems. So why bother?" Is Word really any worse than any other Microsoft applications under Windows? Don't they all suffer from Registry rot?
Various versions of Word aren't 100% compatible. Dvorak and some editors tried to use the change-tracking markup, and "we had a huge mess." What was this mess? He didn't specify.
He doesn't like the warning when you save to an older.DOC file format.
HTML files created by Word are full of useless junk. (Absolutely true, of course.) He says something hand-waving-ish about if the HTML is bad, the XML is probably bad, so he's never tried the XML. (If I write about how I've never tried something, can I be a famous pundit too?)
When you save a plain text file, there are too many options in the dialog box.
Based on his conclusions, Dvorak (who is not a software developer himself) has figured out that the Word code base (which he has never seen) should be scrapped. Quote: "There are many more issues than these. It's clear the program is in decline, with too many patches and teams of coders passing in the night. It's about time that it's junked and we get something new. This code can no longer be fixed." How the heck is he qualified to judge whether the code can any longer be fixed? As it happens, I agree that Word ought to get a major overhaul. Instead of pasting more layers of features onto Word, Microsoft ought to spend a bunch of man-years cleaning it up and making it faster. They won't, because that is not considered a profitable approach. (They actually tried something like this once. Eventually, they terminated that project, and just made the Windows code base the baseline for all future versions of Word. I didn't work on that project, but I heard that it was just taking too long and costing too much to clean it up, and people were worried about how long it might take to debug the final result.)
If Dvorak had wanted to do some actual research, and write an essay that would actually be of some value, he could have installed OpenOffice and tested its compatibility with his documents, and then written about that. This essay is awfully light on facts; I think he must have about 20 columns to write every month, and he just needed to bang something out to meet a deadline. (Note that I have no proof and did no research before making that statement. Just like Dvorak! But no one is paying me anything to write this, so I don't feel too bad.)
If you're actually using a finger tip to hit CTRL way down there, then I congratulate you on your flexibility!
In the late 80's, I went out of my way to use one specific keyboard (a clone of the IBM AT keyboard, with the Esc key moved to a sane place). But in 1990, I decided that the 101 keyboard was here to stay, and I decided to learn to use it rather than trying to have my own special keyboard everywhere I went.
I found it isn't hard to train yourself to use the Ctrl key exactly like a Shift key. You use your right pinky to hit the right Ctrl key when you want to press Ctrl+S (your left hand presses the S as usual). You use your left pinky to hit the left Ctrl key when you want to press Ctrl+I. All it takes is a slight change in your hand position; you can leave your index finger and middle finger on the home keys, so you can easily move your hand back after hitting the Ctrl key.
Also, Ctrl+Shift is really easy. You just change your hand position, and use your pinky to mash down both the Ctrl key and the Shift key, at the same time.
So, now I actually like having two Ctrl keys, one under each Shift key. And I'm much faster hitting a Ctrl key combination than an Alt key combination or a function key.
Note that my system leverages your touch-typing skills. If the Ctrl key is to the left of the A, "where it should be" as people say, then you will wind up having to do something unusual to hit Ctrl+A, Ctrl+Z, and Ctrl+Q (since the pinky of the left hand is hitting Ctrl, it can't also hit those keys). With my system, you use the same finger to hit the A in Ctrl+A as you always use when touch typing.
Slashdot Mirror
Several people have already posted comments asking why the world needs yet another Linux distro.
I wrote a Slashdot comment explaining why Ubuntu is interesting. Click here to read it.
A comment by Doc Ruby states that Ubuntu is not package-compatible with Debian. I said otherwise in my comment linked above, but I haven't checked it out for myself yet so I'm probably wrong.
steveha
Take a gasoline powered leaf blower. Scale it up so that it makes a lot of wind. Now build about five of these and attach them to a car-sized vehicle.
If you bring one of those home, I don't want to be your next-door neighbor.
To be fair, Moller is proposing an engine that would be more efficient (and thus quieter) than the cheap two-stroke engine in a leaf blower. But he still needs a boat-load of thrust per engine to lift the car, its fuel, and its passengers and their luggage, plus extra power to handle engine failure.
The last time I looked at Moller's web site (years ago) it said something hand-wavy about active noise cancellation. It sounded good at the time but now I'm dubious; moving as much air as this thing will move, how can you ever get it really quiet? (And how much will the noise cancellation gear weigh?) Noise cancellation works great if you are sitting in a predictable spot, but how can you cancel the noise of the flying car for all listeners in all directions?
I'd love flying cars, with safe autopilots please, but I think that unless someone invents antigravity they won't be practical.
On Earth, anyway... A Miracle of Science showed flying cars inside the domed cities of Luna. With one-sixth the gravity of Earth, flying cars might be more reasonable.
P.S. I'd like to plug A Miracle of Science. It's a nifty hard-science SF story that nonetheless has a sense of wonder. It's one of my favorite web comics.
steveha
we have a pretty damn effective electricity transport system already in place.
Well, to get power to your house, yes. But we don't have a really practical way to run cars with it. Current electric cars have limited range, and when the batteries wear out you have to buy new ones and safely dispose of the old ones.
If BMW has managed to make a practical hydrogen car, good for them and I hope they get rich. But I'm dubious.
steveha
They didn't perform Star Wars but they have done other movies, and my wife and I attended a showing.
http://www.jetcityimprov.com/twistedflicks/
are they adding content which legitimately justifies the price they are charging on its own
They are. That's the whole idea of "Twisted Flicks".
They show the movie without sound, and they take suggestions from the audience. "Okay, I need the name of a place. Sedro Wooley. Okay, now I need a profession. Dressmaker." And so on.
Then improv comedy guys then use the suggestions from the audience to make something up on the fly. They have to think on their feet and make funny stuff up rapidly. What you are really paying for is the improv.
Really, they can use any movie to do this. But there is potentially extra humor value if the movie is well-known; if the movie is itself part of pop culture, then sly pop-culture references might be easy to slip in.
I'm not a lawyer, but I don't think these guys can hide behind the parody protection laws. They aren't really looking to specifically make fun of the movie, they just need a starting place to hang their improv humor.
They are showing a movie in a theatre, so they need permission from whoever owns the movie.
steveha
You have a hydrogen leak, and someone walks in and flips on the light switch...
I hope that the vent system will have a little pilot light or sparker or something, and will burn up the wasted hydrogen. Maybe even run the waste hydrogen through a fuel cell and charge a storage battery?
Except for your enclosed garage scenario, even unburned waste H2 should be safe, because it's lighter than air, so it will disperse quickly. If it settled to the ground in a pool, that would be bad. But even gases like propane, which can settle in a pool, rarely explode because they dissipate quickly. But yeah, contain them and ignite them and you are having a bad day.
I hope, if they're going to do this, they're at least going to have the sense to perfume the hydrogen, like they do natural gas
I don't think this will work. The whole point of the liquid hydrogen is that it is much much much more compact than gaseous hydrogen. Is there a scent you could add, that would be potent enough you could smell it when one milliliter of liquid hydrogen vaporises into one liter of gas? And would this potent smell chemical cause problems for the fuel cells, or make nasty chemicals when burned in the engine?
We have absolutely no way of getting hydrogen, outside of fossil fuels, that doesn't use up more electricity than we put into it. I've never heard of any way even proposed to get said hydrogen.
The FA proposes to use solar power to get the hydrogen out of water. This will work. You can put large solar plants in places where the sun shines a lot, and transport the hydrogen to where you need it.
The problem is that I'm dubious about liquid hydrogen as the way to transport it. Suppose you build a bunch of solar collectors out in, say, Mexico... or maybe you make solar collectors that just float on the ocean in an out-of-the-way place. Great. But can you keep it cold enough to still be liquid all the way from there, to the local fuel station where the customer buys it? Will you lose a bunch to boil-off as the liquid warms, or perhaps burn up a bunch to run refrigerators to chill it? Will you lose so much that it drives the price up? In short, is this really practical?
The other problem is that if this becomes really popular, I'm not sure solar can scale up enough to provide all the hydrogen people will want. Well, there's always the nuclear option.
I keep hoping someone will invent some magical fuel tank that will somehow really lock the hydrogen in. Maybe little microcells that each store a small quantity or something.
Our biggest problem isn't getting energy, it's getting energy exactly where you want it. If hydrogen becomes practical, you could put a power plant just anywhere and use hydrogen to bring the power wherever it's needed. And the hydrogen part wouldn't pollute anything.
steveha
I'm not any kind of chip expert; I only know what I have read here and there. You sound like you know what you are talking about better than I do.
steveha
The little "wires" connecting things inside a chip are called "traces", and are the thinnest things you can draw on a chip. So "trace size" is used as a yardstick to compare chips for how small their various parts can be. (You will often see the word "process", as in "90 nanometer process" instead of "90 nanometer traces"). The smaller the trace size, the less power dissipation. Both of these new chips have a small trace size, although the articles didn't say exactly how small.
If you make a chip, and then you improve your chip-making technology so that you can draw thinner traces, you can perform a "die shrink": you produce a similar chip design, using the smaller traces, and the whole chip takes up a smaller amount of space. You can buy some graphics cards that are basically an older GPU with a die shrink; they dissipate very little heat and are inexpensive. A GeForce4 MX (budget card) was basically a GeForce 3 with a die shrink, IIRC.
Chips are made from silicon wafers. One whole wafer is "fabbed" (made in a chip fabrication plant), then cut up for individual chips. The more chips per wafer, the cheaper each chip is. (This is all the more true because flaws can happen during the fabbing process; if one flaw means one dead chip, then more chips per wafer means a similar number of flaws results in a lower percentage of dead chips made, and thus lower costs.)
A smaller trace size makes it easier to push the clock rate higher. But GPUs are definitely clocked lower than CPUs, so that helps them dissipate less heat. If you are pushing a Pentium 4 at 3 GHz, and the GPU is only doing 0.7 GHz, clearly that helps the GPU dissipate less heat.
Smaller trace sizes make it harder to make the chip work right; the smaller the traces, the more problems you might have electrically (I don't fully understand the details). Also, you need to be more careful with cooling; a hot chip with a tiny die size needs a really good heatsink, and there is less margin for error. The old, relatively large chips like the 486 family were easy to cool in comparison with today's chips.
Summary:
Smaller trace size means less heat
Lower clock rate means less heat
steveha
To me the big deal is that I'm getting some CDs in the mail that I can give away to people that I don't have to burn.
That's not a big deal to me, because my CD burner works well and I buy 100-packs of CDs. I can burn as many CDs as my friends will ever need.
What's a big deal to me is a distribution that will be good for newbies, but still 100% free software, so I'm allowed to give away CDs. Since I use Debian, I want a Debian-based distro that is both 100% free and newbie-friendly; it looks like this is it.
steveha
It looks to me that Debian is cracking under it's own weight.
I don't agree. Debian works about the same as it always has. The people, like me, who use Debian are all as happy as we have always been.
My point was that Ubuntu is a great complement to the Debian project. Nothing is perfect for everyone; Debian isn't perfect for everyone. Ubuntu will be better for many people.
Debian seems to have too many packages, too many developers
These go together. Most of the Debian developers have a very small number of packages to manage, which is why they are willing to do it for free.
In my opinion, that might make Debian work.
I wouldn't put it this way, since Debian works quite well for me. But we agree that Ubuntu will be a good thing for many people.
steveha
Well, OOo is now a major component of a modern, powerful Linux desktop, so it should be taken into account.
My point is that it does no good to complain to the GNOME developers that OOo is slow. The GNOME developers do not develop OOo.
Re: WinXP, yes, out of the box it's a pig. But just defaulting to classic theme, turning off system restore, disabling the theming and file-cataloging background processors and it's very fast. Even just about usable in 64M!
Well, if you run IceWM, and use Rox Filer for your file manager, and run AbiWord and Gnumeric instead of OOo, I'll bet you could be happy in 64M. You would still be "using GNOME" (the libraries anyway) to run your GNOME apps.
Or you could just buy some more RAM and run full throttle GNOME, which is my choice.
steveha
Who cares about Yet Another Distro? What's different about this one?
It's Debian, in a friendly wrapper, free and for free.
You can get Debian in a friendly wrapper by buying Xandros, or Linspire. They include nonfree software, and Linspire hooks you in to a software distribution scheme that costs a minimum of $50 per year.
You can get Debian free if you are a Linux expert. Get a Debian installer and have fun. However, Debian has 10,000 packages, and you need to know enough to pick and choose which ones you want. Ubuntu has 1,000 packages, and they have made default choices for you. (Want something Ubuntu doesn't offer? Grab it from the main Debian distribution; it will work.)
Also, Debian comes in three major branches: stable, unstable, and testing. Stable is really stable, but only updates every two years or so. Unstable updates daily but can be unstable. Testing updates automatically from unstable when the packages appear stable (a week goes by without major bugs posted against the unstable package, IIRC). Ubuntu on the other hand is promising a six-month release cycle; if you use Ubuntu, you should have a nice stable system, but you will get new packages much more often than if you use Debian Stable.
Ubuntu will occupy a similar niche to Fedora, but Red Hat makes all the decisions for Fedora while Ubuntu will have a community process.
The closest distro to Ubuntu is probably Bruce Perens's UserLinux project. But UserLinux is focused squarely upon business, whilst Ubuntu seems to be more focused on individual users.
Ubuntu should preserve all the things I like best about Debian, while being more friendly to newbies and offering a much fresher stable release. There isn't another distro quite like it.
I'm downloading it now and I look forward to trying it out.
steveha
when are they going to concentrate on performance and memory usage? Right now it's _terrible_ - just as bad as Windows XP.
Old rule of thumb: first you make it work, then you make it faster.
GNOME has been following that rule. Nautilus used to be slow, but in recent versions it has become quite snappy.
GNOME has also done their homework; their architecture is pretty much sound. They don't add bloat just because it would make their life easier in some way. For example, GNOME doesn't have its own file open/file save dialogs; it just uses the GTK ones. And DBUS will let them do all sorts of cool stuff (like knowing when a CD is inserted into a drive) without polling the hardware.
OpenOffice.org is slower and heavier than MS Office.
I can't argue with that. But what does that have to do with GNOME? OOo has nothing to do with GNOME. The "GNOME Office" would be AbiWord, Gnumeric, and other GNOME native apps, not OOo.
OOo started out huge and slow. The OOo guys are doing good work on it (compare OOo with StarOffice 5.x sometime!). But the GNOME apps are faster and lighter than OOo.
I mean, I can run Office, IE and Outlook together SMOOTHLY on a WinXP box with 128M RAM.
This does not match my experience. I helped a friend who was running XP on a 128MB computer and it wasn't smooth. Maybe, once everything was started up, and the system had settled down, maybe then. Maybe. (But then run something and the disk thrashes. 128MB isn't enough for XP.)
One thing to realize, however: the people here like to ridicule MS, but they have some good software, and in particular their compilers are good. Their compilers make tighter, faster code than GCC does, and this is a huge advantage when trying to hit a target like "run a bunch of software in 128MB".
There are LOADS of people with 64 and 128M boxes out there who can't run a modern, desktop Linux effectively
Are there really? Last time I checked, RAM was cheaper than Windows XP. I would encourage those people to buy a 256MB stick of RAM and speed up their systems.
GNOME is working on cool new features. They do work on speed and size to some extent, and GCC keeps getting better, and RAM is still cheap. All the new computers, with 2 GHz processors and tons of RAM, will run GNOME very well. In short, I think the GNOME guys are doing a fine job and they do not need to change their priorities at all.
steveha
You know what? My first draft null pointer checks. I took them out because they weren't funny.
steveha
What a waste of valuable CPU cycles! Here's a speedup that does the same thing much faster:That should run much faster -- standard library functions are always well-optimized.
Just doing my part for data security.
steveha
the realist in me says that once Firefox really takes off, we can look forward to people finding security exploits in it too.
But it will still be better.
Internet Explorer is designed to install software on your computer (or else Windows Update wouldn't use it). Firefox is just an application and doesn't have the ability to install software. This, alone, makes a huge difference for spyware.
Firefox does have the ability to install Firefox extensions, which are little scripts that can make Firefox do things. We will need to be vigilant to make sure that there isn't any script-based spyware.
However, I don't think you can get Firefox to install an extension without popping up a dialog box asking for permission, so it's still better than IE.
steveha
I don't know if this will work for you, but it's cheap and not risky.
Shut down and unplug the equipment. Crumple up sheets of newspaper, and pack them into the open space of the equipment. Seal it up and give it a day or two, then pull out the newspaper and discard it.
Why do I think this might work? I once bought a used refrigerator. It had been stored with the door closed and the power off, and its inside smelled very bad. I scrubbed the insides with cleaner and let it run for a few days, and it still smelled bad. Someone told me to try stuffing it full of newspaper, close it up, and leaving it for a day. I didn't really think it would work, but it did; the odor was gone. Just gone.
I think that the way it works is just that the fibers of the paper soak up the smell.
Now, if the meat-smelling equipment has actual deposits of meaty chemicals (pork fat or whatever) then I'll bet you will need to wash the equipment with alcohol or something to get rid of it. But if you just have odors soaked into various plastic things, this might work, just as it worked on my old refrigerator.
If you try it and it does work, do please let me know.
steveha
Are voucher systems somehow the silver bullet
t ml
They sort of are.
The real silver bullet is an effective system of negative feedback. When the schools do a bad job, they need to be punished, and when they do a good job, they need to be rewarded. A simple idea.
Simple, yes, but hard to do in real life. Teachers' unions, educational bureaucracies all the way up to the federal level, politicians making promises... all of these things can complicate the school system to the point where incompetence isn't punished, nor excellence rewarded. And attempts to use standardised tests to guarantee that kids are taught well, just mean that teachers will wind up "teaching the test".
The best thing you can hope to do is to allow parents to move their kids around to the best schools. This will not, itself, fix the problem instantly; but it will introduce an element of feedback into the system. Over time, this will inevitably force the schools to improve.
If a restaurant has poor food, people will take their business to other restaurants. It doesn't matter what kind of union the cooks have, it doesn't matter what kind of promises politicians might have made, etc. If the customers vote with their feet, the better restaurants will prosper and the worst ones will have to close. The same thing would happen with schools, but it would take longer (people eat several meals per day, but they would probably leave their kids in any particular school for at least a few months before deciding to move the kids somewhere else).
I have debated this issue in the past with some people who claimed that parents must not be trusted to choose schools for their kids. That's lunacy. There will be a few bad parents, but by far most parents really want what is best for their kids. The parents and kids together are the best judges of how well a school is serving them.
Note that middle-class and upper-class parents already have some freedom to pick schools; I know my parents, whenever we moved, would carefully consider what the schools were like, and they would only move someplace where the schools were decent. The poorest people, who are trapped in the bad part of town (no money to move somewhere else), those are the ones who really want school vouchers.
By the way, public school systems spend a lot of money per student. The vouchers are generally for less than the public school system would have spent on a student. If a student takes a $3000 voucher and goes to a private school, that is usually a net profit for the public school. In my state, the average per-student spending is $9,454 per year.
For more on vouchers, click here: http://www.cato.org/research/education/vouchers.h
steveha
Your ideas about grammatical rules is slightly primitive: it's *hard* to check grammar.
My idea was a bit hand-wavy, I admit, but I still think it could work. For coding work, you need experienced software guys; but if you set up the project right, you can have language contributions from non-geeks. Start with existing natural-language tools like NLTK. If you need a dictionary, import one, such as Wiktionary. Set up an easily-parsed format for the rules, and let people submit rules.
The result would be about as good as the Word grammar checker; still brain-dead, still offering comments about things that might be wrong because it can't be sure, but by golly a grammar checker.
You figure it's ten years of work. If you can take advantage of work that has already been done (NLTK), and get help from many people (writing rules), it could shape up much quicker than that.
steveha
I looked at all the screenshots, and nothing on there jumped out and bit me and yelled "Windows! IE!" I have no idea what FlipmodePlaya is complaining about.
It looks to me like it's just the GNOME 2.x that I know and love, with subtle, very incremental bits of polish. FlipmodePlaya, perhaps you could be a bit more specific?
P.S. I'm really looking forward to some of the new features, specifically Volume Manager and the new MIME handlers. GNOME 2.8's MIME features won't just be easier to use than previous GNOME versions--they will actually be easier to use than Windows's application association system.
steveha
I want a PDA that looks exactly like a Palm Tungsten T2, but has the full-size screen like a T3. (To say the same thing another way, I want a Palm Tungsten T3 with buttons and 5-way navigator exactly like a T2. But I want more battery life than the T3 currently gets.) The PDA screen should be visible in bright light (a "transflective" screen like the T2 and T3 already have).
To go along with the PDA, I want a cell phone with Bluetooth. The cell phone will be clipped in a holster on my belt, just like my ancient StarTac is now. With a hands-free earphone/microphone, and using the PDA to dial via Bluetooth, I could make phone calls without removing the phone from the holster. The phone should be a folding phone, about the same size as my StarTac. (Perhaps a little bit smaller. I don't ever seem to wish that the StarTac were smaller; it seems like the perfect size. But some people might like it better if it were smaller.)
The phone, by virtue of its relatively large battery, should also double as a WiFi relay, talking to the PDA with Bluetooth. If I'm in a place with WiFi I can use that to connect to the Internet, and if not I can use the cell phone.
I also want a Bluetooth modem gadget. I wouldn't carry this everywhere I go, but it would let me connect to the Internet where there is no WiFi but there is a phone jack. For total nirvana, this gadget should also have an Ethernet jack, so that it can double as an Ethernet connector. (You can already carry a small Bluetooth USB dongle, and use that to take advantage of an Ethernet network -- as long as there is a PC there, for which you have permission to plug in a USB gadget and permission to install appropriate drivers. An actual Ethernet jack would be easier.) Again for total nirvana, the modem gadget should use the same exact batteries as the cell phone, and be rechargeable with a Palm travel charger (or you should be able to remove the battery and use the cell phone travel battery charger). The modem gadget should also serve as a WiFi relay (although since I carry a cell phone all the time, I really want the WiFi in the cell phone).
I don't actually want the cell phone to be a modem or Ethernet relay, unless they really could do that somehow without making it bigger or heavier. Maybe with optional plug-in attachments.
I'd also like a Bluetooth keyboard for the PDA, with a little easel stand that will let you conveniently use the PDA in either portrait or landscape mode.
For use in the landscape mode, I want an ssh client that can toggle its display between proportional font and fixed-width font. Most of the time, the proportional font would be just fine, and would allow fitting more text on a line with a bigger font size; for times when you need to look at "ASCII art" (such as ncurses programs that draw boxes on the screen) you can toggle back into fixed-width.
I'd like a nice digital camera that has a motion video recording mode, and uses an SD/MMC card (same as the PDA). The PDA should have software for viewing photos and video from the card.
I'd like a GPS that uses Bluetooth to talk to the PDA. You guessed it -- it should use the same battery as the cell phone.
As long as I'm dreaming, I'd like a "sled" that you can dock the PDA into, which would have one or two cell phone batteries docked into it (and could charge the PDA from them), and a CompactFlash slot that would work with all CompactFlash devices. Come to think of it, this could be an easy way to get Ethernet, phone modem, and WiFi with just one gadget!
I'd like the sled to have another slot for new and interesting PDA sled devices, such as an autoranging Volt-Ohm meter, various and divers data collection gadgets, maybe even an oscilloscope (not that I personally need that one). I always thought the Visor's "Springboard" expansion slot was a great idea; this is an improved version. Springboard was too tightly coupled to the particular microcontroller the Visor used; this should be more universal. Maybe just US
Writing a program to check grammar is hard. You will notice that it will frequently flag some bit of grammar and say that it may be wrong, with an explanation of how to tell. Obviously that means it will give you false positives, where it complains about something that is perfectly correct. In fact, the better your grammar in the first place, the more often the checker is wrong, and the more false positives you will have to ignore. (The worse your grammar, the more useful you will find the grammar checker, too.)
If you think the Word grammar checker is so bad, tell me something: what grammar checker is better? MS bought Grammatik and bundled that with Word, and Grammatik is the best grammar checker I have ever used. (It's still brain-dead, with lots of false positives; it's just that I have never used anything better. I'll admit I haven't used many grammar checkers anyway.)
Here's a web page, from a company that sells a grammar checker called Grammar Slammer. I've never used it. But this page talks about what you can reasonably expect a grammar checker to do for you, and it's worth reading:
http://englishplus.com/news/readthis.htm
In short, Word's grammar checker is kind of dumb, but I don't think it's really much different from other available grammar checkers.
P.S. What would it take to make a free, open source software grammar checker? When I think about it, it actually seems very doable. You need an "engine" that can read in the text, do some parsing, and apply rules; and where a rule matches, pop up a dialog. Then you just need a whole bunch of rules, and an open source development process would allow many people to contribute rules. Potentially, once the engine is done, you will get dozens to hundreds of rules and it will quickly become just as good as the proprietary, closed-source grammar checkers.
You could easily make multiple rule sets: English (American), English (Queen's), French, Klingon, etc.
Heck, how about English (Klingon): "There is no honor in passive voice! Rewrite NOW!"
steveha
School Tool is a system specifically designed for running a school. It's written in Python, and it's free, open-source software.
http://www.schooltool.org/
steveha
Word's on-the-fly "auto-formatter" will detect when the user does your 1. and 2., and will convert the text into appropriate formatting.
3. through 5. would require that they replace "Clippy" with "Cujo" or perhaps "Ted" (as in "Bundy").
Actually, bad jokes aside, a really styles-driven word processor would force people to make decent-looking documents, if they would actually use such a word processor. Take a look at Yeah Write.
steveha
Something is wrong with Word, as currently installed on Dvorak's computer. He would rather describe the symptoms in detail than fix it by, say, reinstalling Word. Direct quote: "I suppose I should reinstall Word, but other people have told me they have the same problems. So why bother?" Is Word really any worse than any other Microsoft applications under Windows? Don't they all suffer from Registry rot?
Various versions of Word aren't 100% compatible. Dvorak and some editors tried to use the change-tracking markup, and "we had a huge mess." What was this mess? He didn't specify.
He doesn't like the warning when you save to an older .DOC file format.
HTML files created by Word are full of useless junk. (Absolutely true, of course.) He says something hand-waving-ish about if the HTML is bad, the XML is probably bad, so he's never tried the XML. (If I write about how I've never tried something, can I be a famous pundit too?)
When you save a plain text file, there are too many options in the dialog box.
Based on his conclusions, Dvorak (who is not a software developer himself) has figured out that the Word code base (which he has never seen) should be scrapped. Quote: "There are many more issues than these. It's clear the program is in decline, with too many patches and teams of coders passing in the night. It's about time that it's junked and we get something new. This code can no longer be fixed." How the heck is he qualified to judge whether the code can any longer be fixed?
As it happens, I agree that Word ought to get a major overhaul. Instead of pasting more layers of features onto Word, Microsoft ought to spend a bunch of man-years cleaning it up and making it faster. They won't, because that is not considered a profitable approach. (They actually tried something like this once. Eventually, they terminated that project, and just made the Windows code base the baseline for all future versions of Word. I didn't work on that project, but I heard that it was just taking too long and costing too much to clean it up, and people were worried about how long it might take to debug the final result.)
If Dvorak had wanted to do some actual research, and write an essay that would actually be of some value, he could have installed OpenOffice and tested its compatibility with his documents, and then written about that. This essay is awfully light on facts; I think he must have about 20 columns to write every month, and he just needed to bang something out to meet a deadline. (Note that I have no proof and did no research before making that statement. Just like Dvorak! But no one is paying me anything to write this, so I don't feel too bad.)
steveha
If you're actually using a finger tip to hit CTRL way down there, then I congratulate you on your flexibility!
In the late 80's, I went out of my way to use one specific keyboard (a clone of the IBM AT keyboard, with the Esc key moved to a sane place). But in 1990, I decided that the 101 keyboard was here to stay, and I decided to learn to use it rather than trying to have my own special keyboard everywhere I went.
I found it isn't hard to train yourself to use the Ctrl key exactly like a Shift key. You use your right pinky to hit the right Ctrl key when you want to press Ctrl+S (your left hand presses the S as usual). You use your left pinky to hit the left Ctrl key when you want to press Ctrl+I. All it takes is a slight change in your hand position; you can leave your index finger and middle finger on the home keys, so you can easily move your hand back after hitting the Ctrl key.
Also, Ctrl+Shift is really easy. You just change your hand position, and use your pinky to mash down both the Ctrl key and the Shift key, at the same time.
So, now I actually like having two Ctrl keys, one under each Shift key. And I'm much faster hitting a Ctrl key combination than an Alt key combination or a function key.
Note that my system leverages your touch-typing skills. If the Ctrl key is to the left of the A, "where it should be" as people say, then you will wind up having to do something unusual to hit Ctrl+A, Ctrl+Z, and Ctrl+Q (since the pinky of the left hand is hitting Ctrl, it can't also hit those keys). With my system, you use the same finger to hit the A in Ctrl+A as you always use when touch typing.
steveha