That extra layer of parsing makes it much more difficult to touch type while looking at a source document, especially with the stock (ahem, Microsoft) IME.
One example, using Japanese and the typical Romaji (latin) input method filters for the word "apple".
English, well, there it is: "apple". four different keys, one repeated stroke, and space or punctuation to delimit -- six strokes, and, with practice, you don't have to look at either the keyboard or the screen.
The Japanese word Latinizes to "ringo", and that is what you usually type when using the IME in Romaji mode. But (problem 1) there is no standard small set of delimiters. So you reach for a conversion (henkan) key, which is usually either the space bar or a small key next to a shrunken space bar.
Some methods have several henkan keys, depending on whether you want to just dump the conversion buffer (probably hiragana) or force to the other kana (probably to katakana) or get a list of candidate Kanji (Han characters). More often, there is only one (effective) conversion key that pops up a list of candidates, and the method assumes, for each candidate vocabulary element, a preferred conversion which it puts at the top of the list.
So most people will end up having to check the list of candidates and make sure the desired one is selected. If not, the conversion key is repeated to select the next in the list. (Cursor keys can be used to scroll through the list in many IMEs.)
Oh, and, incidentally, you need to be able to decide for yourself whether the current reference to "apple" is best represented with hiragana (for usual words in modern times), katakana (for foreign words, or emphasis, or to indicate that there is something special -- foreign? -- about this apple), or Kanji (in many cases, you may have an option on which Kanji). Or, maybe you actually want to use Romaji for some reason, although, in that case, you might have typed "appuru" instead). Options, options, options.
Does this sound like something you're going to have easy time of touch typing?
It gets worse. In order to improve efficiency, the method "learns" from the user and reorders the candidate list for you.
Professional data entry operators have professional input methods that do allow touch typing, but they take a lot of learning and training.
Oh, there is kana mode for the keyboard, where the 46 (erm, 50 plus or minus) kana are layed out on the qwerty keyboard, on the right-hand side of the keys (I ought to put a link to something here, but I won't.) If you noticed that we have just laid kana out where there are numbers, that is correct. I am trying to learn to touch type the kana keyboard, but there are several versions with minor variations between them, and, really, the prevailing common sense is to use Romaji mode.
In kana mode, "ri-n-go" is three keys, but you have to hit a modifier key after the "ko" to voice it (to "go"), so that's four keystrokes. Even if I do learn to touch-type in kana mode, it is not that much more efficient. I'm just being a typing geek trying to learn to do that.
The efficiencies are eseentially leveled by the conversion step.
Chinese is getting a phonetic conversion, but, historically, a stroke-radical input method has been preferred. Kanji (Hanji or Xanji when talking about Chinese) are constructed of moderately regular parts (radicals), but there are around 300 of those. That list of 300 is broken down for the keyboard. (Not unreasonable, most of the radicals are composites of simpler stroke sets.)
But you still end up de-parsing at the character and word level, where in Latin you mostly de-parse at the word level.
I'm not sure about China, but the typical Japanese attitude towards computer keyboards is that they would rather write and edit on paper and then type stuff in when they've got it fixed so that they can minimize typing. It takes considerable experience to get past the perceived inconveniences.
One of the reasons the English context worked well in developing computers was the paucity of characters. A small set of glyphs is generally an advantage, even at the cost of overloading the punctuation and such.
Of course, the typical denizen of slashdot (such as myself) is of two minds on problems like this.
The geek recognizes the simplicity of keeping the existing solution, but also recognizes the inherent attraction of the difficult problem of going with the technilogical solution.
I thought about this problem a lot about 25, 30 years ago. How do patent examiners recognize innovation unless they already know pretty much everything under current development in each particular field?
I think some ambulance-chaser-turned-congress-critter must have been thinking about the problem and recognized the implications back then, got himself on the right committee, and started deliberately underfunding the PTO and doing other things to keep them from bringing in competent examiners. I do recall reading about examiners complaining that they couldn't get enough other qualified examiners, and I'm pretty sure I recall reading about one who says they more or less decided that working the backlog down was more important that making sure the patents were valid.
I also spent a bit of time thinking about the database that the patent office should have been building. Too bad I spent more time finding problems than solving them, or even than talking about them.
If you really aren't a troll (pardon me for not taking your assertions at face value), well, I'm not sure how to explain this if you haven't participated in meetings where the issues which are supposed to be discussed have already been discussed and decided on by a faction which is determined to control the results to the detriment of those who were left out of the "real" discussion.
Apparent conflicts of interest are dead easy to manufacture.
Anyway, others have pointed this out, but the conflict of interest only exists if Oracle insists that the fork is the enemy. That was not a necessary assumption until this meeting made it clear that that is the point of view Oracle has decided to take.
Other reasons for a fork, in absence of a license change?
Management issues would be one. If Oracle is not able to deal with the open source style of management, it would be easier for Oracle to manage their product project (although Star Office is already in place) and let people more used to open source manage a new, external project. This does not require two enemy camps. In fact, if Oracle recognized the value of the open source methodology, they would want to keep as many doors open to dialog as possible, including within the community board.
Oracle apparently does not recognize the value of open source methodology. (Perhaps they think Open Office is a valuable fluke?)
Objectives would be another. Oracle may want to gain market value from the brand, and Star Office may not get them the (perceived) value they want. In that case, they might want to start re-packaging open office for the business market.
Again, the re-packaging does not require the two projects to take be at enmity. They could still cooperate to a great extent, simply splitting things where Oracle needs the thing to be oriented towards business.
But Oracle has (by any reasonable reading of the chat meeting record) declared themselves to be in competitive (as in war) stance with the erstwhile community.
This is not an implication, it's an equivalence. Declaring the other side to be in COI if they remain is saying that the other side is not on your side any more.
Kind of like watching the scene where the former girlfriend is telling the geek goodbye, and you see her new boyfriend in the background with bulging, er, wallet.
And she has way too much makeup, way more than she used to, and the way it's caked on (cake, get it? Can't have your cake and eat it too.), the way it's caked on, you gotta wonder whether it's hiding bruises. And you know, the way these plots play out, that's exactly what the makeup is hiding.
You don't know how to read between the lines in this kind of meeting, I'd say.
I've seen enough of these kinds of meetings to see the evidence of backroom deals. (As I noted above, the jammed input on the COI loop is one obvious bit of evidence.)
I was a little surprised to see evidence in the logs that the new employer was exercising so much influence over (former) friends: "COI.... COI.... COI.... COI.... COI.... COI...." Sounded like a broken record. (Or should I say it looks like the input device for the end condition got stuck.)
The first version of the GPL kind of went nowhere.
Getting some kind of gentleman's agreement in place for hardware, where you can really use the hardware you've bought, is taking a bit longer than the mucking with the copyright law we call the GPL, and even the GPL is going to need tuning again after the next round of legal semantic shifting.
Except that it is more than just a statistical anomaly.
Especially when devices fail, or fail to work to spec, or the owner discovers some neat thing that ought to work except there's some legal tangle-up, or,... .
That section said there was no problem with giving facts (such as listing which operating systems the hardware was tested/designed for), but the problem was the use of the logos that would promote the proprietary.
Still, I'm not sure that curing the general problem of endorsement with endorsements is a great idea. (Then again, the GPL uses the features of copyright law, so I'll have to read this and think about it.)
Anyway, I don't think binge drinking is in the same class of activities as consuming fats, unless you are trying to drag the topic from reasonable fat consumption to binge eating.
One of the problems with boh studies and meta-studies is the definition of reasonable. It seems to vary from person to person, whether you are talking about subjectively reasonable or objectively reasonable.
Until the fix the underlying issue, which is bandwidth for the control functions. (The post just above in my browser mentions something about dynamic control channels.)
Both on the part of the people and of the companies.
Seriously, people in Japan just work around the government's attempts at restrictions. That's why they don't really understand the fundamental issues of freedom, such as self-determination. It looks to your novce manager like the ideal place to manage, until you try to get people to do something new or unusual. (Propaganda does work, but it also takes a while.)
The reason people want money is because they think money will give them power.
The reason they want power is that they don't have control over themselves.
No amount of money will bring you real power, just facades and illusions of power.
No amount of power, whether illusion or real, will bring you control.
No amount of control over other things, even if such a thing could possibly be anything other than an illusion, will bring you control over yourself. (Generally gets in the way, in fact.)
That's why rich people and powerful people never seem to be able to get enough.
That's why this story repeats itself every few years. No, much more often than that. Same story, different players, maybe a different market, etc. Details change, but it's always looking for whatever you want to call it in all the wrong places.
Slashdot Mirror
Playing the devil's advocate here, but are you sure that English can be written strictly in ASCII?
That's what the Unicode consortium wants you to believe.
(Yeah, I beg to differ. I work with this stuff. There are issues in ideographs that the Unicode Consortium is still either ignoring or not aware of.)
That extra layer of parsing makes it much more difficult to touch type while looking at a source document, especially with the stock (ahem, Microsoft) IME.
One example, using Japanese and the typical Romaji (latin) input method filters for the word "apple".
English, well, there it is: "apple". four different keys, one repeated stroke, and space or punctuation to delimit -- six strokes, and, with practice, you don't have to look at either the keyboard or the screen.
The Japanese word Latinizes to "ringo", and that is what you usually type when using the IME in Romaji mode. But (problem 1) there is no standard small set of delimiters. So you reach for a conversion (henkan) key, which is usually either the space bar or a small key next to a shrunken space bar.
Some methods have several henkan keys, depending on whether you want to just dump the conversion buffer (probably hiragana) or force to the other kana (probably to katakana) or get a list of candidate Kanji (Han characters). More often, there is only one (effective) conversion key that pops up a list of candidates, and the method assumes, for each candidate vocabulary element, a preferred conversion which it puts at the top of the list.
So most people will end up having to check the list of candidates and make sure the desired one is selected. If not, the conversion key is repeated to select the next in the list. (Cursor keys can be used to scroll through the list in many IMEs.)
Oh, and, incidentally, you need to be able to decide for yourself whether the current reference to "apple" is best represented with hiragana (for usual words in modern times), katakana (for foreign words, or emphasis, or to indicate that there is something special -- foreign? -- about this apple), or Kanji (in many cases, you may have an option on which Kanji). Or, maybe you actually want to use Romaji for some reason, although, in that case, you might have typed "appuru" instead). Options, options, options.
Does this sound like something you're going to have easy time of touch typing?
It gets worse. In order to improve efficiency, the method "learns" from the user and reorders the candidate list for you.
Professional data entry operators have professional input methods that do allow touch typing, but they take a lot of learning and training.
Oh, there is kana mode for the keyboard, where the 46 (erm, 50 plus or minus) kana are layed out on the qwerty keyboard, on the right-hand side of the keys (I ought to put a link to something here, but I won't.) If you noticed that we have just laid kana out where there are numbers, that is correct. I am trying to learn to touch type the kana keyboard, but there are several versions with minor variations between them, and, really, the prevailing common sense is to use Romaji mode.
In kana mode, "ri-n-go" is three keys, but you have to hit a modifier key after the "ko" to voice it (to "go"), so that's four keystrokes. Even if I do learn to touch-type in kana mode, it is not that much more efficient. I'm just being a typing geek trying to learn to do that.
The efficiencies are eseentially leveled by the conversion step.
Chinese is getting a phonetic conversion, but, historically, a stroke-radical input method has been preferred. Kanji (Hanji or Xanji when talking about Chinese) are constructed of moderately regular parts (radicals), but there are around 300 of those. That list of 300 is broken down for the keyboard. (Not unreasonable, most of the radicals are composites of simpler stroke sets.)
But you still end up de-parsing at the character and word level, where in Latin you mostly de-parse at the word level.
I'm not sure about China, but the typical Japanese attitude towards computer keyboards is that they would rather write and edit on paper and then type stuff in when they've got it fixed so that they can minimize typing. It takes considerable experience to get past the perceived inconveniences.
One of the reasons the English context worked well in developing computers was the paucity of characters. A small set of glyphs is generally an advantage, even at the cost of overloading the punctuation and such.
Of course, the typical denizen of slashdot (such as myself) is of two minds on problems like this.
The geek recognizes the simplicity of keeping the existing solution, but also recognizes the inherent attraction of the difficult problem of going with the technilogical solution.
(Did I just say that?)
Yeah, just how different is a current router from the old ham data relays?
I thought about this problem a lot about 25, 30 years ago. How do patent examiners recognize innovation unless they already know pretty much everything under current development in each particular field?
I think some ambulance-chaser-turned-congress-critter must have been thinking about the problem and recognized the implications back then, got himself on the right committee, and started deliberately underfunding the PTO and doing other things to keep them from bringing in competent examiners. I do recall reading about examiners complaining that they couldn't get enough other qualified examiners, and I'm pretty sure I recall reading about one who says they more or less decided that working the backlog down was more important that making sure the patents were valid.
I also spent a bit of time thinking about the database that the patent office should have been building. Too bad I spent more time finding problems than solving them, or even than talking about them.
If you really aren't a troll (pardon me for not taking your assertions at face value), well, I'm not sure how to explain this if you haven't participated in meetings where the issues which are supposed to be discussed have already been discussed and decided on by a faction which is determined to control the results to the detriment of those who were left out of the "real" discussion.
Apparent conflicts of interest are dead easy to manufacture.
Anyway, others have pointed this out, but the conflict of interest only exists if Oracle insists that the fork is the enemy. That was not a necessary assumption until this meeting made it clear that that is the point of view Oracle has decided to take.
Other reasons for a fork, in absence of a license change?
Management issues would be one. If Oracle is not able to deal with the open source style of management, it would be easier for Oracle to manage their product project (although Star Office is already in place) and let people more used to open source manage a new, external project. This does not require two enemy camps. In fact, if Oracle recognized the value of the open source methodology, they would want to keep as many doors open to dialog as possible, including within the community board.
Oracle apparently does not recognize the value of open source methodology. (Perhaps they think Open Office is a valuable fluke?)
Objectives would be another. Oracle may want to gain market value from the brand, and Star Office may not get them the (perceived) value they want. In that case, they might want to start re-packaging open office for the business market.
Again, the re-packaging does not require the two projects to take be at enmity. They could still cooperate to a great extent, simply splitting things where Oracle needs the thing to be oriented towards business.
But Oracle has (by any reasonable reading of the chat meeting record) declared themselves to be in competitive (as in war) stance with the erstwhile community.
This is not an implication, it's an equivalence. Declaring the other side to be in COI if they remain is saying that the other side is not on your side any more.
Everything good Sun had, everything of value even in the corporate world, is now forked, or soon will be.
Trying times ahead, while the money that evaporated recondenses and precipitates somewhere else.
In addition to what others note, the Sun, now Oracle, employees are not the only ones who have been working on it since before Sun bought Star Office.
(But, given the way you framed that, I suspect you are a troll.)
Kind of like watching the scene where the former girlfriend is telling the geek goodbye, and you see her new boyfriend in the background with bulging, er, wallet.
And she has way too much makeup, way more than she used to, and the way it's caked on (cake, get it? Can't have your cake and eat it too.), the way it's caked on, you gotta wonder whether it's hiding bruises. And you know, the way these plots play out, that's exactly what the makeup is hiding.
Is that what you mean?
You don't know how to read between the lines in this kind of meeting, I'd say.
I've seen enough of these kinds of meetings to see the evidence of backroom deals. (As I noted above, the jammed input on the COI loop is one obvious bit of evidence.)
precisely.
I was a little surprised to see evidence in the logs that the new employer was exercising so much influence over (former) friends: "COI .... COI .... COI .... COI .... COI .... COI ...." Sounded like a broken record. (Or should I say it looks like the input device for the end condition got stuck.)
User?
Manufacturer?
Purchaser?
(erm, government?)
One, they are still taking feedback (as the article says), two, compatibilities are allowed to be listed as facts.
Or use your brain to read it.
The first version of the GPL kind of went nowhere.
Getting some kind of gentleman's agreement in place for hardware, where you can really use the hardware you've bought, is taking a bit longer than the mucking with the copyright law we call the GPL, and even the GPL is going to need tuning again after the next round of legal semantic shifting.
Except that it is more than just a statistical anomaly.
Especially when devices fail, or fail to work to spec, or the owner discovers some neat thing that ought to work except there's some legal tangle-up, or, ... .
That section said there was no problem with giving facts (such as listing which operating systems the hardware was tested/designed for), but the problem was the use of the logos that would promote the proprietary.
Still, I'm not sure that curing the general problem of endorsement with endorsements is a great idea. (Then again, the GPL uses the features of copyright law, so I'll have to read this and think about it.)
Bing drinking is bad for you.
erm, binge.
Anyway, I don't think binge drinking is in the same class of activities as consuming fats, unless you are trying to drag the topic from reasonable fat consumption to binge eating.
One of the problems with boh studies and meta-studies is the definition of reasonable. It seems to vary from person to person, whether you are talking about subjectively reasonable or objectively reasonable.
It's actually been known for an awfully long time that a rumor, once started, is impossible to call back.
This is something children used to learn in Sunday School.
I guess we need to be careful when we throw out all the false religion, not to through out the good stuff, too.
(Since people complain if the subject is part of the post, and the filter complains if I repeat it verbatim:)
It had it's own paragraph above
$000000000000
?
Until the fix the underlying issue, which is bandwidth for the control functions. (The post just above in my browser mentions something about dynamic control channels.)
Both on the part of the people and of the companies.
Seriously, people in Japan just work around the government's attempts at restrictions. That's why they don't really understand the fundamental issues of freedom, such as self-determination. It looks to your novce manager like the ideal place to manage, until you try to get people to do something new or unusual. (Propaganda does work, but it also takes a while.)
The reason people want money is because they think money will give them power.
The reason they want power is that they don't have control over themselves.
No amount of money will bring you real power, just facades and illusions of power.
No amount of power, whether illusion or real, will bring you control.
No amount of control over other things, even if such a thing could possibly be anything other than an illusion, will bring you control over yourself. (Generally gets in the way, in fact.)
That's why rich people and powerful people never seem to be able to get enough.
That's why this story repeats itself every few years. No, much more often than that. Same story, different players, maybe a different market, etc. Details change, but it's always looking for whatever you want to call it in all the wrong places.